THE BINARY AND TERNARY SYSTEMS FORMED BY CALCIUM FLUORIDE,
LITHIUM FLUORIDE AND BERYLLIUM FLUORIDE; PHASE
DIAGRAMS AND ELECTROLYTIC STUDIES

By
John L . S p e irs

A THESIS

Subm itted to th e School o f Graduate S tu d ie s o f M ichigan
S ta te C ollege o f A g ric u ltu re and A pplied Science
i n p a r t i a l f u lf illm e n t o f th e requirem ents
f o r th e degree o f
DOCTOR OF PHILOSOPHY
Department o f C hem istry

1952

ProQuest Number: 10008473

All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.

uest
ProQuest 10008473
Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author.
All rights reserved.
This work is protected against unauthorized copying under Title 17, United States Code
Microform Edition © ProQuest LLC.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106-1346

ACKNOWLEDGMENT
The w r ite r w ishes to ex p ress h is s in c e re
a p p re c ia tio n o f th e p a tie n t guidance o f
D r. Laurence L . Q u ill i n a l l phases o f th e
in v e s tig a tio n .
Thanks are a ls o due to D r. Elmer L e in in g er
f o r h is h e lp f u l advice on a n a ly tic a l p ro ­
ced u res .

**•
-X -

345842

THE BINARY AND TERNARY SYSTEMS FORMED BY CALCIUM FLUORIDE,
LITHIUM FLUORIDE AND BERYLLIUM FLUORIDE: PHASE
DIAGRAMS AND ELECTROLYTIC STUDIES

By
John L . S p e irs

AN ABSTRACT

Subm itted to th e School o f Graduate S tu d ie s o f M ichigan
S ta te C o lleg e o f A g ric u ltu re and A pplied Science
i n p a r t i a l f u lf i llm e n t o f th e req u irem en ts
f o r th e degree o f
DOCTOR OF PHILOSOPHY
D epartm ent o f C hem istry
Year

Approved

19^2

John L. S p e irs

THESIS ABSTRACT
P ro d u c tio n o f elem en tal f lu o r in e from calcium f lu o r id e by a
d i r e c t method in v o lv in g e l e c t r o l y s i s o f fu sed f lu o r id e m ix tu re s was
c o n sid e re d .

A te r n a r y m ix ture was re q u ire d to o b ta in th e low ering in

th e m e ltin g p o in t o f calcium f lu o r id e e s s e n tia l to give th e p o s s i b i l i t y
o f f lu o r in e reco v ery a f t e r e l e c t r o l y t i c e v o lu tio n .
L ithium and b e ry lliu m f lu o r id e s were chosen from among th e lim ite d
number o f p o s s ib le a d d itiv e s f o r th e p u rp o se .

The b in a r y system phase

diagram s were determ ined f o r th e component systems*

calcium f l u o r i d e —

lith iu m f lu o r i d e , calcium f lu o r i d e —b e ry lliu m f lu o r id e and lith iu m
f lu o r i d e —b e ry lliu m f l u o r i d e .

For th e l a t t e r system th e r e s u l t s were

compared w ith th e c o n f lic tin g r e p o r ts o f two e a r l i e r i n v e s t i g a t i o n s .
Methods o f s y n th e s is o f two compounds o c c u rrin g in th e se system s were
d isc u sse d and X -ray exam inations re p o rte d f o r calcium f lu o b e r y l la t e ,
CaB.eF4 , and lith iu m f lu o b e r y lla te , LigB eF^
The phase diagram o f th e te r n a ry system calcium f lu o r id e —lith iu m
f l u o r i d e —b e ry lliu m f lu o r id e was determ ined.

A re g io n o f s u ita b ly low

tem p eratu re was lo c a te d i n w hich, however, th e calcium f lu o r id e c o n te n t
was v ery s m a ll.
E le c tr o ly s e s o f m ix tu res in t h i s com position re g io n o f th e te r n a r y
system produced in c o n c lu siv e r e s u l t s .

Evolved f lu o r in e re a c te d w ith

th e anode p re v e n tin g any reco v ery of gaseous p ro d u c t.

I t was n o t

p o s s ib le to dem onstrate th e cath o d ic d e p o s itio n o f calcium and th u s
determ ine w h eth er, e f f e c t i v e l y , calcium f lu o r id e was b e in g decomposed
r a t h e r th a n one o f th e a d d itiv e su b sta n c e s .

TABLE OF CONTENTS
PAGE
INTRODUCTION..........................................................................................................................

1

GENERAL AND THEORETICAL........................................................................................

5

Sources o f F lu o rin e and Methods o f P ro d u c tio n ..............................
5
T h e o re tic a l and P r a c t i c a l Problem s o f Fused S a lt E le c tr o ly s is 8
R e la tiv e D ecom position P o te n tia ls and Choice o f Com ponents.. 16
E f f e c t o f L iq u id M etal C athodes on P o t e n t i a l ................................ 27
A v a ila b le In fo rm a tio n on Phase R e la tio n s and A n tic ip a te d
B eh a v io u r.................................................................................................. 31
The Phase R u le ............................................................................................ 33
bZ

EXPERIMENTAL...............................................................................................................

P r e p a ra tio n o f M a te r i a ls ...................................................................... h2
E quipm ent
..................................................................................................... $1
G eneral P ro c e d u re ..................................................................................... 56
THE SYSTEM CALCIUM FLUORHJE-LITHIUM FLUORIDE.......................................

67

Phase D iagram s............................................................................. .................69,70
THE SYSTEM CALCIUM FLUORIDE-BERYLLIUM FLUORIDE..................................

7h

Phase Diagram....................................................................
THE SYSTEM LITHIUM FLUORIDE-BERYLLIUM FLUORIDE.................................

78
86

Phase Diagrams ( r e s u i t s o f t h i s w o rk )......................................... 91,93,9k
Comparison w ith P re v io u s ly R eported R e s u lts .............................. 97
Phase Diagrams ( r e s u l t s o f e a r l i e r w o rk e rs)...........................101,103
THE TERNARY SYSTEM CALCIUM FLUORIDE-LITHIUM FLUORIDE-BERYLLIUM
FLUORIDE..........................................................................................................109
P a r t A — G en eral..................................................................
109
P a r t B — The L iF-R ich P o rtio n o f th e T ern ary System ........... 116
Phase Diagram ( p a r t i a l ) ....................................................117
P a r t C — The BeF2-R ich P o r tio n o f th e T ern ary System ......... 121
Phase Diagram ( p a r t i a l ) ....................................................122
P a r t D — The L igB eF ^R ich P o r tio n o f th e T ernary S y s te m .... 127
Phase Diagram ( p a r t i a l ) ....................................................130

TABLE OF CONTENTS - Continued

PAGE

P a r t E — The System CaF2-LiF-BeF2 as a W hole.................................... Ik8
Phase Diagram .................................................................................. Ik9
ELECTROLYTIC INVESTIGATIONS IN THE TERNARY SYSTEM...................................... 153
SUMMARY AND CONCLUSIONS.............................................................................................165
APPENDICES........................................................................................................................ 166
1 . Summarized R eferen ces on Phase R e la tio n s o f Calcium
F lu o rid e , L ith iu m F lu o r id e , B eryllium F lu o rid e and
Analogous Compounds............................................. . ................

166

2 . P h y s ic a l and Thermal P r o p e rtie s o f Calcium F lu o rid e ,
Lithium F lu o rid e and B eryllium F lu o r id e ..........................................170
3 . Summarized In fo rm atio n on F lu o b e ry lla te s and Analogous
Compounds......................................

176

k . In fo rm atio n on A llo y System s.................................................................178
5 . G eneral Phase R e la tio n s i n T ernary S ystem s................................... 180
6 . Sample Numbering System s.......................... . ............................................ 201
7 . R e s u lts o f Thermal

A nalyses o f CaF2-LiF System .......................... 203

8 . R e s u lts o f Thermal

A nalyses o f CaF2-BeF2 System ........................205

9 . R e s u lts o f Thermal

A nalyses o f LIF-BeF2 System .......................... 209

1 0 . Calcium F lu o b e r y lla te ; R e s u lts o f Wet and Dry S y n th eses,
O p tic a l and X -ray D a ta .....................................................

217

1 1 . L ithium F lu o b e r y lla te ; R e s u lts o f S yntheses o f Monoly d r a te d and Anhydrous Compounds, X -ray D a ta ...............................227
1 2 . R e s u lts o f ThermalA nalyses

o f CaF2-LiF-BeF2 System ...................23k

13. Q u a lita tiv e X -ray D ata f o r T ernary

System Sam ples......................2k9

l k . A n a ly tic a l M ethods.......................................................................................257
1 5 . A n a ly tic a l R e s u l t s ..................................................................................... 266
P a r t A — A n a ly sis o f Calcium F lu o r id e ................................. 266

TABLE OF CONTENTS - Continued

PAGE

P a r t B — A n aly sis o f B eryllium F lu o rid e and
"B eryllium C a rb o n a te "................................................ 268
P a r t C — A n a ly sis o f Ammonium F lu o b e ry lla te
P r e p a r a tio n s ................................................................... 270
P a r t D — A n aly sis o f B eryllium B asic A c e ta te
P r e p a r a tio n s ................................................................... 271
P a r t E — A n aly sis o f LiaBeF4 .lH 20 P r e p a ra tio n and
D eterm in atio n o f S o lu b ility o f L i2BeF4

273

P a r t F — D isc u ssio n o f A n a ly tic a l Accuracy and
C o rre c tio n o f C a lc u la te d C om positions

277

P a r t G — System CaF2-LiF A n a ly se s.......................................... 279
P a r t H — System CaF2-BeF2 A n a ly se s........................................ 282
P a r t I — System LiF-BeF2 A n aly ses.......................................... 287
P a r t J — System CaF2-LiF-BeF2 A n aly ses................................297
P a rt K
1 6 . Conductance
LITERATURE

— A n aly sis o f E l e c tr o ly tic Sam ples....................301
M easurem ents.................

30lt

CITED.........................................................................................................307

vi

LIST OF FIGURES

FIGURE

PAGE

1 ............................................................................................................................ 69
2 ............................................................................................................................ 70
3 ............................................................................................................................ 73
it ............................................................................................................................ 78
5 ............................................................................................................................ 91
6 ............................................................................................................................ 93
7 ............................................................................................................................ 9k
8
101
103
9 ..............................
1CL...............................................................................................................................113
I I ...............................................................................................................................117
12* .................................................................................................................................................................................... 122
13v ...............................................................................................................................130
II I ...................................................................................................................... 138,200
15 ...............................................................................................................................1U9
16 .............................................................................................................................. 155
1 7 , 1 8 , 19
181
20 ...............................................................................................................................185
21 , 22
189
23, 2it ...................................................................................................................... 191
25 ...............................................................................................................................193
26 ...............................................................................................................................197
27 ............................................................................................................................ 251+
*v'Uote - L a rg er s c a le re p ro d u c tio n s o f th e se
f ig u r e s a re to be found i n th e p o ck et i n ­
s id e th e back c o v e r.

INTRODUCTION

1

INTRODUCTION
The req u irem en ts o f in d u s try f o r f lu o r in e and f lu o r in a te d p ro d u cts
have been g r e a tly stim u la te d in th e l a s t decade, p a r t ic u l a r l y by th e
uranium iso to p e s e p a ra tio n program .

As an a d d itio n a l r e s u l t o f t h i s

i n t e r e s t , th e r e has been a marked in c re a se in th e amount of re s e a rc h i n
j*
t h i s p re v io u s ly r a th e r n e g le c te d f i e l d .
D esp ite th e p ro g re ss th a t has been made, th e p r in c ip le d i f f i c u l t y
co n tin u es to be th e r a th e r expensive and d i f f i c u l t g e n e ra tio n o f elem en tal
f lu o rin e .

The p re se n t method o f p ro d u c tio n i s an in d ir e c t one which

in v o lv es two main s te p s .

F i r s t , anhydrous hydrogen f lu o r id e i s p rep ared

by th e a c tio n o f s u lf u r ic a c id on calcium f l u o r i d e .

In th e second ste p

e l e c t r o l y t i c decom position o f th e hydrogen f lu o r id e i s accom plished a t
about 100°, u sin g a m ixture o f potassium f lu o r id e and hydrogen f lu o r id e
w ith th e approxim ate com position KF*2HF.
The p r in c ip a l g e n e ra tio n p ro cess th e r e fo r e c o n tin u e s to be e s s e n t i a l l y
o n ly a m o d ific a tio n , due to LeBeau and Damiens (9 U ,9 5 ), o f th e c l a s s i c a l
M oissan method w ith which elem en tal f lu o r in e was f i r s t is o la te d (1 1 3 ).
The h i s t o r i c a l development o f th e p ro c e ss has been ad eq u ately d e sc rib e d
by Damiens (20) .

More re c e n t in fo rm atio n and p r a c t i c a l p ro d u c tio n methods

have been summarized i n a c o lle c tio n o f symposium r e p o r ts (lii-U) and by
Simons e t a l . (1 5 0 ).
A d i r e c t method f o r th e p ro d u c tio n o f f lu o r i n e , from i t s p re s e n t
n a tu r a l so u rce —calcium f l u o r i d e , m ight have co n sid e ra b le economic

2

ad v an tag e.

The extrem e e le c tro n e g a tiv e p o s itio n o f f lu o r in e among th e

elem ents makes e l e c t r o l y s i s th e o n ly p r a c t i c a l method o f l i b e r a t i o n , no
chem ical p ro c e ss having ever been found cap ab le o f l i b e r a ti n g th e f r e e
elem en t.

Thermal methods o f su p p ly in g th e energy n e cessary fo r f r e e

f lu o r in e e v o lu tio n a re u n s a tis f a c to r y f o r one o r th e o th e r o f two re a s o n s .
Those compounds capable o f ev o lv in g f lu o r in e by h e a tin g to moderate
tem p eratu re re q u ire f lu o r in e f o r t h e i r p r e p a r a tio n .

O ther compounds,

p re p a ra b le w ith o u t th e use o f th e f re e elem en t, th e rm a lly l i b e r a t e f lu o r in e
only a t such high tem p eratu res t h a t a tta c k on th e c o n ta in e r i s p r o h ib i­
ti v e and no p r a c t i c a l y ie ld i s o b ta in e d .
H i s t o r i c a l l y , th e d i r e c t e l e c t r o l y s i s o f calcium f lu o r id e has r e ­
c eiv e d s c a n t a t te n tio n as a means o f f lu o r in e g e n e ra tio n .

Two o f th e

e a r l i e s t w orkers i n f lu o r in e ch e m istry , Freray (U9) and Gore (53) both
p ro b ab ly accom plished th e ••liberation'* o f f lu o r in e in t h i s way, b u t
n e ith e r was ab le to i s o l a t e th e p ro d u c t.

Fremy a c tu a lly observed th e

e v o lu tio n o f gas from th e anode i n th e e l e c t r o l y s i s o f fu sed calcium
f lu o r i d e , b u t th e experim ent was q u ic k ly term in ated by a tta c k o f li b e r a te d
calcium m etal on th e p latin u m c o n ta in e r .

Had he been ab le to c o n tin u e ,

th e phenomenon of "m etal fog" would undoubtedly have reduced th e y ie ld to
p r a c t i c a l l y n o th in g i n a v ery s h o rt tim e .
Subsequent to th e work o f th e se two i n v e s ti g a to r s , th e re i s no pub­
lis h e d re c o rd o f any f u r th e r attem p t to o b ta in f lu o r in e d i r e c t ly by th e
e l e c t r o l y s i s o f calcium f l u o r id e , or of i t s m ixtures w ith o th e r compounds.
The o r ig i n a l o b je c tiv e o f t h i s d i s s e r t a ti o n was to in v e s tig a te th e

3

p o s s i b i l i t y o f such a method and to determ ine a t th e same tim e i f i t was
rea so n ab ly p r a c t i c a l .
The development o f a p r a c t i c a l method must overcome two m ajor d i f f i ­
c u ltie s .

The p rim ary d i f f i c u l t y i s th e h ig h m eltin g p o in t o f calcium

f l u o r i d e , approxim ately 11*18° C-., which makes i t n ecessary t h a t s u ita b le
a d d itiv e s be found which w i l l p erm it low ering o f th e m eltin g p o in t to a
p r a c t i c a l tem p eratu re ra n g e .

Such a d d itiv e s must meet a c r i t e r i o n imposed

by th e secondary d i f f i c u l t y , namely t h a t f lu o r in e i s th e most e l e c t r o ­
n e g a tiv e elem ent and th a t calcium ranks high among th e e le c tr o p o s itiv e
m e ta ls .

F or a p r a c t i c a l p ro c e s s , such a d d itiv e s should be s u b s ta n tia lly

u n a ffe c te d by th e e l e c t r o l y s i s , o r the main purpose would be d e f e a te d ,
even though f lu o r in e were ev olved.
As f u r t h e r d is c u s s e d , in l a t e r s e c tio n s , o th e r d i f f i c u l t i e s must be
overcome, b u t p rim a rily th e study involved th e choice o f s u ita b le a d d itiv e s
and th e r e s u l t a n t tem p eratu re-co m p o sitio n in v e s tig a tio n to determ ine th e
phase diagram .

The a d d itiv e s chosen were lith iu m f lu o r id e and b e ry lliu m

f lu o r id e f o r reaso n s to be more f u l l y d is c u s s e d .

At t h i s p o in t , however,

i t may be noted th a t th e choice o f a d d itiv e s w ith re s p e c t to m etal c a tio n s
i s very l i m i t e d , and w ith re s p e c t to simple anions th e re i s no c h o ic e .
Since th e a d d itio n o f lith iu m f lu o r id e alone would not be expected to
produce s u f f i c i e n t lo w e rin g , an o th er component was needed.

The b e ry lliu m

f lu o r id e was in te n d ed to f u rn is h t h i s a d d itio n a l lo w erin g , b u t f u n c tio n ­
in g as a complex an io n r a th e r th a n as a sim ply a d d itiv e component.
Subsequent to th e work on th e phase diagram s o f th e b in a ry and
te r n a r y sy stem s, and o th e r problems which were su g g ested d u rin g th e

h

in v e s t i g a t i o n , a t t e n t i o n was d ir e c te d toward th e o r ig i n a l problem of
f lu o r in e g e n e ra tio n .

Crude com parative conductance measurements and sm all

s c a le e le c tr o ly s e s were c a r r ie d o u t on m ix tu res i n th e te r n a ry system to
d eterm in e th e p r a c t i c a l i t y o f th e method f o r th e g e n e ra tio n and reco v ery
o f f lu o r in e from calcium f lu o r id e .

GENERAL AND THEORETICAL

5

GENERAL AND THEORETICAL
Sources o f F lu o rin e and Methods o f P ro d u ctio n
F lu o rin e has been o b ta in e d i n th e p a s t from i t s only abundant con­
c e n tr a te d s o u rc e , th e m in era l f lu o r s p a r .

D eposits o f t h i s n a tu r a l form

of calcium f lu o r i d e a re o f r e l a t i v e l y high p u r it y , o r may be b e n e f ic ia te d
r e a d i l y to o b ta in a s a t i s f a c t o r y grade o f m a te r ia l.

The s t e e l in d u s tr y 's

demand f o r u se as a f lu x fo rm erly was th e la r g e s t a p p lic a tio n , b u t sin c e
1939 an enormous in c re a s e has occu rred i n the demand f o r th e b e s t grades
o f f lu o r s p a r f o r h y d ro flu o ric a c id p r e p a ra tio n .

The prim ary cause of

t h i s in c re a s e was th e u se o f h y d ro flu o ric acid by th e petroleum in d u s try
i n 11a lk y la tio n " p ro c e sse s f o r a v ia tio n gas p ro d u c tio n .
was th e use o f h y d ro flu o ric a c id ,

A secondary f a c to r

f lu o r in e and f lu o r in a te d organic com­

pounds i n th e uranium iso to p e s e p a ra tio n program .
As th e se chem ical a p p lic a tio n s have grown i t has become ap p aren t
t h a t th e Mabundance" o f th e supply of f lu o r s p a r was only ap p aren t and
t h a t th e c o n c e n tra te d o re d e p o s its a re r a p id ly b ein g d e p le te d , le a v in g
o n ly m arginal and low grade d e p o s its i n re s e r v e .

The p ro sp e c tiv e exhaus­

t i o n o f th e economic d e p o s its i n a m a tte r o f y e a r s , r a th e r th a n d ecad es,
has s tim u la te d development o f a new s o u rc e .
The l a r g e s t p o t e n t i a l source suggested (62) i s th e phosphate ro ck
p ro c e ss in g in d u s tr y .

Although f lu o r in e i s p re s e n t as o n ly a minor

c o n s titu e n t (ap p ro x im ately 3.5$) i n phosphate ro c k , th e huge tonnages
handled a n n u a lly make a v a ila b le a la rg e p o t e n t i a l supply o f f lu o r in e ,

6

p r in c ip a ll y i n th e form o f hydrogen f lu o r id e w aste g a se s.

The to x i c i t y

o f f lu o r in e i n b oth f e r t i l i z e r s and c a t t l e food supplem ents make i t
e s s e n t i a l to remove i t i n la r g e p a r t du rin g th e m anufacturing p ro c e s s .
As a w aste g a s , hydrogen f lu o r id e i s a s e rio u s atm ospheric p o l l u t a n t .
The com bination o f circu m stan ces would seem to in d ic a te t h a t in th e n ear
fu tu r e th e r e w i l l be a s u b s ta n tia l c o n tr ib u tio n o f f lu o r in e from t h i s
so u rc e , i n a d d itio n to th e c u r re n t p a r t i a l reco v ery in th e form of
flu o s ilic a te s .

Recovery i n th e form o f a very pure grade o f calcium

f lu o r id e has been d e sc rib e d (6 9 ,1 0 ), which r e q u ire s only sim ple equipment
and a cheap raw m a te r ia l, lim e sto n e .
The d i r e c t demand c re a te d by th e m anufacture o f f lu o r in e and f l u o r i n a te d org an ic compounds i s s t i l l s m a ll.

The p r o p e r tie s of flu o ro carb o n s

are so u n u s u a l, p a r t i c u l a r l y w ith r e s p e c t to s t a b i l i t y , t h a t i t m ight
rea so n ab ly be ex pected t h a t t h e i r com m ercializatio n would be q u ite r a p id .
One ite m , th e c o s t o f f lu o r in e g e n e ra tio n and h a n d lin g , has been th e
p r in c ip a l d e t e r r e n t .

I n g e n e ra l th e h ig h ly f lu o r in a te d compounds have

o n ly been e f f e c t i v e l y p rep a re d by th e u se o f f lu o r in e '’c a r r i e r s " , i . e .
h ig h e r in o rg a n ic f lu o r id e s such as CoF3 , and AgF2 , which re q u ire th e use
o f elem en tal f lu o r i n e I n t h e i r p r e p a r a tio n .

A b r i e f exam ination o f th e

most im p o rtan t f lu o r i n e p ro d u c tio n system w il l in d ic a te th e in h e re n t
re aso n s f o r th e h ig h c o s t.
The l a r g e , modern f lu o r in e c e l ls ,o p e r a ti n g a t about 100° C.on a mix­
tu re having th e approxim ate com position KF:2HR, have been improved to a
p o in t o f r e l i a b i l i t y f a r removed from th e la b o ra to ry c e l l s o f th e p a s t.
However d e s p ite h igh e f f i c i e n c i e s and c e l l c u rre n ts o f 2000 to 3000 amperes

7

a t d e n s it i e s o f app ro x im ately 75-100 a m p ./s q .f t. , th e p ro d u c tio n r a t e s
are o n ly app ro x im ately 3 to 5 pounds p er hour on q u ite s iz a b le u n i t s .
A com parison may be made w ith th e c h lo r in e - c a u s tic in d u s try where a
tremendous in v estm en t i n e l e c t r i c a l equipm ent, c e l l s and b u ild in g s i s
r e q u ire d to o b ta in tonnage p ro d u c tio n .

The high c o s t of f lu o r in e genera­

t i o n i s u n d e rsta n d a b le when th e raw m a te ria l c o s ts (b rin e v s . anhydrous
hydrogen f l u o r i d e ) , c o rro s io n d i f f i c u l t i e s and d i f f i c u l t i e s o f p ro d u ct
shipm ent a re compared.
The e le c tro c h e m ic a l f i e l d has b u t one s o lu tio n to the problem of
low u n i t p ro d u c tio n , namely to in c re a s e th e c u r r e n t.

I n th e h ig h e r

tem p eratu re fu se d s a l t m e lts i t i s custom ary to use c u rre n t d e n s itie s as
la r g e as 1,000 a m p s ./ s q .f t . , w ith c e l l c u rre n ts of 50,000 am peres, and a
consequent in c re a s e i n u n i t o u tp u t.
The absence o f d a ta on any a ttem p ts a t th e d i r e c t p ro d u c tio n of
f lu o r in e by e l e c t r o l y s i s o f th e fu sed raw m a te r ia l, calcium f lu o r i d e ,
prompted t h i s approach to th e problem .

The p o s s i b i l i t y was p re s e n t th a t

th e obvious d i f f i c u l t i e s m ight be s o lv e d , w ith some p o s s ib le g ain in
e f f ic ie n c y because o f th e cheaper raw m a te r ia l and p o s s ib le h ig h er u n it
p ro d u c tio n r a t e .
As an o th er approach to t h i s problem , th e p ro c e ss p a te n te d by Simons
(1 5 2 ), should be m entioned.

E s s e n tia lly th e g e n e ra tio n and h an d lin g of

f lu o r in e i s minim ized by th e a d d itio n to th e KF:2HF m elt o f th e organic
m a te r ia l which i s to be f lu o r in a te d .

Upon a p p lic a tio n of th e c u r re n t i t

i s claim ed t h a t f lu o r i n a t i o n occurs w ith o u t a c tu a l g e n e ra tio n of f l u o r i n e .
R eg a rd less o f th e c o rre c tn e s s o f th e claim th e p ro c e ss i s being

8

com m ercially developed and w i l l undoubtedly be a more s a ti s f a c t o r y method
f o r th e p ro d u c tio n o f some m a te r ia ls .

Problems such as i n s o l u b i l i t y o f

r e a c t a n t s , “frag m en tatio n " o f m o lecu les, and p ro d u c tio n of d i f f i c u l t l y
se p a ra b le m ix tu res w i l l p ro b ably r e s t r i c t i t s a p p lic a tio n to many p ro d u c ts .
T h e o re tic a l and P r a c t ic a l Problems of
Fused S a lt E le c tr o ly s is
C e rta in d i f f i c u l t i e s p e c u lia r to t h i s p a r t ic u la r case and o th e rs
w hich are in h e r e n t f o r most high tem perature fu sed s a l t e le c tr o ly s e s
should be m entioned.

The tem perature o f th e m elt from th e sta n d p o in t o f

f lu o r in e r e a c t i v i t y , th e “ anode" e f f e c t , and problem s of m etal recovery
a t th e c a th o d e , in c lu d in g “m etal fog" form ation must be c o n s id e re d .

Also

th e problem o f r e l a t i v e decom position p o te n tia ls i s o f c o n tr o llin g im­
p o rta n ce .
A t th e s t a r t o f t h i s work i t was judged th a t no calcium f lu o r id e
m ix tu re m e ltin g h ig h er th a n 6d0° would have any chance o f evolving
f lu o r in e sin c e no m e ta llic cathode m a te ria l was known cap ab le o f w ith ­
sta n d in g m olecu lar f lu o r in e a t h ig h e r te m p e ra tu re s.

The e v o lu tio n of

f lu o r in e in atom ic form a t th e e le c tro d e su rface would be expected to be
even more a c tiv e ly c o r ro s iv e , so th a t to be p r a c tic a l a co n sid e ra b ly
low er tem p eratu re would p ro b ab ly be n e c e s sa ry .

These c o n s id e ra tio n s

a f f e c te d th e ch o ice o f components to be added to th e calcium f lu o r id e to
reduce i t s normal m e ltin g p o in t of approxim ately lifL8° C .

O ccasio n ally

non-metal 1jc co n d u cto rs have been used as e le c tr o d e s , b u t t h i s p o s s i b i l i t y
was n o t in v e s tig a te d as such a substance would have to be a f lu o r i d e , to
be f r e e from f lu o r in e a t t a c k , and would p robably be i t s e l f so lu b le in th e
m e lt.

9

In o rd er to produce th e low ering o f tem perature e s s e n ti a l to p o s sib le
re co v e ry o f f lu o r in e evolved from th e e l e c t r o l y s i s o f calcium f l u o r i d e ,
ch o ice o f added su b stan ces must be r e s t r i c t e d to th o se which are n o t
th em selves decomposed by th e e l e c t r o l y s i s .

This r e s t r i c t i o n g r e a tly

l i m i t s th e p o s s i b i l i t i e s , p a r t i c u l a r l y sin c e th e f lu o r id e io n i s th e only
sim ple anio n a d d itiv e which i s com patible w ith t h i s re q u ire m e n t, a l l
o th e r sim ple an io n s b ein g more r e a d ily o x id iz a b le .

ho r e p o r t to th e con­

t r a r y i s on r e c o rd , nor would i t be expected sin c e any evolved f lu o r in e
would be cap ab le o f o x id iz in g reduced form s, such as o th e r a n io n s, because
o f i t s extrem e p o s itio n as th e most e le c tr o n e g a tiv e elem ent.
W ith re s p e c t to p o s s ib le c a tio n a d d itiv e s m eeting th e above r e q u ir e m ent, t h a t i s th o se more d i f f i c u l t l y re d u c ib le th an th e Ca

io n , i t must

be ad m itted th a t th e p re se n t s t a t e o f th e a r t does n o t p erm it a very
sy stem atic o r s c i e n t i f i c d e c is io n on th e m a tte r.

The l i t e r a t u r e on th e

s u b je c t i s r e l a t i v e l y m eager, a c o n d itio n which i s u n d erstan d ab le from a
c o n s id e ra tio n o f th e h i s t o r i c a l developm ent.

S u ccessfu l a p p lic a tio n s o f

fu sed s a l t methods have been few and i n g en e ra l have been r e s t r i c t e d to
a few companies t h a t have not been noted f o r t h e i r l i b e r a l i t y in p u b lic a ­
tio n o f re s e a rc h r e s u l t s .

Academic i n t e r e s t in th e s u b je c t has been low

and l a r g e l y co n fin ed to a few German i n v e s tig a to r s .
C o lle c te d in fo rm atio n on th e s u b je c t o f fu sed s a l t e l e c t r o l y s i s i s
n o n -e x is te n t in E n g lish t e x t s , o th e r th a n as a b r i e f survey o f the f i e l d
i n very s h o rt c h a p te rs .

I n German, two modern p u b lic a tio n s are a v a ila b le ,

one te c h n ic a l i n n a tu re e d ite d by E n g elh ard t (liO) , th e o th e r a t h e o r e t i ­
c a l approach by D rossbach (2 7 ), based on a c o lle c ti o n of th e van Laar

10

thermodynamic p ap ers on th e s u b je c t,

in fo rm a tio n co n tain ed in th e form er

work w ith r e s p e c t to e x a c t v a lu e s o f decom position p o te n tia ls i s probably
w o rth le ss i n th e l i g h t o f D ro ssb ach 's c r itic is m s (2 8 ), except as a g en eral
in d ic a tio n o f th e approxim ate o rd e r o f decom position p o t e n t i a l s .
I n b r i e f , th e s e c r itic is m s concerned f a i l u r e to o b ta in e q u ilib riu m
c o n d itio n s , to i s o l a t e anode and cathode p ro d u cts from each o th e r , and
f a i l u r e to i d e n t i f y a c tu a l e le c tro d e p ro d u c ts .

The o n ly e x te n siv e survey

made on th e s u b je c t (1 1 8 ), has been w idely c r i t i c i z e d (135) ( 2 9 ) , and th e
v a lu e s shown to be im possible from a thermodynamic v iew p o in t, and i n ­
com patible w ith th e r e s u l t s o f o th e r i n v e s tig a to r s ,

M hile i t might be

tho u g h t t h a t any thorough stu d y would be made in such a way as to o b v iate
th e above c r i t i c i s m s , th e p r a c t i c a l d i f f i c u l t i e s w ith r e s p e c t to i s o l a t i o n
o f th e p ro d u c ts by means o f s u ita b le diaphragms are such t h a t on ly a very
few s tu d ie s have been made which a re f r e e of such d e fe c ts in experim ental
te c h n iq u e .

These have la r g e ly been co n fin ed to low tem perature s tu d ie s

o f tr a c t a b l e s a l t s , w ith low c o rro s iv e p r o p e r tie s , in g la ss and p o rc e la in
ap p aratu s (99) ( 30 ) .
¥ i t h re s p e c t to d a ta on o p e ra tin g c e l l s n a tu r a lly no very in fo rm ativ e
d a ta on r e l a t i v e decom position p o te n t ia l s i s o b ta in e d , sin c e such c e l l s
norm ally o p e ra te a t c o n sid e ra b le o v erv o lta g e s which are dependent upon a
m u ltip lic ity of v a ria b le s .
The c o lle c te d in fo rm a tio n given by E ngelhardt ( U0), was however
v a lu a b le f o r i t s q u ite com plete review o f p u b lish ed work on a l l phases o f
fu se d s a l t e l e c t r o l y s i s .

I t c o n s titu te d a la rg e source of " n e g a tiv e 11

in fo rm a tio n , t h a t i s on th e s u b je c t o f a d d itiv e s which could not be expected

11

"to work s a t i s f a c t o r i l y w ith calciu m f lu o r i d e .

T his circum stance aro se

from th e f a c t t h a t , a t one tim e or a n o th e r, calcium f lu o r id e has been
ap p lie d a s an a d d itiv e to m ix tu res in ten d ed f o r th e d e p o s itio n of alm ost
a l l m etals s u ite d to e l e c t r o l y t i c p r e p a ra tio n .
W hile th e thermodynamic approach to th e problem o f decom position
p o t e n t i a l , as ex em p lified by th e van L aar th e o ry (2 7 ), w ill r e q u ire much
more d a ta to p e rm it g e n e ra l a p p l i c a b i l i t y , i t i s mentioned h ere as i t
re p re s e n ts th e o nly sy ste m a tic approach which has shown i t s e l f capable
o f p roducing d a ta in agreem ent w ith c a r e f u l experim ental work.
E s s e n tia lly th e th eo ry d e a ls w ith th e behaviour o f s o lu tio n s as
a f fe c te d by th e “in te rm o le c u la r" fo rc e s e x e rte d by th e components.

In

th e absence o f s u ita b le q u a n tita tiv e th e o r ie s o f th e liq u id s t a t e and
fo rc e s o f i n t e r a c t i o n , van L aar used a d e s c r ip tio n o f th e tem p eratu revolume dependence i n term s o f th e van d e r Waals e q u a tio n of s t a t e .

He

proposed s u ita b le a d d itiv e v a lu e s of th e c o n s ta n ts a^ and b^ (th e
c o n s ta n ts a t th e c r i t i c a l p o in ts ) f o r a la r g e number o f the chem ical
elem ents as elem ents and i n “m olecular" (n o n -io n ic ) compounds, and f o r
a few elem ents i n io n ic compounds.
Very complex ex p re ssio n s are th e n developed f o r th e m olar thermo­
dynamic p o t e n t i a l s o f su b stan ces i n b in a ry m ix tu re s , th e c o n s ta n ts of
which may be e v a lu a te d from a knowledge o f such in fo rm atio n as th e p a r t i a l
m olal volumes and h e a ts o f fu s io n ( a t th e tem perature i n q u e s tio n ) , which
may be o b ta in e d from d i f f e r e n t i a l h e a t o f s o lu tio n m easurem ents.

A nother

approach in v o lv in g th e use o f m olar thermodynamic p o te n tia ls f o r p r e ­
d ic tio n o f m e ltin g p o in t low ering i n m ixtures may be re v e rs e d to o b ta in

12

approxim ate v a lu e s o f th e m o lecu lar' in te r a c tio n c o n sta n ts of the eq u atio n
from th e ex p erim en tal d a ta on m e ltin g p o in t cu rv es o f m ix tu re s .
Such e x p re s s io n s , when developed f o r h ig h e r th an

b in a ry system s,

become extrem ely cumbersome and th e approxim ate forms are lim ite d to
d i l u t e s o lu tio n s w ith sm all v a lu e s o f m e ltin g p o in t lo w erin g .
c o r r e la tio n s are given by D rossbach ( 31 )

E x c e lle n t

f o r th e decom position p o t e n ti a ls

o f some a l k a l i m etal h a lid e s , b u t th e la c k of s u f f ic ie n t therm al d a ta
g e n e ra lly r e s t r i c t s th e a p p lic a tio n o f th e th e o ry and i t d id not f a c i l i t a t e
a t h e o r e t ic a l approach i n t h i s work.
S e v e ra l o th e r s im p lifie d t h e o r e tic a l tre a tm e n ts and a number o f
e m p iric a l c o n tr ib u tio n s have been made w ith r e s p e c t to decom position
p o t e n t i a l s i n fu sed s a l t e l e c t r o l y s i s .

Among th e form er may be mentioned

the work o f Newman and Brown (119) , D rossbach (32) , and Thomson (1 6 8 ).
E s s e n tia lly t h e i r work was based on th e Gibbs-Helmholtz r e l a t i o n ,
E = -A F /n F , f o r th e r e a c tio n

M etal s a l t —

elem en ts, w ith i n g e n e ra l

some atte m p t to ap p ly th e p ro p er value o f AF ( i . e . a t th e p a r t ic u la r
tem p eratu re in v o lv ed ) and w ith c o r re c tio n o f th e

AF v alu e in th e case

where a secondary r e a c tio n o ccu rred a t th e e le c tr o d e .
Such s im p lif ie d approaches n a tu r a lly f a i l to tak e in to account th e
p o s s ib le change in environm ent which occurs when one s a l t i s d isso lv e d
in a n o th e r.

To assume la c k o f i n t e r a c t io n and com plete d is s o c ia tio n

in to independent io n s o r m olecules i s h a rd ly com patible w ith th e evidence
f o r s tro n g a s s o c ia tio n o f th e components which fo llo w s from a c o n sid e ra ­
t i o n o f th e la r g e d if f e r e n c e s between th e m eltin g and b o ilin g p o in ts of
fu se d s a l t s .

13

Of th e e m p iric a l r e l a t i o n s th e only re fe re n c e s p e c i f i c a l l y p e r ta in ­
in g to a l k a l i and a lk a lin e e a r th h a lid e s i s th a t o f B alce ( 5 ) , claim ing
f a i r agreem ent f o r a r e la tio n s h ip between th e change in volume p er u n i t
volume (on fo rm a tio n o f th e s a l t from th e elem ents) and th e decom position
p o t e n t i a l and h e a t o f fo rm a tio n .
C e r ta in problem s which a re p e c u lia r ly im p o rtan t in fu sed s a l t
e le c tr o ly s e s e s s e n t i a l l y in v olve th e i s o l a t i o n o f th e cathode and anode
p ro d u cts so t h a t reco m b in ation and consequent e f fic ie n c y lo s s e s may be
p re v e n te d .

T em perature, d e n s ity and s o lu b i l i t y f a c to r s must be tak en

in to c o n s id e ra tio n .
I n g e n e ra l ad h eren t d e p o s its o f m etal on th e cathode are n o t o b ta in ­
ab le under c o n d itio n s where th e lo c a l tem perature of th e b a th i s so low
t h a t th e m etal does n o t i n i t i a l l y form a m olten m ass.

Thus b a th tem pera­

t u r e s below t h i s m e ltin g p o in t ten d to form lo o se powdery d e p o s its which
may become m ech an ically suspended in th e m elt so t h a t tr a n s f e r to the
v i c i n i t y o f th e anode becomes p o s s ib le .

In cases where th e liq u id m etal

d e n s ity i s l e s s th a n t h a t o f th e fu sed s a l t , a common procedure i s to
re c o v e r th e m eta l in th e form o f a " c a r r o t” by g ra d u a lly r a is in g a w ater
cooled cathode from th e m e lt.

O p eratio n a t high c u r re n t d e n s ity w ill

produce s u f f i c i e n t l o c a l h e a tin g o f th e m elt so t h a t a f lo a ti n g puddle
o f th e m etal w i l l be p r e s e n t, which s o l i d i f i e s i n an adherent manner on
th e r i s i n g cathode (9) (127) .
A nother method o f a tta c k on th e problem i s th e use o f a liq u i d
m e ta llic cathode which may be th e d e p o site d m etal i t s e l f provided th e
d e n s ity r e la tio n s h ip p e rm its i t to rem ain on th e bottom o f the c e l l .
example o f t h i s would be th e p ro c e ss used in aluminum m anufacture in

An

Ik

which th e p ro d u c t i s c o lle c te d on th e bottom o f th e c e l l and tapped o f f
a t p e rio d ic i n t e r v a l s .

A m o d ific a tio n o f t h i s system i s th e d e p o s itio n

o f a m etal ( l e s s dense th a n th e m elt) on a l iq u id cathode composed o f a
more dense m etal w ith which i t may th e n form an a llo y having a d e n s ity
s u f f i c i e n t to h old i t on th e bottom o f th e c e l l .
Such a system w i l l b e - f u r th e r co n sid ered i n connection w ith i t s
e f f e c t on decom position p o t e n t i a l s , b u t i t should be m entioned a t t h i s
p o in t t h a t such a method p re s e n ts d i f f i c u l t y because o f r a p id s a tu r a tio n
of th e s u rfa c e o f th e cathode by d ep o site d m e ta l.

D iffu s io n r a t e s in

such a llo y s are v ery slow and means must be pro v id ed to produce f re s h
s u rfa c e s f o r d e p o s itio n i n o rd e r to o b ta in s a tis f a c to r y o p e ra tio n ( lil t) .
On th e o th e r extrem e, h ig h te m p e ra tu re s , or tem p eratu res much in
excess o f th e m e ltin g p o in t o f th e c a th o d ic a lly d e p o site d m e ta l, cause
an o th er s e rio u s d i f f i c u l t y , commonly r e f e r r e d to as "m etal fo g " .

Under

such circu m stan ces s o lu tio n s ( 1 9 ), or d is p e rs io n s o f m etal may be produced
i n th e m elt which have been d e sc rib e d as c o l l o i d a l , probably because of
th e v a r ie ty o f c o lo rs o b ta in e d , i n analogy w ith such aqueous c o llo id a l
d is p e rs io n s as gold s o ls (1 0 0 ,6 ).

A lk a lin e e a rth m etals are q u ite su b je c t

to t h i s b e h a v io r in h a lid e m e lts .
W hile a t f i r s t t h i s e f f e c t was a ttr i b u te d to v a p o riz a tio n and con­
d e n s a tio n o f th e d e p o site d m etal (1 0 1 ,1 0 2 ), subsequent p r e p a ra tio n of
s o - c a lle d "su b -h a lid e " compounds such as "CaCl" (17^,107) le a d to th e
b e l i e f t h a t such io n s as Ca+1 were produced a t th e cathode by prim ary
o r secondary re d u c tio n p ro c e sse s (3 3 ).

More re c e n t s tu d ie s on th e e f f e c ts

o f e le c tr o n ex cesses and d e f ic ie n c ie s in s o lid s and p a r t ic u la r l y a

15

thorough stu d y made by C u b ic io tti (18) o f the calcium -calcium c h lo rid e
system le a d to th e b e l i e f t h a t “double io n s" such as Ca2++ are formed.
R eg ard less o f th e p a r t i c u l a r reduced form o f th e m etal p r e s e n t, i t
i s c l e a r t h a t th e problem o f i s o l a t i o n o f cathode pro d u ct from th e anode
becomes most d i f f i c u l t and very poor o r zero e f f ic ie n c ie s may be expected
under such c irc u m sta n c e s .
One o f th e s e rio u s problem s i n th e e l e c tr o ly s is o f fu sed s a l t s , th e
"anode e f f e c t " , has been sp e c u la te d upon and d isc u sse d by numerous i n ­
v e s tig a to r s (1+8,165,115,3-b) .

This phenomenon, p a r t ic u la r l y noted w ith

f lu o r id e m e lts , i s e x h ib ite d as a sudden r i s e in c e l l r e s is ta n c e
accompanied by an ap p aren t f a i l u r e o f th e m elt to wet th e anode.

Even

though th e c u r r e n t th e n f a l l s s h a rp ly , th e common p r a c tic e of arran g in g
c e l l s i n a s e r i e s c i r c u i t imposes h ig h e r v o lta g e on th e f a u lty anode-m elt
in te rfa c e .

T his probably ag g rav ates th e source o f d i f f i c u l t y by causing

f u r t h e r lo c a liz e d h e a tin g o f a film o f gas b e lie v e d to be e f f e c tiv e ly
i n s u la tin g th e anode.
The d i f f i c u l t y p robably norm ally a r i s e s from a d e p le tio n of th e con­
c e n tr a tio n o f th e component norm ally o x id ized a t th e anode so th a t th e
wrong c o n s titu e n t commences to d e p o s it, f o r example, f lu o r in e which in
th e case o f carbon anodes may th e n r e a c t to form flu o ro carb o n s such as
te tra flu o ro m e th a n e which in s u la te th e anode.

In th e normal c a s e , th e

remedy i s u s u a lly found i n r e s t o r a t i o n of th e proper com position of th e
m elt and re d u c tio n o f a p p lie d p o t e n t i a l .

However in th e case of a c e l l

in ten d e d f o r f lu o r in e d e p o s itio n th e f i r s t remedy i s not a v a ila b le and
some d i f f i c u l t y m ight be a n tic ip a te d i f evolved flu o r in e re a c te d w ith

16

th e carbon anodes.

The e f f e c t has been observed on m e ta llic e le c tro d e s

as w e ll, where th e remedy has u s u a lly been to d ecrease th e c u rre n t
d e n s ity o r to r e p le n is h th e b a th c o n s titu e n t which a c ts to d e p o la riz e
th e anode.
I n c o n tr a s t to t h i s b e h a v io u r, th e u su a l KF:2HF type of low tempera­
t u r e f lu o r in e c e l l s give b e s t r e s u l t s when th e anode i s not w etted by th e
e le c tro ly te .

At tem p eratu res below 100° and w ith h ig h e r hydrogen

f lu o r id e c o n te n ts th e anodes are w e tte d , become sw o llen , and d i s in t e g r a te ,
a p p a re n tly as a r e s u l t of th e fo rm atio n o f th e compound (CF) . (lf?l) .
R e la tiv e D ecom position P o te n tia ls and
Choice o f Components
As w i l l be se e n , th e d a ta on fu sed s a l t decom position p o te n tia ls
p re s e n t numerous c o n tra d ic tio n s and u n c e r t a i n t i e s .

I t may be h e lp f u l,

th e r e f o r e , to c o n sid e r th e r e l a t i v e l y w e ll e s ta b lis h e d d ata f o r gaseous
io n iz a tio n and aqueous e le c tro d e p o te n t ia l s f o r p e r tin e n t c a tio n s , befo re
d is c u s s in g t h e i r behaviour i n fused s a l t s .
From in s p e c tio n o f th e fo llo w in g ta b le i t may be observed t h a t th e re
i s no sim ple r e la tio n s h ip betw een th e io n iz a tio n p o te n tia ls o f gaseous
atoms and t h e i r s in g le e le c tro d e p o te n tia ls i n aqueous s o lu tio n , as th e
h e a ts o f h y d ra tio n o f th e gaseous ions may e x e rt a com pensating e f f e c t (9 2 ).
Thus th e e a s i ly io n iz a b le Cs atom has th e same value o f aqueous e le c tro d e
p o t e n t i a l as th e more d i f f i c u l t l y io n iz a b le L i because of th e la r g e con­
t r i b u t i o n o f th e h e a t o f h y d ra tio n o f th e L i+ io n to th e o v e r - a ll
r e a c tio n .

17

Element

I o n iz a tio n
P o t e n t ia l o f
Gaseous Atom,
v o lt s

Heat o f Hydra­
t io n o f
Gaseous I o n s ,
K ca l .

E lectro d e P o te n tia l
fo r r e a c tio n in
aqueous s o lu tio n
M -*M +
+ e ~)
v o lt s

5 .3 6
5.12
5.32
5.16
3.87

123.
97.
77.
70.
63.

3.02
2.71
2.92
2.99
3.02

Li
Na
K
Rb
Cs

Ion ic
Radius
o f Metal^
cm.x 10"

0.60
0 .9 5
1.33
1.58
1.69

Somewhat th e same e f f e c t must be p re s e n t i n fused s a l t m edia, sin c e
f o r example potassium s a l t s may be used as a d d itiv e s in lith iu m p ro d u c tio n ,
as a m ix tu re o f equal p a r ts o f L iC l and KC1 w ill d e p o sit predom inately
lith iu m on e l e c t r o l y s i s (6 0 ).

In view of t h i s apparent r e v e r s a l of th e

p o s itio n s i n aqueous s o lu tio n s , i t might be concluded t h a t e ith e r th e
h ig h ly h y d rate d lith iu m io n i s r e l a t i v e l y more so lv a te d in fused m ix tu res
or t h a t th e more p o o rly hydrated potassium io n i s r e l a t i v e l y even le s s
so lv a te d in fu sed m ix tu re s.
A s im ila r r e la tio n s h ip e x is ts f o r Group I I m e ta ls , as noted in the
fo llo w in g t a b l e , except t h a t th e e f f e c t o f h y d ra tio n i s even more im­
p o r ta n t. (93)
With Group I I I m e ta ls , th e even h ig h er io n iz a tio n p o te n tia ls are
so c o u n te ra c te d by

l a r g e r h e a ts o f h y d ra tio n t h a t , i n g e n e ra l, th e n e t

e f f e c t cau ses th e e le c tro d e p o te n ti a ls o f m etals in a l l th re e groups to
lie

la r g e ly in th e range o f 2 to 3 v o l t s .
S im ila rly th e

in d ic a tio n s are

t h a t in fu sed s a l t e l e c t r o l y s i s th e

r e l a t i v e s in g le e le c tro d e p o te n tia l s of th e m etals (a s judged by r e l a t iv e

18

Element

Be
Mg

Ca
Sr
Ba

I o n iz a tio n P o te n tia l
o f Gaseous Atom,
1 s t. d- ,
2nd e “ ,
v o lts
v o lts
9.28
7.61
6.09
5.67
5.19

18.1
lip .96
11.82
10.98
9.95

H eat o f Hy­ E lectro d e P o te n tia l Io n ic
Radius
d r a tio n o f
f o r r e a c tio n i n
of
M etal,
Gaseous Ions , aqueous s o lu tio n
M -*■ M++ + 2 e~, cm.x 10"
K dal.
v o lts
860.
395.
355.
305.

1.70
2.38
2.87
2.89
2.90

0 .3 1
0 .65
0.99
1.13
1.35

decom position p o t e n t i a l s o f t h e i r compounds) lik e w ise l i e w ith in an ap p ro x i­
m ately s im ila r v o lta g e ra n g e .

R e la tiv e p o s itio n s i n th e electro m o tiv e

s e r ie s a r e , however, somewhat s h if te d and th e ty p es o f anions p re s e n t in
m ix tu res appear to e x e rt more o f an e f f e c t , a p a rt from cases where s ta b le
complex fo rm atio n would be ex p ected .
From th e above c o n s id e ra tio n s i t would seem to be ap parent t h a t
i n te r a c tio n between sp e c ie s p re s e n t in fu sed s a l t m ix tu res must be tak en
in to account in any th e o r e t ic a l approach to th e problem o f p r e d ic tio n of
decom position p o t e n t i a l s .

I n a l l p r o b a b ility such in te r a c tio n i s as con­

t r o l l i n g i n fu sed s a l t s as i n aqueous s o lu tio n s .

The very hig h conductance

shown by fu se d s a l t s , which has commonly been assumed to be due to n e a rly
complete io n iz a tio n o f th e components, i s no c e r ta in in d ic a tio n o f th e
f a c t as i t i s p o s s ib le t h a t o th e r modes of e f f e c tiv e e le c tr o n t r a n s f e r
may be p r e s e n t.
I n approaching th e problem o f th e s e le c tio n of s u ita b le a d d itiv e s ,
n o t decomposed under c o n d itio n s where calcium i s d e p o s ite d , th e l i t e r a t u r e

19

was surveyed w ith re s p e c t to r e p o rts o f experim ental and commercial
o p e ra tin g r e s u l t s on m ix tu re s c o n ta in in g calcium io n s , w ith most emphasis
on f lu o r id e m e lts c o n ta in in g calcium f lu o r id e .

For purposes o f a g en eral

in tr o d u c tio n o f th e t o p i c , th e w idely c r i t i c i z e d d a ta of Neumann and
R ic h te r (118) i s p a r t i a l l y reproduced i n th e fo llo w in g t a b l e .

I t is

emphasized t h a t th e se v a lu e s are pro b ab ly not even r e l a t i v e l y c o r r e c t,
b u t th ey a re i l l u s t r a t i v e o f th e e f f e c t s produced by phenomena such as
m etal fo g s , r e a c tio n s o f anode gas w ith the anode, and the a c c e le r a tio n
o f such sid e r e a c tio n s by in c re a s e in te m p eratu re.

The f i r s t mentioned

su b stan ce o f a m ixture i s th e one which th e in v e s tig a to r s b e lie v e d they
were decomposing.
In s p e c tio n o f th e ta b le in d ic a te s a number o f ap parent p e c u l i a r i t i e s .
The decom position p o te n tia ls o f f lu o r id e s are probably a l l in th e wrong
r e la tio n s h ip s to th e c h lo r id e s , i . e . th e form er should be th e h ig h e r,
b u t because of d e p o la riz a tio n e f f e c ts and re a c tio n s w ith th e anode an
e f f e c tiv e u n d erv o ltag e i s o p e ra tin g to reduce th e decom position p o t e n t i a l .
The n e g a tiv e tem p eratu re e f f e c t i s so extreme th a t i t would appear t h a t
w ith b u t l i t t l e more e le v a tio n o f tem perature th e compounds would decom­
pose sp o n tan eo u sly , an e f f e c t which i s h a rd ly r e a liz a b le in p r a c ti c e .
The g e n e ra l o b se rv a tio n th a t a l k a l i m etals are h ig h , a lk a lin e e a r th
m eta ls s l i g h t l y lo w er, and Group I I I m etals s t i l l lower i n th e s in g le
p o t e n t i a l f o r m etal d e p o s itio n i s s u b s ta n tia lly c o r r e c t as judged by
o th e r r e s u l t s .

The e x a c t o rd er and v o lta g e se p a ra tio n s between elem ents

i s n o t as in d ic a te d , acco rd in g to th e r e s u l t s o f o th e r in v e s tig a to r s work­
in g w ith a more lim ite d s e r ie s o f compounds and m ix tu re s.

20

questionable values of decomposition potential of

NEUMANN AND RICHTER

Compound or M ixture
_____________

LiF ( + 1.5% KCl)
NaF ( + KC1)
KF
BeF2 ( + KCl)
MgF2 ( + KCl)
CaF 2 ( + KCl)
SrFa ( + KCl)
BaF 2 ( + KCl)
a i f 3 ( + KF + KCl)
LiC l
NaCl ( + 50 mole % KCl)
KC1 ( + £0 mole % NaCl)
CsCl
RbCl
BeCl2
MgPl2
CaCl2
BaCl2
A1C13
CeCl3
LaCl3
NdCl3
P r € l3

Decomp. P o t e n t ia l a t Temp.
V o lts
°C .

1.26
0 .9 6
1.30
0.70
0.1*3
0 .7 5
0.91
0.99
0.53
2.59
2.75
2.90
3.28
2.93
1.57
2.23
2.82
3.06
1.1*9
2.10
1.65
1.55
1.1*5

787
827
782
737
738
783
783
783
783
61*7
71U
71U
657
763
737
783
638
657
650
738
783
800
737

Decomp. P o te n tia l a t Temp.
V o lts
°C .
872
872
872
836
805
872
872
872
855
737
872
872
737
872
827
881
773
755
756
882
863
855
872

0.95
0.77
0.92
0.52
0.16
0.38
0.55
0 .6 3
0 .3 2
2.57
2.51
2.66
3.15
2.80
1.38
2 .1 6
2.72
2.99
1.37
1.83
1.50
1.55
1 .0 2

In order to somewhat l i m i t the f i e l d o f t h i s d is c u s s io n , i t may be
sa id th a t the p o s s ib le ch o ice o f s u ita b le a d d itiv e s i s e s s e n t ia ll y lim ite d
to p e r io d ic Groups l a , I l a , I l i a and p o s s ib ly IVa.

Subsequent elem ents and

t r a n s it io n s e r ie s elem ents o f b —sub—groups e ith e r form q u ite v o l a t i l e
flu o r id e s or th e m eta ls have r e l a t i v e l y low p o t e n tia ls req uired fo r d ep o si­
t i o n , so th a t th ey are in a p p lic a b le fo r th e p resen t pu rp ose.

Indeed the

l a t t e r as a r u le c o n s t it u t e a very troublesom e c la s s o f im p u r itie s in m elts
in ten d ed fo r the p rep aration o f th e Groups l a , I l a and I l i a m eta ls fo r two

21

reaso n s.

Not o n ly do th ey te n d to d e p o s it as im p u r itie s , b u t th e y may

ta k e p a r t in a s e r i e s o f p a r a s i t i c o x id a tio n and re d u c tio n r e a c tio n s
w ith in th e c e l l , to th e e x te n t th a t y ie ld s may approach z e ro , because
of t h e i r a b i l i t y to assume d i f f e r e n t o x id a tio n s ta t e s ( 23 ) ( 21;) ( 1;1 , 35 ) .
The p o s s i b i l i t y o f d e p o s itio n o f calcium from a s o lu tio n in one o f
the v o l a t i l e flu o r id e s was con sid ered i n i t i a l l y , but the r e a c t iv it y o f
the h alogen f lu o r id e s , boron t r i f l u o r i d e , and o th ers i s so great th a t
no calciu m m etal y i e l d might be e x p ected .

Thus fa r calcium flu o r id e i s

known to be so lu b le in only one o f the v o l a t i l e f lu o r id e s , boron t r i ­
f lu o r id e , w ith form ation o f a compound CaF2:BF3 , which decomposes on
moderate h e a tin g (175) .
In th e fo llo w in g t a b le , th e m eltin g p o in t data fo r flu o r id e s o f the
Groups l a to IVa elem ents have been ta b u la ted in order th a t the p o s s i­
b i l i t i e s on ch o ice o f a d d itiv e s may be n oted .
substan ces have been in clu d ed fo r co m p leten ess.

Some liq u id and gaseous
L ikew ise any a v a ila b le

v a lu e s fo r c e r ta in elem ents such as rubidium , cesium , scandium, y ttriu m ,
lanthanum and hafnium have been shown although th e ir s c a r c it y p r o h ib ite d
th e ir c o n sid e r a tio n as p r a c tic a l a d d it iv e s .
As may be se e n , th e data fo r Groups I and I I i s reasonab ly complete
although th e v a lu e s may not be w e ll e sta b lis h e d in a l l c a s e s .

About tw o-

th ir d s o f th e s e v a lu es are given erro n eo u sly in current handbooks and
some o f th e remainder have not been checked by modern w orkers.

With

r e s p e c t to Groups I I I and IV , data i s incom plete and somewhat c o n tr a d ic to r y .
T h is may have been the r e s u lt o f no e f f o r t , or in p a r t i t i s probably due
to th e f a c t th a t th e s e compounds may sublim e in ste a d o f m e ltin g .

Thus in

Group I I I , p o s s i b i l i t i e s o f d im er iz a tio n e x i s t which may le a d to su b lim a tio n .

22

MELTING POINT DATA FOR FLUORIDES OF THE ELEMENTS OF
GROUPS l a , I l a , I l i a and IVa

Compd.

M. P t . ,
°G .

HF
LiF
NaF
KF
RbF
CsF

-83
81+5
992
857
775
682

BeF 3
MgF 2
CaF 2
SrF 3
BaF 2
RaF2
Note:

800?
1263
llpL8
lilOO
1353
(11+27)

R eference

(11+3)
(79)
(79)
( 11 )
( 11 )

Compd.

M. P t . ,
°C .

b f3
a if3
ScF 3
if3
LaF 3

-127
1272
(1227)
(1387)
(11+27)

cf4
SiF 4
TiF 4
ZrF 4
HfF 4
ThF4

(See note)
(79)
(116)
( 11 )
( 11 )

(ID

Reference

-183
-90
B .P t. 281+
( S u b l. 927)
( S u b l. 927 )
1027

( 11 )
( 11 )

(ID
( 11 )
( 11 )
( 11 )
( 11 )
( 11 )
( 11 )
( 11 )

BeF2 i s polym eric and t h i s value i s only an approximate
s o fte n in g p o in t.
V alues in p a re n th e se s are u n c e r ta in or e stim ated v a lu e s .

From th e above in fo rm a tio n i t would appear th a t Group I l i a and IVa
f lu o r id e s would p ro b ab ly not be s u c c e ss fu l a d d itiv e s as such because o f
i n s t a b i l i t y , w ith th e ex cep tio n o f A1F3 .

I f however th e y were complexed

by th e fo rm atio n i n th e m elt o f f lu o —
a n io n s, such as f l u o - t i t a n a t e s and
f lu o - z ir c o n a te s , i t would be expected t h a t th e y would be s ta b le .

Of

th e se elem en ts, aluminum, zirconium and p o s s ib ly tita n iu m might be s e r v ic able i n th e form o f complex a n io n s .

Because o f th e experim ental d a ta

a v a ila b le on aluminum p ro d u ctio n from complex flu o r id e m e lts , i t was
decided t h a t complex f lu o r id e s o f th ese two groups would probably have
too low a decom position p o t e n ti a l f o r th e complexed m etal ( 1+2 , 12 ) .

23

The therm al red u c tio n s tu d ie s on some o f th e se m etals le a d to s im i­
la r c o n c lu sio n s ( 3 9 ) .

Thus Marden and Rich (lOli.)

reduced potassium

flu o -z ir c o n a t e w ith aluminum, R uff and B r in tz in g e r (136)

reduced

zirconium oxid e w ith a m ixture o f calcium and sodium and de Boer and
F a st (2 2 )

reduced th e oxide w ith magnesium and sodium.

W hile such

s tu d ie s are not c o n c lu siv e w ith r e s p e c t to r e la t iv e e l e c t r o p o s i t i v i t y ,
th ey g e n e r a lly in d ic a te a low er decom position p o t e n tia l than th a t o f
calcium or an i n s t a b i l i t y w ith r e sp e c t to i t as th e m etal.
Of th e Group l a and I l a m e ta ls , rubidium, and cesium were ru led out
as im p r a c tic a l, even fo r sm all s c a le t e s t work, because o f c o st co n sid era ­
t io n s .

T heir r e l a t i v e l y low gaseous io n iz a t io n p o t e n t ia ls would In d ic a te

th e ir p o s s ib ly high p o s it io n i n th e e l e c t r o - p o s i t i v i t y s e r ie s in fu sed
s a lt s as w e ll as in aqueous s o lu t io n s .
a v a ila b le (l[|3>,lU6,ll47)

However, the s l i g h t in form ation

would in d ic a te th a t cesium i s low er than barium

(and probably calcium ) in fu sed s a l t d e p o s itio n p o t e n t ia l.
The rem aining p o s s i b i l i t i e s , L i , Na and K in Group I and B e, Mg, Ca,
Sr and Ba in Group I I were co n sid ered w ith r e s p e c t to th e ir order o f
d e p o s itio n , p a r tic u la r ly from flu o r id e m e lts .

Three o f th e se may be

excluded from c o n sid e r a tio n fo r th e fo llo w in g rea so n s.
Magnesium d e f i n i t e l y ranks lower than calcium in d e p o s itio n p o t e n tia l
from a c o n s id e r a tio n o f the widespread use o f calcium flu o r id e as an
a d d itiv e to m e lts used fo r magnesium p ro d u ctio n .

C oncentration s o f calcium

f lu o r id e up to 20% by w eight are reached b efo re the magnesium product i s
a p p reciab ly contam inated w ith calcium ( U l ) .

In co rp o ra tio n o f sodium and

potassium s a l t s in magnesium c e l l m ixtures in co n cen tra tio n s as high as

2h

90% KC1-10^ M ^!l3 and 70% NaCl-30$ M^312 d e p o sit s a t i s f a c t o r i l y pure
magnesium a t low p o t e n t ia ls .

However, w ith normal o p era tin g p o t e n t ia ls

and g r ea ter cu rren t d e n s it ie s the sodium and potassium co n ten ts must be
co n sid e ra b ly low ered to p reven t a lk a li m etal im purity d e p o s itio n (l|.l) .
In c o n tr a s t to some o f th e se r e s u l t s , R uff and Busch (135)

found th a t

in flu o r id e m elts the order o f in c r e a s in g d e p o s itio n p o t e n t ia l was

K,

Na, Mg, B a, L i and Ca, and th ey p a r tic u la r ly condemned th e in c lu s io n o f
Na and Ba flu o r id e s i n magnesium b a th s.

They found th a t in co rp o ra tio n o f

LiF and CaF3 p erm itted produ ction o f com m ercially pure magnesium d ir e c t ly .
From th e above, d e s p ite some apparent c o n tr a d ic tio n , i t i s c le a r th a t
magnesium has a low er d e p o s itio n p o t e n tia l than calcium in general and
in p a r tic u la r from flu o r id e m e lts .

A co n sid era b le body o f inform ation

s u b s ta n tia te s t h is c o n c lu s io n , fo r example th e work o f Grube and J a is le

(56)

and Grube and Henne (55)

may be m entioned.

With r e s p e c t to strontium and barium flu o r id e s i t should be noted
th a t th e m e ltin g p o in ts o f l i |00° and 1353° r e s p e c t iv e ly make them in ­
h e r e n tly somewhat u n d esirab le as a d d itiv e s sin c e th e p ro sp ects o f o b ta in ­
in g a low m e ltin g e u t e c t ic m ixture w ith calcium flu o r id e are p o o r.

Their

d e p o s itio n p o t e n t ia ls have been rep o rted by Arndt and W illn er ( 3) to be
so c lo s e to th a t o f calcium in c h lo r id e m elts th a t the d iffe r e n c e i s
n e g lig ib le in p r a c t ic e .

Barium flu o r id e has been used in terch an geab ly

w ith calcium flu o r id e in some experim ental e le c t r o ly s e s ( 56) (55) ( 66)
( 57) ( 6l ) as an a d d itiv e w ithout s ig n if i c a n t d iffe r e n c e s being noted .
Troutz (170)

has shown th a t, in c h lo r id e m e lts , strontium (91%)

d e p o s its from a 32% KC1-68# SrC l3 m ix tu re.

Neumann and Bergve ( 1 1 7 ) ,

25

working w ith a 26% K C l-7l$ SrC l2 m elt,con firm ed t h is b eh aviou r, but
under comparable c o n d itio n s ob tain ed no y ie l d o f barium m e ta l.

Probably

th e b e s t c o n c lu sio n th a t may be drawn from th e se r e s u lt s i s th a t K, Ca,
Sr and Ba are c lo s e ly grouped w ith r e sp e c t to decom position p o t e n t ia l,
and th a t in a l l lik e lih o o d Sr and Ba do not rep resen t s u ita b le a d d it iv e s ,
p a r t ic u la r ly i f a co n sid era b le low erin g in m elt temperature i s n e c e ssa r y .
Inform ation on th e r e la t iv e d e p o s itio n p o t e n t ia l o f sodium i s
r e l a t i v e l y c le a r cu t as th e g rea t m a jo rity o f r ep o rts show th a t i t
d e p o s its ahead o f calciu m , from most m e lts , and from flu o r id e m elts in
p a r t ic u la r .

P e r tin e n t in form ation i n t h i s r e sp e c t i s given by numerous

in v e s tig a to r s (135) (56) ( 5 5 ) .
and von Kuegelgen (21)

The work reported by D anneel, Stockem

i s s l i g h t l y c o n tr a d ic to r y in th a t i t rep o rts an

interchange o f the normal p o s itio n s o f sodium and calcium when tempera­
tu r e s exceed 800° in c h lo r id e m e lts .

The evidence is ,h o w e v e r ,q u ite

c o n c lu siv e th a t sodium flu o r id e should not c o n s titu te a s u ita b le a d d itiv e
fo r the purpose o f t h is work.
The in form ation w ith regard to the r e la t iv e p o s it io n o f potassium
i s con fu sed , some r e p o r ts in d ic a tin g th a t i t f a l l s below calcium ( 135)
( i d ) , o th er s in d ic a tin g approxim ately th e same range fo r both (llU ) ( U3)
( 2 1 ) , and s t i l l o th ers in d ic a tin g th a t potassium has the h ig h er d ep o si­
t io n p o t e n t ia l ( il l; ) (117) .

The apparent c o n f l i c t s are probably caused
t

In p a rt by tem perature and com position f a c t o r s , but the evidence in d ic a te s
th a t calcium and potassium are c lo s e to g eth er w ith r e s p e c t to d e p o sitio n
p o t e n t i a l , w ith potassium probably s l i g h t l y the high er o f the tw o.

The p o s itio n o f lith iu m i s lik e w ise u n c e r ta in w ith r e s p e c t to calcium .
Less in fo rm a tio n i s a v a ila b le , and the co n clu sio n i s t h a t calcium and

26

lith iu m are lik e w is e very c lo s e to g e th e r , w ith somewhat l e s s d is a g r e e ­
ment among the r e p o r ts th a t p la c e lith iu m s l i g h t l y above (21) (118) and
s l i g h t l y below calcium (135) .

The work o f Guntz (60) in d ic a te s th a t

lith iu m and potassium are c lo s e to g e th er w ith lith iu m b ein g predom inantly
d e p o site d from c lilo r id e m e lts .

The commercial p rep a ra tio n o f a llo y s o f

lith iu m and calcium in w id ely varying p rop ortion s by s in g le step
e le c tr o ly s is

o f fu se d lith iu m -c a lciu m h a lid e m ix tu r es, as noted by

Osberg (122)

and V a l’dman (171)

i s fu r th e r in d ic a tio n o f th e s im ila r

e l e c t r o l y t i c behaviour o f th e s e elem en ts.
The rem aining elem en t, b e r y lliu m , has s p e c ia l c h a r a c t e r is t ic s th a t
d is tin g u is h i t from th o se p r e v io u s ly m entioned.

Probably b e r y lliu m , in

an uncomplexed form., would d e p o sit a t a lower p o t e n tia l than calciu m , as
one in v e s tig a to r has p la ced i t below sodium in the p o t e n tia l s e r ie s (25) ,
and i t s p rep aration from m elts c o n ta in in g barium flu o r id e has been com­
m e r c ia liz e d .
The com plexing te n d en cie s o f b ery lliu m are q u ite str o n g , however,
and i t shows co n sid era b le a b i l i t y to form sta b le complex h a lo -b e r y lla te s
such as m| BeX4

and

BeX4 ,.

I t was observed th a t a t lower tem p eratures,

such as near th e m eltin g p o in t o f Na2BeF4 a t about 7 0 0 -800°, o n ly cathodic
d e p o s itio n o f sodium was ob tain ed {kk ) , whereas in th e case o f h igh er
m e ltin g MgBeF4 or BaBeF4 a t about 1000° co n sid era b le d is s o c ia t io n apparent­
l y occu rred , and B e, rath er than Mg or B a, was ob tain ed in the cathod ic
d e p o s it .
In view o f the absence o f h ig h ly s u ita b le a d d itiv e s , i t was decided
th a t b e r y lliu m flu o r id e should be t r ie d as a c o n s titu e n t o f the m elt on

27

the th eo ry th a t a t low er tem peratures i t might be complexed so co m p letely
th a t i t would be p r e se n t in s u b s ta n tia lly o n ly the a n io n ic form.

An

a d d itio n a l advantage a n tic ip a te d was th a t th e a d d itio n o f a substance
forming another s p e c ie s o f anion would be more e f f e c t iv e in producing the
d e sir e d lo w erin g than would the a d d itio n o f a substance c o n tr ib u tin g o n ly
flu o r id e a n io n s .
From th e above in form ation i t was concluded th a t potassium and
lith iu m were th e b e s t a v a ila b le sim ple c a tio n ic a d d it iv e s , although
t h e ir d e p o s itio n p o t e n t ia ls so c lo s e ly approach th a t o f calcium th a t in
p r a c tic e s e l e c t i v e d e p o s itio n o f calcium might not be p o s s ib le .

As d i s ­

cu ssed in th e next s e c t io n , a method was sought by which the e f f e c t i v e
d e p o s itio n p o t e n t ia ls might be s h if t e d to favor calcium d e p o s itio n .
E ffe c t o f L iquid M etal Cathodes on P o t e n tia l
The in form ation in the p reced in g s e c tio n refer re d to d e p o sitio n
p o t e n t ia l r e la t io n s h ip s o f the m eta ls in s u b s ta n tia lly pure form.

D ep osi­

t io n o f a m etal upon th e su rfa ce o f another liq u id m etal w ith form ation
o f an a llo y might be expected to decrease th e e f f e c t iv e d e p o sitio n
p o t e n tia l by an amount dependent upon the h eat o f form ation o f the a l l o y .
Thus i t might be p o s s ib le to s e l e c t i v e l y d e p o s it calcium in the form o f
an a llo y w ith some liq u id m e ta llic cathode w ithout s u b s ta n tia l d e p o s itio n
o f a d d itiv e c a tio n s such as lith iu m or p otassium .
The co n scio u s use o f t h i s method fo r the purpose o f s h if t in g d e p o si­
t io n p o t e n t ia ls in commercial fu sed s a l t e l e c t r o l y s i s has not been found
in the l i t e r a t u r e , but a few experim ental s tu d ie s have been made which

28

appeared en cou ragin g.

J e llin e k and C zerw inski (77)

working w ith m ixtures

o f varyin g p rop ortion s o f sodium c h lo r id e w ith a lk a lin e ea rth c h lo r id e s ,
showed th a t up to approxim ately te n tim es as much o f the Group I I m etal
was s e l e c t i v e l y d e p o site d in a m olten lea d cathode fo r approxim ately
equirriolar m ixtures o f th e s a l t s .

For m ixtures c o n ta in in g approxim ately

90 mole p e r ce n t o f sodium c h lo r id e , the w eight p ercen ts o f sodium and
Group I I m etal d e p o site d were approxim ately e q u a l.
C onsiderable commercial p rep a ra tio n o f ca lciu m -lea d a llo y s has been
c a r r ie d out in m e lts u sin g a d d itiv e s o f sodium c h lo r id e or potassium
c h lo r id e , u s u a lly to calcium c h lo r id e m e lts which used m olten le a d ca th o d es.
Numerous r e fe r e n c e s are given fo r t h i s p ro cess by Englehardt (1+5).

While

such p r o c e sse s are operable i t has been noted th a t i t i s very e s s e n t ia l
to provide la r g e cathode a r e a s , p r e fe ra b ly w ith a g ita t io n and co n tin u a l
p r e se n ta tio n o f fr e s h lea d s u r fa c e s .

The extrem ely slow d if f u s io n r a te s

o f th e d ep o sited calcium and th e compound Pb3Ca apparently are c o n tr o l­
l i n g and i f the p ro cess i s not a s s is t e d in some way the su rface becomes
sa tu r a te d w ith calcium and th e d e p o s itio n p o t e n tia l i s s h ift e d to a more
normal valu e ( l l l i ) .
Inform ation on v a rio u s common low m eltin g m etals was c o lle c t e d w ith
r e s p e c t to the a llo y system s which th ey form w ith p o t e n t ia l c o n s titu e n ts
o f a calciu m flu o r id e m e lt.

Based on la t e r inform ation as to the probable

o p era tin g tem perature c e r ta in e lim in a tio n s were p o s s i b l e .

C o n sid era tio n s

such as r e la t iv e a v a i l a b i l i t y and a llo y in g te n d en cie s w ith calciu m ,
lith iu m , p otassiu m , and b ery lliu m in d ic a te d th a t probably le a d was g en era l­
l y most s u it a b le , p a r t ic u la r ly as the in form ation was l e s s complete fo r

29

o th e r p o s s ib le a llo y in g m etals such as cadmium, z in c , t i n and bism uth.
R eferen ces to p e r tin e n t a llo y system s are given in Appendix i*.
C alcium -lead a llo y s have been p re p a re d from b a th s c o n ta in in g sodium
c h lo rid e o r p o tassiu m c h lo rid e as a d d itiv e s (112) (1 3 9 ), and th e p a r t i a l l y
s e le c tiv e d e p o is tio n o f calcium on a m olten le a d cathode ahead o f sodium
from c h lo rid e m e lts had been dem onstrated by J e llin e k and C zerw inski (77) .
I t was decid ed to t r y th e com bination o f calcium , lith iu m and b ery lliu m
f lu o r id e s i n c o n ju n ctio n w ith a le a d cathode to see whether calcium would
be d e p o site d even more s e le c tiv e ly i n th e presence o f lith iu m .

I n th e

event t h a t t h i s proved n o t to be th e c a s e , i t was f e l t th a t potassium
f lu o r id e m ight be s u b s titu te d f o r th e lith iu m f lu o r id e , to see whether
th e normal o rd e r o f in c re a s in g d e p o s itio n p o te n tia l o f p o tassiu m , lith iu m
and o alciu m , as given by R uff (135) f o r flu o r id e m e lts , could be s h if te d
to fav o r calcium d e p o s itio n .
I n s u f f ic ie n t h e a t o f fo rm atio n d a ta were a v a ila b le to perm it e stim a­
t i o n o f th e r e l a t i v e e f f e c t o f a llo y fo rm atio n on the d e p o s itio n p o te n tia ls
o f lith iu m and calciu m .

For lith iu m -le a d compounds no in fo rm atio n was found.

For calciu m , th e h e a ts o f fo rm ation o f th e compounds C a ^ b and CaPb have
been re p o rte d to be h7 said 25 K cal./m ole r e s p e c tiv e ly (88) , b u t no d a ta
a re a v a ila b le f o r th e compound CaPb3 which i s of most i n t e r e s t .

An e stim a te

o f 10-15 K cal./m ole f o r a minimum v alu e o f th e h e a t o f fo rm atio n o f th e
l a t t e r compound would p ro b ab ly be re a s o n a b le .

The lith iu m —
le a d compounds

a re c o n s id e ra b ly low er m e ltin g th an th e calcium compounds which might
be in d ic a tiv e o f somewhat low er h e a ts o f fo rm atio n f o r th e lith iu m a llo y s .

30

The phase diagram o f th e ca lciu m -lea d system (8 8 ) i s given below to
i l l u s t r a t e the p o in t th a t o p e r a tio n i s not p o s s ib le a t very low tempera­
tu r e s w ith very a p p reciab le calciu m c o n te n ts .

11001000-

70 0 -

500'

300 -

90

CaPb

CaPb

Ca,Pb

31

A v a ila b le Inform ation on Phase R e la tio n s
and A n ticip a te d Behaviour
At the s t a r t o f t h i s work, no in form ation was a v a ila b le on the
b in a ry system s and the tern a ry system formed by calcium f lu o r id e ,
lith iu m flu o r id e and b e r y lliu m f lu o r i d e .

A survey o f inform ation on

analogous system s was made to lea rn whether th ere might be any c o r r e la ­
t io n which would perm it p r e d ic tio n o f th e general behaviour o f the system s
o f in t e r e s t in t h i s in v e s t ig a t io n .

R eferences to inform ation obtained

in t h i s p relim in a ry su rv ey , and to subsequently p u b lish ed r e p o r ts , have
been summarized in Appendix 1 .
Inform ation on th e thermal and oth er p h y sic a l p r o p e r tie s o f the
th ree c o n s t it u e n t s , calciu m , lith iu m and b erylliu m f lu o r id e s , has been
c o lle c t e d in Appendix 2 .
Since in form ation on compounds formed in the ternary and

binary

system s might have been rep orted in d ep en dently o f any phase diagram
s t u d ie s , a lit e r a t u r e search was made fo r the methods o f p rep a ra tio n and
p r o p e r tie s o f compounds which might be expected to form.

P e r tin e n t in ­

form ation and r e fe r e n c e s have been assem bled in Appendix 3 .
Based on th e above in form a tio n the fo llo w in g behaviour was a n t i c i ­
pated fo r th e

b in ary and tern ary system s.

The calcium f lu o r id e -lith iu m flu o r id e system was exp ected to show
a sim ple e u t e c t ic behaviour w ith no compound form ation.

I t was b e lie v e d

th a t the e u t e c t ic would occur a t approxim ately 30 mole p ercen t o f calcium
flu o r id e a t about 750° .

32

In th e calcium flu o r id e -b e r y lliu m flu o r id e system form ation o f the
compound CaBeF4 was exp ected and two e u t e c t ic s were a n tic ip a te d .

No

r e l i a b l e e stim a te s could be made as to the p o s itio n s or m eltin g tempera­
tu r e s o f the e u t e c t ic s or o f th e compound.
In th e lith iu m flu o r id e -b e r y lliu m flu o r id e system , compound form­
a tio n was a n t ic ip a te d , a t l e a s t fo r L i3BeF4 and p o s s ib ly fo r other
compounds.

The e u t e c t ic in th e r e g io n o f 30 mole p ercent BeF2 was

expected to be rath er sh allow w ith r e sp e c t to th e peak fo r the compound
L i3BeF4 .

The l a t t e r was expected to have a m eltin g p o in t in the range

o f U50 to 5 0 0 °.

Another lower e u t e c t ic was a n tic ip a te d in th e reg io n

o f about 50 mole p ercen t BeF3 a t a temperature o f about ii.00-ij.500 .
In g e n e r a l, the a n tic ip a te d bin ary system s have a f a i r l y c lo s e
resemblance to th o se a c tu a lly found.

However, fo r the a n tic ip a te d ternary

system , the p r o g n o s tic a tio n s proved l e s s a c cu ra te.

No inform ation was

a v a ila b le to in d ic a te th e form ation o f a compound w ith in the ternary
system .

From th e number o f components and the a n tic ip a te d b in ary

in v a r ia n t p o in t s , a minimum o f th ree in v a r ia n t p o in ts .was expected w ith in
the tern a ry system .

These p o in ts were estim ated as probably ly in g out in

th e r e g io n o f 10 to 20 mole percen t CaF3 , a co n sid era b le o v er-estim a te
in th e l i g h t o f subsequent d a ta .

Temperatures o f the in v a r ia n t p o in ts

n a tu r a lly were exp ected to l i e somewhat below the b in ary in v a r ia n t p o in ts ,
but not as low as proved to be th e c a s e .
In th e b ery lliu m flu o r id e r ic h s e c tio n o f the system some d i f f i c u l t y
was expected because o f t h e 'form ation o f g la s s e s rather than c r y s t a ls ,
bu t the s e v e r it y o f th e e f f e c t and i t s rath er wide area o f in flu e n c e was
not a n t ic ip a te d .

33

These e stim a te s o f th e p o s s ib le behaviour have been g iv en sin c e
th ey o u tlin e th e p o in t o f view used i n in te r p r e tin g i n i t i a l r e s u l t s .
A p e r f e c t ly unbiased approach to the study o f a tern ary system i s prob—
ab ly as poor as th e o p p o site extreme o f r e ta in in g preconceived id e a s ,
as th e treatm ent i s b e s t handled by v a r ia tio n s to t e s t the c o r r e c tn e ss
o f in t e r p r e t a t io n s .

The Phase Rule
The com p lication o f phase r e la t io n s caused by the in tro d u c tio n o f
a th ir d component, w ith the consequent p o s s i b i l i t y o f in te r a c tio n s
between tern ary compounds, b in ary compounds and the th ree components,
in c r e a s e s the p o s s ib le perm utations in to th e order o f hundreds fo r on ly
m oderately com plicated system s.

C onsequently no t e x t d e a lin g w ith phase

r e la t io n s may reason ab ly hope to ex p la in and i l l u s t r a t e more than a very
sm all f r a c tio n o f the p o s s ib le c a s e s .
U n fortu n ately in most t e x t s in E n g lis h , th e su b ject o f th ree com­
ponent system s has u n t i l q u ite r e c e n tly r e c e iv e d o n ly a very sketchy
treatm en t, w ith the e x ce p tio n o f th e important system s formed w ith w a ter.
W hile th ere i s no form al d iffe r e n c e between aqueous and non-aqueous s y s ­
tem s, th ere i s the p r a c t ic a l d iffe r e n c e th a t most o f the work on aqueous
system s i s p resen ted in isoth erm al s t u d ie s , so th a t com plete d e s c r ip tio n s
o f behaviour over a wide range o f temperature are a r a r it y .

Hence such

d e s c r ip tio n s are u s u a lly not h e lp fu l fo r the non -isotherm al s tu d ie s o f
fu sed s a l t sy stem s.

3k

The unrecognized c o n tr ib u tio n s to Phase Rule th eo ry by Gibbs ( 5 l ) ,
were r ev iv e d in an ex h a u stiv e treatm ent in th e German language by
Roozeboom ( 1 3 3 ) .

Based on th e l a t t e r work, Masing (109) wrote a much

l e s s com plete t e x t i n German, which has been tr a n s la te d in to E n g lish by
Rogers ( 1 3 0 ) .

A lso in E n g lish th ere i s a treatm ent by Marsh ( 1 0 5 ) , based

in p a rt on Masing*s t e x t .

U n til the appearance during the p a st y ear o f

s e v e r a l new and r e v is e d t e x t s , th ose o f Rogers and Marsh were th e o n ly
a v a ila b le h e lp fu l summaries in E n g lish on the su b jec t o f ternary system s.
In d is c u s s in g the su b jec t o f th e Phase Rule and phase diagrams
c e r ta in c o n v en tio n a l assum ptions w i l l be made.

The system s w i l l be

tr e a te d as “condensed systems'* u n a ffe c ted by normal pressu re changes,
magnetic and g r a v ita tio n a l f i e l d s , e t c .

In a d d itio n , ex cep t where s ta te d

to the c o n tr a r y , th e absence o f s o l i d s o lu tio n s w i l l be assumed fo r two
reason s.

F i r s t l y , th e presence o f s o lid s o lu tio n s g r e a tly com p licates

the phase diagrams and seco n d ly , no evidence was found in the presen t
work fo r th e occurrence o f any s u b s ta n tia l mutual s o l u b i l i t y o f th e s o lid
p h a se s.

N a tu ra lly such a complete la c k o f m is e i b i l i t y o f th e s o lid s i s

im p o ssib le .

However, i f i t i s sm all in e x te n t, i t p r a c t ic a lly reduces

to the case o f com plete im m is c ib ilit y , w ith c e r ta in ex ce p tio n s as w i l l
be n o ted .
I t i s a ls o im p lic it ly assumed th a t equilib riu m c o n d itio n s p r e v a il
and th a t a l l phase tran sform ation s occur in a com p letely th e o r e t ic a l
manner, although i t w i l l be shown th a t such i s not always the ca se in
th e system under d is c u s s io n .

35

The Phase R ule o f Gibbs d e a ls w ith th e e q u ili b r i a e x is tin g among
th e homogeneous p o r tio n s o f a heterogeneous system .

The homogeneous

p o r ti o n s , u s u a lly c a lle d p h a s e s , a re re g io n s having uniform p h y s ic a l
p r o p e r tie s and co m p o sitio n s, se p a ra te d from each o th e r by d i s t i n c t phy­
s i c a l b o u n d a rie s .

P re ss u re and tem perature are assumed to be uniform

th ro u g h o u t th e heterogeneous system .

The fundam ental h y p o th esis form ing

th e b a s is o f th e Phase Rule i s t h a t th e p r o p e rtie s o f a phase are
u n iq u e ly determ ined by sp e c ify in g values f o r a minimum number o f in d e ­
pendent v a r ia b le s , an e m p iric a lly v e r if ie d f a c t .
I t has been determ ined th a t f o r a phase made up of "nn components,
th e e q u a tio n o f s ta t e re q u ire d to com pletely determ ine th e p r o p e rtie s
o f th e phase must c o n ta in n + 2 v a r ia b le s in th e case of non-condensed
system s.

I n such c a s e s , two o f th e se v a r ia b le s must be tak en from th e

t r i o of in te rd e p e n d e n t v a r ia b le s , p , T and v ( s p e c if ic volume)j w hile n - 1
c o n c e n tra tio n v a r ia b le s must be chosen from th e n p o s sib le c o n c e n tra tio n
v a lu e s d e s c rib in g th e com position o f th e ph ase.

Then s p e c if ic a tio n o f

any p a i r o f th e p , v , T v a r ia b le s a u to m a tic a lly s p e c if ie s th e t h i r d , and
th e c o n c e n tra tio n o f th e Mn th M component i s a u to m a tic a lly determ ined
by s p e c if ic a tio n o f n - 1 c o n c e n tra tio n v a r ia b le s .

Hence i t i s ap p aren t

t h a t th e a r b i t r a r y choice o f 2 + ( n - l) = n+1 v a ria b le s w i l l i n th e o ry
p erm it c a lc u la tio n o f th e »n th w v a ria b le from th e eq u atio n of s t a t e ,
i . e . th e phase p r o p e r tie s a re com pletely f ix e d .
Thus f o r a homogeneous phase made up o f n components, w ith an
e q u a tio n o f s t a t e o f th e form —
f [ C i,C 2 , .......... c n - l , P , T( ° r v )] = 0 ,

36

th e s p e c if ic a t io n o f T and p (o r v ) and n-1 co n cen tra tio n s a u to m a tica lly
determ ines th e s p e c if ic volume (o r T or p ) and oth er p r o p e r tie s o f t h is
p h a se .
Some ex p la n a tio n o f the meaning o f 11components" i s n e c essa ry .

The

number o f components o f a heterogeneous system i s the minimum number o f
su b stan ces whose c o n c en tr a tio n s in the d if f e r e n t phases may be v a ried
in d ep en d en tly o f each o th e r , under the c o n d itio n s o f the experim ent.
The

s ig n if ic a n c e o f th e l a t t e r phrase may be c l a r i f i e d by se v e r a l

exam ples.
A c lo se d con tain er having liq u id water and water vapor p resen t at
ordinary tem peratures and p r e ssu r e s would be considered a one component
system .

At much high er tem peratures and p ressu res the same system would

be a two component system sin c e d if f e r e n t ia l s o l u b i l i t y e f f e c t s would cause
the com position o f th e p a r t ia l l y d is s o c ia t e d gas phase to be d if f e r e n t
from th a t o f the i n i t i a l s t a r t in g m a te r ia l, w ater.

L ikew ise the composi­

tio n o f the liq u id phase would be d if f e r e n t from the s ta r tin g m a te r ia l,
w a ter, fo r the same r ea so n s.

Thus even though a l l d erived from a s in g le

su b sta n ce, the system must be tr e a te d as a two component system , sin c e
a d d itio n o f water would not change the r e la t iv e co n cen tra tio n s o f H2 , 0 2
or H30 in th e two p h a se s.

Hence an a d d itio n a l component i s n ecessa ry to

perm it independent v a r ia tio n o f the co n cen tra tio n s in th e two phases and
th e system i s one o f two components under th ese c o n d it io n s .
I f H3 , 0 3 and w ater ( liq u id and vapor) are p resen t in a con tain er
a t low enough tem perature so th a t r e a c tio n does not o ccu r, the system
must be tr e a te d as one o f th ree components.

Here a d d itio n o f water would

37

in no way a l t e r th e com position o f th e gas p h a se, nor would a d d itio n o f
0 2 or

perm it independent a lt e r a t io n o f the co n cen tra tio n s in the

liq u id phase (H30 + d is s o lv e d 0 3 and H2) , co n sta n t t o t a l p ressu re b ein g
assumed.

I t i s apparent th a t in order to vary th e co n cen tra tio n s in d e­

p en d en tly in both p h a se s, a l l th ree substances must be c la s s e d as com­
ponents fo r th e se circu m stan ces.
I f the above system were heated to higher tem peratures where r e ­
a c tio n and d is s o c ia t io n occur in accordance w ith th e Mass A ction Law,

the system becomes a two component one ( i f e n t ir e ly gaseous) .

T his

fo llo w s , sin c e s p e c if ic a t io n o f any two o f the co n cen tra tio n s determ ines
the t h ir d , and con seq u en tly th ere must be two components whose concen­
tr a tio n s are ind ep en dently v a r ia b le to perm it s p e c if ic a t io n o f a system
o f any a r b itr a r y co m p o sitio n .

Thus th e c o n d itio n s and any mass a c tio n

r e la t io n s h ip s must be con sid ered in determ ining the number o f components
o f a p a r tic u la r system .
For a non-condensed system th e Phase Rule i s u s u a lly w r itte n as
P + F = n + 2 where P s i g n i f i e s th e number o f phases p r e s e n t, n the number
o f components and f th e number o f degrees o f freedom.

The 2 a r is e s from

th e assumed p o s s i b i l i t y o f temperature and p ressu re v a r ia tio n .

In the

even t th a t e ith e r or both tem perature and p ressu re are a r b it r a r ily h eld
con stan t th e e x p r essio n has the forms

38

P + F = n + 1 ,
P + F * n

fo r con stan t T or p .
, fo r con stan t T and p .

Xn th e p r e se n t work, the con v en tio n a l condensed system i s assumed w ith
co n sta n t p r e ss u r e , i . e . the e f f e c t o f v a p o r iz a tio n i s assumed to be
n e g lig ib le and on ly liq u id and s o lid phases are presumed to a f f e c t the
behaviour o f the tern ary system .
For a condensed system o f n components and P phases the Phase Buie
may be d erived in the fo llo w in g manner.

I f the n components can be

p r e se n t in each p h ase, th en sin c e n - 1 co n cen tra tio n v a r ia b le s are
s u f f i c i e n t to d e fin e the com position o f each p h a se, a t o t a l o f P(n - l )
co n c en tr a tio n v a r ia b le s are needed to s p e c ify th e com positions o f P
p h a ses.

In a d d itio n 1 must be added fo r the temperature v a r ia b le ,

pressure b ein g considered f ix e d .

Then th e t o t a l number o f v a r ia b le s

th a t need to be s p e c if ie d in order to d e fin e the s t a t e o f th e system are
equal to

P(n - l ) + 1 .

The chem ical p o t e n t ia l, or escapin g tend en cy, o f each component o f
a phase i s equal a t eq u ilib riu m to i t s chem ical p o t e n tia l in a l l other
phases o f th e system .

I f th e v a r ia b le s o f temperature and com position

are d efin ed c o m p le te ly , th en th e chem ical p o t e n tia l o f each component
i s t h e o r e t ic a lly e x p r e s s ib le as a fu n c tio n o f th ese v a r ia b le s .

Then,

w ith r e s p e c t to a phase co n ta in in g a l l th e components in eq u ilib riu m
w ith P - 1 oth er p h a se s, th ere are a v a ila b le P - 1 r e la t io n s fo r each
component showing the e q u a lity o f th e chem ical p o t e n tia l fo r each com­
p on en t.

For a l l n components th ere are n(P - l ) equations a v a ila b le

which in t e r r e la t e the com position s o f a l l p h a se s.

39

For a system r e q u ir in g th e s p e c if ic a t io n o f P(n - l ) + 1 v a r ia b le s
to d e fin e th e system , fo r which on ly n(P - l ) eq u ation s are a v a ila b le ,
th e d iffe r e n c e o f
[F (n - 1) + 1] - [n(p - 1 ) ]

= n - P +1 = F

i s th e number o f p o s s ib le independent changes o f s t a t e , a lso c a lle d
"degrees o f freedom ".

These rep resen t the changes in co n cen tra tio n s or

tem perature which are p o s s ib le w ith in the system w ithout v io la t io n o f
th e n(P - l ) equations r e la t in g th e com positions in a l l P p h a se s.

In

order to co m p letely d e fin e the system , F v a r ia b le s must be a r b it r a r ily
f ix e d so th a t th e rem aining v a r ia b le s w i l l be determined by th e e q u ilib r ia
e x is t in g among components in the d if f e r e n t p h a se s .
Thus fo r a condensed system , in g e n e r a l, F = n - P + 1 and fo r a
three component or tern ary system
F = 3 “ P + 1 = 1+ - P
Then fo r a tern ary system , the presence o f two phases perm its two degrees
o f freedom , i . e . f ix i n g o f one com position v a r ia b le and the temperature
i s n ecessa ry in order to d e fin e the system .

For th ree phases p r e se n t,

i t i s o n ly n e c essa ry to f i x one v a r ia b le , e . g . the tem perature, in order
to d e fin e th e system .

F in a lly fo r fou r phases p r e se n t, th ere are no

p o s s ib le d egrees o f freedom , i . e . t h is circum stance occurs as a sin g u la r
p o in t a t o n ly one temperature and s e t o f com position v a lu e s , and the
system i s c a lle d in v a r ia n t s in c e one o f th e phases must vanish b efore
any change i s p o s s ib le among the temperature and com position v a r ia b le s .
C onfining th e d is c u s s io n to ternary condensed system s in which one
o f th e phases i s a liq u id and th e o th ers s o l i d s , two ty p es o f in v a r ia n t

bo

p o in ts are p o s s ib le .

Sym bolizing the phases as

m e lt, A, B and C, two

p o s s ib le ty p es o f e q u ilib r ia may be w r itte n —
m elt
and

A + B +C

m elt + A <— - B + C

The f i r s t r e la t io n i s th a t o f the tern a ry e u te c tic p o in t at wliich a
m e lt, sim u lta n eo u sly sa tu ra ted w ith r e sp e c t to A, B and C s o lid p h a se s,
e x i s t s a t eq u ilib riu m w ith them a t a p a r tic u la r o v e r - a ll com position value
and a f ix e d tem perature.

By w ithdrawal o f h e a t, the eq u ilib riu m i s d riven

to th e r i g h t , u n t i l f i n a l l y w ith th e disappearance o f the l a s t o f the
m e lt, th e system becomes u n iv a ria n t and temperature or com position v a r ia ­
t io n o f the rem aining phases becomes p o s s ib le .
The second r e la t io n i s c h a r a c t e r is tic o f the ternary p e r it e c t ic
p o in t a t which a m elt sa tu ra ted w ith A reaches a p a r tic u la r com position
and temperature at which i t i s a lso sa tu ra ted w ith r esp e c t to B and C,
Upon withdrawal o f h e a t, i t i s commonly explain ed th a t the m elt '‘reacts"
w ith

A to form Band

upon d e p o s itio n o f

C.

A b e tte r way o f s ta tin g th e s itu a t io n

i s th a t

B and C th e m elt becomes un saturated w ith r e sp e c t

A and some o f i t d i s s o l v e s .

to

On c o n tin u a tio n o f t h i s process i t can be

seen th a t one o f two c o n c lu sio n s may be reached, depending upon the
r e la t iv e amounts o f A and m elt which are p r e se n t.

E ith er A may be in

e x ce ss in which case com plete s o l i d i f i c a t i o n occurs a t the tern ary p e r i­
t e c t i c p o in t , or the m elt may be in e x ce ss in which case A v a n ish es and
th e system made up o f m e lt, B and C i s again a u n i v a ria n t system fr e e
to undergo change o f tem perature or com position v a r ia b le s .

u

Analogous s in g u la r p o in ts occur in b in a ry condensed systems except
t h a t th e number o f s o lid phases p a r tic ip a tin g i s reduced by one.
I t should be noted th a t in p r a c tic e p e r i t e c t ic 11r e a c tio n s'1 may not
proceed as com p letely as th eory r e q u ir e s , because o f i n s u f f i c ie n t time
to a t t a in eq u ilib r iu m , e f f e c t iv e i s o l a t i o n o f p r e v io u sly d ep o sited s o lid
by a c o a tin g o f a new p h a se, and sim ila r d i f f i c u l t i e s .

L ikew ise the

form ation o f a s o lid phase may occur in a n o n -th e o r e tic a l manner in th a t
su p erco o lin g may occur to sm all or la r g e e x te n ts so th a t therm al an alyses
may y ie ld d e c ep tiv e r e s u l t s .
C e rta in g e n e r a liz a tio n s and s p e c if ic ca ses d e scr ib in g th e behaviour
o f tern ary system s have been given in Appendix $ .

The in te r p r e ta tio n o f

the experim ental r e s u lt s depends upon th e proper a p p lic a tio n o f th ese
id eas which are d eriv ed from gen eral Phase Rule theory and th e geom etrical
system o f r e p r e se n ta tio n which i s u se d .

EXPERIMENTAL

b2

EXPERIMENTAL
P rep aration o f,M a te r ia ls
P rep aration o f the m e lts req u ired fo r the study o f the b in ary and
tern ary system s fu rn ish ed a se r io u s problem in one r e s p e c t.

W hile

calcium flu o r id e and lith iu m flu o r id e were r e a d ily a v a ila b le , b ery lliu m
flu o r id e was more d i f f i c u l t to o b ta in and con seq u en tly much o f t h i s
m a teria l was prepared or reclaim ed by the methods to be d e scr ib e d .
Through th e c o u r te sy o f Mr. C arl Anderson o f the Ozark-Mahoning
Company, th er e were provided se v e r a l l o t s o f high grad e, hand s e le c t e d ,
n atu ral flu o r sp a r o r e .

One l o t o f t h is m a te r ia l, o f a q u a lity which

would be c a lle d " o p tic a l grade", was used alm ost e x c lu s iv e ly during t h is
work.

The tran sp aren t m a ter ia l was again hand s e le c t e d to remove p ie c e s

with any v i s i b l e in c lu s io n s o f fo r e ig n m a tter.
crushed and mixed to form a homogeneous sam ple.

The m a te r ia l was then
The standard method o f

a n a ly s is fo r low er grade flu o r sp a r was a p p lied ( 103) , but the in h eren t
in a c c u r a c ie s o f the method are such th a t th ey showed e s s e n t i a l l y no
im p u r itie s p r e s e n t , except fo r p o s s ib ly 0 .1% o f a c id so lu b le ca rb o n a te,
w ith n e g lig ib le tr a c e s o f im p u r itie s such as s i l i c a and heavy m e ta ls .
The calcium flu o r id e was th e r e fo r e con sid ered to be n ea rly 9 9 .9 % pure
and in a l l com putations was co n sid ered to be 100^ p u re.

D e ta ils o f

a n a ly se s and r e s u lt s have been c o lle c t e d in Appendix 1 5 , Part A.
Lithium flu o r id e used was J . T. Baker Analyzed grade and no attem pt
a t p u r if ic a t io n was made.

The t o t a l o f l i s t e d im p u r itie s in d ic a te d by

th e ir a n a ly s is was approxim ately 0.1^>.

h3

L ithium c a rb o n a te , J . T, Baker Analyzed g rad e, was used in s y n th e tic
wet p r e p a r a tio n o f L i2BeF4 .

In d ic a te d im p u ritie s o f approxim ately 0.5%

o f o th e r a l k a l i m etals p ro b ab ly d id n o t in tro d u c e any s ig n if ic a n t
im p u rity in to th e p ro d u c t.
Ammonium f lu o r id e and b if lu o r id e used i n th e p r e p a ra tio n o f (NH4) 2BeF4
were o f v a rio u s degrees o f p u r it y .

Im p u ritie s and p re p a ra tiv e methods

were such t h a t i t was u n lik e ly t h a t th e y c o n trib u te d im p u ritie s which
were n o t e lim in a te d d u rin g one o f th e p r e p a ra tiv e s te p s or i n th e v a rio u s
re c la m a tio n s te p s e a r r ie d o u t on th e m other liq u o r re s id u e s from th e
(NH4) 3BeF4 p r e p a ra tio n .
The supply o f th e n e c e ssa ry b e ry lliu m f lu o r id e p re se n te d much more
d iffic u lty .

One commercial l o t o f m a te r ia l ( l pound) was o b tain ed from

th e Brush B ery lliu m C o rp o ra tio n .

This b e ry lliu m f lu o r i d e , s ta te d to have

approxim ately 99% p u r i t y , ten ded to form s l i g h t l y cloudy m elts w ith th e
o th e r f lu o r i d e s .

A n aly sis confirm ed th e p u r ity as 99% c a lc u la te d from th e

a n a ly s is f o r b e ry lliu m m etal as th e o x id e .

(See Appendix 1 $ , P a r t B .)

Subsequent a ttem p ts to o b ta in a d d itio n a l q u a n tit ie s were u n s u c c e s s fu l, b u t
a f t e r th e co n clu sio n o f th e t h e s is work, th e American B ery lliu m C o rp o ra tio n
in d ic a te d t h a t i t co u ld th e n make a v a ila b le a p robably b e t t e r grade o f
m a t e r ia l .
B ery lliu m f lu o r id e used i n about th e f i r s t q u a r te r o f th e work on
b e ry lliu m f lu o r id e system s were p rep a re d from an o ld l o t o f about i|00 g.
o f Kahlbaum " b e ry lliu m carb o n ate" which was a v a i l a b l e .
m a te r ia l i s covered i n Appendix 15, P a r t B.

A n aly sis o f t h i s

kb

In th e very l a s t sta g e s o f th e work, some s o - c a lle d b e r y lliu m b a s ic
carbonate was ob tain ed from the A. D. MacKay Company.

The m a te r ia l was

n ot a n a ly z e d , but was put through the reg u la r b a s ic a c e ta te p u r if ic a t io n
p ro cess d escrib ed b elow .

C onsiderable q u a n tit ie s o f ir o n were removed

and o n ly a $0% y ie ld o f t h e o r e t ic a l was obtained in d ic a tin g th a t the
o r ig in a l m a te r ia l did not correspond to the co n v en tio n a l b a s ic carbonate
form ulated as (B e0)s :C02:5H20 .
Other than by th e r e a c tio n o f the e lem en ts, b ery lliu m flu o r id e in
pure anhydrous form has app arently o n ly been prepared by the c l a s s i c a l
method o f LeBeau (9 6 ) in which anhydrous (NH4)gBeF4 i s heated in a p ro­
t e c t iv e atmosphere to d rive o f f 2MH4F le a v in g the anhydrous BeF2 .

W hile

the method appears q u ite e le g a n t and no d i f f i c u l t i e s are d escrib ed in the
lit e r a t u r e , the procedure on a la b o ra to ry s c a le q u ick ly g e ts in to d i f f i ­
c u l t i e s , although i t i s app arently used by both la r g e commercial producers
in t h is cou n try.

B r i e f l y , as NH4F i s evolved th e mass o f m a ter ia l m e lts

to a th in liq u id which becomes th ic k e r and th ic k e r as i t approaches th e
com position o f BeF2 .

The com bination o f v is c o u s liq u id and. e v o lv in g NH4F

causes form ation o f p rod igiou s q u a n tit ie s o f foam resem blin g ,,Foamglasn .
The p rep a ra tio n in q u a n t it ie s , even as sm all as 5 to 10 grams, r eq u ir es
m anipulations th a t are h ig h ly incom p atible w ith the requirem ents fo r
carryin g out th e r e a c tio n in an atmosphere o f dry carbon d io x id e .
Although the b ery lliu m flu o r id e fo r the i n i t i a l in v e s t ig a t io n o f the
CaF2-BeF2 system was prepared in th e above manner by h ea tin g (NH ^gBeF^
on subsequent work i t was found th a t th e NH4F co u ld be r e li a b ly evolved
in a co n sid e ra b ly smoother manner i f e ith e r calcium or lith iu m flu o r id e s

US

were p r e se n t to low er the v i s c o s i t y o f the liq u id m ass.

R easonably good

r e s u lt s were obtained by h ea tin g such m ixtures in a platinum d is h under
an in v e r te d fu n n e l through which dry carbon d io x id e was passed a t a high
r a te .

Q u a n tita tiv e checks in d ic a te d th a t a ll the NH4F was evolved a f t e r

th e m elt cea sed rapid e v o lu tio n and th e o n ly v i s i b l e gas bu bbles were a
few adhering to the bottom o f the d is h .

This was true a t tem peratures

as low as U00°, whereas w ith (NH4) 2BeF4 a lo n e , co n sid era b ly high er
tem peratures or evacu ation i s n ecessa ry to remove the l a s t o f th e NH4F .
B erylliu m flu o r id e i s q u ite hygroscopic and unusual w ith r e s p e c t
to i t s high co o rd in a tin g a b i l i t y .

I t s aqueous s o lu tio n s w i l l y i e l d b a sic

s a l t m ixtures u n le s s the c o o rd in a tio n requirem ents are s a t i s f i e d by some
oth er means.

The p r o c ess o f s o lu tio n in w ater i s v ery slow in reach in g

eq u ilib riu m s in c e b ery lliu m flu o r id e i s e s s e n t i a l l y a polym er.

On reach ­

in g s a tu r a tio n , however, i t i s found th a t alm ost 18 m oles o f BeF2 per l i t e r
o f 'o r ig in a l water are p r e se n t, which has le a d to the p roposal th a t a
H20 X / F
stru ctu re l i k e
^,Be
i s p resen t ( 9 8 ) .
H20
Since p rep aration from aqueous s o lu tio n le a d s to in d e f in it e o x y flu o r id e
m ixtu res, an attem pt was made to determ ine whether the u se o f anhydrous
liq u id hydrogen flu o r id e might be more s u c c e s s f u l.

Sin ce no o x y g en -free

b erylliu m s a l t was a v a ila b le , b e r y lliu m carbonate was added slo w ly to a
c o n sid era b le e x c e ss o f anhydrous hydrogen f lu o r id e .

Upon removal o f e x c e ss

hydrogen f lu o r id e , however, a mixed gummy o x y flu o rid e was o b ta in e d , in d ic a t ­
in g th a t th e b ery lliu m flu o r id e formed (n o t s o lu b le in the medium) was
ab le to r e t a in the w ater evolved i n the r e a c tio n d e s p ite th e extreme
dehydrating e f f e c t o f th e anhydrous hydrogen f lu o r id e .

U6

Subsequent p rep a ra tio n s were con fin ed to the ordinary wet methods
in v o lv in g the form ation o f (NH4) 2BeF4 .

The u se o f NH4HF2 i s much more

econom ical than th e method u sin g WH4F as may be seen from the s t o ic h io ­
m etric r e la t io n s beloxj, but the p u r ity o f the a v a ila b le NH4HF2 i s con­
s id e r a b ly p o o rer.

Both methods were used depending on a v a ila b le

m a t e r ia l.
E s s e n t ia lly th e method c o n s is te d o f p r e c ip it a t in g , f i l t e r i n g and
washing th e hydrated b e r y lliu m h yd roxid e, d is s o lv in g i t in a s o lu tio n o f
e ith e r NH4F or NH4HF2 , f i l t e r i n g to remove im p u r itie s , ev a p o ra tin g ,
c o o lin g , and c r y s t a l l i z i n g out (NH4) 2BeF4 .

The l a t t e r was f i l t e r e d and

washed w ith d il u t e (50%) and then more con cen trated (80$) eth a n o l-w a ter
s o lu tio n to remove mother liq u o r , and then d ried a t 110° to g iv e the
anhydrous p ro d u ct.

The r e a c tio n s in v o lv ed are:

Be++: x H20 + 2NH40H ----- >- Be(0H)2: x H20

+ 2NH4 +

Be(OH) 3: x H^O + 2NH4HF2 ---------- (M 4) 2BeF4 + (2 + x) H20
or

Be(OH)2: x H20 + 1+M4F

(NH4) 2BeF4 + 2NH3

+ (2+x) H^O

The p rep aration has been d escrib ed by Hay (128) .
C ertain p r a c t ic a l d i f f i c u l t i e s should be m entioned.

I f the s t a r t in g

q u a n tity o f b ery lliu m i s not a c c u r a te ly known, e x ce ss o f NH4F w i l l tend
to low er the p u r ity o f the p rod u ct.

I f an attem pt i s made to govern the

amount o f NH4F by the a d d itio n o f ju s t enough to d is s o lv e the Be(QH)2 , the
attainm ent o f com p letion o f th e r e a c tio n i s slow , or too much NH4F may
be added i f sm all amounts o f in s o lu b le cloudy m a ter ia l are p r e se n t (a s
was th e ca se in th e reclaim ed b ery lliu m s o lu tio n s o f th e p r e se n t work) .
On th e o th er hand th e hydrated b ery lliu m o x id e can a ls o be p e p tiz e d by

hi

the s tr o n g ly aramoniacal s o lu t io n .

I f e x ce ss NH4F i s p r e se n t i t does not

u s u a lly s u b s t a n t ia lly a f f e c t th e p u r ity o f th e f i r s t c ro p , but very
d e f i n i t e l y does tend to be p r e se n t in subsequent crops o f c r y s t a l s .
The extreme s o l u b i l i t y o f (NH4) 2BeF4 (probably c o n sid e ra b ly g rea ter
than 100 g./lOO g; o f H20 )

makes the recovery p r o c ess a slow on e.

L ikew ise th e i n i t i a l w ashing o f the c r y s t a lli n e crop must not be c a rr ie d
ou t w ith too con cen trated an ethan ol s o lu tio n l e s t the adherent mother
liq u o r be su b jec te d to a s a lt in g out e f f e c t .

On th e oth er hand the

lo s s e s by e f f e c t i v e w ashing are u s u a lly very s e v e r e , and con seq u en tly th e
e n tir e p r o c e ss i s a lo n g drawn-out one sin c e the mother liq u o r may o n ly
be reco n cen tra ted on the steam bath because o f the ,,bumping,, d i f f i c u l t i e s
produced by f a s t e r evap oratio n .
Proper adjustm ent o f th e eth an ol co n ten t o f th e wash s o lu tio n p er­
m its a co n sid era b le red u ctio n in th e s o l u b i l i t y o f th e (MH4) 2BeF4 .

Some

workers have used eth an ol a d d itio n as a s u b s titu te fo r ev a p o ra tio n , but
a t r i a l o f t h is method in d ic a te d th a t the y ie ld had somewhat low er p u r it y .
In g e n e r a l, p rep a ra tio n s o f (NH4) 2BeF4 , made during many recovery sta g e s
in t h i s work, ranged from about 99 to 89% p u re.

The b a la n c e , c o n s is t in g

o f NB4F , had o n ly one harmful e f f e c t , i . e . i t made more d i f f i c u l t the
d eterm in ation o f th e amount o f BeF2 b ein g introduced in to a m e lt.

See

Appendix 1 9 , Part C .
As th e b e r y lliu m co n ten t o f the m e lts was salvaged and reused many
tim es during the work, a d e s c r ip tio n o f the recla m a tio n p r o c ess i s in
o rd er.

The p u lv e r iz e d m e lts were tr e a te d w ith 80% p e r c h lo r ic a cid -w a ter

m ixtures and b o ile d to d rive o f f a l l flu o r in e as HF.

(Use o f s u lfu r ic

U8
a c id was n o t s a t is f a c t o r y because o f form ation o f in s o lu b le calcium s u l­
f a t e .)

The p e r ch lo ra te s o lu tio n was th en d ilu te d and th e b ery lliu m

p r e c ip it a t e d as th e hydroxide w ith e x c e ss ammonium h y d ro x id e.

N ic k e l

p r e se n t as an in p u r ity (from th e n ic k e l s t ir r in g rods used in th e m e lts)
was sep arated as th e so lu b le ammonia complex.

Calcium and lith iu m trapped

i n th e p r e c ip it a t e were removed by a d d itio n a l r e p r e c ip it a t io n s .

O ccasional'

1 y th e recovered b ery lliu m hydroxide was given an 8-hydroxyquinoline p u r i­
f i c a t i o n treatm ent to remove ir o n and aluminum im p u r itie s which might
oth erw ise have tended to accumulate ( 7 0 ) .
I n th e l a t t e r p a r t o f th e work, when the inadequacy o f th e a n a ly t ic a l
method fo r b e r y lliu m made i t v e ry d e s ir a b le to o b ta in 100% pure (NH4) 2BeF4 ,
an attem pt was made to a r r iv e a t t h i s by preparing a pure b ery lliu m
compound, the b a s ic a c e t a te , or Be40(CH3C 00)6 .

I t was hoped th a t w ith

t h i s as a s t a r t in g m a ter ia l i t might be p o s s ib le to a v o id any e x c e s s o f
ammonium flu o r id e i n the (NH4) aBeF4 .
The p rep a ra tio n o f Be40(CH3C00) 6 in v o lv ed a d d itio n o f th e p u r ifie d
p r e c ip ita te d hydroxide to g la c ia l a c e t ic a c id and evap oration to dryness
on th e steam b a th .

F i lt r a t io n p r io r to eva p o ra tio n removed some s i l i c a

and p o s s ib ly a d d itio n a l in s o lu b le r e s id u e s r e s u lt in g from a tta c k on th e
glassw are used in th e p e r c h lo r ic a c id treatm ent where HF was e v o lv ed .
The Be40(CH3COO)6 was r e d is s o lv e d in b o ilin g g la c ia l a c e t ic a c id and
f i l t e r e d h o t w ith th e removal o f a c o n sid era b le amount o f troublesom e
in s o lu b le m a te r ia l, th e id e n t it y o f which was unknown, but which was
t e n t a t iv e ly a scrib ed to s i l i c a or to the p o s s i b i l i t y o f c r o s s -lin k e d b a s ic
a c eta tes.

The product was sep arated by f i l t r a t i o n from th e c o o l s o lu t io n

and was tw ic e r e c r y s t a l liz e d from g la c ia l a c e t ic a c id and f i n a l l y d ried
a t about l £ 0 ° .

h9

T h is Be40(CH3C00) 6 p r o d u c t was o f h ig h p u r i t y su c h t h a t th e
i n a c c u r a c i e s o f t h e b e r y ll iu m a n a l y s i s d id n o t p e r m it d i s t i n g u i s h i n g i t
from 100$ p u r i t y .

See A p p en d ix 1 5 , P a r t D .

The h o p e d - f o r im provem ent i n

t h e (NH4) 2BeF4 p u r i t y was o n ly p a r t l y a t t a i n e d , h o w e v e r, f o r v a r i o u s
r e a s o n s su ch a s th e u n c e r t a i n t y i n c o m p o s itio n o f t h e NH4HF2 u s e d , e t c . ,
b u t t h e m ethod was r e t a i n e d th r o u g h o u t t h e b a la n c e o f t h e w ork a s a
m eans o f p r e v e n t i n g r e c y c l i n g and b u ild u p o f i m p u r i t i e s .
The r e c r y s t a l l i z a t i o n o f b e r y ll iu m b a s i c a c e t a t e h a s b e e n d e s c r ib e d
by B o o th ( 8 ) .

The e a r l i e r w ork on i t s p r e p a r a t i o n and p r o p e r t i e s

f o r a to m ic w e ig h t p u r i f i c a t i o n p u rp o s e s was d e s c r ib e d b y P a r s o n s (1 2 6 )
and m ore r e c e n t l y b y H Snigschm id (7 3 ) .
F o r p r e p a r a t i o n o f (NH4) 3BeF4 , t h e b a s i c a c e t a t e was c o n v e r te d to
n i t r a t e b y b o i l i n g w ith n i t r i c a c i d a n d , c o n t r a r y to l i t e r a t u r e r e p o r t s ,
t h e t r e a tm e n t m u st b e made q u i t e d r a s t i c , a s ju d g e d from th e p ro lo n g e d
p e r io d o f a c e t i c a c i d e v o l u t i o n , i f freed o m fro m a c e t a t e i s to b e a t t a i n e d .
P r e c i p i t a t i o n o f th e h y d ro x id e fro m t h e n i t r a t e s o l u t i o n was fo llo w e d b y
c o n v e rs io n t o th e c h l o r i d e w hich g iv e s a more w a s h a b le ty p e o f h y d ro x id e
p re c ip ita te .

The l a t t e r was th e n u s e d i n th e (NH^^BeF^. p r e p a r a t i o n

p ro c e ss p re v io u s ly d e s c rib e d .

The e l e c t r o l y t i c p r e p a r a t i o n o f a

c r y s t a l l i n e ” fo rm o f b e r y l l i u m h y d ro x id e h a s b e e n d e s c r ib e d b y B o o th (8 )
a s th e b e s t way to o b t a i n a p r o d u c t f r e e o f o c c lu d e d f o r e i g n e l e c t r o l y t e s .
I n a n o th e r a t te m p t to o b t a i n a m e lt o f d e f i n i t e l y known c o m p o s itio n ,
th e g e n e r a l m ethod d e s c r ib e d by Bay (1 2 8 ) w as t r i e d i n w h ich L i-jB eF ^lH gO
w as p r e p a r e d b y th e a d d i t i o n o f L i 2C03 t o a s o l u t i o n o f (NH4) 2BeF4 .

When

50

t h i s was t r i e d i n a m o d if ie d way u s i n g h o t s o l u t i o n s and r a p i d e v a p o ra ­
t i o n , a p o o r lo o k in g p r o d u c t w ith i n d i c a t i o n s o f th e p r e s e n c e o f oxy­
f l u o r i d e s w as o b t a i n e d .

A n o th e r a t te m p t was made u s in g a c o l d s o l u t i o n

o f LijjBeF.^ p r e p a r e d b y d i s s o l v i n g th e p r o d u c t o f a d r y f u s i o n o f th e
s t o i c h i o m e t r i c q u a n t i t i e s o f L£F an d (NH4) 3BeF4 .

The r e s u l t i n g s o l u t i o n

was th e n e v a p o r a te d a t room te m p e r a tu r e ( o r lo w e r) i n a l a r g e vacuum
d e s ic c a to r.

Upon r e a c h in g th e volum e a t w h ich s a t u r a t i o n w ith L ig B e F ^ lH ^ )

s h o u ld h av e o c c u r r e d ( a c c o r d in g to R a y ) , no s o l i d s w ere p r e s e n t .

A fte r

s e v e r a l m ore w eeks o f e v a p o r a tio n t r a n s l u c e n t c r y s t a l s , w ith a n in d e x o f
r e f r a c t i o n a lm o s t i d e n t i c a l w ith t h a t o f w a te r , commenced to fo rm .

A fte r

a b o u t a n o th e r week o f e v a p o r a tio n , t o re d u c e th e volum e t o t h a t am ount
w h ich s h o u ld h av e b e e n e a s i l y c a p a b le o f h o ld in g an y s o lu b le i m p u r i t i e s ,
th e s o l i d s w ere e q u i l i b r a t e d f o r s e v e r a l w eeks w ith th e s o l u t i o n .

The

p r o d u c t was s e p a r a t e d b y f i l t r a t i o n and was w ashed s p a r i n g l y w ith c o ld
w a te r and th e n w ith a v e r y l i t t l e

e t h a n o l f o llo w e d b y e t h e r .

A n aly sis o f th e r e s u l t i n g p roduct showed n e a rly th e th e o r e tic a l
co m p o sitio n .

See Appendix 1 5 , P a r t E.

A c tu a lly th e e r r o r s o f a n a ly s is

were such t h a t th e p u r ity p ro b ab ly was 100$ w ith th e p o s s ib le e x c e p tio n
t h a t th e e x te n t o f h y d ra tio n m ight n o t have been e x a c tly according to
th e o ry .

T h is p ro d u c t, when h e a te d to c o n s ta n t w eight to d riv e o f f th e

w a te r, gave a p e r f e c t l y c le a r m elt w ith which th e m e ltin g p o in t o f L i 2BeF4
could be determ ined w ith more c e r t a i n t y .
The a t t r a c t i v e n e s s o f t h e Ray L i 2C03 m ethod i s somewhat d e c r e a s e d b y th e
know ledge o f th e r e l a t i v e s o l u b i l i t i e s in v o lv e d .

I n o r d e r to p r e p a r e one

51

m ole o f LigB eT^ ( 9 8 .9 g . ) , i t w ould b e n e c e s s a r y to s t a r t w ith a b o u t 5
l i t e r s o f w a te r i n o r d e r t o d i s s o l v e th e LigC03 .

The s o l u b i l i t y o f a b o u t

1 3 .2 g. p e r l i t e r c la im e d b y Hay (1 2 8 ) f o r LigB eF4 w ould make n e c e s s a r y th e
vacuum e v a p o r a tio n o f a p p r o x im a te ly k . 9 l i t e r s o f w a te r i n o r d e r to o b t a i n
a b o u t 90 % y i e l d .

A c t u a l l y t h e L igB eF 4 i s c o n s id e r a b ly more s o l u b l e , a

ro u g h f i g u r e o f 23 g . p e r l i t e r b e in g o b ta in e d i n t h i s w ork fro m a n a l y s i s
o f th e u n d i l u t e d m o th e r l i q u o r .

(S e e A p p en d ix 1 5 , P a r t E) .

E q u ip m en t

The p r i n c i p a l eq u ip m en t u s e d i n th e w ork c o n s i s t e d o f a s p e c i a l l y
c o n s t r u c t e d f u r n a c e and a p o t e n t i o m e t e r - r e c o r d e r f o r a u to m a tic p l o t t i n g
o f t h e tim e - te m p e r a tu r e c u r v e s .
S in c e some o f th e w ork in v o lv e d te m p e r a tu r e s o f lliOO0 an d h i g h e r ,
th e u s u a l t y p e s o f w ir e wound f u r n a c e s w ere n o t s u i t a b l e .

F o r t h i s s tu d y

a f u r n a c e was c o n s t r u c t e d u s in g C arborundum " G lo b a r” r e s i s t a n c e h e a t in g
u n i t s w h ich a r e l i m i t e d t o a recom mended maximum te m p e r a tu r e o f ll i 50 ° .
E ig h t' o f t h e s e h e a t i n g u n i t s w ere d i s t r i b u t e d s y m m e tr ic a lly i n a c i r c l e ,
s u r ro u n d in g and p a r a l l e l to a v e r t i c a l 3" 1 .D . '’Alundum” t u b e .

S u rro u n d ­

i n g th e c i r c l e o f h e a t i n g e le m e n ts w as a s h e l l o f c a s t r e f r a c t o r y c l o s e d
a t to p and b o tto m e n d s b y h e a v y c a s t r e f r a c t o r y p i e c e s th r o u g h w h ich th e
c o o l e r end s e c t i o n s o f th e h e a t i n g e le m e n ts p a s s e d .

E l e c t r i c a l c o n ta c ts

w ere made t o th e e le m e n ts o n t h e o u t e r f a c e s o f th e r e f r a c t o r y end p l a t e s .
The a c t i v e h e a t i n g r e g i o n o f t h e e le m e n ts , a b o u t 6W i n l e n g t h , was t h u s
p e r m i t t e d t o d i r e c t l y r a d i a t e t o th e c e n t r a l tu b e w here t h e sam ple
c o n t a i n e r was l o c a t e d .

H e a t l o s s e s from t h e e le m e n ts w ere h e ld t o

52

reason ab le v a lu e s by about iu11 th ic k n e ss e s o f r e fr a c to r y on the ends and
l a t e r a l l y by about 5" o f r e f r a c to r y , p lu s about 2" o f e x f o lia t e d verm icul i t e p la ced between the r e fr a c to r y and the ou ter s h e l l .

The ends o f the

"Alundum" tube were c lo se d by about If1 o f r e fr a c to r y ex cep t fo r a sm all
h o le to perm it o b se r v a tio n , s t ir r in g o p e r a tio n s , e t c .
The r e fr a c to r y used was d e lib e r a t e ly chosen to o b ta in minimum
p o s s ib le heat storage c a p a c ity com patible w ith good in s u la tio n q u a l i t i e s .
Johns-M anville "Light "Weight F ir e c r e te " was used even though the
d e te r io r a tio n o f the r e fr a c to r y was severe during the sh o r t p eriod o f use
at tem peratures h igh er than i t s recommended peak se r v ic e temperature o f
1300° .

Power supply was c o n tr o lle d by a U5 ampere c a p a c ity "Powerstat"
autotransform er w ith v o ltm eter and ammeter in d ic a tio n o f the secondary
power demands.

C onnection to the elem ents was made thru a bus-board

arrangement p e r m ittin g a lt e r a t io n o f th e s e r ie s - p a r a lle l c i r c u i t s as the
elem ents in c r e a se d in r e s is ta n c e xtfith a g ein g .
A Leeds and horthrup "Micromax" Potentiom eter-R ecorder was used in
con ju n ction w ith a p latinu m -platinu m , 10$ rhodium therm ocouple.
speed o f 6" per hour was u se d .

A ch art

The fix e d s c a le o f the record er was

ad ju sted during c a lib r a t io n , and a l l subsequent c r i t i c a l temperature
measurements were made from the f ix e d s c a le r a th e r than the chart r ec o r d ,
as the l a t t e r i s su b je c t to e x p a n sio n -co n tr a ctio n and tr a c k in g e r r o r s .
By u sin g th e g la s s fa c e o f th e instrum ent as a mirror and lin in g up th e
p o in te r w ith the p u p il o f the eye i t was p o s s ib le to e stim a te tempera­
tu re s to l e s s than 1° over th e 1600° ran ge.

53

The same thermocouple was used throughout the work, although
g ra d u a lly shortened by numerous breakages and rew elding o p e r a tio n s .
Leeds and Northrup com pensating le a d s were used between the thermocouple
and th e recorder which was equipped w ith a c o ld -ju n c tio n compensator fo r
ambient tem perature c o r r e c tio n .

In te r n a l accuracy o f the instrum ent was

e x c e lle n t and o th er u n c e r ta in tie s are b e lie v e d to have been la r g e r than
any in stru m en ta l e r r o r .

A u to -sta n d a rd iza tio n occurred a t regu lar 55

minute in t e r v a ls a g a in s t the in t e r n a l standard c e l l and fo r c r i t i c a l
measurements the instrum ent was hand-standardized to remove s l i g h t erro rs
due to d r i f t o f th e op eratin g dry c e l l v o lt a g e .
The thermocouple was c a lib r a te d on numerous o c ca sio n s a t a v a r ie ty
o f fix e d p o in ts over th e range o f 100° to about 1100° , and once the
fix e d s c a le had been ad ju sted no fu r th e r change was found n e c e ssa r y .
The c a lib r a tio n p o in ts used were the fo llo w in g ( 9 l ) :
B o ilin g w ater 100°
F reezin g t i n 231.85°
F reezin g le a d 327.5°
M elting potassium dichromate 397.5°
F reezin g z in c 519.55°
F reezin g antimony 630.5°
M eltin g potassium ch lo r id e 770 .3°,r
M eltin g sodium ch lo r id e 8 0 0 .5 °*
M eltin g potassium s u lfa t e 10 69.1°
The u n c e r ta in t ie s in the d eterm in ation o f the fix e d p o in ts were v a r i­
a b le , u s u a lly from 1 to 2° , but i n th e ca ses in d ic a te d w ith an a s t e r is k
agreement was b e t t e r , probably to w ith in l e s s than 1 ° .

In g en era l th e r e ­

fo r e a i l ° erro r in a b so lu te accuracy was p o s s ib le , but the r e la t iv e
err o r s were probably i n th e order o f 1 3 /5 ° or l e s s .

With r e s p e c t to the

h ig h e s t ranges above 900° no d e f in it e statem ent can be made as to the

5h

accuracy although i t i s b e lie v e d th a t ±2° would be a reason ab le e stim a te
o f th e probable e r r o r .

An attem pt was made to o b ta in a c a lib r a tio n w ith

n ic k e l a t lii5 2 ° , bu t the attem pt was u n su c c e ssfu l and was not rep eated
because o f th e extreme overload p la ced upon the furnace and power supply
a t t h i s tem p erature.

The check obtained on th e calcium flu o r id e m e ltin g

p o in t w ith th e valu e given by N aylor (116) in d ic a te s an error o f no
grea ter than 5° a t th e llf.00° l e v e l .

T his i s q u ite understandable in

view o f th e la rg e r a d ia tio n e rro rs which may occur a t t h i s tem p erature.
C a lib r a tio n s in which m etals were used n e c e s s ita te d th e use o f a
thermocouple w e ll ( g la s s or g rap h ite), but in a l l o th er c a s e s , and during
a l l o f th e phase diagram work, th e couple was immersed d ir e c t ly i n the
m e lt.

Breakage was more severe a t high tem p eratures, but below 800° a

s in g le weld would fr e q u e n tly l a s t th e eq u iv a len t o f 300 to 500 hours o f
continuous o p e r a tio n .

The use o f bare therm ocouples was not o n ly a con­

v e n ie n c e, but was a n e c e s s it y due to th e r e la t iv e sh a llo w n ess o f the m e lts
u se d .

Use o f p r o te c tiv e w e lls in th e phase diagram work would have

created la r g e o p p o r tu n itie s fo r conduction e r r o r s .

No e f f e c t o f d ir e c t

immersion upon the w elds was ob served , breakage norm ally o ccu rrin g a t a
p o in t where c o r r o sio n tended to be most s e v e r e , i . e . ju s t a t the su rfa ce
o f the m e lt .
In use th e couple was immersed to a p o in t s l i g h t l y above the bottom
o f th e c o n ta in e r , in a manner comparable to th a t used in c a lib r a t io n .
Almost a l l m e lts were h a n d -stir r e d w ith a n ic k e l ro d , e ith e r c o n tin u o u sly
or a t s u f f i c i e n t l y sh o rt in t e r v a ls so th a t temperature d i f f e r e n t i a l s
w ith in the m e lts were minimized a t a l l p o in ts o f c r i t i c a l i n t e r e s t .

As

55

■unstirred m e lts i n some o f the system s gave e r r a tic r e s a l t s , s e v e r a l
m echanical a g it a t io n system s were t r i e d , in c lu d in g p r o p e llo r s and sp r in g load ed p lu n g e rs.

None were very s u c c e s s fu l and breakage o f the bare

therm ocouple was freq u en t in th e fr e e z in g m e lt.
As th e b e n e f it s o f o b se rv a tio n and " feel" during s t ir r in g were con­
s id e r a b le , hand s t ir r in g was used in most o f the work.

N ick el w elding

ro d s, which are r e l a t i v e l y poor conductors o f h ea t and f a i r l y s ta b le to
a tta ck by the m e lts , were u sed .
As sample c o n ta in e r s , carbon c r u c ib le s (about 3" O.D. and 2" I .D .
by 2 1/2" deep) were machined from carbon e le c t r o d e s .

These were used

fo r most o f th e work on the CaF2-LiF system , but on subsequent work
sh allow platinum d ish e s (about J4O m l. c a p a c ity ) were u se d , p r im a r ily
because th ey perm itted a more s a t is f a c t o r y check o f the changes in th e
w eight o f th e m e lt.
For e l e c t r o l y t i c work a d ir e c t current supply was c o n str u c te d .
A selenium r e c t i f i e r , arranged fo r f u l l wave r e c t i f i c a t i o n was fe d by
the secondary o f a Pow erstat a u to -tra n sfo rm er.

The D.C. output was

f i l t e r e d by a double LC type o f f i l t e r , u sin g high c a p a c ity e l e c t r o l y t i c
condensers and s p e c ia lly co n stru cted low r e s is ta n c e (ir o n co re) choke
c o ils .

With t h i s arrangement low r ip p le cu rren ts o f 30 amperes a t low

v o lta g e (10 V) cou ld be p rovid ed , as w e ll as sm all cu rren ts (1 to 2 A)
w ith n e g lig ib le r ip p le a t high er v o lta g e s such as 50 V.
E le c t r o ly t ic apparatus i s d escrib ed in th e s e c tio n co v erin g th a t
phase o f th e experim ental work.

56

General Procedure
C erta in p e r tin e n t comments and the gen eral procedure fo llo w e d in
th e work on the phase diagrams are given below in some d e t a il fo r the
d if f e r e n t system s stu d ie d .
M ixtures fo r the p rep ara tio n o f the m e lts were u s u a lly made w ith
oven -d ried m a te r ia ls weighed to w ith in the n e a r e s t 0.0005’ g .

The t o t a l

w eigh ts used v a r ied c o n sid e r a b ly , th e p r in c ip a l determ ining fa c to r b ein g
the requirem ents fo r b ery lliu m f lu o r id e .

Since th e l a t t e r was d i f f i c u l t

to p rep are, p a r t ic u la r ly in la r g e q u a n t it ie s , m e lts c o n ta in in g la r g er
p rop ortion s o f t h i s m a ter ia l were u s u a lly made up to t o t a l w eig h ts o f
about 30 g«

Other m e lts req u irin g l e s s b ery lliu m flu o r id e were made

on a somewhat la r g e r s c a le , but the s iz e o f the a v a ila b le platinum d is h e s
r e s t r ic t e d the maximum amounts u se d .
The b e r y lliu m flu o r id e used was obtained in th ree gen eral w ays.
For the CaF2-BeF2 system , th e i n i t i a l s e t o f runs was made u sin g b ery lliu m
flu o r id e prepared by decom position o f (NH4) 2BeF4 i n la r g e carbon c r u c ib le s
in th e absence o f oth er flu o r id e s

A few runs made on t h is system a t a

much la t e r date u t i l i z e d b e r y lliu m flu o r id e prepared in the presence o f
calcium f lu o r id e .

For work on about tw o -tiiir d s o f th e LiF-BeF2 system

and th e tern ary system , b ery lliu m flu o r id e obtained from th e Brush
B erylliu m C orporation was u s e d .

T his m a te r ia l had been p la c e d i n a la r g e

number o f sm all ja r s u sin g dry box te c h n iq u e , and the ja r s were kept in
a d e s ic c a to r over magnesium p e r c h lo r a te .

During w eig h in g , th e m a ter ia l

was exposed to th e atmosphere as l i t t l e as p o s s ib le and the m ixture was

57

im m ediately p la ced in the fu rn ace to avoid m oisture p ick u p .

For the

b alan ce o f the work, the b ery lliu m flu o r id e was prepared by the decompo­
s i t i o n o f weighed q u a n tit ie s o f (NH4) 2BeF4 in the p resence o f e it h e r or
both o f the o th er f lu o r id e s .

T his o p e r a tio n was c a r r ie d out under an

in v e r te d fu n n el su p p lied w ith a rapid flo w o f dry carbon d io x id e .

A fter

th e e v o lu tio n p r o c ess was com plete the platinum d ish and lo a d were
weighed to perm it c a lc u la tio n o f th e a c tu a l w eight o f b ery lliu m flu o r id e
p r e s e n t.

C a lc in a tio n o f th e calcium and lith iu m f lu o r id e s showed s lig h t

l o s s e s over the oven -d ried w eigh ts and sm all c o r r e c tio n s were made to
th e weighed q u a n tit ie s to compensate fo r such l o s s e s where u n ca lcin ed
m a te r ia ls were u se d .
A ll m e lts were th en p o s itio n e d in the same r e la t iv e v e r t i c a l p o s it io n
in the fu rn a ce , u sin g as a support a number o f in v e r te d p o r c e la in Gooch
c r u c ib le s in an attem pt to m inimize th e heat sto ra g e o f m a te r ia l in
immediate c o n ta c t w ith th e co n ta in er and m e lt.

The top opening o f the

c e n tr a l "Alundum" tube o f the furnace was c lo se d w ith a s p l i t r e fr a c to r y
p lu g le a v in g on ly a sm all a ccess h o le fo r the suspended thermocouple and
s t ir r in g r o d .

A slow stream o f th orou ghly d ried carbon d io x id e was

passed in through a copper i n l e t tube p r o je c tin g down in to t h i s opening.
M elts were g e n e r a lly heated 100 to 150° above the a n tic ip a te d reg io n
o f in t e r e s t and thorougiily m ixed.

The c o o lin g r a te was c o n tr o lle d by

the app rop riate s e t t in g o f th e power in p u t which was v a ried in accordance
w ith th e p a r tic u la r range o f o p era tio n so th a t c o o lin g r a te s were roughly
com parable.

The approximate tim e in v o lv ed per c y c le averaged about 1 .5

h ou rs, fig u r e d from peak tem perature to peak tem perature on su c c e s siv e

58

m e lt s .

U su a lly about 20-25 m inutes o f c o o lin g occurred p r io r to an

in v a r ia n t p o in t.

The h old in g tim e a t such p o in ts was about 15-20 m inutes

depending on e u t e c t ic co n ten t o f the m e lt.

Subsequent c o o lin g and r e h e a t­

in g u s u a lly req u ired about J4O m in u te s.
During c o o lin g the m elt was s t ir r e d o c c a s io n a lly and the temperature
read ing from the f ix e d s c a le o f the recorder was noted on the c h a r t.

In

the r e g io n o f p o in ts o f in t e r e s t s t ir r in g was u s u a lly made as contin uous
as p o s s ib le and, p a r t ic u la r ly during the in v a r ia n t p r o c e s s , s t ir r in g was
c a r r ie d on as lo n g as th e m elt was liq u id enough to perm it m ixing o f the
p la s t ic m ass.

A c tu a lly , however, judging from the thermal h old in g tim e s ,

about 25- 30$ o f th e p rocess remained to be accom plished when the m elt
was too s o lid fo r fu r th e r s t i r r i n g , due to a la r g e amount o f non-apparent
liq u id p r e se n t in th e s o lid m ass.
The therm al e f f e c t s observed, p a r t ic u la r ly w ith r e sp e c t to primary
c r y s t a lli z a t i o n o f a s in g le component were v a r ia b le .

In r e g io n s o f

shallow s lo p e o f the temperature v s . com position cu rv e, the e f f e c t s o f
primary and secondary c r y s t a l li z a t i o n s showed up unm istakably.

In o th er

r eg io n s o f much ste ep er s lo p e , where very l i t t l e primary s o lid s might be
d ep o sited during a drop o f 100° or more, the e f f e c t s v a ried from reason­
ab ly d e te c ta b le to no ob servab le therm al e f f e c t , in the sen se o f a
r e l a t i v e l y good in f l e c t i o n p o in t on th e c o o lin g c u rv e.

The f a c t th a t the

furnace was b u i l t and in s u la te d fo r work a t high tem peratures in p art
aggravated t h i s c o n d itio n a t the low er tem peratures.
Because o f th e u n ce r ta in ty in a few o f the r e s u lt s o f the CaF2-BeF2
system , o b se rv a tio n s were made on the c o o lin g m elt and i t was found th a t

59

a v i s i b l e e f f e c t occurred on the su rfa ce j u s t p r io r to obvious p o in ts o f
c r y s ta lliz a tio n .

T his was a ttr ib u te d a t th e tim e to the removal o f sm all

su rfa ce d u st p a r t ic le s during the form ation o f the f i r s t su rfa ce c r y s t a l s .
On subsequent work in th e b in a ry LiF-BeF2 and tern ary system s, when thermal
e f f e c t s became u n c e r ta in , o b se rv a tio n u sin g a pow erful s p o tlig h t was used
and i t was found p o s s ib le in most c a se s to d e te c t the i n i t i a l c r y s t a l l i z a ­
t io n from su rfa ce e f f e c t s ju s t p r io r to observable thermal e f f e c t s .

In

the c a se o f calcium or lith iu m flu o r id e s appearing as the f i r s t c r y s t a ls ,
the e f f e c t was q u ite sharp, as i f suddenly a c o a tin g o f tin y sh in in g o i l
d r o p le ts developed on the s u r fa c e .

W hile q u ite s im ila r in t h e ir i n i t i a l

appearance, th e behaviour o f the "droplets" d iffe r e d s u f f i c i e n t l y during
the next minute or two o f c r y s t a l li z a t i o n so th a t the two could be
d if f e r e n t ia t e d .

In the ca se o f L i2BeF4 c r y s t a l l i z a t i o n , the su rfa ce o f

the m elt suddenly developed r a p id ly growing needle c r y s t a ls which would
cover the su rfa ce in a very few seco n d s.

Both the above e f f e c t s preceded

the thermal e f f e c t s (where observed) by about 1 to 2° depending on the
c o o lin g r a t e , th e su rface o f the m elt tend ing to be s l i g h t l y c o o le r due
to r a d ia tio n , d e s p ite c o n sta n t s t i r r i n g .
The primary c r y s t a l l i z a t i o n o f CaBeF4 was not as a r u le observab le
by a su rface e f f e c t .

A lthough the m a te r ia l i s n eed le shaped and might

be expected to i n i t i a l l y appear on th e surface i f th e oth er c r y s t a lli n e
forms c o u ld , i t c h a r a c t e r is t ic a lly seemed to form below the su rfa ce and
sometimes would be attach ed somewhat to the platinum d is h .

T h is would

r e s u lt in a g r it t y f e e l i n g as the hand m anipulated s t ir r e r passed a c ro ss
th e bottom o f th e platinum d is h .

In c e r ta in c a se s where the i n i t i a l

60

c r y s t a l l i z a t i o n o f the CaBeF4 was accompanied by no apparent therm al
e f f e c t s t i l l another teclm ique was t r ie d .

I t was found somewhat

a c c id e n ta lly th a t i f a c o ld 1 /8 H n ic k e l rod was plunged to the bottom o f
the m elt and im m ediately withdrawn, sm all n e e d le s were observab le s t ic k in g
ou t from th e th in s h e l l o f fr o z e n m a ter ia l on the end o f the rod.

T his

method o f d e te c tio n o f CaBeF4 d e p o s itio n had to be used in a sm all r e g io n
o f the m elt where no oth er methods worked.

The r e p r o d u c ib ility o f t h i s

method was checked and in some c a se s was found to be good to w ith in
about S° , but in oth er ca ses i t was poorer, p o s s ib ly because o f o th er
in t e r f e r e n c e s .

These r e s u lt s a cco rd in g ly were tr e a te d as d o u b tfu l.

No

d i f f i c u l t y was occasion ed by the f a c t th a t both calcium and lith iu m
flu o b e r y lla t e s have n e e d le -lik e sh a p es, sin ce th a t o f lith iu m i s not
s ta b le above approxim ately U60o and c h a r a c t e r is t ic a lly i t always appears
f i r s t as su r fa c e c r y s t a ls .
For the bin ary system s, CaF2-LiF and CaF2-BeF2 , m ixtures were p re­
pared w ith com position s spaced c l o s e l y enough to adequately d e fin e the
liq u id u s curves o f th e system .

The in g r e d ie n ts o f the m e lts were weighed

p erm ittin g c a lc u la tio n o f the com p osition , and g e n e r a lly the r e s u lt in g
m elt was afterw ards sampled and analyzed fo r purposes o f check ing the
c a lc u la te d com p osition .
The b in ary LiF-BeF2 and ternary system s were la r g e l y in v e s tig a te d
jo in tly .

The d i f f i c u l t y in th e p rep aration o f b ery lliu m flu o r id e and i t s

poor handling p r o p e r tie s made i t d e s ir a b le to economize on e f f o r t and
avoid exposure o f th e m a ter ia l to the atmosphere by many w eighing opera­
tio n s .

C onsequently the gen era l procedure used was to prepare a m elt on

61

th e LiF-BeF2 s id e o f the tern a ry diagram and determ ine i t s thermal
b eh aviou r.

Then a weighed amount o f calcium flu o r id e was introd uced

in to t h i s m e lt, s h if t in g th e com position along a s tr a ig h t l in e toward
the o p p o site calcium flu o r id e v e r te x o f the ternary diagram.

R e p e titio n

o f t h is p ro cess p erm itted 11sectio n in g " o f the tern ary system along t h i s
lin e .

D e v ia tio n s may occur a f t e r se v e r a l such ste p s fo r a number o f

reason s to be d is c u s s e d .

C onsequently the ternary m elt was sampled a t

in t e r v a ls and the m elt reused u n le s s the temperature l e v e l or in d ic a ­
t io n s o f h y d r o ly s is made fu rth e r use u n d e s ir a b le .
one la r g e advantage in a d d itio n to economy.

Such a method had

T his advantage was th a t

r e la t iv e tem perature v a lu es were obtained w ith more c e r ta in ty than i f
two sep arate m e lts were prepared, sin c e th ere was very l i t t l e u n c e r ta in ty
in the r e l a t i v e com position s o f two m elts d if f e r in g o n ly by a sm all
a d d itio n o f th e calcium f lu o r i d e .

In general the method worked q u ite

s a t i s f a c t o r i l y , but in one r eg io n the r e s u lta n t u n c e r ta in tie s undoubtedly
caused as much work as the method saved in oth er ways.
D e v ia tio n s o f the m elt com position from th a t c a lc u la te d from the
in g r e d ie n ts occurred fo r a number o f rea so n s.

The p r in c ip a l ones were as

fo llo w s :
1) Small v o l a t i l i t y lo s s e s from the i n i t i a l in g r e d ie n ts on f i r s t
h e a tin g .

These were compensated fo r by c o r r e c tio n s based on

ex p erien ce o f previou s l o s s e s .
2) General v o l a t i l i t y lo s s e s can be q u ite sev ere from such m elts
p a r t ic u la r ly a t tem peratures above 850 d e g r e e s.

L o sses due to

t h i s item might be as la r g e as l / k to 1/2% by w eight in the

62

course o f s e v e r a l therm al c y c le s .

The p o s s ib le fr a c tio n a t io n

e f f e c t produced i s unknown.
3) Creepage l o s s e s rep resen t a d i f f i c u l t problem in th e CaF2-LiF
system and in trod u ce some u n c e r ta in tie s in th e oth er system s.
G enerally creepage would occur up to the rim and o c c a s io n a lly
over th e edge and p a r t ia l l y down the s id e o f the d is h .

From

exp erien ce gained, th e m elt was removed and the m a ter ia l
broken o f f the s id e s and e x te r io r b efore i t had a chance to
com p letely e sca p e.
To compensate fo r lo s s e s on item s 2) and 3) the m elt was withdrawn
and weighed w ith the con tain er a t freq u en t in t e r v a ls , sometimes a ft e r
each run in the la t e r sta g e s o f the work.

Allowance was then made in the

c a lc u la tio n o f th e com position fo r a l l such lo s s e s on the assumption th a t
the com position o f the lo s s was r e p r e s e n ta tiv e o f the m e lt.

Allowance

was made a lso fo r the w eight and ch aracter o f any samples removed.

By

means o f such compensated c a lc u la tio n s agreement was u s u a lly obtained w ith
the r e s u lt s o f a n a ly s is to w ith in 1 mole p e r c e n t.

Sin ce the o v e r - a ll

a n a ly t ic a l accuracy was probably no b e tte r than t h i s , th e c a lc u la te d v a lu e s
o f com p osition were u s u a lly con sid ered more tru stw o rth y , u n le s s s p e c ia l
reasons e x is t e d fo r b e lie v in g o th e r w ise .
w h ile th e above u n c e r ta in t ie s were sm a lle r , or not o f g rea t s i g n i f i ­
can ce, in o th er p o r tio n s o f the tern ary system , th ey p resen ted ser io u s
problems in th e v i c i n i t y o f the L i2BeF4 p o r tio n o f th e system where much
l e s s u n c e r ta in ty was d e s ir a b le .

Here se v e r a l v a r ia tio n s were used in an

attem pt to decrease the r e la t iv e com p osition u n c e r t a in t ie s .

63

In a d d itio n to s e c tio n s d ir e c te d toward the calcium flu o r id e corner
o f th e t r ia n g l e , s e v e r a l oth er ty p es o f s e c tio n s were employed.

M aster

b atch es o f m e lts were prepared a t about the 20 mole p ercen t .CaF3 p o in t
o f the CaF2-LiF bin ary system and a t about the 50 mole p ercen t CaF2 p o in t
o f the CaF2-BeF2 b in ary system .

A d d ition o f p o r tio n s o f one or the o th er

o f th ese b a tch es to m e lts in the LiF-BeF2 system caused com position
s h i f t s d ir e c te d along s e c tio n s term in a tin g a t th e com position s o f th e se
b a tc h e s .

In a d d itio n m aster b atch es o f m elt were prepared, w ith the

approximate com position o f L i2BeF4 , which were then s h if t e d in e ith e r
d ir e c t io n along the b in ary LiF-BeF2 sid e o f the m elt by th e a d d itio n o f
one or th e oth er o f th e pure components.

A g rea t many c r is s c r o s s in g

s e c tio n s were thus ob tain ed along which were d is tr ib u te d a g rea t many
p o in t s , sometimes as c lo s e to g e th er as 0 .1 mole p e r c e n t.

W hile t h is

method fu rn ish ed d e t a ile d in form ation about the contours alon g a given
s e c t io n , d isco rd a n t r e la t iv e p o in ts on d if f e r e n t s e c tio n s were p o s s ib le
because o f th e r e l a t i v e l y se r io u s e f f e c t in t h i s reg io n o f sm all erro rs
in com p osition and tem perature.
For m e lts w ith o r ig in a l com p osition s fa r from th a t o f an in v a r ia n t
p o in t , th e therm al a n a ly s is was o n ly c a r r ie d as fa r as the primary or
secondary c r y s t a l l i z a t i o n break, sin c e in v a r ia n t r e a c tio n s in the presence
o f to o much p r e v io u s ly d e p o site d s o lid s tend to y i e l d low tem perature
v a lu e s .

The d e te c tio n o f secondary c r y s t a l l i z a t i o n p o in ts fu r n ish e s

a d d itio n a l in form ation provided the i n i t i a l l y c r y s t a lli z e d s p e c ie s i s
known, s in c e the p o in t o f secondary c r y s t a l l i z a t i o n l i e s on a s tr a ig h t
l i n e d ir e c te d away from th a t s p e c ie s .

The in t e r s e c t io n s o f such l i n e s ,

6k

fo r m e lts o f d if f e r e n t o r ig in a l com positions but the same secondary
c r y s t a l l i z a t i o n tem p eratures, serv e to lo c a te p o in ts along l i n e s o f two­
f o ld s a tu r a tio n fo r th e sim pler tern a ry c a s e s .
For m e lts w ith com position s c lo s e to th a t o f an in v a r ia n t p o in t the
therm al a n a ly s is was fo llo w ed through th e primary and secondary c r y s t a l­
l i z a t i o n s and through the in v a r ia n t p o in t.

In gen eral the p ro cess was

th en fo llo w e d below t h is p o in t to an e x te n t s u f f i c i e n t to in d ic a te the
r e l a t i v e com pleteness o f the s o l i d i f i c a t i o n a t the in v a r ia n t p o in t.
Because such a v a r ie t y o f w eig h ts and c o o lin g c o n d itio n s were u se d , the
r e s u lt in g data were not s u s c e p tib le to the s im p lifie d procedure o f tabu­
la t in g " r ea c tio n tim es" and "holding times" on a "per gram" b a s is .
In p la c e o f t h is method, a d ir e c t comparison was maae o n ly between
c lo s e ly r e la t e d m e lts co o led under sim ila r circu m sta n ces.

For example

su c c e s siv e m e lts o f a given s e c tio n d iffe r e d in w eight o n ly by the sm all
amount o f the a d d itiv e component and were co o led under s im ila r c o n d itio n s
and hence t h e ir therm al behaviours were alm ost d ir e c t ly comparable on
the b a s is o f th e ch art r e c o r d s.

Other m elts o f d if f e r e n t com positions

(b ut the same w eight) cooled from much d if f e r e n t temperature l e v e l s were
not d ir e c t ly com parable.
Samples o f th e s o l i d i f i e d m elts were examined from two sta n d p o in ts,
th a t i s w ith r e s p e c t to gross com position and the d eterm in a tio n o f the
phases p r e se n t.

The samples were analyzed g e n e r a lly in groups o f 10 to

20 fo r c a tio n c o n te n t, no attem pt b ein g maae to determ ine flu o r in e
c o n te n t.

B r i e f l y th e com plete a n a ly ses covered removal o f flu o r in e as

hydrogen f lu o r id e by p e r c h lo r ic a c id trea tm en t, fo llo w e d by p r e c ip it a t io n

65

o f B e(0H )2 which was ig n ite d and weighed as the o x id e , BeO.

The f i l t r a t e

was tr e a te d to p r e c ip it a t e calcium as CaC30 4 which was sep arated and
determ ined v o lu m e tr ic a lly w ith standard KMrD4 .

The f i l t r a t e o f the c a l ­

cium d eterm in ation was tr e a te d to s u b s tit u te c h lo r id e fo r o th er anions
and remove ammonium s a l t s .

A volum etric determ ination o f lith iu m was

th en performed on t h i s resid u e based on the p r e c ip it a t io n o f a complex
lith iu m p e r io d a te .

Iodine lib e r a te d by th e r e a c tio n o f the p e r io d a te

w ith potassium io d id e was determ inea w ith standard sodium t h i o s u l f a t e .
More com plete d e t a i l s are giv en in Appendices lU and 15 concerning the
methods ana r e s u lt s o f th e se a n a ly s e s .
O p tic a l and X-ray methods were used fo r the d eterm in ation o f phases
p r esen t in the sam ples.

A p o la r iz in g m icroscope and s u ita b le immersion

media p erm itted g en eral i d e n t if ic a t io n o f the c o n s titu e n ts where the
h a b its ( i . e . c u b ic , n eed le-sh a p ed , or g la s sy ) alone were in s u f f i c i e n t fo r
t h i s c h a r a c te r iz a tio n .

For most c a se s s p e c ia l immersion liq u id s were

prepared u sin g g ly c e r in e -w a te r m ixtures as most o f th e phases o f in t e r e s t
have u n u su a lly low in d ic e s o f r e f r a c t io n .
For X-ray a n a ly s is , powdered samples were mounted i n Lindemann g la s s
c a p i l l a r i e s and exposed to Copper

r a d ia tio n , u sin g a North American

P h i l l i p s X-ray D if f r a c t io n apparatus w ith 11.1+59 cm. diam eter cameras.
Measurement o f th e p a tte r n or d ir e c t comparison was s u f f i c i e n t to determ ine
th e presence or absence o f phases p resen t in the d if f e r e n t r eg io n s o f
t h i s tern ary system .

The s e n s i t i v i t y o f t h i s method appeared to be high er

than normal as i t i s b e lie v e d as l i t t l e as 2% o f a p a r tic u la r phase was
d e te c ta b le , w ith th e e x ce p tio n o f b ery lliu m f lu o r i d e .

The l a t t e r ,

66

appearing o n ly as a c e n tr a l halo on the p a tte r n ( t y p ic a l o f the behaviour
o f a semi-amorphous s u b s ta n c e ), probably was n o t d e te c ta b le in amounts
l e s s than k%.

The g e n e r a lly high s e n s i t i v i t y i s b e lie v e d to be due to

the low ab sorp tion power o f th e c o n s t it u e n t s , s in c e g e n e r a lly th e atom ic
numbers o f th ese elem en ts were low er than th o se p resen t in u su a l m ix­
tu r e s .

A ls o , as a consequence o f t h i s f a c t , i t was not n ecessa ry to

d ilu t e th e samples in the customary way.

THE SYSTEM CALCIUM FLUOEIDE-LITHIUM FLUORIDE

67

THE SYSTEM CALCIUM FLUORIDE-LITHIUM FLUORIDE
The CaF2-LiF system was th e f i r s t to he s tu d ie d .

The r e la t iv e

s im p lic it y o f th e system and th e g r ea ter a v a i l a b i l i t y o f th e components
p erm itted n e a r ly complete c h a r a c te r iz a tio n o f the system on the f i r s t
a ttem p t.

There rem ained, however, a s l i g h t u n c e r ta in ty as to the ex a ct

e u t e c t ic com p osition and th e ex a ct lo c a t io n o f the s te e p ly slo p in g
liq u id u s li n e im m ediately ad jacen t to the e u te c tic on th e calcium flu o r id e
r ic h s id e .

C onsiderably l a t e r , rep eat d eterm in ation s were made on th ese

p o in ts u sin g v is u a l o b se rv a tio n as w e ll as thermal a n a ly s is , and th e un­
c e r t a i n t i e s were g r e a tly reduced.
In co n n ectio n w ith t h is study the m eltin g p o in t o f calcium flu o r id e
was red eterm in ed .

A value o f approxim ately llt l3 ° was o b ta in e d , but b e­

cause o f th e la c k o f c a lib r a tio n in t h is range and the p r o b a b ility o f
la r g e r e r r o r s , the l i i l o 0 value ob tain ed by h aylor ( 1 1 6 ) , has been used
in th e fig u r e d e sc r ib in g the system .

L ikew ise the fig u r e o f 8iib° fo r

th e fr e e z in g p o in t o f lith iu m flu o r id e given by oth er workers (I ii3 )( l6 6 )
has been used in s te a d o f the v a lu e o f 850° ob tain ed in t h i s work.
The p r in c ip a l d if fe r e n c e s in procedure from th a t g e n e r a lly d escrib ed
in the p reced in g s e c tio n in v o lv ed the amounts o f m elt prepared and the
c o n ta in e r s u se d .
100

For t h is work the w eig h ts were la r g e r , g e n e r a lly about

g . , and la r g e carbon c r u c ib le s were u sed .

The creep in g te n d en cie s

o f th ese m e lts are f a i r l y grea t and th e litliiu m flu o r id e has a p p recia b le
v o l a t i l i t y a t th e higher tem p eratures, both o f which e f f e c t s tend to

66

cause some discord ance between the c a lc u la te d and analyzed c o m p o sitio n s.
H y d ro ly sis became more sev ere a t th e se h igh tem peratures than was the
c a se w ith th e o th e r system s, and a f a in t odor o f hydrogen flu o r id e and
tr a c e s o f o x id e - lik e m a ter ia l on the w a lls above th e m elt were n o tic e a b le .
For t h i s b in ary system a n a ly se s were a t f i r s t made o n ly fo r calciu m ,
and because o f d i f f i c u l t y in removing the m elt from th e carbon c r u c ib le
w ith ou t con tam in ation , no grea t r e lia n c e was p la ced on the a n a ly ses fo r
the f i r s t s e r ie s o f r u n s .

On the subsequent check d eterm in ation s

platinum c o n ta in er s were used and a n a ly ses were made fo r both calcium and
lit h iu m .
On th e l a t e r s e t o f runs a m aster batch o f m ixture was prepared which
was s p l i t and c a r r ie d (by a d d itio n o f one or the oth er component) in both
d ir e c t io n s from the e u t e c t ic reg io n in order to more a c c u r a te ly o b ta in
the r e la t iv e tem perature-com position d a ta .
The r e s u lt s o f th e therm al a n a ly ses fo r the two s e r ie s o f runs have
been ta b u la te d , to g e th e r w ith th e c a lc u la te d co m p o sitio n s, in Appendix 7 .
The a n a ly t ic a l r e s u lt s and comparisons o f the c a lc u la te d and analyzed
m elt com p osition s have been ta b u la ted in Appendix 1 5 , P art G.
In F igu res 1 and 2 , which f o llo w , the r e s u lt s fo r th e binary CaF2-LiF
system are shown g r a p h ic a lly .

F iv e v a lu es ob tain ed on th e f i r s t t r i a l s ,

concerning which th er e was grea t u n c e r ta in ty , have been om itted where the
subsequent d eterm in ation s in d ic a te d th a t th ey were in g ro ss e r r o r .

The

rem aining v a lu e s which d e v ia te from the smooth curves would probably be
improved by a red eterm in a tio n , but t h is was not b e lie v e d n ecessa ry in view
o f t h e ir r e la t iv e unimportance fo r th e main purpose o f th e problem.
in g the f ig u r e s a ta b u la tio n i s g iv en o f th e p lo tte d v a lu e s .

F ollow ­

418
1400

1300

200

Hioo

H000

FIGURE I
SYSTEM CaF,-LiF

r900

850

800

1-800

750

750
MOLE % Ca&

70

875i

m

FIGURE 2
SYSTEM C a F g - L iF
(LiF — RICH REGION)
8 5 OH

825H

800

775

750
LiF

MOLE % CaF.

100%

71

TABULATION OF VALUES PLOTTED IN FIGURES 1 AND 2

Mole P ercen t CaF2

Tamp

C.

Mole P ercent CaF2

Temp

C.

815

2 2 .5

803

3 .0 (?)

85 l

2 2 .8 (?)

801 (? )

5 .9 (?)

839

2 5 .0

819

8 .6 (? )

823

25.0

826

1 1 .1

808

2 5.9

850 (?)

1 3 .5

812

26.0

850 (?)

i5 .o

79J4-

27.0

855

1 8 .0

780

29.2

861

1 8 .3

779

32 .1

898

1 9 .3

775

3 8 .0

955 (? )

1 9 .6

773

5 3 .0

993

2 0 .0

777

5 7 .0

1018

2 0 .0

775

50.7

1035

20 .8

782

60 .0

1115 (?)

2 1 .5

787

70 .5

1180 (? )

22.0

795

100 .0

0

N ote;

lia s

(? ) s i g n i f i e s more than normal amount o f u n ce r ta in ty
as to com position or tem perature. Other v a lu es prob­
ab ly are c o r r e c t to w ith in about t.3% and - 2 ° . In
th e r e g io n from 18 to 22% the accuracy i s probably
good to w ith in ± ,1% and 1 1 °.

72

The system as shown i s a sim ple e u t e c t ic whose o n ly unusual
fe a tu r e i s a ra th er sharply descending slo p e im m ediately to th e r ig h t o f
the e u t e c t i c i which i s somewhat out o f l in e w ith the b alan ce o f the slo p e
from the calcium flu o r id e m eltin g p o in t.

The e u te c tic i s b e lie v e d to be

lo c a te d a t 1 9 .5 t 0 .2 mole p ercen t o f calcium flu o r id e w ith a s o l i d i f i c a ­
t io n tem perature o f 773 i 1° C .
The behaviour o f th e system i s s u b s ta n tia lly the same as would be
exp ected from th e lim ite d amount o f data on analogous system s g iv en in
Appendix 1 .

For purposes o f comparison rough p l o t s fo llo w in Figure 3

showing the r e la t io n s o f the known LiF-Group I I m etal flu o r id e s and CaF2Group I m etal flu o r id e s system s.

73

X .M

UF

LiF

L iF

KF

FIGURE 3

THE SYSTEM CALCIUM FLUOHIDE-BERYLLIUM FLUORIDE

7h

THE SYSTEM CALCIUM FLUORIDE-BERYLLIUM FLUORIDE
The work to o b ta in th e phase diagram o f the CaF3-BeF2 system was
performed in two p a r ts .

In the f i r s t phase o f th e work the nature o f

the system was not w e ll e lu c id a te d , and many o f the r e s u lt s appeared
to be in c o n c lu s iv e .

The p e c u l i a r i t i e s o f the system are such th a t they

le a d to r e s u lt s which d e c e p tiv e ly resem ble "delayed" c r y s t a ll i z a t i o n
e ffe c ts.

A lso in th e BeF2-r ic h p o r tio n o f th e diagram no d e f in it e r e ­

s u lt s were o b ta in a b le because o f genuine delayed c r y s t a lli z a t i o n e f f e c t s .
At t h is sta g e th er e fo re the system was not c le a r ly d e fin e d .
Subsequent to th e above work, the system was in clu d ed among the
to p ic s p resen ted in a paper by Osborn C12U ), but no in form ation has been
r e c e iv e d and no p u b lic a tio n on the work has appeared.
In la t e r work on th e ternary system , s u f f i c i e n t in fo rm a tio n was
ob tain ed to perm it a b e tt e r understanding o f the bin ary CaF2-BeF2 system .
T h ereafter a new a tta ck was made on the system which more c l o s e ly d efin ed
i t in the r e g io n s where c r y s t a llin e products are o b ta in a b le .
For the p rep aration o f th e s e m elts the b erylliu m flu o r id e was
ob tain ed e x c lu s iv e ly from the decom position o f (KH4) 2BeF4 .

For th e f i r s t

phase o f th e work t h i s decom position was c a rr ie d out in la r g e carbon
c r u c ib le s and the phase diagram runs were maae in the same c o n ta in e r s .
Due to th e o x id a tio n o f the c r u c ib le s t h i s method did not perm it main­
ta in in g an accu rate check on the v o l a t i l i z a t i o n l o s s e s .

Creepage and

o x id a tio n o f m a te r ia l on th e upper w a lls o f the c r u c ib le s caused

75

d iffic u ltie s .

In th e la t e r in v e s t ig a t io n o f the sy stem , platinum d ish e s

were used as c o n ta in e r s .

The (NH4) 2BeF4 could be decomposed s u c c e s s fu lly

in th e s e sm aller co n ta in er s In the p resence o f calcium f lu o r id e , ana the
com p osition s and v o l a t i l i z a t i o n lo s s e s could be fo llo w e d more s a t i s f a c t o r i l y .
V o l a t i l i z a t i o n , e it h e r r e a l or the r e s u lt o f h y d r o ly sis r e a c tio n s , becomes
very a p p recia b le from th e se m e lts a t tem peratures above 900° so th a t
com p osition u n c e r ta in t ie s become l a i g e r .
Because o f the d i f f i c u l t i e s in the p rep a ra tio n o f th e b ery lliu m
f lu o r id e , th e se m e lts were reused s e v e r a l tim es by the a d d itio n o f p o r tio n s
o f calcium flu o r ia e to s h i f t th e com positions toward th a t s id e o f the
b in ary phase diagram.

On too many such r e p e t it io n s , p a r t ic u la r ly a t the

higher tem p eratures, th e se m elts became s e r io u s ly contam inated w ith
b ery lliu m o x id e s .

The presence o f the l a t t e r m a ter ia l probably accounts

fo r some d isco rd a n t tem perature r e s u lt s ob tain ed fo r an in v a r ia n t p o in t
lo c a te d near the cen te r o f the system .
The r e s u lt s o f the thermal a n a ly ses have been ta b u la ted in Appendix
8 , o m ittin g th ose runs which gave no c o n c lu siv e r e s u lt s and which were
la t e r r ep e a te d .

The co m p lica tio n s o f the c o o lin g curves p reven t an en­

t i r e l y adequate d e s c r ip t io n o f the r e s u lt s in tab u lar form , but an attem pt
w i l l be made to d e sc r ib e th ese in general terms in the d is c u s s io n which
f o ll o w s .
The r e s u lt s o f th e an a ly ses o f samples o f m e lts have been assembled
in Appendix 1 5 , P art H.

The c a lc u la te d com positions based on in p u t and

v o l a t i l i z a t i o n lo s s e s are a ls o in d ic a te d .

F in a lly a ta b u la tio n i s given

o f the c a lc u la te d com position o f th e m e lts based on the r e s u lt s o f the

76

a n a ly s e s , in t e r p o la tio n b ein g n ecessa ry in the case o f in term ed ia te
samples to take in to account v a rio u s l o s s e s .
A n a ly tic a l r e s u lt s fo r the v a r io u s p rep a ra tio n s o f (i\H4) 2BeF4 are
g iv en in Appendix 1 5 , P art C.
W ith in t h is system the b in a ry compound calcium flu o b e r y lla te occurs
and a c o n sid era b le amount o f work was done on the wet p rep a ra tio n and
X-ray c h a r a c te r iz a tio n o f t h i s m a te r ia l.

S y n th etic work i s covered in

Appendix 9, and the X-ray in v e s t ig a t io n i s d isc u sse d in Appendix 1 3 .
W ith r e s p e c t to the l a t t e r m a te r ia l i t may be remarked th a t i t was not
p o s s ib le to in d ex th e l i n e s o f the powder d if f r a c t io n p a tte rn or to
determ ine the u n it c e l l dim ensions o f the compound.

Apparently i t i s a

member o f one o f th e more complex c r y s t a lli n e c la s s e s such as m onoclinic
or t r i c l i n i c which could not be ch a r a cter iz ed w ith th e a v a ila b le equip­
ment .
O p tic a l exam inations o f the phases p resen t in the m e lts are d isc u sse d
i n co n n ectio n w ith the X-ray work.

The calcium flu o r id e occu rs as an

is o t r o p ic p h ase, 'while the b ery lliu m f lu o r id e , tends to form a str a in e d
is o t r o p ic g l a s s .

The compound CaBeF4 occu rs almost e x c lu s iv e ly as prism a­

t i c m a te r ia l so o r ie n te d on th e m icroscopic s lid e th a t i t i s a n is o tr o p ic
when examined w ith p o la r iz e d l i g h t .

When b erylliu m oxid e was p r e se n t i t

was r e a d ily ob servab le as sm all hexagonal prisms o f much h igh er r e f r a c t iv e
in d ex than the oth er components o f the m e lts .

Q u a lita tiv e d eterm in ation

o f th e ph ases p resen t was th er e fo re q u ite r e a d ily accom plished w ith
o p t ic a l m ethods.

77

The ph ases were examined w ith r e s p e c t to th e constancy o f r e f r a c ­
t iv e in d ex u s in g m icroscop ic immersion m ethods.

In gen eral no good

in d ic a tio n s o f changes o f t h i s p rop erty w ith m elt com p osition were
o b ta in e d , but the s tr a in e d ch a ra cter o f the b ery lliu m flu o r id e prevented
good o b s e r v a tio n s .
In X-ray exam inations fo r the d eterm in ation o f c o n s titu e n t p h a se s,
a s l i g h t apparent change o f aboutO.Ol A° shrinkage o f the l a t t i c e plane
se p a r a tio n s o f CaBeF4 was noted as the com position was s h if t e d away from
the BeF2- r ic h s id e o f the sy stem .

Whether t h is i s due to s o lid s o lu tio n

o f BeF2 in CaBeF4 i s not known and probably i s not determ inable due to
the g la s s form ing p r o p e r tie s o f m e lts in t h i s r eg io n where the BeF2
i t s e l f does not c r y s t a l l i z e .

The d iffe r e n c e ofO.OlA0 i s however la r g e r

than the normal measuring error ana hence i s b e lie v e a to be a t l e a s t
p a r tly r e a l .
I n s p e c tio n o f the ta b u la ted r e s u lt s o f th e thermal a n a ly se s in
Appendix 8 , w i l l show co n sid era b le discordance among the v a lu e s , in
p a r tic u la r th o se o f th e in v a r ia n t p o in t o ccu rrin g near the cen ter o f the
system .

The r e s u lt s w i l l be d isc u s se d w ith r e sp e c t to c e r ta in reg io n s

o f th e diagram and an attem pt w i l l be made to e x p la in th e p e c u l i a r i t i e s
o f the system .

The phase diagram o f the system i s g iv en i n Figure h

fo r a s s is ta n c e in fo llo w in g th e d is c u s s io n .

A ta b u la tio n o f the e x p e r i­

mental v a lu e s used in p lo t t in g th e diagram fo llo w s the f ig u r e .

78

FIGURE 4

SYSTEM
CaF0— BeF,
-1300

-1200

-1100

-1000

-9 0 0

-8 0 0

900-

700
BeF.

CaBeR

CaF,

MOLE % CaF,

79

TABULATION OF COMPOSITIQN-TMPERATUAE VALUES FOE FIGURE 6
C om position,
Mole P ercen t CaF2
ca.

ca.
ca.

ca.
ca.
ca.

3 .3
6 .0
6 .2
7 .1
10 .1
1 3 .1
1 5 .8
1 9 .2
2 0 .5 (? )
2 5 .2
29 .0
37 0
3 9 .5
1;0 .0
UO.O

•I4.2 .0
6 3 .6
c a . 6 6.1
6 6 .1
65 .2

Tem perature,

°c.
772
712
702
716
750 (?)
71+0
750
757
751
780
770
800
813
818
819
828
833
833
833
836

C om position,
Mole P ercen t CaF2
6 5 .8
6 6 .6
67 .6
1+7.7
6 8 .6
kb. 8
6 9 .0
5 0 .6
50 .5 (? )
51.0
52 .0
5 3 .7
5 5 .6
5 6 .8
5 7 .0
6 1 .0
6 6.5 (? )
6 6 .0
7 3.0
7 5 .0

78 .o(?)

Temperature,
°c.

636
835
832
836
836
c a . 890
836
838
860 (?)
831
859
876
907
831
833
c a . 980
900
ca.1075
831
ca .1160
831

N ote: l ) Symbol "ea" s i g n i f i e s approxim ately ±1/2 mole p ercen t composi­
t io n u n c e r ta in ty .
2) Symbol (?) s i g n i f i e s g rea ter than normal u n c e r ta in ty , p o s s ib ly
± 1 mole p ercen t or more.
3) I t w i l l be noted th a t on a number o f th e m elts above $3 mole
p e r ce n t the i n i t i a l c r y s t a l l i z a t i o n was u n d etected and the
e u t e c t ic tem perature i s g iv en as the f i r s t observed thermal
e ffe c t.
6) V alues fo r seven m e lts , probably contam inated w ith BeO im­
p u r it y , have been om itted from the p lo t te d v a lu e s .

80

As p r e v io u s ly mentioned th e b ery lliu m flu o r id e norm ally e x i s t s in
a polym eric g la s s y form w ith ou t a tru e m eltin g p o in t .

At 800 0 , the s o -

c a lle d " m elting p o in t ", th ere i s no la r g e ob servab le change in a gradual
s o fte n in g p r o c e ss which probably i s n o tic e a b le as low as 700° .

At 800°

th e m a ter ia l i s extrem ely v isco u s and not u n t il about 900° does the
m a te r ia l e x h ib it a reason ab le resem blance to a l i q u i d .

V o l a t i l i t y o f th e

b e r y lliu m flu o r id e i s s u f f i c i e n t a t t h i s temperature to be v i s i b l y
e v id e n t .
The l e f t hand p o r tio n o f the bin ary system from 0 to about 20 mole
p ercen t CaF2 i s not determ inable on the b a s is o f the r e s u lt s o f t h i s work.
In t h i s r e g io n th e e f f e c t o f th e a d d itio n o f calcium flu o r id e i s m erely
to r e l a t i v e l y low er the gradual t r a n s it io n zone o f the pure b ery lliu m
f lu o r id e .

The m e lts are p r o g r e s s iv e ly l e s s v isc o u s fo r the same tempera­

tu re bu t no d e f i n i t e liq u id u s l in e i s determ ined.
I t i s b e lie v e d th a t a e u t e c t ic t h e o r e t ic a lly occurs in the v i c i n i t y
o f 10 mole p ercen t CaF2 but no good in d ic a tio n s o f i t s p o s it io n were ob­
ta in e d .

C onsequently t h i s reg io n o f th e phase diagram i s shown by the

d o tted l i n e s which g iv e estim ated gen eral behaviour i f the m elts would
c r y s ta lliz e .

T his " th e o r e tic a l" e u te c tic might p o s s ib ly be as h ig h as

750° based on s o l i d i f i c a t i o n tem peratures o f m elts w ith high er calcium
flu o r id e c o n te n ts .
In th e gen era l r e g io n o f 20 to $0 mole p ercen t CaF2 the i n i t i a l
work on th e system showed th e fo llo w in g c h a r a c t e r is t ic s .

As the calcium

flu o r id e c o n te n t was in c r e a se d the p o in t o f i n i t i a l c r y s t a l l i z a t i o n rose
to a maximum o f approxim ately 836° .

The thermal e f f e c t s observed s h ift e d

81

from sm all f l a t p la te a u s to much lo n g e r p la tea u s showing a v ery f l a t
i n i t i a l p o r tio n fo llo w e d by a gradual slo p e and th en a rap id temperature
drop which approxim ately c o in c id e d w ith th e disappearance o f th e l a s t
apparent l i q u i d .

The g en eral behaviour th er e fo re g r e a tly resem bled an

in v a r ia n t c r y s t a l l i z a t i o n p ro cess in which about o n e -h a lf to tw o -th ir d s
o f th e m elt s o l i d i f i e d a t n ea rly co n sta n t temperature and th en the
b alan ce s o l i d i f i e d over a much la r g e r temperature ran ge.

For com position s

o f about 55 mole p ercen t CaF2 , the m e lts appeared to la r g e ly s o l i d i f y
w ith in about a 10° ran ge, w ith ap p aren tly about 80$ o f the p r o c e ss o ccu r­
r in g w ith in a 1° range.

To add to th e co n fu sio n t h i s s o l i d i f i c a t i o n

p r o c e ss d id not occur a t a p a r tic u la r temperature fo r m elts o f n ea rly
th e same co m p o sitio n , but in s te a d in about o n e -h a lf th e c a se s occurred a t
about 633° , w h ile the b alan ce o f the s o l i d i f i c a t i o n p r o c esses were
s c a tte r e d over lower tem peratures down to about 820° .

(The l a t t e r r e ­

s u l t s are now b e lie v e d to have been due to the e f f e c t o f contam ination
o f th e m elt by b ery lliu m o x id e .)

From the r e s u lt s ob tain ed a t t h i s sta g e

o f th e work, i t was s tr o n g ly su sp ected th a t most o f the therm al e f f e c t s
ob tain ed in the 20 to 50 mole percen t CaF2 reg io n might a l l rep resen t
th e same in v a r ia n t eq u ilib riu m p r o c e s s , d is to r te d however by hindered
c r y s t a l l i z a t i o n e f f e c t s which were dependent on com p osition .
In th e r e g io n o f 50-100 mole p ercen t CaF2 the i n i t i a l r e s u lt s were
g e n e r a lly more c o n c lu s iv e but' a ls o in p a rt c o n fu sin g .

M elts in the r eg io n

o f 50 to 55 mole p ercen t showed th e same therm al e f f e c t s as b e fo r e , w ith
no a p p a r e n t - in it ia l c r y s t a l l i z a t i o n a t h ig h er tem peratures than about
833°•

Suddenly fo r m e lts o f about 60 mole p ercen t and h ig h e r , apparent

82

i n i t i a l c r y s t a l l i z a t i o n appeared a t tem peratures o f about 975° and
h ig h e r .

S u b seq u en tly, by the tim e th e low er in v a r ia n t p o in t o f about

833 was reach ed , e it h e r very l i t t l e or no apparent liq u id rem ained.

T his

p e c u lia r b eh a v io u r, in d ic a t iv e o f a e u t e c t ic lo c a te d to the r ig h t o f the
compound CaBeF4 , was app arently not confirmed by m e lts in th e gen eral
r e g io n o f $0-55 mole p e r ce n t CaF3 .

In ste a d the l a t t e r e x h ib ite d much the

same therm al e f f e c t s as m e lts to th e l e f t o f 50 mole p e r c e n t.

(Here

ag a in the p resen ce o f b e r y lliu m oxid e which i s in c r e a s in g ly produced a t
h ig h er tem peratures added to the c o n fu sio n by c o n tr ib u tin g a number o f
r e s u lt s in th e 820° r a n g e.)
At t h i s sta g e th ese r e s u l t s , i f th ey cou ld be b e lie v e d , in d ic a te d
a e u t e c t ic a t about 50 mole p ercen t CaF2 , w h ile the p reced in g work in d i­
c a te d , i f an yth in g, a p e r it e c t ic r e a c tio n in t h is v i c i n i t y .

Work was

then d isc o n tin u e d on t h i s system to see i f l a t e r r e s u lt s on the tern ary
system would c o n tr ib u te anything to the understanding o f t h i s system .
I t i s now b e lie v e d th a t the gen era l d e c ep tiv e n e ss o f the system w ith a
tendency fo r q u ite f l a t p la te a u s (fo llo w e d by gradual slo p e s) i n the 20
to 50 mole p ercen t reg io n a r is e s from two c a u se s.

The f i r s t cause i s

b e lie v e d to be th a t the system a c tu a lly has a rather sh allow slo p e o f
the liq u id u s l i n e , which i n th e v i c i n i t y o f kS to 50 mole p ercen t becomes
p r a c t ic a lly h o r iz o n ta l.

The second cause i s th a t c r y s t a l l i z a t i o n i s

i n i t i a l l y s l i g h t l y d elayed and c o n seq u e n tly , even when s ta r te d , i t cannot
proceed r a p id ly enough to r a is e the su p erco o led m elt tem perature to th e
proper v a lu e .

T h is r e s u lt s in a f la t t e n in g - o u t o f th e upper p o r tio n o f a

slo p in g thermal a n a ly s is curve which r e a l l y r ep re sen ts a u n iv a ria n t p r o c e s s .

83

A fte r work on th e tern ary system showed approxim ately the same
behaviour in th e v i c i n i t y o f th e bin ary CaF2-BeF2 system , except w ith
b e tt e r in d ic a tio n s o f a e u t e c t ic p r o c e s s , a r e in v e s t ig a t io n was made o f
the c e n te r o f th e b in ary system .

Improved tech n iq u es perm itted a b e tt e r

d eterm in a tio n o f th e phase r e la t io n s , w h ile the use o f platinum c o n ta in e r s
seemed to g r e a tly d ecrease te n d e n c ie s fo r b ery lliu m o x id e form ation
which p r e v io u s ly had co n trib u ted discordance to the r e s u l t s .
I n t h i s la t e r work a m aster batch o f m elt was prepared w ith 1+0 mole
p ercen t CaF2 co m p o sitio n , and the batch was s p l i t in two p a r ts which
were c a r r ie d up to and beyond 50 mole p ercent by the a d d itio n o f por­
t io n s o f calcium f l u o r i d e .

C leaner m e lts p erm itted v is u a l o b serv a tio n

and more c e r t a in t y as to co m p o sitio n .

The fo llo w in g r e s u lt s were ob­

ta in e d .
The i n i t i a l c r y s t a l l i z a t i o n p o in t, as in d ic a te d by the appearance
o f su r fa c e n e e d le s o f CaBeF4 and the s t a r t o f a therm al p la te a u , was
g r a d u a lly s h if t e d toward h igh er tem peratures w ith in c r e a s in g CaF2 con­
te n t.

The p la t e a u s , a t f i r s t slo p in g s l i g h t l y , again became r e l a t i v e l y

f l a t above 1+5 mole percen t CaF2; w h ile th e p o in t o f f i n a l disappearance
o f apparent liq u id ro se from about 770° to about 8 1 0 °.

At 1+7.6 and 1+7.7

mole p ercen t no i n i t i a l c r y s t a l l i z a t i o n o f CaF2 was observed p r io r to
th e p la t e a u .

On a subsequent m elt c a lc u la te d to be 1+8.8 mole p e r c e n t,

p o s s ib le CaF2 d e p o s itio n was observed a t about 890° but t h i s may have
been im p u rity .

Another m elt a t 1+8.6 mole p ercen t d id n o t , however, show

t h is b eh aviou r, showing o n ly an 83I+0 p la te a u .
On th e next two m e lts a t 5 3 .7 and 5 5 .6 mole p e r c e n t, i n i t i a l
d e p o s itio n o f CaF2 was observed a t 87l+° and 90 7° r e s p e c t iv e ly .

These

85

liq u id u s p o in ts were co n sid e ra b ly d if f e r e n t in lo c a t io n from th e p r io r
d ou b tfu l o b se r v a tio n a t 1+8.8 mole p e r ce n t and more w eight was placed
upon them as th ey were in b e tt e r agreement w ith o th er d a ta .

Above

t h i s r e g io n th e liq u id u s curve was a ste ep slo p e le a d in g upward toward
th e CaF3 m e ltin g p o in t .
The behaviour i s in te r p r e te d as th a t o f a somewhat d e fe c tiv e bin ary
e u t e c t ic r e a c tio n i n th e v i c i n i t y o f about 52 mole p ercen t CaF2 .

The

f a c t th a t some liq u id was apparent below the in v a r ia n t p o in t fo r m elts
w ith g rea ter than 50 mole p ercen t CaF2 co n ten t i s not ex p la in a b le on
e it h e r a p e r i t e c t i c or e u t e c t ic b a s is fo r a compound o f th a t co m p o sitio n ,
excep t as a s l i g h t l y incom plete phase r e a c tio n .

Only s l i g h t evidence

fo r a dip o f about 2 to 5° e x is t s fo r th e v a lle y ly in g between 50 and
52 mole p e r c e n t.

A s im ila r dip exten d s out in to the tern a ry system fo r

some d is ta n c e and, as d isc u s se d th e r e , the evid en ce as a whole p o in ts
toward a b a r e ly con gru en tly m eltin g CaBeF4 w ith a c lo s e l y a d jo in in g
e u t e c t ic a t about 52 mole p ercen t GaF2 .
The peak value obtained fo r CaBeF4 d e p o s itio n was 836° and th e
in d ic a te d value o f th e e u te c tic was about 8 3 2 °, the l a t t e r value g e n e r a lly
f a l l i n g in the range 83O to 833° ( a l l tem peratures based on c o o lin g c y c le s ) .
Based on h e a tin g c y c l e s , which tend to be somewhat h igh er fo r v ariou s
r e a so n s, th e in d ic a te d v a lu e s would be l e s s than 81+0° fo r th e CaBeF4 peak
and g r ea ter than 836° fo r th e b in a ry e u t e c t ic p o in t.
The f a i l u r e to observe the e u t e c t ic which should e x i s t in the range
o f 0 to 50 mole p ercen t CaF2 , probably around the 10-15 mole p ercent
r e g io n , i s probably a r e s u lt o f the i n a b i l i t y o f th e m e lts to overcome

85

th e high v is c o u s r e s is ta n c e to c r y s t a lli z a t i o n a t th ese low er tempera­
tu r e s.

There were s l i g h t in d ic a tio n s th a t th e e u t e c t ic l i e s as high as

about 770° , s in c e in one c a se a very sh o r t p la tea u was ob tain ed a t 775°
on a 37 mole p ercen t m ix tu re, and in many ca ses apparent s o l i d i f i c a t i o n
occurred by about 770 to 780° .
In a l l p r o b a b ilit y , because o f the com plexity o f b ery lliu m flu o r id e
and i t s s o lu tio n s in the r e g io n o f 0 to 30 mole p ercen t CaF2 , the system
i s one o f many components r a th e r than b in a r y , w ith th e c o n c en tr a tio n and
ty p es o f s p e c ie s b e in g a fu n c tio n o f the calcium flu o r id e c o n te n t.
Under such circum stances ordinary phase r u le r e la t io n s h ip s would n o t be
exp ected to be co m p letely v a lid fo r p r o c e sse s ta k in g p la c e in r e l a t i v e l y
f i n i t e tim e in t e r v a ls during which eq u ilib riu m i s approached a t th e
slow er r a te s c h a r a c t e r is t ic o f g l a s s e s .

THE SYSTEM LITHIUM FLORIDE-BERYLLIUM FLUORIDE

86

THE SYSTEM LITHIUM FLUORIDE-BERYLLIUM FLUORIDE

The in v e s t ig a t io n o f th e LiF-BeF3 system was c a r r ie d out in con ju n ction
w ith the stud y o f th e tern ary sy stem .

In g e n e r a l, a f t e r u se o f a m elt to

determ ine a p o in t on the liq u id u s l in e o f the b in ary system , calcium
f lu o r id e (o r a s u ita b le m ixture o f calcium flu o r id e and lith iu m or b ery lliu m
f lu o r id e ) was added to s h i f t th e com position o f the m elt in to th e tern ary
system alon g a " section " d ir e c te d toward th e com position o f th e a d d itiv e .
T his procedure reduced th e n ecessa ry handling and p rep a ra tio n o f b ery lliu m
flu o r id e to a minimum.
In th e reg io n o f the compound lith iu m f lu o b e r y l la t e , LigBeF.^, th e
above procedure cou ld not r e li a b l y fu r n ish r e la t iv e com position s w ith the
accuracy n e c e s s ita te d by th e p e c u l i a r i t i e s o f the b in a ry system .

Here i t

was n ecessa ry to proceed by a method in v o lv in g s u c c e s s iv e a d d itio n s
( u s u a lly o f lith iu m flu o r id e ) to an i n i t i a l m elt in the bin ary system , to
cause s h i f t s in com position acro ss the reg io n o f compound o ccu rren ce.
W hile t h is procedure fu rn ish ed q u ite s a t is f a c t o r y r e la t iv e d a ta , a n a ly t ic a l
err o r s made th e data i n s u f f i c i e n t l y accu rate on an a b so lu te b a s i s .

In

attem pts to overcome th e l a t t e r problem attem pts were made to s y n th e siz e
th e pure compound, LigBeF4 , to remove the a b so lu te u n c e r t a in t ie s .

(See

Appendix l l ) .
B efore commencing th e work on t h i s b in a ry system the rep o rts o f rec e n t
work by T h ilo and Lehmann (166) and Roy, Roy
a v a ila b le .

and Osborn (13^) became

These in v e s t ig a t o r s rep orted q u ite d iv erg en t r e s u lt s which

87
c o in c id e d w ith t h e ir d if f e r e n t co n cep ts o f th e s i l i c a t e system s fo r which
LiT-BeF3 should c o n s t it u t e a Goldschmidt (52) "model** sy stem .

The com­

p a r a tiv e r e s u lt s w i l l be d isc u sse d a t the end o f t h is s e c t io n , but a t
t h is p o in t i t may be mentioned th a t in th e p r e se n t work good agreement
was ob ta in ed w ith about o n e -h a lf o f the work

o f each group.

These d iv e r ­

gent r e s u lt s la r g e l y were r e sp o n sib le fo r th e somewhat e x c e s s iv e i n v e s t i ­
g a tio n o f t h is system in the p r e se n t work, although accompanying d i f f i ­
c u l t i e s in th e tern a ry system a lso caused a d d itio n a l work.
The system may be broken down in to fo u r p o r tio n s fo r purposes o f
d is c u s s io n o f procedure and r e s u l t s .
In th e r eg io n o f 0 to 25 mole p ercen t BeF3 i t may be seen in F igu re 5
th a t th e slo p e o f the liq u id u s l i n e i s r e l a t i v e l y gradual and co n seq u en tly
i n t h i s r e g io n s a t is f a c t o r y thermal e f f e c t s were obtained upon the i n i t i a l
c r y s t a l l i z a t i o n o f lith iu m flu o r id e and no s p e c ia l d i f f i c u l t i e s were en­
cou n tered .
In th e reg io n o f 25 to 36 mole p ercen t BeF2 , the system i s p e c u lia r
in s e v e r a l r e s p e c ts and the in t e r p r e ta tio n o f r e s u lt s i s su b jec t to a
number o f d i f f i c u l t i e s .

In the r e g io n o f 25 to 30 mole p e r ce n t th e

liq u id u s becomes s te e p e r and o n ly s l i g h t therm al e f f e c t s o f i n i t i a l
c r y s t a l l i z a t i o n were o b ta in e d , which had to be confirmed by v is u a l de­
t e c t io n o f th e i n i t i a l c r y s t a l l i z a t i o n .

In th e reg io n from 30 to about

3 3 .5 mole p ercen t th e slo p e o f approxim ately 50° per mole p ercen t becomes
so ste e p th a t no r e li a b le therm al e f f e c t was ob tain ed on c o o lin g when
the liq u id u s l i n e was reached.

Here v is u a l d e te c tio n had to be used

e x c lu s iv e ly to determ ine the tem perature o f i n i t i a l d e p o s itio n o f lith iu m

88

f lu o r i d e .

Although v is u a l d e te c tio n was rea so n a b ly s a t is f a c t o r y because

o f the c h a r a c t e r is t ic " oily" appearance o f th e f i r s t lithium , flu o r id e
su rface c r y s t a l s , s l i g h t in a c c u r a c ie s o f com p osition r e s u lt e d in d isco rd a n t
tem perature v a lu e s .

By c o n tr a s t, in the r e g io n from about 3 3 .5 to 36 mole

p ercen t th e system i s q u ite f l a t , and the i n i t i a l d e p o s itio n o f litliiu m
f lu o b e r y l la t e , L i3BeF4 , in c h a r a c t e r is t ic n eed le form,was accompanied by
a stro n g therm al e f f e c t during which a l l m e lts in t h is r e g io n seemed to
s o l i d i f y a t p r a c t ic a lly the same tem perature.
Here a g a in , as in the CaFa-BeF3 system , th e p r a c t ic a l c o in cid en ce o f
an in v a r ia n t p r o c ess and a compound w ith a very blu n t peak on th e liq u id u s
curve c a u ses c o n sid e ra b le d i f f i c u l t y in in t e r p r e ta tio n .

As th ere are th ree

p o s s ib le in v a r ia n t p r o c e sse sj i . e . p e r i t e c t i c w ith incongruent compound,
e u t e c t ic w ith congruent compound, and th e sin g u la r c a se o f an "exactly*1
con gru en tly m e ltin g compound, co n sid era b le e f f o r t and c o n sid e r a tio n was
giv en to t h i s r e g io n as d isc u sse d la t e r in t h i s s e c t io n .
In the reg io n o f 36 to about 53 mole p e r ce n t BeF3 i t may be seen in
F igure 5 th a t th e liq u id u s has a v ery gradual slo p e to a bin ary e u te c tic
p o in t a t about 3 5 8 °.

In t h is r e g io n i n i t i a l c r y s t a ll i z a t i o n produced stron g

thermal e f f e c t s and no d i f f i c u l t i e s were encountered ex cep t in the g en era l
v i c i n i t y o f 38O to 36O0 where th e v i s c o s i t y o f th e m e lts caused th e i n i t i a l
c r y s t a l l i z a t i o n to become somewhat s lu g g is h .

I t should be n o te d , however,

th a t a l l c o o lin g curves o f m elts in the 36 to 53 mole p ercen t reg io n showed
i n i t i a l l y a f l a t p la te a u , th e le n g th o f which decreased g ra d u a lly u n t i l i t
became l i t t l e more than an i n f l e c t i o n p o in t in th e r e g io n o f U5 to 53 mole
p e r c e n t.

T h is behaviour i s worthy o f n o te in co n n ectio n w ith the evidence

89

o f T h ilo and Lehmann fo r compound occurrence w ith th e com position
MgBegF.? a t I4O mole p ercen t BeF2 .
In th e r e g io n o f 5? to 100 mole p ercen t BeF2 the g l a s s - l i k e c h a r a cter ­
i s t i c s o f b ery lliu m flu o r id e co m p letely prevented s a t is f a c t o r y determ in­
a tio n o f th e liq u id u s curve , i f in d eed one a c t u a lly e x i s t s .

A b in a ry

e u t e c t ic a t about 358° and $2 mole p ercen t was apparent, but the balance
o f the BeF2- r ic h m elts showed o n ly gradual th ic k e n in g .

The broken curve

in t h i s r e g io n o f F ig u re 5 in d ic a te s t h is g en era l behaviour by p o in ts o f
g r e a te s t change in th e v i s c o s i t y o f the m e lts .
The o th er in v e s t ig a t o r s o f the b in ary system a p p aren tly had l i t t l e
b e t t e r su c ce ss w ith t h i s p o r tio n o f the system .

T h ilo and Lehmann gave

o n ly a fragm entary p o r tio n o f the curve and made no extravagant claim s
as to i t s s ig n if i c a n c e .

Osborn e t a l . showed a com plete liq u id u s l in e

o b ta in e d by the ’'quenching1' te ch n iq u e, b u t th e r e s u l t s are i n com plete
c o n f l i c t w ith th e p r o p e r tie s o f b ery lliu m flu o r id e and i t i s b e lie v e d
th a t th e curve has no r e a l s ig n if ic a n c e oth er than as a measure o f s o lid
phase r e a c tio n r a te as a fu n c tio n o f tem perature and com p osition .
The r e s u lt s
on

o f th e LiF-BeF2 b in a ry system work are shown g r a p h ic a lly

the fo llo w in g f i g u r e s .

given in F igure 5 .

The g en era l p ic tu r e o f the

e n tir e system i s

I t should be emphasized th a t the shape o f th e liq u id u s

curve in th e 30 to 35 mole p ercen t r e g io n was drawn from a c o n sid e r a tio n
o f the more accu rate data given in succeeding f ig u r e s , ra th er than on
se v e r a l r e l a t i v e l y in a ccu ra te p o in t s , shown in t h is r e g io n on Figure 5 .
The r e s u lt s
F ig u res 5 , 6 and

o f therm al a n a ly ses used in p lo t t in g
7 are ta b u la te d in Appendix 1 0 .

The

the diagrams o f
v a lu e s i n F igure 5

90

are g iv e n in the ta b u la tio n im m ediately fo llo w in g F igu re $ .

The v a lu es

fo r F ig u r e s 6 and 7 are a v a ila b le in seg reg a ted form in Appendix 1 0 .

The

a n a ly t ic a l r e s u l t s } which in some c a se s are the b a ses fo r the ta b u la te d
com p osition s are g iv e n in Appendix 1 5 , P a rt I .

91

850*i

800H

750

FIGURE 5
SYSTEM
LiF — BeE,

700

650

600

55<H

500H

450H

400H

350

LiF

BeR

92

TABULATION OF VALUES USED IN FIGURE 5

C om position,
Tem perature,
C om position,
Temperature,
Mole % BeF2____________ ^C._______________ Mole %_BeF2_________ °C .

ca.
ca.
ca.

ca.

l+.l
8 .1
12 .0
1 5 .2
1 9 .0
2 2 .1
2 3 .0
21+.3
2 7 .5
2 9 .5
30 .1+
32 .1+
3 2 .5
3 3 .3
3b.o
3 5.9
3 6 .0
38.0

833
808
777
756
715
687
ca. 660
c a . 61+5
600
1+98
ca. 521
1+86
1+60?
1+1+8
1+52
1+51
UU5
1+1+5

l+o.o
l+o.o
1+2 .0
1+3.0
1+5.0
1+6 .1+
1+8.1
c a . 50.0
5 i.o
5 3 .6
56.7
6 0 .0
6 5 .0
70.0
80.0
90.0
95.0

1+37
1+1+1
1+27
1+20
1+03
392
382
372
366
c a . 36O
—

ca.
ca.
ca.
ca.
ca.
ca.

Note — ttca." s i g n i f i e s more than normal u n c e r ta in ty .
Normal u n c e r ta in ty approxim ately * 0 .5 mole % and
t 2° .

1+75 (t h ic k
m
500
n
600
n
700
n
720
n
800

93

FIGURE 6 a

T SERIES
LiF— BeF0
550-

500-

45<H

T

T

T

T

T

T

T

T

T

T

MOLE % BeF2
500

450

MOLE % BeF,
V

FIGURE 6 b
SERIES & MELT IS
LiF — B eF2

94

500-

450MOLE % Be
500-

450-

MOLE % BeR

FIGURE 7 b
V SERIES, L iF -B e F ,

95
The s p e c ia l s tu d ie s o f the 30 to 35 mole p e r ce n t r e g io n are shown
in F ig u r es 6 a , 6b, 7a and 7b to a v o id co n fu sio n o f th e p o in t s .

These

curves are s u b s t a n t ia lly superim posable w ith in l i m i t s o f about ± 1 /2
mole p ercen t or l e s s .

The s l i g h t l y low er tem perature v a lu e s in the

r e g io n o f 33 ^c 35 mole p ercen t are n o t s ig n if i c a n t but are probably
th e r e s u lt o f s l i g h t im p u r it ie s .
be th e more

The h ig h er v a lu e s are co n sid ered to

r e lia b le o n es.

Two o f th e se c u r v e s , F ig u r es 6a, and 6b were ob tain ed u s in g syn­
t h e t ic L i3BeF4 as a s t a r t in g m a te r ia l.

In o n ly one c a s e , th a t o f F igure

6 a , was th e product co n sid ered to be v ery s a t i s f a c t o r y , but a n a ly t ic a lly
th e m a te r ia l d id not show th e e x a c t ly t h e o r e t ic a l co m p o sitio n .

The

product used fo r F igu re 6b, on th e c o n tr a r y , was poor in appearance, but
did have th e proper a n a ly t ic a l co m p o sitio n .
The r e s u l t s shown in F ig u res 7a and 7b were ob tain ed from m e lts
prepared from Brush BeF2 and L iF .

The r e s u lt s o f F igu re 7a were o b ta in ed

by s h i f t i n g th e com position o f two m e lts , w h ile th o se o f F igu re 7b
rep re sen t a s e r ie s o f in d ep en d en tly prepared m e lts which co n seq u en tly
show s l i g h t l y grea ter r e la t iv e d e v ia t io n s .
The most accu rate data on the 30 to 35 mole p e r ce n t reg io n i s th a t
o f F igu re 6a.

The data i s b e lie v e d a ccu rate to w ith in a t l e a s t 1 l / h

mole p ercen t and th e tem perature to t. 1° .

Some o f th e oth er f ig u r e s are

based on e q u a lly good r e la t iv e data but th e a b so lu te v a lu e s o f com position
are l e s s c e r ta in and th e tem peratures g e n e r a lly were s e v e r a l degrees
lo w e r .
The o n ly p o r tio n o f th e b in ary system r e q u ir in g d e t a ile d co n sid e ra ­
t io n i s th e r e g io n im m ediately ad jacen t to the compound L i2BeF4 .

Here

96

a knowledge o f the e x a c t type o f b in ary in v a r ia n t r e a c tio n would f a c i l i ­
ta t e th e in t e r p r e ta tio n o f the tern a ry system .

S o l i d i f i e d m e lts c o n ta in ­

in g as l i t t l e a s I4 mole p ercen t BeF2 showed f a i n t l y th e c h a r a c t e r is t ic
X-ray powder p a tte r n o f L i2BeF4 in a d d itio n to stron ger l i n e s o f L iF .

As

the BeF3 con ten t was in crea sed th e r e la t iv e i n t e n s i t i e s were rev ersed
and the d u ration o f th e in v a r ia n t p ro cess i n the reg io n o f b$0 to I|60o
was in c r e a se d to a maximum a t approxim ately the com position o f L i2BeF4 .
M elts c o n ta in in g very l i t t l e l i q u i d upon rea ch in g th e in v a r ia n t p o in t
n a tu r a lly tended to e x h ib it abnorm ally low in v a r ia n t tem p era tu res.
S o l i d i f i c a t i o n was ap p aren tly com plete a t t h i s p o in t fo r m e lts c o n ta in in g
l e s s than 3 3 .3 mole p e r c e n t BeF2 .

During t h is s o l i d i f i c a t i o n p r o c e s s ,

the L i2BeF4 c r y s t a lli z e d in q u ite m assive forms resem bling columnar
prism s and th ere was no apparent tendency to form any f i n e l y c r y s t a lli z e d
" c l a s s i c a l H e u t e c t ic m ixtu re.
In th e r e g io n o f 3 3 .3 ^0 3^ .5 mole p e r c e n t, th ere were no r e li a b l e
in d ic a tio n s o f th e e x is te n c e o f liq u id below the in v a r ia n t tem perature
e ith e r during c o o lin g or h e a tin g , but t h i s did not preclu de the p o s s ib le
e x is te n c e o f very sm all amounts o f liq u id w ith in the s o l i d m a ss.
At about 3 b . 5 mole p ercen t and t h e r a f t e r , the e x is te n c e o f liq u id
below th e in v a r ia n t tem perature was d e f in it e and a t 35 mole p ercen t or
more the low er in v a r ia n t e u t e c t ic was a lso observab le a t approxim ately
3 5 8 °.

Such m e lts tended to show f a i r l y f l a t therm al p la te a u s accompanying

the i n i t i a l d e p o s itio n o f L iaB eF ^ probably as th e r e s u lt o f s lig h t
i n i t i a l u n d ercoolin g o f th e m e lt, combined w ith th e e f f e c t o f a r e l a t i v e l y
gradual tem p eratu re-com p osition s lo p e .

No d is c o n t in u it ie s were observed

97

o th er than th e i n i t i a l p la te a u which g ra d u a lly slo p ed o f f to a curve f a l l ­
in g sm oothly to the' low er in v a r ia n t p o in t a t 358° , and a c c o r d in g ly no
a d d itio n a l compound form ation was in d ic a te d in th e 3 3 .5 to 53 mole p ercen t
r e g io n .
I n s p e c tio n o f th e v ariou s liq u id u s cu rves o f F ig u r e s 6a , 6b , 7 a and 7b
shows a q u ite c o n s is t e n t tendency fo r the p r o je c te d ste e p slo p e to s t r ik e
th e f l a t slo p e a t about 3 3 .5 to 3h.O mole p ercen t BeF3 .

There i s no

apparent s l i g h t d e p r e ssio n in th e p la te a u to in d ic a te : a ;sm a ll bin ary
e u t e c t ic v a lle y to th e l e f t o f th e compound lo c a t io n .

The type o f c r y s t a l­

l i z a t i o n a t th e in v a r ia n t p o in t i s more c h a r a c t e r is t ic o f a p e r i t e c t i c
than a e u t e c t ic r e a c tio n .

Hence on th e b a s is o f th e in fo rm a tio n obtained

in t h i s work, th e in v a r ia n t p o in t must be in te r p r e te d as a p e r i t e c t i c
r e a c tio n lo c a te d a t a com position o f approxim ately 3 3 .5 mole p ercen t BeF3 .
Because o f a n a ly t ic a l u n c e r ta in t ie s i t i s not im p o ssib le th a t th e compound
i s one which i s e x a c t ly con gru en tly m e ltin g , but th ere i s no in d ic a tio n
o f a e u t e c t ic r e a c tio n i n t h i s r e g io n .
R eference i s now made to th e work o f T h ilo and Lehmann (166) and o f
Roy, Roy and Osborn (l3ij-) fo r comparison o f the r e s u lt s rep orted fo r t h is
system .
8 and 9 .

The phase diagrams which th ey rep o rted are reproduced i n F ig u res
Both s e t s o f in v e s t ig a t o r s were stu d yin g the system fo r t h e o r e t i­

c a l rea so n s connected w ith th e resem blance o f the system to c e r ta in
s i l i c a t e sy stem s.
A ccording to the Goldschmidt th eory o f "model" system s (52) a
"weakened11 mo.del system i s ob ta in a b le by the s u b s t it u t io n o f elem en ts hav­
in g s im ila r s i z e s but low er v a le n c ie s .

Such a weakened system would be

98

exp ected to e x h ib it the same g en era l phase r e la t io n s h ip s and have the
same g en era l s tr u c tu r e s , but would be much l e s s s ta b le and hence s u s c e p t­
i b l e o f in v e s t ig a t io n a t much low er tem peratures than a corresponding s i l i ­
c a te system .

T his g en era l th eory due to Goldschmidt was a p p a ren tly f i r s t

t e s t e d by 0 ‘D a n iel and T s c h e is c h w ili (121) and i s c u r r e n tly b e in g i n v e s t i ­
gated by T h ilo and Osborn as judged by more r e c e n t a r t i c l e s (167) (125) .
W hile th e r e s u l t s have been o f in t e r e s t in t h e ir p a r t ia l v e r i f i c a t i o n o f
th e Goldschmidt th e o r y , i t w i l l be seen th a t th e in v e s t ig a t o r s may have
been too q u ic k ly convinced when t h e ir f i r s t e f f o r t s gave th e d e sir e d r e ­
s u l t s needed fo r f u l l agreement w ith th e th e o r y .
To i l l u s t r a t e th e Goldschmidt approach the fo llo w in g ta b le s o f
c r y s t a l r a d ii are reproduced to show some o f th e p o s s ib le s u b s tit u tio n s
p erm itted by t h i s '‘weakened11 model th eo r y .

In the l e f t hand p o r tio n are

shown compounds p r im a r ily o f i n t e r e s t in high m eltin g s i l i c a t e system s
and o p p o site each i s shown a compound o f elem ents o f n e a r ly e q u iv a le n t
s i z e , but decreased v a le n c e , which c o n s t it u t e s the weakened model system .

99

MODEL STRUCTURES
R ein forced
Substance
I o n ic R a d ii , A °.

Weakened
Substance
Io n ic R a d ii » A0 •

S i0 2
MgO
T i0 2
Sn02
Th02
Zn2S i0 4
SrTi03
BaS04
CaMgSi20 6

BeF2
LiF
MgF2

0.31;
0 .7 8
0 .7 8

1 .3 3
1 .3 3
1 .3 3

CaF2
L i2BeF4
KMgF3
RbBF4
N a liB e2F6

1 .0 6
0 .7 8
1 .3 3
1.U9
0 .9 8

1 .3 3
0.3U
0 .7 8
0 .2
0 .7 8

0 .3 9
0 .7 8
0.61;
0.71;
■1.10
0 .8 3
1 .2 7
1.U3
1 .0 6

1 .3 2
1 .3 2
1 .3 2
1 .3 2
1 .3 2
0 .3 9
0.61;
0 .3
0 .7 8

—
—

---------------

—
_ _ _

1 .3 2
1 .3 2
1 .3 2
0 .3 9

1 .3 2

--------

1 .3 3
1 .3 3
1 .3 3
0 .31;

1 .3 3

The above m a te r ia l i s reproduced from th e Osborn a r t i c l e u sin g Goldschm idt’ s
io n ic or c r y s t a l r a d i i .
T h ilo and Lehmann used the P a u lin g io n ic r a d ii v a lu es shown below
and co n seq u en tly a r riv e d a t s l i g h t l y d if fe r e n t c o n c lu s io n s .

Coord. No.
fo r Oxygen

k
6
6
6-8
8
8

S tr u c tu r a l U n it
in S ilic a t e s
0 “2
S i +4
Mg+3
Zn+s
Ca*2
Sr+2
Ba+3

Io n ic R a d ii, S tru ctu r a l U nit
A°
A0
in F lu o b e r y lla t e s
l.UO-

o.ia
0 .65
0.71;
0 .9 9
1 .1 3
1 .3 5

1 .3 6
0 .3 1
0 .6 0

Be+2
L i+i

--------

-----

0 .9 5
—

1 .3 3

Coord. N o.
fo r F lu o r in

h
6
—

6 -8

Na+3-----

—

K+x

8

Goldschmidt o r i g i n a l l y b e lie v e d th a t w ille m it e , Zn2S i0 4 j or p h e n a c it e 3
Be2S i0 4 , should be r e la t e d to Li;gBeF4 , whereas T h ilo and Lehmann b e lie v e d
the M g0-Si02 system should be analogous to the LiF-BeF2 system on the b a s is
o f b e tte r s im ila r it y o f io n ic s i z e s .

Hence f o r s t e r i t e , Mg2S i0 4j and

c 'lin o e n s t a t it e , MgSi03 j would be ex p ected to resem ble LigBeF,^ and LiBeF3 .

100

Osborn e t a l . b e lie v e d th a t w ille m it e , £ n 2S i0 4 , should resem ble L i2BeF4
and th a t th e ZnD-Si02 system should be the r e in fo r c e d model system o f
which LiF-BeF2 was the weakened m odel.
C h r o n o lo g ic a lly , T h ilo and Lehmann p u b lish ed f i r s t and i t i s d o u b tfu l,
because o f th e tim in g , th a t Osborn e t a l . were aware o f t h e ir p r io r p u b li­
c a t io n .
Osborn e t a l . claim ed on the b a s is o f t h e ir r e s u lt s th a t th ere was
a c o n sid e ra b le s im ila r it y between th e system s Z n 0 -S i0 2 and LiF-BeF2 in
th a t th e g en era l fe a tu r e s o f e u t e c t ic s and compound form ation occurred a t
p r a c t ic a lly th e same c o m p o sitio n s.

The fo llo w in g comments may be made

w ith r e s p e c t to t h e ir work, which i s shown g r a p h ic a lly in F ig u re 8 .
For th e m e ltin g p o in t o f lith iu m flu o r id e a v a lu e o f 870° was taken
from th e 1922 e d it io n o f M ellor ( i l l ) .

A s in g le liq u id u s p o in t was ob­

ta in e d in the r e g io n from 0 to 3 3 .3 mole p ercen t BeF2 and based upon t h is
r e s u lt a e u t e c t ic was rep orted a t about 20 mole p e r c e n t.

Both th e p ro­

cedure and th e c o n c lu sio n were erroneous and t h i s r e g io n o f the diagram
may be co n sid ered to be w o r th le s s .
In the r e g io n from 3 3 .3 to 52 mole p e r ce n t BeF2 the Osborn r e s u lt s
and th o se o f the p r e se n t work are in q u ite good agreement fo r the fou r
liq u id u s p o in ts which th ey show.
Over the 52 to 100 mole p ercen t r e g io n th e ir v a lu e s , which r ep re sen t
the liq u id u s fo r th e nlow -q u artzn form o f b ery lliu m f lu o r i d e , le a d to an
e x tr a p o la te d valu e o f 5^-0° fo r th e m e ltin g p o in t o f t h i s form o f b ery lliu m
f lu o r id e .

S in ce th ey ob ta in ed somewhat d isco rd a n t r e s u lt s from a

'’c r y s to b a lite '1. form ob tain ed by the decom position o f (NH4) 2BeF4 , i t i s

101

850i

800

750

FIGURE 8

SYSTEM LF-BeF2

700H

DATA OF
THILO & LEHMANN
650

600

550

500H

450H

400

350H

V

MOLE % BeF-

102

su ggested th a t th e e n t ir e m atter i s u n s e t t le d , p a r t ic u la r ly as th ey g iv e
no in fo rm a tio n on th e X-ray exam inations o f what were probably a t b e s t
o n ly s tr a in e d amorphous samples p o o rly s u ite d to s tr u c tu r a l d e te r m in a tio n s.
The work o f T h ilo and Lehmann (See F igure 9) was a much more thorough
in v e s t ig a t io n in which adequate therm al a n a ly se s were ap p aren tly u sed a t
rea so n a b ly spaced in t e r v a ls , and i n which very e x te n s iv e c r y s ta llo g r a p h ic
and tem perature r e la t io n s h ip s were shown betw een the M g0-Si02 and LiF-BeF2
sy stem s.

Very com plete agreements were ob tain ed and, based upon th e sup­

posed complex form ing a b i l i t y o f flu o b e r y lla t e io n s , an e x te n s iv e analogy
was made w ith s i l i c a t e system s w ith regard to the form ation o f l i n e a r ,
branched and m u ltid ir e c tio n a l a n io n ic com plexes.

D e sp ite the thoroughness

and im agin ation e x h ib ite d by t h i s rep o rt s e v e r a l c r it ic is m s may be made
which would le a d to a d if f e r e n t in t e r p r e ta tio n .
These in v e s t ig a t o r s worked w ith undisturbed m elts ( i . e . u n stir r e d )
and ex p erien ce i n th e p r e se n t work has in d ic a te d th a t la c k o f a p o s it iv e
means o f o b ta in in g tem perature u n ifo r m ity in the m elt norm ally tend s to g iv e
somewhat sp u riou s r e s u l t s , e ith e r h ig h or lo w , depending on th e ex a ct p o s i­
t io n o f the thermocouple w ith in the s o l id if y in g m ass.

D i f f e r e n t i a l s in th e

u n stir r e d liq u id u s u a lly amount to s e v e r a l d egrees even in sm all volum es.
W ith r e s p e c t to the r e g io n o f 0 to about 30 mole p e r ce n t BeF2 q u ite
good agreement e x i s t s between t h e ir v a lu e s and th o se o f the p r e se n t work
fo r the liq u id u s p o i n t s .

Throughout t h i s r e g io n th ere was a gradual in c r e a se

in t h e ir 11e u te c tic " h old in g tim e per gram and th e o n ly p e c u lia r it y was a
v a r ia tio n in th e tem perature o f the " e u te c tic " r e a c tio n in th e order o f
il65>° , !f70°, i|65>° and U 72°. W hile a v a r ia tio n i s norm al, the u su a l tendency

103

870
850-

800FIGURE 9
750-

SYSTEM LiF — BeF2
DATA OF
ROY, ROY & OSBORN

700-

650

600

550

500

EUTECTIC

450

400

350
MOLE % BeF.

io JU

would be to have a gradual in c r e a s e as th e e u t e c t ic com p osition was
approached.
On th e next m elt a t 3 1 .5 mole p ercen t th ey found no i n i t i a l c r y s t a l­
l i z a t i o n i n f l e c t i o n fo r LiF d e p o s itio n , but o n ly a ’’e u t e c t ic ’1 a t J4.620 .
S tr a n g e ly t h e ir h o ld in g tim e per gram a t the e u t e c t ic was about tw o -th ir d s
th a t o f the p reced in g m elt in s te a d o f b ein g g r e a te r .

The discord ance

between the 1|62° valu e and some o f the p rev io u s high er v a lu e s ap p aren tly
d id not bother them and th ey e s ta b lis h e d the ’' e u t e c t ic ’' a t 1|62° and 3 1 .
mole p ercen t BeF2 , in good agreement w ith a e u te c tic in the M g0-Si02 system a t
"about 30 mole percent" according to th e work o f Andersen and Bowen ( 2 ) .
On th e next two m e lts a t 3 3 .3 and 3 6 . mole p ercen t BeF2 th ey ob tain ed
liq u id u s v a lu e s o f 1|75° and 1|75° 3 and th ey gave "holding tim es o f f i r s t
c r y s t a lliz a t io n " which were r e s p e c t iv e ly about 5 and 10 p ercen t l e s s than
th a t o f th e p reced in g " e u te c tic " h o ld in g tim e .

(At t h i s p o in t comparison

i s made w ith the very s im ila r f l a t p la te a u s o f g ra d u a lly d ecrea sin g
le n g th o b tain ed i n t h i s work in th e same r e g io n , which have however a
I4.6O0 maximum v a lu e .)

On t h e ir n ext m elt a t 38 mole p e r ce n t the liq u id u s

value dropped to li53° w ith th e same " holding tim e o f f i r s t c r y s t a lliz a t io n "
as the p reced in g m elt a t 36 mole p e r c e n t.
Up to t h is p o in t t h e ir thermal r e s u l t s c l o s e l y p a r a lle le d th o se ob­
ta in e d in t h i s work, e x ce p t fo r the tem perature v a lu e s ex ceed in g I|.60o .
I t i s su g g ested however th a t t h e ir I4.7O0 fig u r e s show in te r n a l in c o n s is t ­
e n c ie s , as do th e e u t e c t ic h old in g tim e s .

Sin ce th ey r e l i e d co m p letely

on therm al e f f e c t s , the p o s s ib le i n i t i a l d e p o s itio n o f LiF alon g th e very
ste ep liq u id u s curve a t 3 1 .5 mole p ercen t m ight have occurred w ith ou t

io 5

d e t e c t io n , and th e fo r t u it o u s com bination o f a low p o in t o f lj.620 might
be accep ted w ith ou t reason ab le doubt as a probable e u te c tic p o i n t .

If

any tem perature u n c e r ta in ty o f la r g e e x te n t were apparent, however, some
r e in v e s t ig a t io n would seem n e c e ssa r y , bu t th e se workers a p p a ren tly were
s a t i s f i e d d e s p ite the discord ance o f temperature and h o ld in g tim e v a lu e s .
The two su cceed in g v a lu e s o f Ii75° confirm ed t h e ir e u t e c t ic d ia g n o s is ,
although the d ecreased "holding tim es o f f i r s t c r y s t a lliz a t io n " are not
e x p la in a b le on th e same b a s i s .
Summarizing t h is reg io n o f th e T h ilo and Lehmann work, th ere was sub­
s t a n t i a l l y com plete agreement o f t h e ir thermal r e s u l t s w ith th e p r e se n t
work as to g en eral b eh aviou r.

With r e sp e c t to tem perature, t h e ir r e s u lt s

in d ic a te a maximum m e ltin g tem perature o f k7S° fo r L i ^ e F ^ whereas in
t h is work a maximum tem perature o f i|60° was ob tain ed on c a r e f u lly prepared
wet sy n th e siz e d m a te r ia l.

The rem aining d iffe r e n c e s may be due to com posi­

t io n e r r o r s .
Here t h e ir experim ental data i s not s u f f i c i e n t to perm it a d e c is io n .
T heir com p osition s were based e n t ir e l y on the input of- pure LiF and o f
(NH4) 3BeF4 c o n ta in in g some NH4F .

The b e r y lliu m co n ten t o f th e (NH4) 3BeF4

was determ ined by con v ersio n to BeS04 and subsequent r ed u c tio n to BeO.
The c o n tr o l a n a ly se s based on a pure b ery lliu m b a s ic a c e ta te were s ta te d
to f lu c t u a t e , but no in d ic a tio n o f th e p r e c is io n o f th e o v e r - a ll a n a ly s is
i s o b ta in a b le from t h e ir r e p o r t.
With r e s p e c t to the 33 to $0 mole p e r ce n t BeF2 r e g io n th e r e s u l t s o f
the p r e se n t work are in good agreement w ith the rep o rt o f Osborn e t a l . ,
and in f a i r agreement w ith T h ilo and. Lehmann ex cep t fo r t h e ir claim o f the

106

occurrence o f a compound I^B egF ? a t th e I|0 mole p ercen t p o in t .
reasons e x i s t fo r b e lie v in g th a t t h i s assumption i s in c o r r e c t .
e t a l . rep o rted no such in d ic a tio n s in t h e ir work.

S ev era l
Osborn

In th e p r e se n t work a

complete s e r i e s o f g r a d u a lly sh o rten in g p la te a u s were o b ta in ed as the
liq u id u s curve was stu d ie d throughout t h is r e g io n .

T h ilo and Lehmann

a r b it r a r ily assumed th a t because th er e was a la r g e d iffe r e n c e in th e
"holding tim e o f i n i t i a l c r y s t a lliz a t io n " between t h e ir 38 and l+O mole
p ercen t r e s u l t s , th a t such a compound e x is t e d .

T heir liq u id u s curve a lso

shows an i n f l e c t i o n a t th a t p o in t which would be absent i f t h e ir high
v alu es o f lj.750 were changed to i|60° fo r the 3 3 .3 and 3 6 . mole percen t r e ­
s u lts .

They fr a n k ly adm itted th a t o n ly L i2BeF4 sep a ra tes alon g th e curve

throughout th e reg io n 33 .3 to $0 mole p ercen t and th a t L i3Be2F7 was not
d e te c te d in any o th er manner.

They s t a t e th a t the c o o lin g curves in the

r e g io n are anomalous and o f f e r an e x p la n a tio n based on gradual exotherm ic
decom position o f h igh er complex io n s to perm it fu r th e r d e p o s itio n o f
L i2BeF4 .

The "anomalous" c u r v e s , although in a d eq u a tely d e s c r ib e d , would

appear to have been p u rely due to p a r t ia l s o l i d i f i c a t i o n fo llo w e d by
u n dercoolin g and th en renewed s o l i d i f i c a t i o n , a com bination which was
sometimes observed in the p r e se n t work when the m elt was not s u f f i c i e n t l y
a g it a t e d .
The claim o f T h ilo and Lehmann fo r the e x is te n c e o f th e compound
LiBe? 3 i s not based on any d e f i n it e therm al d a ta , but rather on th e i s o l a ­
t io n o f a d if f e r e n t typ e o f c r y s t a l li n e product b e s t o b ta in ed by c h i l l i n g
m elts w ith th e r a tio L iF/BeF2 = h . 9 / 5 .1 in a m od erately r a p id manner;
oth er c o o lin g r a te s y ie ld in g L i2BeF4 rath er than LiBeF3 .

These c r y s t a ls

107

o f ’’L iB eF ^ showed a str o n g resem blance to m on oclin ic p y ro x en e, presum­
a b ly in th e form o f c l i n o e n s t a t i t e , or e n s t a t i t e , MgSi03 .

The is o la t e d

m a te r ia l co u ld not be analyzed s a t i s f a c t o r i l y u s in g t h e ir s u lf a t e method,
and hence was c h a r a cter iz ed o n ly by X-ray powder p a tte r n .

T h ilo and

Lehmann gave a comparison o f X-ray l i n e s fo r th re e sam ples:

a) L igB eF^

b) Sample o f s lo w ly s o l i d i f i e d m elt w ith o v e r - a ll com position o f LiBeF3 ,
and c) "LIBeF^ is o la t e d by the c o n tr o lle d c h i l l i n g p r o c e s s .

T his tabu­

l a t i o n i s reproduced i n Appendix 1 3 , to g e th e r w ith v a lu e s o b ta in ed fo r
L i2BeF4 in t h i s work.
In c o n sid e r in g th e T h ilo and Lehmann data the fo llo w in g p o in ts
should be c o n sid e r e d .

A ccording to them th e product o f b) would show

o n ly the l i n e s o f L i2BeF4 ra th er than LIBeF3 .

They s t a t e th a t b) v a lu e s

are in term ed ia te between th ose o f a) and c ) , which c o n c lu sio n i s h a rd ly
confirm ed by in s p e c tio n o f the r e s u l t s .

From th e manner o f arrangement

of the 52 l i s t e d l i n e s , 16 are a p p a ren tly supposed to be unique fo r LiBeF3 ,
19 fo r L i2BeF4 , and 17 l i n e s common to both compounds.

By comparison o f

t h e ir b) and c) d a ta , however, i t i s apparent th a t even more s im ila r it y
e x i s t s , 27 v a lu es o f l i n e s b ein g th e same fo r both b) and c ) , le a v in g
on ly 6 l i n e s as -unique fo r c) i f the au th or’s im p lied p r e c is io n i s
accep ted .
I f , how ever, th ere i s a s h i f t in th e p o s it io n o f th e L i2BeF4 l i n e s ,
as in d ic a te d by comparison o f t h e n ) and b) v a lu e s , e ith e r r e a l or due
to experim en tal e r r o r , th e claim ed d iffe r e n c e between th e b) and c) v a lu es
may be p u r e ly im aginary.

On t h i s b a s is probably 31 or 32 o f th e t o t a l

o f 33 l i n e s are common to both b) and c ) , th e rem aining d if fe r e n c e s b ein g

108

la r g e l y in e stim a te d i n t e n s i t y .

Such a com plete s im ila r it y o f th e X-ray

powder p a tte r n s o f two r e la t e d compounds i s not p r o b a b le , excep t fo r the
l a t t e r o n e -th ir d o f the l i s t where th e l i n e s caused by th e sm all separa­
tio n s o f p la n es o f high in d ex v a lu e s n a tu r a lly ten d to show c lo se ly spaced v a lu e s .
The g en eral c o n c lu sio n w ith r e s p e c t to t h i s data i s th a t the r e a l i t y
o f any d if fe r e n c e between L i2BeF4 and "LiBeFQ11 rem ains undem onstrated, or
th a t a t b e s t some d is t o r t io n o f th e L i2BeF4 str u c tu r e o ccu rs as a fu n c tio n
o f co m p o sitio n .

The dem onstration o f the Goldschmidt ''model” behaviour

i s reduced to a much sm aller degree o f s im i la r it y between the LiF-BeF2 and
the M g0-Si02 sy stem s.

U n til th e l a t t e r has been ad eq u ately in v e s t ig a t e d ,

the an alogy must remain in com p lete.
On the b a s is o f th e p r esen t work i t i s concluded th a t th e LiF-BeF2
system e x h ib it s two in v a r ia n t p o in ts and one in co n g ru en tly m e ltin g com­
pound, L i2BeF4 .

The low er in v a r ia n t p o in t , r ep r e se n tin g th e e u t e c t ic

m ixture o f L i2BeF4 and BeF2 occurs a t approxim ately $2 mole p ercen t BeF2
and a t about 3 5 8 °.

The upper in v a r ia n t p o in t occu rs as a p e r i t e c t i c

r e a c tio n a t approxim ately 3 3 -5 mole p e r c e n t BeF2 and a t about 558-i|60o .
The region from. 52 to 100 mole p e r ce n t BeF2 y ie ld e d no d e f in i t e liq u id u s
l i n e in fo rm a tio n .
No s im ila r it y e x i s t s betw een the 2n D -S i02 and LiF-BeF2 system s as
proposed by Osborn e t a l .

The e x te n t o f s im ila r it y betw een th e M g0-Si02

and LfF-BeF2 system s as proposed by T h ilo and Lehmann i s a c t u a lly much
l e s s than th ey b e lie v e d , and such s im ila r it y as th er e i s may be co n sid ered
fo r tu ito u s on th e b a s is o f p r e se n t know ledge.

THE TERNARY SYSTEM CALCIUM FLUORIDE—
LITHIUM FLUORIDE—BERYLLIUM FLUORIDE

109

THE TERNARY SYSTEM CAECUM 'FLUORIDE—
LITHIUM FLUORIDE—BERYLLIUM FLUORIDE
P art A — General
The tern a ry system was in v e s tig a te d co n c u r re n tly w ith th e work on
th e b in ary LiF-BeF2 system .

During much o f th e work on th e tern ary s y s ­

tem th e e x a c t natures o f the in v a r ia n t r e a c tio n s were u n c e r ta in in two
o f the con tigu ou s binary system s near th e compounds Li^BeF,^ and CaBeF4 .
The s l i g h t com p osition d iffe r e n c e s sep a ra tin g th e in v a r ia n t p o in ts and
th e compound lo c a tio n s c re a te d d i f f i c u l t i e s in determ ining whether a
p e r it e c t ic or e u t e c t ic phase r e a c tio n was in v o lv e d .

As a consequence o f

t h i s u n c e r ta in ty , th e alread y la r g e number o f p o s s ib le v a r ia tio n s w ith in
the tern ary system was g r e a tly in c r e a s e d .
For a three, component system , whose contiguous b in a ry system s show
two a d d itio n a l compounds and no s o lid s o lu t io n s , th re e in v a r ia n t p o in ts
w i l l occur w ith in the tern ary diagram.

For sy stem s, such as th o se o f

the p r e se n t work, in which none o f th e b in a ry system s show a continuous
s e r ie s o f s o lid s o lu t io n s , one o f the tern a ry in v a r ia n t p o in ts must be
a tern ary e u t e c t i c .

I t i s not p o s s ib le to p r e d ic t th e nature o f th e

oth er two tern ary p o in t s , even w ith f u l l knowledge o f th e b in ary system s.
These two p o in ts m aybe e it h e r p e r i t e c t i c and p e r i t e c t i c , or p e r it e c t ic
and e u te c tic in c h a r a c te r .
The th ree c o n s titu e n t bin ary system s have been d e scrib ed in the
precedin g s e c t io n s .

The CaF2-LxF system i s p u rely e u t e c t i c .

The System CaF2-

BeF2 shows two e u t e c t ic s w ith a compound which b a r e ly m e lts c o n g ru en tly .

110

The LiF-BeF3 system shows both p e r i t e c t i c and e u t e c t ic r e a c tio n s w ith a
b a r e ly in c o n g r u e n tly m eltin g compound.
From th e r e la t iv e tem perature l e v e l s o f the s ig n if i c a n t p o in ts o f
the binary system s i t i s p o s s ib le to make some e stim a te s as to the prob­
ab le b ehaviou r.

Thus, th e l i n e s o f tw o -fo ld s a tu r a tio n ex ten d in g inward

from th e in v a r ia n t p o in ts o f the b in ary system s must descend toward low er
tem peratures w ith in th e tern a ry sy stem .

In a d d itio n , in th e absence o f

s o lid s o lu tio n r eg io n s exten d in g f a r in to the tern a ry system , phase r u le
th eory p erm its drawing c o n c lu sio n s as to which l i n e s o f tw o -fo ld sa tu ra ­
t io n cannot i n t e r s e c t .

The th eo r y , however, p erm its no ch o ic e between

oth er a lt e r n a t iv e arrangements o f th e paths and in t e r s e c t io n s o f the
l i n e s o f tw o -fo ld s a tu r a tio n which are not in v io la t io n o f th e th eo r y .
According to th e assumed arrangements and p o s it io n s o f such in t e r s e c t io n s ,
however, th e th eo ry w i l l d escrib e th e r e s u lta n t phase r e a c tio n s along such
l i n e s and a t th e p o in ts o f in t e r s e c t io n .
B efore r e f e r r in g to th e experim ental r e s u lt s i t seems b e s t to give
some c o n s id e r a tio n t o th e t h e o r e t ic a l p o s s i b i l i t i e s in order th a t the
r e s u lt s may be in te r p r e te d from the sta n d p o in t o f t h e ir agreement w ith
th ese p o s s i b i l i t i e s .

The g en era l th eo ry i s covered more co m p letely in

Appendix $ , hence th e d is c u s s io n here w i l l be more s p e c i f i c .

The e f f e c t

o f s o l i d s o lu tio n s i s om itted from the c o n s id e r a tio n s , not because th ese
s o lu tio n s are n o t p r e s e n t, but fo r th ree main r e a so n s.

F i r s t l y , th e i n ­

v a r ia n t p o in ts are so d is ta n t from a l l e x ce p t one o f th e compounds th a t
extreme m i s c i b i l i t y ranges would be n e c essa ry to a f f e c t the in t e r p r e ta tio n
of r e s u lts .

S eco n d ly , no such s o lid s o lu tio n form ation was apparent from

Ill

o p t ic a l and X-ray exam ination.

F i n a l l y , in th e case o f th e compound

L i3BeF4 , which i s very c lo s e in com p osition to one o f th e in v a r ia n t
p o in t s , s o l i d s o lu tio n form ation cannot be excluded e n t i r e l y .

H ere,

the therm al r e s u lt s and phase r e a c tio n s show no in d ic a tio n s th a t s o lid
s o lu tio n s need be taken in to account in the in t e r p r e ta tio n .
I f the b in a ry system s are p la ced in th e u su a l tr ia n g u la r arrangem ent,
w ith th e bin ary phase diagrams in d ic a te d by p r o je c tio n along the s id e s o f
the tr ia n g le o f F ig u re 1 0 , c e r ta in forb id d en and p erm itted p o s s i b i l i t i e s
may be d e sc r ib e d .

For purposes o f d is c u s s io n , th e BeF2 r ic h p o r tio n s o f

the b in ary system s are shown as approximate e s tim a te s , although th e ex a ct
nature o f th e se r e g io n s was n o t determ ined.

The tem peratures correspond­

in g to th e b in ary in v a r ia n t p o in ts are a ls o shown.

C onsiderable d i s ­

t o r t io n o f th e diagrams w ith r e sp e c t to com position has been made fo r
c l a r i t y , in order to sep arate c lo s e ly spaced p o in t s , o f in t e r e s t fo r t h i s
d is c u s s io n .

Assumed l i n e s o f tw o -fo ld s a tu r a tio n w ith in the tern a ry

system are shown p r o je c te d upon the te r n a r y t r ia n g le .
As f u l l y d isc u sse d in Appendix 5 , i t i s not p o s s ib le fo r the p a ir o f
tw o -fo ld s a tu r a tio n l i n e s from F and G (o r from H and I ) to in t e r s e c t
p rio r to t h e ir in t e r s e c t io n w ith o th er tw o -fo ld s a tu r a tio n l i n e s , in th e
absence o f e x te n siv e s o l i d s o lu tio n r e g io n s .

I t fo llo w s a s a consequence

th a t the l i n e from H must in t e r s e c t the l in e from G a t p o in t K.

K must

a lso be a tern a ry e u t e c t ic p o in t s in c e G has th e lo w e st b in a ry in v a r ia n t
tem perature.
In a d d itio n , because p o in t F and th e observed tern ary in v a r ia n t
p o in ts have tem peratures c o n sid e ra b ly below th o se o f p o in ts w ith in the

112

t r ia n g le AEC, t h i s reg io n may be excluded from the p r e se n t d is c u s s io n .
The p o s s i b i l i t i e s then remain th a t the th ree l i n e s o f tw o -fo ld sa tu r a ­
t io n le a d in g from F , I and J may in t e r s e c t in a v a r ie t y o f w ays.

These

v a r ia tio n s may in c lu d e both the order o f th e in t e r s e c t io n s and th e
lo c a t io n o f th e in t e r s e c t io n s among th e th ree t r ia n g l e s AED, BDE and CDE.
One such com bination i s i l l u s t r a t e d in th e fig u r e w ith the l i n e s
from J and F in t e r s e c t in g a t the p e r i t e c t i c p o in t M, fo llo w e d by i n t e r ­
s e c tio n o f the l i n e from M w ith th a t from I a t another p e r i t e c t i c p o in t
L.

From L the l i n e o f tw o -fo ld s a tu r a tio n f a l l s to th e f i n a l tern a ry

e u t e c t ic p o in t K.

Had another e q u a lly p o s s ib le com bination been chosen

in which p o in ts M and L were lo c a te d w ith in th e com position t r ia n g le s
ACD and CDE r e s p e c t iv e ly , the two in v a r ia n t r e a c tio n s a t th e s e p o in ts
would have been e u t e c t ic ra th er than p e r i t e c t i c .
N orm ally, such a la r g e number o f p o s s i b i l i t i e s does not g r e a t ly
hamper a n a ly s is o f th e phase diagram when th erm al-com p osition data have
been o b ta in e d .

In the p r e se n t c a s e , how ever, th e in v a r ia n t p o in ts M and

L were v ery c lo s e to each oth er in tem perature and com p osition .

A ls o ,

because o f t h e ir c lo s e proxim ity to p o in t D th ey were n e c e s s a r ily c lo s e
to th e d iv id in g l i n e s DC and DE.

T h is com bination o f circu m sta n ces,

to g eth er w ith the accompanying u n c e r ta in t ie s o f the nature o f th e r e ­
a c tio n s in th e bin ary system s a t p o in ts F and I , made th e in t e r p r e ta tio n
o f the behaviour in t h i s r e g io n ra th er d i f f i c u l t .

In a d d itio n , e x p e r i­

mental e rro rs w ith r e s p e c t to the d eterm in a tio n o f tem perature and
com position approached too c l o s e l y th e m agnitudes o f the d iffe r e n c e s b e ­
tween th e p o in ts M and L .

'< $ •

113

900'

FIGURE 10
DISTORTED VIEW
TERNARY SYSTEM
CaFp- LiF — Be Fp

Ilk

For c l a r i t y th e d is c u s s io n o f th e tern a ry system r e s u l t s i s d iv id e d
in to fou r p a r ts corresponding approxim ately to th e fo llo w in g r eg io n s:
(S ee F igure 10)
P art B) R egion w ith in ADMJ and to a lim ite d e x te n t above l i n e
MJ; i . e . th e L iF -r ic h p o r tio n .
P art C) R egion w ith in BHILKG and to a lim ite d e x te n t above li n e
IL; i . e . th e BeF3- r ic h p o r tio n .
P art D) R egion w ith in FGKLM and th e im m ediately adjacent a rea s;
i . e . th e L i3BeF4 - r ic h p o r tio n .
P art E) The tern a ry system as a w h o le.
The area ly in g in gen era l above a li n e from I to J was n ot i n v e s t i ­
gated fo r s e v e r a l reasons o f which th e fo llo w in g are most p e r tin e n t .
Sin ce th ere was no reason to expect co m p lex ity w ith in the upper p o r tio n
o f th e diagram, and no evid en ce was found fo r tern a ry compound form ation
t h e r e , the liq u id u s su rfa ce may rea so n a b ly be ex tr a p o la te d from th e
b in ary c u r v e s.

The high tem perature in v o lv ed (from about 800° to II4JOO°)

made th e r e g io n o f no i n t e r e s t fo r the o r ig in a l purpose o f the i n v e s t i ­
g a tio n .

The c o s t , as w e ll as th e u n c e r ta in ty o f r e s u lt s due to v o l a t i l i ­

z a tio n and h y d r o ly s is r e a c t io n s , made i t seem unwise to in v e s t ig a t e t h i s
r e g io n .
The th erm al-com p osition data and a n a ly t ic a l r e s u l t s have been
assembled in Appendices 12 and 1 5 .

The thousands o f item s o f data i n ­

volved make i t im p o ssib le to d e sc r ib e th e r e s u lt s o f th e e x c e s s iv e ly la r g e
number o f thermal s tu d ie s excep t as g e n e r a liz a tio n s .

The r e s u lt s are b e s t

fo llo w ed by c o n s id e r a tio n o f the tren d s o b ta in ed as th e tern ary system was
" s e c tio n e d ” along l i n e s , u s u a lly extendin g from the LiF-BeFa s id e toward

115

some p o in t such as th e CaF2 corner o f the diagram.

U su a lly the procedure

o f making s u c c e s s iv e a d d itio n s o f a component was fo llo w e d as a means o f
o b ta in in g b e tte r r e l a t i v e com position v a lu e s .

116

P art B — The L iF-R ich P o r tio n o f th e Ternary System
The g en eral reg io n covered by t h is s e c tio n i s shown in F igure 1 1 , a
p a r t ia l phase diagram o f t h i s corner o f th e ternary system .
tem peratures are shown ad jacen t to the com position p o i n t s .

The liq u id u s
The tempera™

tu re contour l i n e s have been drawn on th e b a s is o f both liq u id u s tempera­
tu r e s and in form ation ob tain ed by c o n sid e r a tio n o f th e tem peratures o f
secondary i n f l e c t i o n p o in t s , or observed changes in th e ch a ra cter o f the
c r y s ta lliz a tio n p r o c e sse s.

The p o in ts may be i d e n t if i e d w ith p a r tic u la r

m e lts l i s t e d in Appendix 12 from the l e t t e r (in d ic a te d on the LiF-BeF2
edge) from which a s e r ie s o f m e lts proceeds out along a s tr a ig h t l i n e
in to th e tern a ry system in th e sequence A - l , A -2, e t c .

The tem perature

v a lu e s along th e LiF-CaF2 and LiF-BeF3 edges have been taken from the
smooth curves drawn fo r the liq u id u s l i n e s o f the two b in a ry sy stem s.
R e fer r in g to F igu re 1 1 , the tem perature contour l i n e s show in g en era l
the behaviour to be exp ected fo r the r e g io n .

They in d ic a te a drop from

the LiF peak toward th e low er tem perature r e g io n s o f the v a ll e y le a d in g
down from the IdF-CaF2 b in ary e u t e c t ic p o in t, i n the d ir e c t io n o f the
compound L i2BeF4 .

On the oth er s id e o f the v a lle y the slo p e clim bs

s te e p ly toward the CaF2 peak.

In th e l e f t h a lf o f t h is r e g io n no d i f f i ­

c u l t i e s were p resen ted and the therm al r e s u lt s were in good agreement
w ith th e exp ected b eh aviou r.

The o n ly p e c u lia r it y i s th e ra th er sh allow

slope o f th e bottom o f th e v a ll e y .
I n th e r ig h t hand p o r tio n o f th e f i g u r e , a t about m elt F - l and sub­
sequent m e lts o f s e r ie s G, H, e t c . , th ere i s l e s s and l e s s tem perature

17

o ct

tn cd

CD U)

118

drop alo n g each s e c t io n and, as w i l l be s e e n , th ere i s su b seq u en tly an
experim ental r i s e in tem perature as the tern a ry system i s e n te r e d .

T his

f l a t r e g io n , exten d in g in from the LiF-BeF2 s id e , e x p la in s the sh a llo w
slop e o f the v a lle y le a d in g down in to i t , s in c e a p p a ren tly the a d d itio n
o f CaF2 i s not e x e r tin g a normal low erin g e f f e c t on the LiF-BeF2 bin ary
system .

The exact lo c a t io n o f t h i s v a lle y becomes d i f f i c u l t because o f

shallow s lo p e s p r io r to the v a lle y and the very steep slo p e on the CaF2
s id e .

For t h is slo p e thermal liq u id u s e f f e c t s were poor and o b serv a tio n

o f i n i t i a l CaF2 c r y s t a lli z a t i o n became n e c essa ry to r e l i a b l y determine
the liq u id u s tem p eratu re.
Subsequent to the H s e r ie s o f m elt an. a d d itio n a l co m p lica tin g fe a tu r e
appeared.

In th a t reg io n o f th e diagram, a p r e c ip ito u s slo p e occu rs d i­

r e c te d toward th e L i2BeF4 r e g io n , corresponding to the ste e p slop e in
the a d ja cen t p art o f the LiF-BeF2 bin ary system .

I n i t i a l d e p o s itio n o f

LiF cou ld th en o n ly be d e te c te d by o b serv a tio n as the therm al e f f e c t s
became poorer and p o o r e r.

T his s te e p drop near the LiF-BeF2 sid e i s

accompanied by a corresponding in c r ea se d slo p e o f the v a lle y from the
LiF-CaF2 bin ary e u t e c t ic w ith i t s s te e p ly slo p in g CaF2 s id e .

Along the

s e c tio n l i n e s th e system changes from r e l a t i v e l y f l a t to a c o n d itio n in
which s e c tio n s show a sm all i n i t i a l r is e o f about 5 to 10° , w hile p a s s ­
in g over a s lig h t promontory exten d in g in the gen eral d ir e c t io n o f the
compound L i2BeF4 .

A fte r p a ssin g over t h i s hump th e therm al s e c tio n s

show a s l i g h t and g r a d u a lly d ecrea sin g f a l l in to the v a lle y from the
LiF-CaF2 e u t e c t i c , so th a t the bottom o f the v a lle y becomes more p o o rly

119

d e fin e d .

The fa r s id e o f the v a l l e y 3 where CaF2 d e p o s its i n i t i a l l y ,

becomes somewhat l e s s ste e p as th e L i2BeF4 reg io n i s approached.
At t h i s p o in t th e v a lle y from the LiF-CaF2 b in a ry e u te c tic ( o r ig ­
i n a l l y about 20 mole p ercen t CaF2) has extended to a p o in t in c lo s e
p roxim ity to th e LiF-BeF2 s i d e , th a t i s o n ly about 5 mole p ercen t CaF2
a t about 30 mole p ercen t BeFa '

The balance o f t h i s r e g io n i s more ap­

p r o p r ia te ly d isc u sse d in P art D on a la r g e r com position s c a l e .

The m e lts

ly in g alon g l i n e s i d e n t i f i e d by ,feM and 111" were handled in a somewhat
d if f e r e n t manner, b ein g s h if t e d toward the p o in ts L iF , CaF2 or L i2BeF4 as
in d ic a te d by the path o f t h e ir guide l i n e s .
Summarizing th e r e s u l t s in t h i s s e c tio n o f the diagram, i t may be
sa id th a t th e behaviour i s th a t which might be g e n e r a lly exp ected from a
c o n sid e r a tio n o f the b in ary sy ste m s.

The r e l a t i v e l y sm all low erin g

e f f e c t o f CaF2 on the LiF extends out in to the system and becomes pro­
g r e s s iv e ly p o o rer.

The moderate low erin g e f f e c t o f the L i2BeF4 on th e

LiF exten d s w ith in th e tern ary system u n t i l a c r i t i c a l BeF2 co n ten t i s
reached a t about 25 mole p e r c e n t, f o llo id n g which the low erin g e f f e c t o f
the L i2BeF4 becomes much more d r a s t i c .
W ith r e s p e c t to observed s o lid p h a se s, L iF , CaF2 and L i2BeF4 were found
in s ig n if i c a n t amounts.

Sm all tr a c e s o f BeF2 in crea sed in the r ig h t hand

p o r tio n as a r e s u lt o f an incom plete phase r e a c tio n in the v i c i n i t y o f
L i2BeF4 .

A ll m e lts in t h i s reg io n seemed to be co m p letely s o l i d i f i e d

p r io r to or on reach in g an in v a r ia n t r e a c tio n a t about kSk° as more f u l l y
d isc u sse d in P art D.

Thermal r e s u lt s o c c a s io n a lly in d ic a tin g to the

contrary were probably due m ain ly to a r r iv a l a t the in v a r ia n t p o in t w ith

120

in s u f f i c i e n t liq u id to produce a good p la te a u , or to s l i g h t l y incom plete
con version o f LiF to L i2BeF4 .
In g en eral th e tem perature contours may .be drawn w ith out too much
o m is s io n o f d isco rd a n t p o i n t s .

Beyond th e tern ary e u te c tic v a lle y

(d escen d in g from the CaF2-LLF e u t e c t ic ) the s it u a t io n becomes poorer as
com position u n c e r ta in ty and m elt degrad ation in c r ea se w ith the r a p id ly
r is in g tem perature l e v e l s o f th a t r e g io n .

121

P art C — The BeF2-R ich P o r tio n of* the Ternary System

As p r e v io u s ly m entioned, the p o r tio n o f th e system w ith very high
BeF2 c o n ten t s o l i d i f i e s in g la s s y form w ith no d e f in it e liq u id u s
temperature a t reason ab ly low c o o lin g r a t e s .

C onsequently the r e g io n o f

Figure 12 c o n ta in in g the broken contour l i n e s i s on ly a rough e stim a te
which may be e n t ir e ly w ith out r e a l i t y , the contou rs b ein g in s e r te d
p rim a rily fo r c l a r i t y in v is u a liz in g the phase diagram.
The same gen eral '•sectio n in g '1 procedure was fo llo w e d on most o f the
m e lts in t h i s reg io n and the p o in ts are id e n t if ia b le in a s im ila r manner
as were th o se in th e p reced in g s e c t io n .

The tem peratures along the

BeF3-CaF2 s id e are taken from the liq u id u s curve o f th a t bin ary system .
As surging th e presence o f a v a lle y le a d in g down from an assumed
b in ary e u t e c t ic in the BeF2-CaF2 system ( a t about 10 mole p ercen t CaF2) ,
n e ith e r o f which phenomena were d e f i n i t e l y observed, i t may be s a id th a t
no good liq u id u s p o in ts were determ ined between t h i s v a lle y and the
LiF-BeF2 sid e o f the diagram .

On the oth er sid e o f the v a l l e y , and

d e f i n i t e l y w ith in about 5 mole p e r ce n t from i t , CaBeF4 was i n i t i a l l y de­
p o s ite d w ith apparent therm al e f f e c t s .

T h e r e a fte r , w ith in c r e a se in

CaF2 c o n te n t, very stron g therm al e f f e c t s were ob tain ed upon CaBeF4
d e p o s itio n over most o f th e r e g io n , p r im a r ily because o f the r e l a t i v e l y
shallow slo p e o f the space su rfa ce r ep r e se n tin g sa tu r a tio n w ith t h is
component.

A corresponding shallow slo p e i s p r e se n t in the b in a ry system

fo r the liq u id u s curve o f t h i s compound.

When th e space su rfa ce o f CaF2

d e p o s itio n was encountered ( a t a v a lle y le a d in g down from th e CaF2-CaBeF4

122

650

,0T1

6oo
775

750

Vr

71S
700

O
ro

Larger
F I GURE

scale

copy

of

this

IN REAR POCKET*

123

e u t e c t i c ) , t h is fa v o ra b le thermal e f f e c t va n ish ed .

Beyond t h is v a lle y ,

i n i t i a l d e p o s itio n o f CaF3 had to be d e te c te d v i s u a l l y .
R e fer r in g to F igu re 1 2 , co v erin g t h is s e c tio n o f the ternary diagram,
the r e s u l t s may be g e n e ra lize d b r i e f l y as f o ll o w s , d e t a ile d r e s u lt s b ein g
given in Appendix 1 2 .
S e c tio n s p a ssin g in to the tern a ry system near the BeF2-CaF2 sid e
showed a g ra d u a lly r is i n g s lo p e , w ith s l i g h t evidence o f p a ssin g over a
sm all hump exten d in g out from the v i c i n i t y of th e compound Cai3eF4 .
T h e r e a fte r , th ey f e l l very s l i g h t l y to a v a lle y le a d in g down from the
CaF2-CaBeF4 e u t e c t ic a t approxim ately 51 mole p ercen t CaF2 .

Beyond t h i s

v a lle y , th e slo p e o f i n i t i a l CaF2 d e p o s itio n r o se very s te e p ly and much
more d iscord an t v a lu e s were o b ta in ed as a r e s u lt o f the h ig h tem perature
degradation o f the m e lt s .
S e c tio n s more removed from the BeF2-CaF2s id e , such as the Y, X and
W s e r i e s , showed th e same i n i t i a l slow r i s e o f tem perature fo r CaBeF4
d e p o s itio n w ith stron g therm al e f f e c t s .

They d id not however show

in d ic a tio n s o f p a ssin g over a sm all r i s e p r io r to rea ch in g the v a lle y
le a d in g down from the CaF2-CaBeF4 e u t e c t ic .

A very s l i g h t v a lle y i s

in d ic a te d by th e contour arrangement (s in c e i t i s t h e o r e t ic a lly n e c e s s a r y ),
but th e in t e r s e c t io n s o f contour l i n e s a t the v a lle y appear to become
in c r e a s in g ly o b liq u e .

T h e r e a fte r , the v a lle y on ly made i t s e l f apparent

as a stron g therm al e f f e c t fo llo w in g i n i t i a l CaF2 d e p o s itio n .
For th e se m e lts (lo c a te d reason ab ly d is t a n t from th e BeF2-CaF2 sid e
o f the system ) i t became ex p erim en ta lly p o s s ib le to fo llo w th e s o l i d i f i c a ­
tio n p r o c e ss to com pletion in order to determ ine the u ltim a te d e s tin a t io n

1214

of s o lid if y in g m ix tu res.

Of th e m e lts which were th us c o o le d , i t was

found th a t one or th e oth er o f two d e s tin a tio n s were reached.
th ese was th e lo w e st tem perature p o in t o f the system .

One o f

T his was the

tern ary e u t e c t ic p o in t (ap proxim ately 350° ) in the near v i c i n i t y o f the
binary L i2BeF4-BeF2 e u t e c t ic (about 52 mole p ercen t BeF2) .

The oth er

a lte r n a tiv e d e s tin a t io n was found to be a t a h igh er tem perature near
I4500 a t a lo c a t io n c lo s e r to the L i1BeF4 reg io n as fu rth er d isc u sse d in
P art D .
S ig n if ic a n t ly , th e f i r s t d e s tin a tio n a t th e lower temperature was
shown by m e lts i n i t i a l l y d e p o s itin g CaBeF4 and a lso by th o se very c lo s e
to the LiF-BeF2 s id e which had no d e f in it e liq u id u s p o in t .

The second

(h igh er tem perature) d e s tin a tio n was shown by th ose m e lts i n i t i a l l y de­
p o s itin g CaF2 , i . e . th ose beyond the e u t e c t ic v a lle y r e p re sen tin g the
tw o -fo ld sa tu r a tio n l i n e fo r CaBeF4 and CaF2 .

T his behaviour and th e

liq u id u s thermal e f f e c t s were used in e stim a tin g the lo c a t io n o f the
v a lle y .
Between th ese two extrem es o f b eh aviou r, th ere was a t h ir d c la s s
rep resen ted by m e lts which had com p osition s somewhat below the CaF2
con ten t o f th e e u t e c t ic CaBeF4-CaF2 v a l l e y .

Such m elts e x h ib ite d a

second i n f l e c t i o n a t tem peratures in term ed ia te between 35° ° and US0° .
E ven tu a lly th ey reached th e 350° d e s tin a tio n w ith apparent therm al e f f e c t ,
u n le ss too sm all an amount o f liq u id rem ained.

These two i n f l e c t i o n s ,

due to i n i t i a l CaBeF4 d e p o s itio n and subsequent in t e r s e c t io n w ith a
v a lle y r e p r e se n tin g a tw o -fo ld sa tu r a tio n lin e f o r CaBeF4 and L i2BeF4 ,
w i l l be fu r th e r d isc u sse d in P art D .

125

For m elts s t i l l fa r th e r from the BeF2-CaF2 sid e o f th e system , such
as th e V s e r i e s , th e slo p e r e p r e se n tin g CaBeF4 d e p o s itio n was i n i t i a l l y
ste e p e r and in d ic a te d a rapid d e c lin e toward the tern ary e u te c tic
lo c a te d near the 52 mole p ercen t BeF2 r e g io n .

This i n i t i a l l y steep slo p e

fo r CaBeF4 d e p o s itio n in crea sed in abruptness fo r the r eg io n s to th e l e f t
o f t h i s s e r ie s so th a t therm al e f f e c t s were not ob tain ed and s p e c ia l
technique was n e c essa ry in d e te c tin g CaBeF4 d e p o sitio n as fu rth er d isc u sse d
in P art D.

Thermal e f f e c t s were obtained o n ly a ft e r the i n i t i a l slo p e

had undergone an in f l e c t i o n to form a sh o rt slo p in g p la tea u below the
e u te c tic CaBeF4 -CaF2 ’’v a lle y ” , which extends toward th e L i2BeF4 r e g io n .
In the r e g io n under d is c u s s io n the ttv a lle y ” has extended to about the
com position p o in t o f 20 mole p e r ce n t CaF3 and I4O mole p ercen t BeF2 .
One s e r ie s o f m e lts extends a c ro ss t h is r e g io n along a s tr a ig h t l i n e
from p o in t g-1 (near 55 mole p ercen t CaF2 on the BeF2-CaF2 s id e ) d ir e c te d
toward th e I4.8 mole p e r ce n t BeF2 p o in t o f the LiF-BeF2 s id e .

These p o in ts

were o b ta in ed d if f e r e n t ly from the p r e v io u sly mentioned m elt s e r ie s in
th a t th ey r ep resen t a 11s e c tio n ” surveyed by a d d itio n o f m a ter ia l o f th e
la t t e r com position to o r ig in a l m a te r ia l o f the com position g -1 .

The

general r e s u l t s were in agreement w ith th ose o f the oth er s e c tio n s pre­
pared by a d d itio n o f CaF2 to m elts o f the LiF-BeF3 sid e o f th e system .
Summarizing th e behaviour o f t h i s r eg io n o f the tern a ry sy stem , a
li n e o f tw o -fo ld s a tu r a tio n (r e p r e se n tin g a e u te c tic r e a c tio n form ing CaF2
and CaBeF4) exten d s down from th e corresponding b in a ry e u te c tic p o in t a t
about 51 mole p e r ce n t CaF2 .

Beyond t h i s gra d u a lly slo p in g v a l l e y , a

s te e p ly ascend ing slo p e i s found extendin g a cro ss th e r eg io n (th e surface

126

o f s a tu r a tio n w ith r e s p e c t to CaF2) , in agreement w ith th e s im ila r slo p e
found beyond the LiF-CaF2 e u t e c t ic l in e in P art B.
On the low er s id e o f th e v a l l e y , a broad (g r a d u a lly slo p in g ) promontory
extends inward from the CaBeF4 com position p o in t .

T his a grees w ith the

co n sid era b ly d is s o c ia t e d ch aracter in d ic a te d fo r t h i s compound by the
“r o u n d e d -o ff1 peak which i t shows in th e binary BeF2-CaF2 sy stem .

The

o v e r - a ll e f f e c t i s one produced by a very in c o n se q u e n tia l lo w erin g by
CaF2 o f the'CaBeF4 liq u id u s and, by c o n tr a s t, a f a i r l y stron g low erin g
e f f e c t o f the CaBeF4 upon the CaF2 liq u id u s su r fa c e .
In the r e g io n nearer the compound Li^eF,^ th er e

i s anin d ic a tio n

th a t

su b s ta n tia l low erin g e f f e c t o f the L i2BeF4 upon th e CaBeF4 i s obtained
o n ly a f t e r a f a i r l y c r i t i c a l com position o f the former i s reach ed , fo llo w ­
in g which th e e f f e c t iv e low ering i s v ery a b ru p t.
With r e s p e c t to the BeF2-r ic h p o r tio n adjacent to the LiF-BeF2 s id e
o f the system no r e s u lt s o f s ig n ific a n c e were o b ta in ed .

The low erin g

produced a t the tern a ry e u t e c t ic p o in t, as compared w ith the ad jacen t
L i2BeF4 -BeF2 b in a ry e u te c tic p o in t , was v ery sm all and i s d isc u sse d in
P art D.
In g e n e r a l, throughout the r e g io n covered by t h is s e c t io n th e therm al
r e s u lt s were in reason ab le agreement w ith the in d ic a te d tem perature con­
tour l i n e s .

Some e x c e p tio n s occur p a r t ic u la r ly in the high temperature

region above the CaBeF4 -€aF3 e u t e c t ic v a lle y , p r in c ip a lly because o f
com position u n c e r ta in ty and contam ination by h y d r o ly sis p r o d u cts.

Some

im agination was n e c e s s a r ily employed in e stim a tin g the contour shapes and
lo c a t io n , as t h e ir f in e str u c tu r e cou ld not be ob tain ed ex cep t by an ex ­
c e s s iv e ly la r g e number o f experim ental p o in t s .

127

P art D — The L i2BeF4-R ich P o r tio n o f th e Ternary System

The c e n tr a l p o r tio n o f the tern ary diagram along the LiF-BeF2 sid e
(from 30 to 50 mole p ercen t BeF3) , exten d in g in to the system to a con­
c e n tr a tio n o f about 1$ mole p ercen t CaF2 , w i l l be d isc u sse d in t h i s
s e c t io n .

The p r e v io u s ly d escrib ed p o r tio n s o f th e tern a ry system were

th ose ly in g roughly w ith in th e ranges o f 0 to J>0 mole p e r ce n t BeF2 and
50 to 100 mole p ercen t BeF2 .

In the r e g io n from about 30 to $0 mole p e r ­

cen t BeF2 , th e major experim ental and in t e r p r e tiv e d i f f i c u l t i e s were
encountered.
Because o f the la r g e number o f m e lts o f c l o s e l y spaced com position s
which were prepared in t h i s r e g io n , i t has been n e c essa ry to p r e se n t th e
r e s u lt s on an enlarged s c a l e . W hile the s c a le does not g iv e a f a l s e im­
p r e ss io n o f th e accuracy o f gravim etric p rep aration o f th e m e lts , the
p o s s ib le a b so lu te com position err o r s are much la r g e r than th e s c a le o f
F igure 13 m ight i n d ic a t e .

Because o f the fo rm u la tio n o f th e m e lts from

g e n e r a lly th e same m a te r ia ls in s im ila r manners, th e r e la t iv e p o s it io n s
are t h e o r e t ic a lly known to an accuracy somewhat c lo s e r to th e s c a le o f the
diagram.

The d istu r b in g fa c to r s such as creepage l o s s e s , s l i g h t v o l a t i l i ­

z a tio n and h y d r o ly sis in tr o d u c e , however, much la r g e r u n c e r ta in t ie s o f
probably a t l e a s t ■§ 1 /h mole p ercen t under the b e s t o f circu m sta n ces.
Such u n c e r ta in ty must be kept in mind fo r c lo s e ly spaced m e lts and the r e ­
s u lt s must be in te r p r e te d in terms o f tren d s d e s p ite th e p resen ce o f
o c c a sio n a l d iscord an t v a lu e s .

128

W ith r e s p e c t to tem perature v a lu e s , a somewhat analogous s it u a t io n
e x i s t s , p r im a r ily because o f th e very sm all d iffe r e n c e s (such as 5 to 10°)
in tem perature l e v e l s between s ig n if i c a n t p o in t s .
appear as a d isco rd a n t value .

Here a 1° error w i l l

I t w i l l be a p p recia ted th a t even w ith good

temperature d e t e c t io n , o c c a s io n a l un d ercoolin g or f a i l u r e to d e te c t f i r s t
c r y s t a lli z a t i o n may r e a d ily c r e a te erro rs o f t h is magnitude or g r e a te r .
For la r g e ly s o l i d i f i e d m e lt s , tem peratures r e p re sen tin g the ends o f c e r ta in
therm al e f f e c t s are co n sid e ra b ly more in error because o f th e much poorer
tem perature eq u ilib riu m w ith in the sample c o n ta in e r .
F i n a l l y , and o f even more im portance, th ere i s the discord ance which
may be in trod u ced by any in com p leten ess o f th e phase r e a c t io n s .

P e r ite c tic

r e a c t io n s , o f which two are p r e se n t in t h i s r e g io n , are n o to r io u s ly bad
o ffen d e rs in t h i s r e s p e c t.

C onseq uently, should any e f f e c t iv e i s o l a t i o n

o f a s o lid component occur during s o l i d i f i c a t i o n , such th a t i t i s not in
eq u ilib riu m w ith the m e lt, the m elt w i l l n e c e s s a r ily behave l i k e one o f
d if f e r e n t o v e r - a ll com p osition .

A ll m e lts were mixed as e f f e c t i v e l y as

p o s s ib le u n t i l th e s o l i d i f i c a t i o n p reven ted fu rth e r s t i r r i n g .

As p r e v io u s ly

m entioned, co n sid era b le q u a n tit ie s o f non-apparent liq u id (up to 2$%) might
s t i l l remain as judged by th e continuance o f the therm al e f f e c t .
A ll th e se e f f e c t s must be co n sid ered to g e th er w ith v is u a l ob serv a ­
tio n o f th e ch aracter o f the c r y s t a l l i z a t i o n .

The data have been tab u ­

la te d in Appendix 12 in a way th a t b e s t perm its c o n sid e r a tio n from the
stand point o f the changes which occur throughout a p a r tic u la r s e c t io n .
A great many m e lts must be c o n sid e r e d , o f which about one hundred were
concen trated in the immediate v i c i n i t y o f th e I i 2BeF4 p o in t in an attem pt

129

to ob tain a d e t a ile d p ic tu r e o f the therm al contour c o m p lica tio n s in
t h is r e g io n .
Because o f th e p e c u l i a r i t i e s o f the system in t h i s r e g io n , even t h is
super-abundance o f data does n o t perm it a p u re ly o b je c tiv e c o n c lu s io n .
C onsiderable judgment s t i l l was n ecessa ry to choose among the numerous
p o s s ib le ex p la n a tio n s o f the d a ta , in the l i g h t o f the p r e v io u sly men­
tio n ed u n c e r ta in t ie s as to ex a ct com position s and tem p eratures.
In F igu re 13 a l l r e s u lt s o b ta in ed in t h i s reg io n have been in d ic a te d
by a com p osition p o in t and an a s s o c ia te d temperature value rep re sen tin g
the observed liq u id u s tem perature.
in t h is r e g io n .

S ev era l ty p es o f " s e c tio n s n were made

Some were s h if t e d from the LiF-BeF2 sid e in the d ir e c t io n

o f the CaF3 corner o f the system .

Others were d ir e c te d from the v i c i n i t y

o f the L i2BeF4 com position p o in t toward e ith e r the CaBeF4 com position p o in t
or another p o in t a t approxim ately the lo c a tio n o f the b in ary LiF-BeF2
e u t e c t ic .

L igh t guide l i n e s are in d ic a te d , exten d in g out from the i d e n t i ­

f ie d i n i t i a l m elts on th e LiF-BeF2 s id e , to in d ic a te the path o f each
s e c t io n .

The few ex ce p tio n s to t h i s g en era l proced ure, s e r ie s ’’e" , " t" ,

'fgf and s , were prepared and c a r r ie d alon g oth er s e c tio n s in a manner
which i s apparent from a n .exam ination o f the d ir e c t io n fo llo w e d by t h e ir
r e s p e c tiv e guide l i n e s .

The id e n t if ic a t i o n o f a p a r tic u la r com position

p o in t may be made by r efer en ce to the o r ig in o f i t s guide l i n e , the
numbering f o r example fo llo w in g the sequence :S, S - l , S -2 , e t c . , or 12Q,
12Q-1, 12Q-2, e t c .
In th e d is c u s s io n o f P a rt B i t was noted th a t s ta r tin g a t about
the 30 mole p e r ce n t BeF2 p o in t , a p e r c e p tib le and g ra d u a lly in c r e a s in g

130

O

ro uJ

w
•H
+3 •
+3
!+h ©
O^
o
t’s O
O* Q,
O
O Ps

cd

© ©
rH S4

©
OC
to .h
©

© p-t
bOp
Jm 5 b
aj *rt
i-d

131

i n i t i a l r is e was found as a b in a ry LiF-BeF3 m elt was c a rr ie d out in to
the tern ary system by a d d itio n o f CaF3 .

On such s e c t io n s the sm all i n i t i a l

r i s e was s h o r tly fo llo w e d by a sm all tem perature drop as th e CaF3 con ten t
was in c r e a s e d .

A ll s e c tio n s o r ig in a tin g in the reg io n o f 30 to about

3 2 .5 mole p ercen t BeF3 showed t h is same b eh aviou r, in d ic a tin g th a t the
s e c tio n s were c u tt in g through a sm all promontory exten d in g down toward
the L i3BeF4 p o in t .

The r e la t iv e tem perature l e v e l s o f the s e c tio n s dropped

sh arp ly as the BeF3 con ten t in c r e a s e d .

A cco rd in g ly , throughout t h i s

r e g io n i n i t i a l LiF d e p o s itio n had to be d e te c ted v i s u a l ly fo r la c k o f
therm al e f f e c t s .

The r e l i a b i l i t y o f such a d e te c tio n was probably very

good in t h is r e g io n , and b e tt e r than would be in d ic a te d by some apparent
d isco rd a n t v a lu e s .
Subsequent to th e sm all r is e and f a l l mentioned fo r m e lts c a rried
in to the tern ary system (from the reg io n o f

to 32 mole p ercen t BeF3) ,

a r is e in the liq u id u s temperature was encountered a t c o n c en tr a tio n s o f
approxim ately 3 to h mole p ercen t CaF3 .

Here th ere was a sm all reg io n

where i t was d i f f i c u l t to decide the id e n t if y o f the i n i t i a l l y d ep o sited
s o lid by v is u a l methods, i . e . whether LiF or CaF3 .
d e f i n i t e l y d ep o sited f i r s t .

Up to t h i s reg io n LiF

Beyond t h is reg io n CaF2 was observed as th e

i n i t i a l l y d e p o site d s o l i d , the liq u id u s tem perature r is in g rath er s te e p ly
fo r th ese m e lt s .

The r e g io n o f u n c e r ta in ty corresponds rou gh ly (1 0 .5

mole p ercen t) w ith the path o f the tw o -fo ld s a tu r a tio n lin e fo r LiF-CaF3
d e p o s itio n .

In order to b rin g out t h i s o b serv a tio n the p o in ts o f the

fig u r e have been shown as s o lid c i r c l e s or squares corresponding to the
d e p o sitio n o f LiF or CaF3 r e s p e c t iv e ly as the i n i t i a l s o l i d .

132

The m elts s t a r t in g from the reg io n o f 3 2 .5 to 3 3 .5 mole p e r ce n t
BeF2 on th e LlF-BeF2 sid e lik e w is e showed LiF as th e i n i t i a l s o l i d .

A lso

th ey tended to show l e s s and l e s s i n i t i a l r i s e , u n t il by 3 3 .5 mole p e r ­
cent th e liq u id u s slo p e ex ten d in g in to the tern a ry system was very s l i g h t .
On e x te n sio n to c o n c en tr a tio n s g r ea ter than 2 .5 to 3 mole p ercen t CaF2 ,
th ere was a m od erately slo p in g r i s e in the liq u id u s temperature accompanied
by i n i t i a l d e p o s itio n o f CaF2 .
The above d is c u s s io n r e f e r s to b in a ry m e lts to th e l e f t o f th e Li^eF.^
p e r i t e c t i c p o in t which were s h if t e d in to th e ternary system in th e d i ­
r e c tio n o f th e CaBeF4 com position p o in t.

To the r ig h t o f the L i2BeF4

p e r i t e c t i c p o in t , two m e lts were c a r r ie d along s e c tio n s d ir e c te d toward
the CaBeF4 p o in t .

The fo llo w in g comments r e fe r to th e s e m e lts o f the

12Q- and 13Q- s e r i e s .

The i n i t i a l l y d e p o site d s o li d was L i2BeF4 ( t r i ­

angular symbols) and the liq u id u s tem perature f e l l v ery s l i g h t l y u n t i l the
CaF2 co n c en tr a tio n reached about 3 mole p e r c e n t.

The liq u id u s temperature

then r o se q u ite gra d u a lly accompanied by a s h i f t to CaF2 as th e i n i t i a l l y
d ep o sited s o l i d .

The t r a n s it io n r e g io n corresponds to the tw o -fo ld

sa tu r a tio n l i n e ML which con n ects th e two tern a ry in v a r ia n t p o in ts o f
Figure 10.
Most o f th e m e lts to th e r ig h t o f the L i2BeF4 p e r it e c t i c p o in t ( i . e .
th ose w ith 3 3 .5 to 37 mole p ercen t BeF2) x^ere s h ift e d along s e c tio n s d i­
r ec te d toward the v i c i n i t y o f th e CaF2-LiF b in ary e u t e c t ic p o in t .

The

lo c a tio n o f th e se s e c tio n s was such th a t p r im a r ily th ey were o n ly u s e fu l
in e s t a b lis h in g th e approximate lo c a tio n o f th e tw o -fo ld s a tu r a tio n l i n e
extendin g in from the L i2BeF4 p e r i t e c t i c p o in t .

Thus, as may be seen

133

from F igure 1 3 , a l l th e se m e lts showed i n i t i a l d e p o s itio n o f L i2BeF4
(in d ic a te d Toy a tr ia n g u la r sym bol), w ith o n ly a s l i g h t upward trend o f
the liq u id u s tem perature as the com position was s h if t e d upward toward
th e l e f t .

Upon p a ssin g the tw o -fo ld s a tu r a tio n l i n e , th ere was a s h i f t

to i n i t i a l d e p o s itio n o f LiF accompanied by a r a th er rap id in c r e a se in
the liq u id u s tem perature, in reasonab le agreement w ith the p reviou s
p o in ts o b ta in e d in th a t a r e a .
From th e r e g io n o f 37 to $0 mole p ercen t BeF2 a number o f s e r ie s o f
m e lts were s h if t e d in to the tern ary system along s e c tio n s d ir e c te d toward
the CaF2 v e r te x o f the diagram.

Most o f th e se m e lts showed normal therm al

e f f e c t s , corresponding to th e i n i t i a l d e p o sitio n o f CaBeF4 (open c i r c l e
sym b ols), o n ly a f t e r a minimum co n c en tr a tio n o f about 3 to k mole percent
CaF2 had been reached.

T h erea fter the thermal r e s u lt s showed the presence

o f a gra d u a lly slo p in g p la te a u .

The l a t t e r rep resen ted th e co n tin u a tio n

o f th e p la te a u f a l l i n g from the v i c i n i t y o f the CaBeF4 p o in t as d escrib ed
in F art C o f t h i s s e c t io n .
Nearer the LiF-BeF2 s id e , however, th er e i s a p p a ren tly a very ste e p
s lo p e , r e p r e se n tin g th e drop from the above p la tea u to th e tw o -fo ld
s a tu r a tio n l i n e le a d in g toward th e tern a ry e u te c tic p o in t .

T his drop,

amounting to as much as 1^0° in some p la c e s , occurs in a very narrow
r eg io n lo c a te d betw een 0 and 3 mole p ercen t CaF2 up from th e LiF-BeF2
sid e o f th e diagram.

In t h is r e g io n , the i n i t i a l d e p o s itio n o f CaBeF4

exerted no d e te c ta b le therm al e f f e c t , as o n ly a very sm all amount o f
s o lid formed in dropping through a la rg e temperature r a n g e .

T his

13k

corresponded to th e behaviour in th ree o th er p a r ts o f the tern a ry diagram
where very ste e p slo p e s were encountered.
The s it u a t io n in t h i s r e g io n was w orse, however, s in c e CaBeF4 de­
p o s ite d w ith ou t any su rface in d ic a tio n th a t c r y s t a l l i z a t i o n was o c cu rr in g .
S o l i d i f i c a t i o n occurred as a gradual accum ulation o f n eed le shaped c r y s t a ls ,
c h a r a c t e r is t ic a lly v ery t in y compared to th e u su a l n eed le and p r ism a tic
h a b its o f L i^ eF ,^ .

O c c a sio n a lly the s o l i d i f i c a t i o n was d e te c ta b le by a

sudden g r i t t y f e e l i n g produced as the hand operated s t i r r e r passed over
the bottom o f the platinum sample c o n ta in e r .
For la c k o f a b e tt e r method the fo llo w in g v a r ia tio n o f quenching
technique was u se d , w ith o n ly p a r t ia l s u c c e s s , in an attem pt to d e te c t
i n i t i a l CaBeF4 d e p o s itio n in t h is r e g io n .

By p lu n gin g a c o ld n ic k e l

s t ir r in g rod ( l / 8 n diam .) to the bottom o f the m elt and in s t a n t ly w ith ­
drawing i t , i t was p o s s ib le to withdraw a r a p id ly c h ille d sample from
which th e CaBeF4 p r o je c te d in the fo r m o fa fu z z o f t in y n e e d le s .

The

n e e d le s , u s u a lly about the s iz e o f human h air and from 1/8" to 1/k" lo n g ,
were i d e n t i f i e d by X-ray m ethods.

W hile in some cases the method was

shown to be rep rod u cib le to w ith in 5° as a method o f d e t e c t io n , in oth er
c a se s i t gave o n ly very d isco rd a n t r e s u l t s .

T his might not have been

th e f a u lt o f the method, but rath er a s h i f t in the p o in t o f c r y s t a l l i z a t i o n
to the upper, ou ter edge o f the m elt w ith subsequent adherence to the
platinum c o n ta in e r .

Such a p r e f e r e n t ia l tendency fo r CaBeF4 to form

adherent c r y s t a ls on the platinum m etal d ish was noted in some c a s e s .
Two very d e t a ile d m elt s e r ie s were run in t h is reg io n ( s e r i e s LLand MM-), a t v ery sm all com position in t e r v a ls ( 0 .1 to 0 .2 mole p ercen t CaF2)

135

in an attem pt to c h a r a c te r iz e t h i s ste ep s lo p e , making use o f the
"quenching" te c h n iq u e .

The r e s u lt s were o n ly p a r t i a l l y s u c c e s s fu l in

th a t th ey in d ic a te d the probable lo c a t io n o f the tw o -fo ld s a tu r a tio n
l i n e along th e base o f th e s lo p e , but in no way fu rn ish ed d e t a il s o f the
liq u id u s li n e u n t i l the l a t t e r decreased in slo p e a t h igh er temperature
le v e ls .
Up to t h i s p o in t in P art D, th e d is c u s sio n has been e x c lu s iv e ly con­
fin e d to th e r e s u lt s ob tain ed fo r liq u id u s p o in ts throughout the gen eral
r e g io n o f 30 to $0 mole p e r ce n t BeF2 near the LiF-BeF2 sid e o f th e d ia ­
gram.

I f a l l o f th e se tem perature and com position v a lu e s x^ere p e r f e c t l y

accurate and i f th e system were p e r f e c t ly behaved, th ere would probably
be no need fo r e ith e r a d d itio n a l in form ation o r d is c u s s io n .

C onsidera­

t io n o f th e data obtained and e stim a tio n o f the accuracy (w ith r e la t io n
to the p e c u lia r requirem ents o f the system.) in d ic a te th a t co n sid era b le
supp orting evid en ce i s req u ired to j u s t i f y the proposed phase diagram.
Two a d d itio n a l sources o f in form ation are the X -ray exam inations o f
the s o l i d i f i e d m elt samples and th e supplem entary therm al inform ation
obtained i n con ju n ction w ith the liq u id u s v a lu e s .
The X-ray exam ination o f m elt sam ples from t h i s reg io n i s summarized
and d isc u sse d f u l l y in Appendix 1 3 .

The agreement i s e x c e lle n t w ith

th e proposed phase diagram beh aviou r, co n sid e rin g th e f a c t th a t p e r it e c t ic
r e a c tio n s are in v o lv e d .

D e ta ile d c o n sid e r a tio n o f the phases t h e o r e t ic a ll y

p r e d ic ta b le fo r each o f th e m elts (based on th e lo c a tio n o f t h e ir compo­
s i t i o n w ith in c e r ta in com position t r ia n g le s ) i s in good agreement w ith
th ose a c t u a lly found by X -ray a n a ly s is .

In f i v e c a s e s , the "forbidden"

136

s it u a t io n o f fo u r c o e x is tin g s o l i d phases was found, and th e se p a r tic u ­
la r sam ples are noted to be ones ly in g r ig h t on the b o r d e r lin e s between
two p o s s ib le modes o f b eh aviou r.

In a d d itio n to th e s e c a se s th er e i s

e v id e n c e , throughout a l l th e r e s u l t s , o f a s lig h t degree o f incom p leten ess
o f the p e r it e c t ic r e a c t io n s , s in c e a t l e a s t tr a c e s o f b ery lliu m flu o r id e
were always p r e se n t i n the sam p les.

The X-ray data th e r e fo r e are in

accord w ith th e proposed phase diagram, b u t th ey show th a t the system
does not always behave p e r f e c t l y , a c o n c lu sio n which i s p a r t ic u la r ly
e v id e n t from th e secondary therm al d a ta .
W hile liq u id u s data are e x p ressa b le by one or two p ie c e s o f inform a­
t io n per p o in t , th e secondary thermal data are extrem ely voluminous and
a l s o , in t h i s c a s e , very d e c e p tiv e .

In ta b u la tin g th e m a ter ia l in

Appendix 1 2 , o n ly the more im portant fe a tu r e s o f th e thermal behaviour
could be shown.

Such data as the p o in ts o f f i r s t , second and th ir d

i n f l e c t i o n s , th e ty p es o f s o lid s form ed, and a g en era l d e s c r ip tio n o f
therm al p la te a u s have been in c lu d e d .

The l a t t e r v a lu e s , how ever, are so

dependent upon the p a r tic u la r c o n d itio n s th a t e x i s t fo r each m elt th a t
th ey should o n ly be used fo r comparison w ith very adjacent m e lts o f the
same s e r i e s .

The amount o f s o lid s p r e se n t p r io r to a p la te a u , th e

amount o f liq u id p r esen t a t the end, e t c . , a l l a f f e c t the shape and
tem perature d e s c r ip t io n s .

Worst o f a l l , in the v i c i n i t y o f th e L i2BeF4

p o in t, th ere i s a c l o s e l y spaced com bination o f a ternary p e r i t e c t i c l i n e ,
th ree tern ary e u t e c t ic l i n e s and two tern a ry p e r i t e c t i c p o in t s , in a
narrow range o f tem perature and co m p o sitio n .

137

M ention i s a ls o made o f the phenomenon (e x p la in e d in d e t a i l a t the
end o f Appendix 5) in w hich circum stances perm it a m elt to depart from
a tern ary p e r i t e c t i c l i n e when a b iv a r ia n t c o n d itio n i s c r e a te d .

Under

th e se circum stances the tern a ry p e r i t e c t i c i s not reached and th e m elt
proceeds a c r o ss a b iv a r ia n t space su rfa ce u n t il another tw o -fo ld sa tu ra ­
t io n lin e i s encountered.
T his com bination o f p o s s i b i l i t i e s may r e s u lt in freq u en t t r a n s it io n
from one type o f phase r e a c tio n to another w ithout therm al i n f l e c t i o n .
This b len d in g to g e th e r o f u n iv a ria n t and in v a r ia n t p r o c e sse s c o n s titu te d
the g r e a te s t o b s ta c le to the in t e r p r e ta tio n o f the b eh a v io u r, a sid e
from th e phenomena due to ••forbidden” ty p es o f phase behaviour o ccu rrin g
in t h is same r e g io n .
For convenience in th e fo llo w in g d is c u s sio n Figure li; i s used as
i t s exaggerated p rop ortion s make i t p o s s ib le to d e scr ib e the secondary
thermal behaviour in g en eral terms more r e a d ily than do the d e ta ile d
phase diagram s.

The fig u r e i s the same as F igure 10 except th a t the b in ary

system s have been o m itte d .
In P art C, th e gen eral s o l i d i f i c a t i o n behaviour o f m e lts in the
r ig h t hand h a lf o f the tern ary system was p a r t ia l ly d e sc r ib e d .

For ca ses

where s u f f i c i e n t liq u id remained a t the s t a r t o f the secondary s o l i d i f i c a ­
t io n p r o c e s s e s , i t was p o s s ib le to c l a s s i f y th e subsequent behaviour in to
se v e r a l c a t a g o r ie s .
1 . M elts w ith o r ig in a l com p osition s w ith in th e tr ia n g le CDS
and above l i n e M l, showed a second i n f l e c t i o n a t the p o in t
corresponding to L.

The m e lts s o l i d i f i e d a t t h is p o in t ,

138

Ca Fp
c £

LiF

LLBeR
FIGURE 14
DISTORTED VIEW
TERNARY SYSTEM
CaFp— LiF—BeFp

BeE

139

except fo r m e lts in c lo s e p ro x im ity to p o in t

which are

d isc u sse d l a t e r .

2 . M elts w ith in the area corresponding to the ir r e g u la r fig u r e
EHKE u lt im a te ly s o l i d i f i e d a t th e tern ary e u t e c t ic p o in t K.
These m e lts did not show the req u ired second in f l e c t i o n
p o in t upon reach in g the ternary e u te c tic l i n e HK^ but o n ly
th ick en ed g r a d u a lly .

3 . M elts ly in g between the areas o f ( l ) and ( 2 ) , i . e . in the ir r e g u ­
la r r e g io n EILKE, showed a second i n f l e c t i o n upon rea ch in g the
tern ary e u t e c t ic lin e KL.

T his li n e slo p e s from L , at f i r s t

very g r a d u a lly and th en more s t e e p ly .

C onsequently the e f f e c t

nearer L resem bled a g ra d u a lly slo p in g in v a r ia n t r e a c tio n .

For

m elts read in g LK somewhat nearer p o in t K, the second i n f l e c t i o n
came a t a lower temperature and was n e ith e r as f l a t nor as
(ap p aren tly) prolonged as th ose s t r ik in g LK nearer to L.

These

m elts a l l showed the th ir d i n f l e c t i o n corresponding to the
ternary e u t e c t ic p o in t K.

The e f f e c t here was q u ite dependent

on the amount o f liq u id rem aining sin c e in some c a se s apparent
‘'undercooling" o f 15° to 20° occu rred .

M elts c o n ta in in g

reasonab le amounts o f liq u id s o l i d i f i e d a t K a t f a i r l y c o n s is ­
te n t tem p eratu res } which were p r a c t ic a lly the same as th a t o f
the b in ary e u te c tic a t G.
The m e lts d isc u sse d in P art C show good agreement fo r t h e ir secondary
thermal behaviour w ith th a t req u ired by the proposed phase diagram.

They

llj.0

w ere, however, too remote in o r ig in a l com position to be o f much help in
e s t a b lis h in g th e lo c a tio n s o f s ig n if i c a n t p o in ts in the L i2BeF4 r e g io n .
In P art B, the behaviour o f m e lts in the gen eral reg io n o f area
AFMJ was d is c u s s e d .

As a r u le such m e lts to the l e f t o f 30 mole p ercen t

BeF2 were n ot cooled to th e p o in t a t which th ey might fu rn ish u sab le
in form ation about the behaviour around lin e FM and p o in t M.

These m e lts

showed a m oderately good second i n f l e c t i o n on reachin g lin e JM i f near
the J end o f the l i n e .

I f th ey reached l i n e JM somewhat nearer to M,

the secondary in f le c t i o n s became n i l .

There was a b len d in g to g eth er o f

the primary and secondary thermal e f f e c t s as the space su r fa c e s and space
curve o f tw o -fo ld sa tu r a tio n have q u ite s im ila r slo p es .

;Such m e lts ( i f

allow ed to c o o l s u f f i c i e n t l y ) s o l i d i f i e d a t J , but the temperature v a lu e s
were not c o n c lu siv e as e x c e s siv e s o lid s were p r e se n t on reachin g the
in v a r ia n t p o i n t .
For the secondary therm al behaviour o f m elts in the gen eral reg io n
o f 30 to 50 mole p ercen t BeF2 a la r g e amount o f data i s a v a ila b le ,
p a r t ic u la r ly th a t fu rn ish ed by the d if f e r e n t MQ" s e r ie s o f m e lt s .

M elts

in the l e f t hand p o r tio n o f t h i s r e g io n ( i . e . s l i g h t l y to the l e f t o f
l i n e FM) con tain ed co n sid era b le liq u id on reachin g the tern ary p e r i t e c t i c
li n e FM.

M elts in t h is v i c i n i t y , which were b ein g s h if t e d in th e d ir e c tio n

o f the CaBeF4 p o in t , such as the s e r ie s 1Q-, 3§~, 7Q-, 8Q -, and 10Qshoxjed the fo llo w in g behaviour w ith r e sp e c t to th e second i n f l e c t i o n
tem perature.

For th ese m e lts ( c lo s e to the LiF-BeF2 sid e ) th e second

in f l e c t i o n appeared a t about 1^ 6° in th e form o f a n e a r ly f l a t p la tea u
during which th e la r g e s t p a rt o f the m elt s o l i d i f i c a t i o n o ccu rred .

The

lill

end o f th e p la te a u a t about Ij-520 and I4.I180 was accompanied by an abrupt
drop in tem perature c o in c id e n t w ith the disappearance o f apparent liq u id .
As th e s e same m elt s e r ie s were extended in to the tern ary system , the
secondary i n f l e c t i o n tem perature g ra d u a lly dropped to about 1lU9° fo r
m elts ly in g below th e l i n e extendin g from p o in t A to p o in t M o f F igure l k .
The shape o f th e p la tea u a t f i r s t slo p ed s l i g h t l y more than the preceding
m e lts , but as th e i n i t i a l p la te a u tem perature decreased the slo p e became
fla tte r .

For secondary in f l e c t i o n tem peratures o f )|J|9° the p la tea u

became again very f l a t , term in a tin g i n an abrupt temperature d r o p -o ff.
The abrupt drop^off was c o in c id e n t i n most c a se s w ith the disappearance
o f apparent l i q u i d , but on subsequent reh ea tin g liq u id was u s u a lly observed
in sm all amounts a t tem peratures about 10

o

low er than th a t corresponding

to the disappearance o f liq u id .
Another s ig n if ic a n t o b serv a tio n a t t h is p o in t was the behaviour during
c r y s t a l l i z a t i o n on the p la te a u .

P reviou s m e lts had shown no tendency to

form any c r u s t on the su rfa ce when s t ir r e d in a normal manner.

At t h i s

reg io n a change occurred and the m elt c h a r a c t e r is t ic a lly formed a c ru st
over the su r fa c e and s o l i d i f i e d as an adherent mass on the s t ir r in g ro d .
This s h i f t in behaviour c o in c id e d q u ite w e ll w ith the attainm ent o f m elt
com position s a t or beyond a l i n e from p o in t A to p o in t M, and w ith the
change to a f l a t t e r p la te a u a t about k k 9 ° .

Such a " c ru stin g 11 phenomenon

was unique fo r t h i s r e g io n , and i s b e lie v e d to be due to a p a r tic u la r
mode o f c r y s t a l l i z a t i o n o f th e CaF2 as i t s o l i d i f i e s a t the tern a ry p e r i­
t e c t i c p o in ts near k$3° .

ih2

Due to the p roxim ity o f p o in t M to the l i n e DE th ere was reason to
f e e l th a t t h i s d if f e r e n t type o f c r y s t a l l i z a t i o n might be th e r e s u lt o f
a ternary e u t e c t ic r a th er than a p e r i t e c t i c r e a c tio n .

While a few r e ­

s u lt s seemed confirm atory o f t h i s h y p o th e s is , th e great m a jo rity i n d i ­
cated th a t the r e a c tio n was one in which la r g e c r y s t a ls were formed, un­
l ik e a c l a s s i c a l e u t e c t ic type o f m ix tu re.

The incom plete s o l i d i f i c a t i o n

(in d ic a te d on r eh ea tin g ) was a lso in d ic a tiv e o f a n o n -e u te c tic r e a c tio n .
As th e s e same m e lts o f the s e r i e s 3 ^ -, 7Q-., 8Q- and 10Q- were extended
fa r th e r in to the system the s h i f t to i n i t i a l d e p o sitio n o f CaF2 o ccu rred ,
and the secondary in f l e c t i o n appeared as a very f l a t p la te a u w ith i n i t i a l
tem peratures gra d u a lly s h if t e d downward from 1^9° to 14j.7° .

This behaviour

i s in te r p r e te d as the r e s u lt o f i n i t i a l CaF3 d e p o sitio n which caused the
m elt com p osition to s h i f t u n t i l the tw o -fo ld sa tu r a tio n li n e ML was
reached.

Here th e secondary d e p o s itio n o f L i2BeF4 commenced, and th e m elt

fo llo w e d the 11e u t e c t ic ” lin e ML to p o in t L .

The s o l i d i f i c a t i o n t h e o r e t ic ­

a l l y req u ired a t p o in t L (te r n a r y p e r i t e c t ic p o in t) was outwardly s a t i s ­
f ie d by the disappearance o f liq u id a t

but on r eh ea tin g liq u id was

u s u a lly d e te c ta b le in s l i g h t amounts as much as 5° bo 10° low er than the
apparent s o l i d i f i c a t i o n temperature .
H ere, two a d d itio n a l s e r ie s o f m elts (12Q- and 13Q-) may be included,
in the d is c u s s io n .

These m elts lo c a te d ju s t to the r ig h t o f lin e FM were

also s h if t e d across th e tern ary system toward the CaBeF4 p o in t .

T heir

behaviour was d if f e r e n t from th a t o f the p reced in g m elts s in c e L i2BeF4
was i n i t i a l l y d e p o s ite d .

Here a v ery g ra d u a lly slo p in g p la te a u was

Ik3

obtained as a f i r s t i n f l e c t i o n , the i n i t i a l d e p o s itio n tem perature
grad u ally s h if t in g downward from about I4.58
inward.

as th e m elts were s h ift e d

At a com position valu e on th e l in e ML, t h i s value fo r the 12Q-

s e r ie s had d e c lin e d to Wa$ ° and, fo r th e 13Q s e r i e s , to )|J|)|° 4 T h erea fter
th ese m elts showed i n i t i a l CaF2 d e p o s itio n and th e r e a c tio n along the
lin e ML appeared as a q u ite f l a t p la te a u accompanied by the p r e v io u sly
mentioned " c r u st*1 form ation .
Another s e t o f the m e lts should be mentioned b efo re le a v in g t h is
reg io n ; th a t i s th e Ij.Q-, 6Q -, 9Q- and 11Q- s e r i e s , s ta r te d in the binary
system to the r ig h t o f p o in t F , and s h if t e d in the d ir e c t io n o f th e
CaF2-LiF b in ary e u t e c t ic (p o in t J) .

These m e lts e x h ib ite d a t r a n s it io n

in behaviour when th e lin e FM was reached and th e r e a fte r fo llo w ed the
same g en eral tren d s shown by oth er m elts in the r eg io n to the l e f t o f
li n e FM, as p r e v io u s ly d is c u s s e d .
B efore th ese m e lts (I4Q- e tc ) p a ss the l i n e FM, and come under the
c o n tr o l o f the tern ary p e r it e c t ic r e a c tio n which ta k es p la c e along i t ,
th ey are f r e e to proceed in a v a r ie t y o f d ir e c t io n s .

T heir path during

L i2BeF4 d e p o s itio n i s governed by the r e la t io n o f th e m elt com position
to the L i2BeF4 com position (p o in t D) .

The r e a c tio n path o f the m elt may

be such as to s tr ik e l i n e LK w e ll down toward p o in t K.

There w i l l be a

gradual upward s h i f t o f the secondary i n f l e c t i o n p o in t as th e m elt path
in t e r s e c t s LK a t p o in ts p r o g r e s s iv e ly approaching p o in t L.

In oth er

ca ses the in t e r s e c t io n o f the m elt path w i l l be w ith the l i n e ML, where
a secondary i n f l e c t i o n should occur i f not masked by the very sh allow
slop e o f th e L i2BeF4 primary c r y s t a l l i z a t i o n s u r fa c e .

llili

The g rea t v a r ie t y o f r e s u lt s req u ired by th e se p o s s i b i l i t i e s , may
be seen in th e ta b u la tio n s o f th e data fo r the s e r ie s ! $ - , 6Q-, 9Q- . and
11Q -.

The i n i t i a l d e c lin e to th e tern a ry e u t e c t i c , the g ra d u a lly r is in g

value fo r the tem perature o f apparent s o l i d i f i c a t i o n along the l i n e LK,
th e upper in v a r ia n t r e a c tio n s and oth er required secondary thermal
phenomena, are p r e se n t in s u b s t a n t ia lly good agreement w ith th e proposed
phase diagram .
Up to t h is p o in t the r e g io n s near p o in t L (a lo n g l i n e LI and ju s t
above l i n e LK) have not been co n sid ered in d e t a i l .

The p r in c ip a l sources

o f in form ation fo r the former r e g io n are the m elt s e r ie s 12Q- and 13Q-.
For the l a t t e r r e g io n , bordered by l i n e LK, in form ation i s fu rn ish ed by
the two s e r ie s LL- and MM-.

The two s e t s o f data w i l l be con sid ered

j o i n t l y as th ey are m u tu ally dependent.
The ch a ra cter o f i n i t i a l c r y s t a lli z a t i o n from th e 12Q- and 13Q- s e r ie s
(when extended beyond l i n e ML), the therm al liq u id u s v a lu es (o b ta in ed
e n t ir e ly v i s u a l l y ) , and the contours (above and to the l e f t o f p o in t L)
a l l req u ired th a t th ere be a f a i r l y d e f in it e v a lle y le a d in g up along l in e
L I.

The slo p e o f t h i s v a lle y appeared a t f i r s t to be grad u al, th en

s l i g h t l y ste e p e r as the clim b was made to the slo p in g p la te a u extendin g
downward from the CaBeF4 p o in t along th e lin e L I.
The data fo r th e l a s t members o f th e 13Q- s e r ie s are not co n sid ered
as r e li a b le (See d iscord an t r e s u lt s ) as the data o f the 12Q- s e r i e s .

L ik e­

w is e , in th is r e g io n th ere i s th e p o s s i b i l i t y th a t CaBeF4 may i n i t i a l l y
d e p o sit in c h a r a c t e r is t ic a lly unobservable fa s h io n , so th a t the observed

I ks

surface form ation o f CaF2 corresponded to a secondary r a th e r than a
primary p r o c e s s .
Below th e l i n e LI in the v i c i n i t y o f p o in t L the s it u a t io n w ith
r e s p e c t to r e l i a b l e liq u id u s data i s even p o o rer.

In t h i s reg io n CaBeF4

app arently i n i t i a l l y d e p o s its w ith out d e te c ta b le therm al e f f e c t , the slo p e
o f the liq u id u s su rface b e in g estim a ted to be about 30° per mole p ercen t
change in CaF3 c o n te n t.

U n til a tem perature l e v e l o f about 1+80 to $00°

was reach ed , i t was found th a t even th e "quenching" method o f dipping to
d e te c t CaBeF4 form ation was a f a i l u r e .
On m e lts LL-11 and LL-12 r e s u lt s were obtained in .r e a so n a b le a g ree­
ment w ith th e 12Q- and 13Q s e r ie s r e s u l t s .

For m elts up to and in c lu d in g

LL-8 i t i s f e l t th a t L i^eF ^ was i n i t i a l l y d e p o site d .

T h ereafter i t i s

b e lie v e d th a t th e tern ary e u t e c t ic l in e LK was passed and th a t subsequent
CaBeF4 d e p o s itio n caused the upward displacem ent o f the tem peratures o f
second (? ) i n f l e c t i o n v a lu es ( f o r LL-9 and LL-10) to s l i g h t l y high er v a lu es
rep re sen tin g in t e r s e c t io n s w ith LK a t p o in ts c lo s e r to L.
For the adjacent s e r ie s exten d in g in from m elt MM o f the binary
LiF-BeF3 s e r ie s e n t ir e ly analogous r e s u lt s were o b ta in ed , the l i n e LK
ap p aren tly b ein g p assed a t m elt MM-8 and the f i r s t CaBeF4 b ein g d e te c te d
a t 500° to 515° on MM- 1 0 .

The secondary in f le c t i o n s showed a c o n sid e ra b ly

la r g er r i s e ( a f t e r p a ssin g LK) corresponding to the in c r e a s in g slo p e o f
the l in e LK in t h i s r e g io n .
T his upward tren d in th e secondary i n f le c t i o n p o in ts fo r th e LL- and
MM- s e r ie s has been in d ic a te d i n the phase diagram

(by a sm all arrow

showing d ir e c t io n o f t r a v e l and by e n c lo sin g secondary i n f l e c t i o n tem peratures

lli.6

in p a ren th eses) fo r th ose c a se s where primary i n f l e c t i o n data a re la c k in g .
Over th e balance o f the range from about I4.2 to 52 mole p ercen t BeF2
( i . e . to th e l e f t o f , and c lo s e t o , l in e GK o f Figure llj.) very l i t t l e
a d d itio n a l in form ation was obtained as to the ex a ct lo c a tio n o f l i n e LK.
Here again m e lts s h if t e d in to the tern a ry system o n ly gave inform ation
on the tem perature l e v e l o f L i3BeF4 d e p o s itio n .

CaBeF4 , however, was

not d e te c ta b le u n t i l tem perature l e v e l s o f $00 to $$0° were reached.
From th e o th er contour in form ation and th e se p o in t s , i t i s estim a ted
th a t the slo p e o f the CaBeF4 liq u id u s su rface i s approxim ately 75° per
mole p ercen t change in CaF3 c o n te n t.
e f f e c t s i s u n d erstan d ab le.

The consequent la ck o f thermal

The li n e LK, le a d in g down to th e ternary

e u te c tic p o in t K, i s b e lie v e d to approach w ith in about 1 mole p ercen t o f
the LiF-BeF3 b in ary s id e o f the diagram by the tim e p o in t K i s reached.
M elts w ith in the above r e g io n gave, under fa v o ra b le circu m sta n ces,
v a lu e s o f about 3$0 to 353° degrees on c o o lin g fo r the temperature o f
the tern a ry e u t e c t ic p o in t K.

Because o f the extreme s lu g g ish n e ss o f

m e lts a t th ese low tem p eratu res, th ere was co n sid era b le u n d er-co o lin g o f
the m e lt.

C onsequently a range o f 356° to 358° ob tain ed by c a r e fu l

h e a tin g i s a more r e lia b le tern ary e u t e c t ic temperature fig u r e and an
average valu e o f 357° has been u sed .

T his i s no more (and probably l e s s )

than 1° below th e 358° value found fo r th e binary L i3BeF4-BeF3 e u t e c t ic
p o in t a t G.
In a l l th e p rev io u s d is c u s s io n s o f P a r ts B , C, and D o f t h is Appendix
the v a lu e s g iv en have been in terms o f the r e s u lt s obtained on c o o lin g
c y c le s .

In th e case o f in v a r ia n t r e a c t io n s , i t i s p o s s ib le (by slow

Iii7

h ea tin g and maintenance o f tem perature eq u ilib riu m by a g ita t io n ) to o b ta in
valu es u s u a lly about 2

h igh er than are o b ta in ed on c o o lin g c y c le s .

Since i t was im p o ssib le to make very la r g e numbers o f check reh ea ts a t
such p o in t s , a l l th e rep orted v a lu es are c o n s is t e n t ly based on c o o lin g
c y c le s u n le s s th e contrary i s s t a t e d .
comparable and are so p lo t t e d .

The v a lu e s are th e r e fo r e q u ite

In r ep o r tin g the Important in v a r ia n t

p o in ts o f the diagram i t i s d e s ir a b le to g iv e c o rr e cte d v a lu es as shown in
the summarizing P art E o f t h i s s e c t io n .
In P a rts B , C and D an attem pt has been made to d ig e s t a la r g e amount
o f d a ta .

I t was hoped the d is c u s s io n would be o f a s s is ta n c e to any person

in t e r e s t e d in a s s e s s in g the c o n s tr u c tiv e evidence fo r th e proposed i n t e r ­
p r e ta tio n o f the tern ary system phase r e la t i o n s .

To have endeavored to

have done oth erw ise would have com p letely d efea ted the purpose.

S u f f ic ie n t

emphasis has been p laced upon the u n c e r ta in tie s and the cause o f d i s ­
cordant r e s u lt s so th a t i t should be r e a liz e d th a t th ey are p r e s e n t .
i s f e l t th a t th e se fa c to r s are i n s u f f i c i e n t l y im portant to v i t i a t e the
c o n c lu sio n s based on a preponderance o f the e v id e n c e .

It

Ili8

P art E — The System CaF2-LlP-BeFa as a Whole
In P a r ts B, C and D o f t h i s s e c t io n , the r e s u lt s were d isc u sse d in
general by r e fe r e n c e to th e d is to r te d c o n str u c tio n s o f F igure 10 and 111.
This was con ven ien t because o f the awkwardly la r g e s iz e o f any o v e r - a ll
diagram which might be co n stru cted to show the d e t a il s on th e n ecessa ry
s c a le .

A more r e a l i s t i c d e s c r ip tio n may now be u sed .

In th e sm all s c a le o v e r - a ll view o f th e phase diagram o f F igu re 1 5 ,
which i s shown on the fo llo w in g p a g e, c e r ta in v a lu e s have been g iv en fo r
the in v a r ia n t p o in ts which d if f e r s l i g h t l y from the r e s u lt s which were
g e n e r a lly ob tain ed on c o o lin g c y c l e s .

As noted a t th e c lo s e o f P art D,

v alu es which were approxim ately 2° higher were ob tain ed on c a r e f u lly con­
t r o lle d h e a tin g c y c le s .

This procedure was o n ly a p p lic a b le to in v a r ia n t

p o in ts and n ot to the gen eral liq u id u s su rfa ce v a lu e s .

In th e com plete

system , shown in F igu re 1 5 , th e se s l i g h t l y high er v a lu es have been used
as th ey are not su b je c t to the e rr o r s caused by u n d erco o lin g , and because
the la r g e s c a le makes th e change n e g lig i b le .

In con n ection w ith the con­

tour l i n e s i t should be noted th a t 25° in t e r v a ls d estro y the fin e s tr u c tu r e ,
but any change in contour in t e r v a ls to o b ta in f in e d e t a i l would have to be
made u n iform ly throughout th e diagram to a v o id a d ecep tiv e appearance.
R eference must th e r e fo r e be made to F igu res 1 1 , 12 and 13 fo r d e t a i l s .
The th ree in v a r ia n t p o in ts o f th e tern ary system , corresponding to
p o in ts M, L and K o f Figure lU , are r e s p e c tiv e ly :
1 . The tern ary p e r i t e c t i c p o in t lo c a te d a t approxim ately 3U.3
mole p ercen t BeF3 and 2 .5 mole p ercen t CaF2 a t a temperature

149

150

o f 1+56 .
o c c u r s.

At t h is p o in t the fo llo w in g eq u ilib riu m p ro cess
CaF2 + LiF + m elt

- —»■ Li^eF^j + CaF2 .

2 . The tern ary p e r i t e c t i c p o in t lo c a te d a t approxim ately 36
mole p ercen t BeF2 and 3 mole p ercen t CaF2 a t about bS0°}
a t which p o in t th e fo llo w in g eq u ilib riu m p rocess o ccu rs.
CaF2 + L i2BeF4 + m elt

CaBeF4 + LigBeF^.

3 . The tern ary e u te c tic p o in t a t 357 ° , lo c a te d w ith l e s s c e r ­
t a in t y a t approxim ately 52 mole p ercen t BeF2 and 1 mole
p ercen t CaF2 .

The value w ith r e sp e c t to CaF2 com position

i s the more u n c e r ta in , th e probable valu e being l e s s than
2 mole p ercen t and g r ea ter than 0 .5 mole p ercen t CaF2 .
The eq u ilib riu m a t t h i s p o in t i s presumablyj
CaBeF4 +Li2BeF4 +BeF2-tMelt

> CaBeF4+Li2BeF4 +BeF2

Further summarizing the r e s u l t s , i t was found p o s s ib le to o b ta in
very low m eltin g m ixtures o f CaF2 w ith BeF2 and L iF , co n sid era b ly below
o r ig in a l e x p e c ta tio n s .

However, the CaF2 co n ten ts o f such m elts were

also low and o f th e order o f 1 to 3 mole p e r c e n t, at th e a lte r n a tiv e
temperature l e v e l s o f about 357° and i+50° .
Because o f the c lo s e proxim ity o f the two high er in v a r ia n t p o in ts to
the c r i t i c a l l in e con n ectin g the com position p o in t o f L i2BeF4 w ith the
CaBeF4 p o in t , th ere might be some q u estio n as to whether th e se p o in ts
were tern ary p e r it e c t ic or e u t e c t ic .
not perm it an a b so lu te d e c is io n .

The p o s s ib le a n a ly t ic a l errors do

In one case the in v a r ia n t p o in t was

151

lo c a te d alm ost on th e d iv id in g li n e between the two p o s s ib le b eh a v io u rs,
the oth er b e in g a l i t t l e l e s s than 1 mole p ercen t away from th e d iv id in g
lin e .

The assignm ents o f lo c a tio n s are dependent on a great many

separate m elt com position s so th a t a c o n s is te n t e r r o r , r a th er than a few
chance e r r o r s , would be n ecessa ry to s e r io u s ly b ia s the in t e r p r e t a t io n s .
A confirm atory fa c to r was the s o l i d i f i c a t i o n behaviour o f m e lts
upon reach in g the in v a r ia n t p o in t s .

C erta in m elts were observed to p ass

through th ese p o in ts w ithout s o l i d i f i c a t i o n , in a manner permittevd by
the Phase Rule fo r p a r tic u la r o r ig in a l m elt co m p o sitio n s.

Others appar­

e n t ly s o l i d i f i e d a t the in v a r ia n t p o in ts as required by the Phase Rule fo r
th ose p a r tic u la r circu m sta n ces.

Interm ed iate in behaviour were some c a se s

which ap p aren tly f a i l e d to q u ite meet the com plete s o l i f i c a t i o n t e s t de­
manded by th e o r y .

T his could a r is e from a type o f behaviour d escrib ed at

the end o f Appendix 5 in which i t was shown th a t a p ro cess o f s o l i d i f i c a ­
tio n cou ld depart from a l in e o f ternary p e r it e c t ic behaviour upon ex­
h a u stio n o f the substance b ein g used up by the p e r i t e c t ic p r o c e s s .

In

th ese c a s e s , how ever, i t i s b e lie v e d th a t th e i s o la t io n o f LiF from
equilibrium , w ith th e m elt caused e f f e c t iv e exhaustion to o c c u r .

T his

r e s u lte d In 11forb id d en ” behaviour f o r m e lts w hich, in th eo r y , should have
fo llo w ed the p e r i t e c t i c li n e to a f i n a l s o l i d i f i c a t i o n at the in v a r ia n t
p o in t .
T his “forbidden" behaviour i s c h a r a c t e r is tic o f p o o r ly behaved p e r i ­
t e c t ic p r o c e s s e s .

I t i s , on th e c o n tra ry , d e f in it e ly not c h a r a c t e r is tic

o f the e u t e c t ic type o f p ro cess in a tern ary system where s o l id s o lu tio n s
are not in v o lv e d .

H ere, the p r o c ess i s independent o f p r e v io u sly

152

d ep o sited s o l i d s .

The e u t e c t ic p r o c ess along a tw o -fo ld s a tu r a tio n l in e

i s then " s e l f regu latin g" and the path i s fo llo w e d in a com pletely
t h e o r e t ic a l manner.

To do o th e r w ise 3 the m elt would have to heat i t s e l f

to a high er tem perature w h ile a t the same time i t was d e p o s itin g an ex­
c e s s o f one s o l i d .

Sim ultan eou sly th e s o l u b i l i t y o f t h is s o l i d in the

m elt would be in c r e a s in g as required by the h igh er tem perature.

O bviously

th ere could be no c o n tin u a tio n o f such a thermodynamically improbable
p r o c e s s 3 and a cco rd in g ly e u te c tic l i n e s o f tw o -fo ld sa tu r a tio n show good
agreement w ith Phase Rule req uirem en ts.

The same holds true fo r a tern ary

e u te c tic p o in t r e a c tio n .
I t i s the ap p aren tly d e fe c tiv e nature o f the p r o c e sse s (o ccu rrin g
along the p e r i t e c t i c l in e and a t the in v a r ia n t p o in ts in the L i2BeF4
region ) th a t c o n s t it u t e s the b e s t in d ic a tio n th a t th ey are p e r i t e c t i c
p r o c e sse s rath er than e u t e c t i c .

As a consequence o f the thermal data^ and

because o f th e above c o n s id e r a tio n s } i t i s b e lie v e d th a t the phase diagram
in te r p r e ta tio n i s c o r r e c t as shown in F igure 1$3 d e s p ite the u n c e r ta in tie s
which in d ic a te th a t oth er ex p la n a tio n s o f the behaviour might be p o s s ib le .

ELECTROLYTIC INVESTIGATIONS IN THE TERNARY SYSTEM

153

ELECTROLYTIC INVESTIGATIONS IN THE TERNARY SYSTEM
The phase diagram o f the tern a ry system co n ta in s one r e l a t i v e l y low
temperature reg io n ly in g very c lo s e to the bin ary LiF-BeF2 sid e in the
range o f 3^- "to $2 mole p ercen t BeF2 , xrith a corresponding CaF2 co n ten t
ranging from 3 ^ 1

mole p e r c e n t.

Surrounding t h is reg io n o f low

tem perature (and low CaF2 co n ten t) are s t e e p ly ascending s lo p e s .

The

slop e arrangements are such th a t tem peratures o f 550° to 600° would be
n ecessa ry in order to l i q u i f y m ixtures co n ta in in g about tw ice th e CaF2
con ten t o f th o se which e x i s t as liq u id s in the lower temperature r eg io n
o f about 350° to bS 0°,
D e sp ite the u n favorab ly low CaF2 co n te n ts o f the low er m e ltin g m ix­
tu res i t was decided to t r y them fo r e l e c t r o l y t i c purposes .

T his d e c is io n

was in p a r t governed by the f a c t th a t i t i s p r a c t ic a ll y n ecessa ry to
operate a t tem peratures s l i g h t l y above the m e ltin g p o in ts in order to
avoid o p e r a tin g d i f f i c u l t i e s .

O peration at tem peratures in e x ce ss o f 55>0°

could o n ly be exp ected to r e s u lt in g r e a tly in c r e a se d r e a c t i v i t y o f
evolved f lu o r in e , which promised more d i f f i c u l t y than the o th er a lte r n a ­
t iv e o f o p era tio n u sin g a m elt c o n ta in in g very l i t t l e CaF2 .
The d eterm in ation o f e le c tr o d e products in such a p ro cess p r e se n ts
p a r tic u la r d i f f i c u l t i e s which were more f u l l y d e scrib ed in th e "General
and T h eo retica l" s e c t io n .

The c o ll e c t i o n and d e te c tio n o f evolved

flu o r in e ( i f not used up by a tta c k on the e le c tr o d e ) would not be expected
to be too d i f f i c u l t in a q u a lit a t iv e way.

The d e p o site d m etal from the

l$h

e l e c t r o l y t i c p rocess rep re sen ts a more d i f f i c u l t problem in the case o f
calciu m .

The o p e r a tin g tem perature in q u estio n i s fa r below th a t r e ­

quired fo r form ation o f a coherent m etal d e p o s it .

A s a method o f i s o l a ­

tio n o f th e cathode p ro d u ct, i t was decided to use a m olten le a d ca th o d e.
I f s u c c e s s f u l, t h is would have th e advantage o f p rev en tin g d is s ip a tio n
o f the calcium m etal throughout th e m e lt.

In a d d itio n a fa v o ra b le e f f e c t

on the d e p o s itio n p o t e n t ia l o f calcium might be obtained as more f u l l y
d isc u sse d in e a r lie r s e c t i o n s .
W hile e l e c t r o l y s i s o f fu se d s a l t s on a sm all s c a le b a s is i s p a r tic u ­
la r l y u n d e sir a b le , p r a c tic a l c o n sid e r a tio n s made i t n ecessa ry to work
w ith sm all c o n ta in e r s and amounts o f m elt which could be handled at
mo derate cu rren t l e v e l s .
As a source o f D.C, power fo r t h i s work a fu ll-w a v e , f i l t e r e d o u tp u t,
r e c t i f i e r was c o n str u c te d .

The output o f a selenium r e c t i f i e r , arranged

fo r fu ll-w a v e b rid ge type r e c t i f i c a t i o n , was p assed through a f i l t e r i n g
network to reduce th e r ip p le v o lt a g e .

T his f i l t e r , a double L ty p e ,

c o n s is te d o f la r g e e l e c t r o l y t i c c a p a c ito r s and very h ig h inductance choke
c o ils .

The l a t t e r were s p e c ia lly co n stru cted to have very low r e s is t a n c e s .

The output r ip p le from the f i l t e r v a r ie d from about 2 to 10% depending
upon cu rren t lo a d in g .

W ith t h is arrangem ent, su p p lied by a v a r ia b le au to­

tran sform er, a maximum current o f 30 amperes a t any v o lta g e from 0 to 100 V
was a v a ila b le a t th e output te r m in a ls .
For purposes o f making rough com parative conductance measurements
and fo r e le c t r o ly s e s a sm all c e l l was c o n str u c te d .
showing the c o n s tr u c tio n i s given in F igure 16.

A schem atic drawing

The n ic k e l c r u c ib le

C

Ano d e
Co

nnect

athode

Connection

\ON

SH OwTMG N o t

\

fSK lR T * 4 3 > IA T H R A < jM

Arpno*
ki/OUiP
T ^R M

wueuu

ocoople

I

A node

Cathode

FIGURE 16

CELL
CONSTRUCTION

nsolation

156

c o n s titu tin g the c e l l p o t was in terch a n g ea b le w ith carbon c r u c ib le s o f
s l i g h t l y la r g e r s iz e which were used in some o f the l a t e r work.
c e l l p o t served as th e cathode o f the c e l l .

The

Diaphragms o f d if f e r e n t

le n g th s were prepared to perm it i s o la t io n o f th e anode products in m elts
having d if f e r e n t su rface l e v e l s , the diaphragm b ein g chosen so as to
p r o je c t o n ly s l i g h t l y below the m elt s u r fa c e .

A c e n tr a l anode arrange­

ment was u se d , carbon, n ic k e l and copper rods b eing the a lte r n a tiv e forms
a v a ila b le .

Anode gas t a k e - o f f was by way o f a hollow passage a t the top

o f th e anode, le a d in g to a copper tube m a n ifo ld .
The anode and diaphragm were in s u la te d from each oth er and from the
cover o f th e c e l l .

The anode e l e c t r i c a l con n ectio n was made through the

copper exhaust m a n ifo ld .

The cathode e l e c t r i c a l co n n ectio n was made

e ith e r through another copper li n e connected to the cover o f the c e l l ,
or in o th er c a se s by means o f a copper band attach ed to the carbon
c r u c ib le w a l l s .
For conductance measurements i t was p o s s ib le to determ ine the le a d
r e s is ta n c e s by sh o r tin g the c i r c u i t ( a t a p o in t j u s t above the anode)
w ith th e a id o f a movable sh o r tin g n u t.

These le a d r e s is ta n c e s were de­

ducted to determ ine th e c e l l r e s i s t a n c e .
For r e s is ta n c e m easurements, a t th e very low v a lu e s which were found
to e x i s t , no very s a t is f a c t o r y equipment was a v a ila b le .
Impedance B r id g e , Type 650-A , was u se d .

A General Radio

The source o f the A .c . measuring

v o lta g e was th e s e lf - c o n t a in e d 1000 c y c le microphone hummer o f the
b r id g e .

A 5" DuMont Cathode Ray O s c illo sc o p e was used fo r d e te c tin g the

n u ll a t the balan ced p o s it io n o f the b r id g e .

The measuring error fo r the

157

higher r e s is ta n c e v a lu e s was about 1%) but fo r th o se o f most i n t e r e s t , at
about 1 ohm, th e error was probably a t l e a s t 1 0 $ .

The p o s s ib le e f f e c t

o f in s u la t in g f ilm s on th e e le c tr o d e s u r fa c e s i s not known.
For conductance measurements th ree m e lts lo c a te d in th e bin ary
LiF-BeF2 system were prepared.

T h e r e a fte r , th e s e m elts ( c a lle d 1EL,

2E1 and 3EL) were s h if t e d in to th e tern ary system (to form m e lts 1EL-1,
2EL-1 and 3EL-1) to determ ine whether a d d itio n o f CaF2 caused much
change in th e conductance .

Sin ce BeF2 i s i t s e l f non -conductive as a

liq u id i t was f e l t th a t such rough measurements o f conductance were de­
s ir a b le .

These would be o f some h elp in d e c id in g whether any p e c u lia r ly

high r e s is t a n c e phenomena which m ight be encountered could p rop erly be
assign ed to the m elt or to a p o la r iz a tio n e f f e c t .
As th e r e s u lt s ob tain ed on th e s e conductance measurements were
n e ith e r very s ig n if i c a n t nor c o n c lu s iv e , the data has been r e le g a te d to
Appendix 1 6 .

The o b se rv a tio n s are b r i e f l y summarized by the fo llo w in g

sta te m e n ts .
1 . The two m e lts prepared w ith com position s ( i n th e LiF-BeF2
b in a ry system ) near 33 mole p ercen t BeF2 were approxim ately
th ree to fo u r tim es as condu ctive a s the s in g le m elt w ith a
com p osition o f approxim ately 5k mole p ercen t BeF2 .
2 . The a d d itio n o f CaF2 caused a f a i r l y la r g e in c r ea se in the
conductance o f th e l a t t e r m elt when s h if t e d to a com position
o f about 1 mole p e r ce n t CaF2 and 5k mole p e r ce n t BeF2 .

158

3 . The a d d itio n o f CaF2 to th e oth er two m elts (b y which
th ey were s h if t e d to about 0 .9 mole p e r ce n t CaF2 and
3 2 .1 and 3 3 .6 mole p ercen t BeF2) gave d iscord an t r e s u l t s ,
one in c r e a s in g and th e oth er d ecrea sin g in conductance,
n e ith e r e f f e c t b ein g la r g e .

k. I t was concluded th a t a r e s is ta n c e o f 0 .5 to 1 .0 ohm might
be exp ected from e it h e r the high or the low BeF2 co n ten t
m e lts fo r the c e l l u se d , over th e range o f about 550 to h5^ ° .
The n e g a tiv e temperature c o e f f i c i e n t o f r e s is ta n c e was
la r g e , p o s s ib ly about 1% per d eg ree.
5 . Very ap p reciab le conductance in the s o l i d s t a t e was n o ted .
The conductance valu e noted was somewhat lower than th a t o f most
fu sed s a l t m e lts .

The u su a l range o f v a lu e s from e x c e lle n t to average

conductors i s about 20 to 2 ohms

cm.

fo r the s p e c if ic conductance.

The ir r e g u la r shape o f the c e l l p reven ts more than a rough estim a te th a t
in t h i s case the s p e c if ic conductance co u ld hardly have been h igh er than
0 .1 to 0 .2 ohm^cnC1 .

(See Drossbach (27) fo r a l i s t i n g o f s p e c if ic

c o n d u c tiv ity fo r v a r io u s fu sed s a l t s . )
For th e i n i t i a l e l e c t r o l y t i c work the c e l l d escrib ed in Figure 16
was u sed , tr y in g d if f e r e n t e le c t r o d e s , diaphragms, n ic k e l and carbon
c r u c ib le s , and m olten le a d cathodes as v a r ia tio n s in attem pts to o v er­
come o p e r a tin g d i f f i c u l t i e s .

Sin ce t h is c e l l was c lo se d a t the top i t

was not p o s s ib le to observe the ca u ses o f th e tr o u b le .

The i n i t i a l

159

m elts th a t were used were th o se mentioned a b o v e.
m elts were prepared.

M elt
1
2
3
k
3

EL-1
EL-1
EL-1
EL-1
EL-1

Su bseq uently two oth er

The com p osition s were as fo llo w s :

Mole P ercen t
CaFa
BeF2
LxF

0.8^
0 .8 8
1 .0 k
O.Sii
0 .8 9

32.1
3 3 .6
Sh.O
33.3
2 9 .6

W t. o f M elt,
______ gj_____

66.1
65.5
U5.0
65.9
69.5

100.9
97.9
89.7
101.6
9h.O

The attem pts a t e l e c t r o l y s i s o f th e se m elts are d escrib ed below .
On m elt 3EL-1, a n ic k e l c r u c ib le and carbon anode were te s te d u sin g
the diaphragm shown in F igure 1 6 .
flow when 5V was a p p lie d .

There was o n ly a momentary cu rren t

Upon r a is in g th e ap p lied v o lta g e to 30V cu r­

ren t a g a in flo w e d , but in an ir r e g u la r manner.

No apparent anode gas

product cou ld be d e te c te d alth ough the e l e c t r o l y s i s was continued fo r
about 20 m in u tes.
about 390 to [|8 0 °.

During t h i s tim e th e c e l l temperature ro se from
Upon shutdown and removal from the furnace i t was

found th a t a h o le had been burnt in th e c r u c ib le w a ll.

T h is ap p aren tly

was th e .r e s u lt o f an arcin g sh o r t p a ssin g through a co n sid era b le amount
o f b la c k fr o th y c r u s t on th e su rfa ce o f th e m e lt.

The carbon anode had

been co n sid e ra b ly attack ed and the n ic k e l c r u c ib le had a lso l o s t some
w e ig h t.
M elt 1EL-1 was n ex t t r ie d u sin g a n ic k e l anode and a m olten le a d
cathode a t a tem perature o f 5 0 0 °.

Although th e current flo w could be

m aintained a t low er v o lt a g e , symptoms o f e r r a t ic behaviour caused the

160

t e s t to be h a lte d a f t e r o n ly a few m in u tes.

Upon in s p e c tio n th e same

apparent s h o r tin g e f f e c t through the b la ck fr o th y m a ter ia l was su sp e cte d .
The n ic k e l anode was found to have been s l i g h t l y a tta ck ed .
product was o b served .

No gaseous

The le a d cathode was r eserv ed fo r a n a ly s is

(marked "A fter 1 E L -1").
M elt 2 EL-1 was then used w ith a carbon c r u c ib le , m olten le a d cathode
and carbon anode.

The temperature ro se from 510° to 580° as i t was

attem pted to fo r c e a cu rren t to flo w by the use o f 50-75 v o lt s .

The

current flo w was very e r r a tic and n e a r ly n e g lig ib le p art o f the tim e.
In sp e c tio n in d ic a te d about th e same c o n d itio n s as xfere observed fo r the
precedin g t r i a l s .
Another t r i a l was b r i e f l y made on th e same m elt u sin g a carbon
c r u c ib le } m olten le a d cathode and a n ic k e l anode.
from ii70° to 520° and the

The temperature rose

behaviour observed on m elt 1 EL-1 was rep ea ted .

The diaphragm was th en d isca rd ed fo r a l l subsequent work and the
t r i a l on m elt 2 EL-1 was rep eated u sin g a carbon c r u c i b l e m o l t e n le a d
cathode and n ic k e l anode.

O peration was much more s ta b le and a run o f
o
o
approxim ately 30 m inutes was made a t about 7V and 15A and 525 to 550 .

On shutdown th e le a d cathode (w eig h t 150 g .) was reserv ed fo r a n a ly s is
(marked "A fter 2 EL-1") .

The n ic k e l anode was found to have been

h e a v ily attacked_, b u t a t no time was th ere any evidence o f the e v o lu tio n
o f flu o r in e by way o f th e

anode gas m an ifold .

At t h is sta g e i t wasd ecid ed to tr a n s fe r o p era tio n s to

an open

carbon c r u c ib le (h ea ted by gas) so th a t th e behaviour could be observed
d ir e c t ly .

Only th e c e n tr a l anode-holding p o r tio n o f the c e l l top was

161

u sed , minus the diaphragm.

A carbon c r u c ib le , w ith m olten le a d cathode

(100 g .) was u se d , f i r s t in co n ju n ctio n w ith a n ic k e l anode.

Im m ediately

upon th e a p p lic a tio n o f power a red-orange s o lid spread across th e sur­
fa c e from the anode rod .

T h is changed c o lo r and a b lack fr o th com­

menced to form as a r e s u l t o f r is i n g b u b b les.

The l a t t e r did n o t,

however, escape as the fr o th remained unbroken.

The experim ent was

contin ued w ith the fo llo w in g sch ed u le.
About 5 m inutes a t 10V, 1/2A .
About 6 m inutes a t 15V, 1A.
About 3 m inutes a t 17V, IjA.
About 13 m inutes a t 12V, 10A.
As noted th e r e s is ta n c e dropped somewhat, the temperature r is i n g from
1460° to 5U0° la r g e ly as a r e s u l t o f in t e r n a l h e a tin g .
th ere was no evid en ce o f th e escape o f any f lu o r i n e .

During the t e s t
Upon removal o f

the anode i t was found to have been very s tr o n g ly a tta ck ed .
A carbon anode was s u b s titu te d and i t was found th a t the m elt d id
not wet the e le c tr o d e .

When v o lta g e exceeding about IV was a p p lie d ,

sparks were observed over the su rfa ce o f th e anode.

Upon removal o f

the anode i t was found to be eroded,p»ossibly because o f the sparking
a c tio n .

S u b s titu tio n o f th e n ic k e l anode again p erm itted current flo w

at normalv o lt a g e .

When a copper anode was t r ie d , i t lik e w is e

co n sid e ra b ly a tta ck ed in

a b r i e f p e r io d .

was

The t e s t wasd isco n tin u ed

and the le a d cathode w eighing 150 g . was reserv ed fo r a n a ly s is (marked
HA fte r 3 EL-1") .

162

To determ ine th a t the n o n -w ettin g phenomena fo r th e carbon anode
was n ot a p rop erty o f the contam inated m e lt, a fr e s h m elt 5 EL-1 was
t r ie d .

An open carbon c r u c ib le was u sed w ith a carbon anode, the m olten

lea d cathode b e in g om itted a t t h i s s t a g e .

A v ery sh o rt t r i a l was made

w ith th e same r e s u l t s , th a t is ,s p a r k in g from the anode to the m elt w ith
very e r r a tic cu rren t flo w , th e tem perature o f the m elt b ein g 5 5 ° ° .
A n ic k e l anode was then s u b s titu te d to check on whether the red orange anode product observed fo r a very sh ort in t e r v a l might not be due
to le a d con tam in ation .

T h is was shown to be in c o r r e c t as th e phenomenon

was im m ediately observed in the absence o f the le a d cath od e.

The co lo red

product had o n ly a tr a n s ie n t e x is te n c e and im m ediately disappeared in
b lack fr o th which was n ext produced.
The m olten le a d cathode was then added and a 1$ minute run at 55°
was made a t a cu rren t o f $A, th e v o lta g e b ein g r eg u la ted between 5 an^.
10V in order to keep th e current ste a d y .

The lea d cathode removed on

shutdown (w eight 1$0 g .) was marked "After 5 EL-1" and reserv ed fo r
a n a ly s is .

The n ic k e l anode had su ffe r ed a l o s s o f about 1$% by w eight

during t h i s run, based on the immersed p o r tio n o f the rod
During a l l th e se t r i a l s the o n ly evidence o f anode gas e v o lu tio n
was th a t a few bubbles could be observed r is i n g to form a fr o th w ith the
b lack m a te r ia l o r ig in a t in g a t the anode.

The o th er in d ic a tio n o f the

anode p ro cess was the i n i t i a l co lo red product and the b la ck product which
follow ed i t .

From th ese o b serv a tio n s i t might reason ab ly be concluded

th a t evolved flu o r in e a tta ck ed the anode.

163

For a d eterm in ation o f th e type and amount o f m etal deposited, by
the cathode p r o c e s s , an a n a ly s is o f th e se h e a v ily contam inated m e lts
promised to be very u n s a t is f a c t o r y , when co n sid ered in co n ju n ctio n w ith
the a n a ly t ic a l erro rs found in the e a r li e r work.

I t was th er e fo re

decided to analyze the le a d cathodes on a "before and a fter" b a s is to
determine th e change in co m p o sitio n .

In order to do s o , i t xiras n ecessa ry

to scrape th e surface o f the le a d fr e e o f tr a c e s o f the fused, s a l t mix­
tu re.

T his may have had some b ea rin g on the r e s u l t s o b ta in ed .

The a n a ly t ic a l r e s u lt s have been ta b u la ted in Appendix 1 5 , p a rt K.
The r e s u lt s were q u ite anomalous in th a t th ey in d ic a te d no change in the
a n a ly s is o f the le a d .

The l a t t e r was o r ig in a l ly c h em ica lly pure grade

and the samples o f th e m olten cathodes were o n ly n e g lig ib ly d if f e r e n t
w ith r e s p e c t to Ca, L i and Be co n ten ts than the blank determ inations
shown by the s t a r t in g m a t e r ia l.

O v e r -a ll recovery was checked a t the

same tim e by a le a d determ in ation and the a n a ly t ic a l l o s s e s were found
to be sm a ll.
The a n a ly s is i s f a i r l y d i f f i c u l t because i t in v o lv e s the sep a ra tio n
o f a sm all amount o f "impurity" from the le a d which in t e r f e r e s w ith the
d eterm in ation o f th e oth er m eta ls o f i n t e r e s t .

The method used was a

s p e c ia l on e, but i t i s b e lie v e d th a t i t x^as adequate fo r the purpose.
The o n ly ex p la n a tio n th en rem aining i s th a t th e fo r e ig n m etal content
o f the le a d sam ples, as a n a ly zed , must have been n e g lig ib ly sm a ll.
W hile t h i s c o n c lu sio n appears improbable i t might a r is e from one o f
two p o s s ib le b eh a v io u rs.

P o s s ib ly any cathode product was never d e p o site d

on the le a d ca th o d e, but in ste a d d ep o sited on th e w a ll o f th e c r u c ib le

15k

e l e c t r i c a l l y connected to the le a d .

A lt e r n a t iv e ly , any m a te r ia l de­

p o site d on th e le a d was n ot p resen t a t the time o f a n a ly s is .

Wo in ­

d ic a tio n o f the former p o s s i b i l i t y was observed, but such an o b serv a tio n
would be u n lik e ly .

W ith regard to th e second p o s s i b i l i t y i t i s c o n c e iv ­

a b le , fo r exam ple, th a t CaPb3 formed on the le a d su rface m ight not have
been r e ta in e d in the a n a ly t ic a l sample .
E e fe r r in g to the phase diagram fo r the Ca-Pb system and th e d i s ­
c u ssio n o f th e behaviour o f m olten le a d cathodes g iv e n at the end o f
the s e c tio n on th e " E ffect o f L iquid C athodes, e t c . , " i t appears th a t
such a p o s s i b i l i t y i s c o n c e iv a b le .

E ith e r in the han dlin g and sep ara­

tio n o f the m elt from th e le a d cathode, or in c le a n in g the cathode m etal
to o b ta in a n a ly t ic a l sam ples, any su rfa ce la y e r o f CaPb3 might have been
removed.

As th e s e experim ents were n e c e s s a r ily performed below the

m eltin g tem perature o f th a t compound, the a llo y in g a c tio n w ith the liq u id
lea d depends on ab sorp tion and such r e a c tio n s have been observed to be
very slow (lllj.) .
The e l e c t r o l y t i c t r i a l s were th e r e fo r e f a ilu r e s from two sta n d p o in ts.
The recovery o f flu o r in e was not o b ta in e d , and i t s probable e v o lu tio n
could on ly be in fe r r e d from the form ation o f a r e a c tio n product accompanied
by a strenuous a tta c k on th e anodes u sed .

As no more s u ita b le m a te r ia ls

than n ic k e l, copper or carbon were known, there d id not appear to be any
way o f overcoming the anode c o n str u c tio n problem excep t by- red u cin g the
tem perature,

On the oth er hand i t was not found p o s s ib le to recover the

m etal d e p o site d in order to determine whether CaF2 was decomposed.

The

s o lu tio n to t h i s d i f f i c u l t y ap p aren tly does not l i e i n the same d ir e c t io n ,
but ra th er in th e d ir e c t io n o f r a is in g the tem perature l e v e l .

SUMMARY AND CONCLUSIONS

165

SUMMARY AND CONCLUSIONS
The problems o f d evelop in g a d ir e c t e l e c t r o l y t i c method fo r the
production o f elem en tal flu o r in e from calcium flu o r id e have been d i s ­
cussed and a p o s s ib le method proposed which in v o lv ed the e le c t r o l y s i s
o f fu sed m eta l flu o r id e m ix tu res.
The use o f th e method req u ired a knowledge o f the s o l u b i l i t y r e la ­
tio n s o f th e m olten flu o r id e s o f ca lciu m , lith iu m and b e r y lliu m .

The

bin ary and tern ary system s formed by t h i s com bination o f flu o r id e s
have been in v e s tig a te d and t h e ir phase diagrams d e scr ib e d .
Comparison has been made o f th e r e s u lt s obtained fo r the lithium ,
flu o r id e -b e r y lliu m flu o r id e system w ith th ose rep orted by two previou s
groups o f in v e s t ig a t o r s .

I t was concluded th a t n e ith e r o f th e se r e ­

p orts .was com p letely a c c u r a te .
The sy n th eses and X -ray exam inations o f compounds e x is t in g in the
tern ary system have been d e sc r ib e d .
The most fa v o r a b le r e g io n o f the tern ary system o f calciu m , lith iu m
and b ery lliu m f lu o r id e s was used to t e s t th e proposed method fo r the
decom position o f calcium f lu o r id e .

The r e s u lt s o f e l e c t r o l y t i c i n v e s t i ­

g a tio n s have been d escrib ed in which i t co u ld not be demonstrated th a t
the o r ig in a l purpose was a ch iev ed .
I n a b i l i t y to recover flu o r in e as a product o f the e l e c t r o l y s i s has
been a ttr ib u te d to chem ical r e a c tio n w ith the anode m a te r ia l.

The

p o s s ib le reason s fo r f a il u r e to reco v er and i d e n t if y the c a th o d ie a lly
d ep osited m etal have been c o n sid e re d .

APPENDIX I

166
APPENDIX 1
SUMMARIZED REFERENCES ON PHASE RELATIONS OF CALCIUM FLUORIDE—
LITHIUM FLUORIDE—BERYLLIUM FLUORIDE AND ANALOGOUS COMPOUNDS
A) Calcium F lu o r id e Systems and A n alogs.
CaF2-LiF-MgF2 , r ep o rt o f lo w e st m eltin g tern a ry m ixture as 665°,
no phase diagram s t u d ie s , used fo r magnesium p ro d u ctio n . ( 135)
Ternary r e c ip r o c a l system o f F

and Cl

o f lith iu m and ca lciu m ,

c o n ta in s no r ep o rt o f LiF-CaF2 system . ( 13)
Quaternary CaF2-LiF-BaF2~MgF2 system . — lo w e st m eltin g m ixture
603° . ( 83)
CaF2-NaF-KF, phase diagrams o f b in a r ie s and te r n a r y .
The CaF2-NaF system , sim ple e u t e c t ic m eltin g at 810°.
The CaF2-KF system , two e u t e c t ic s a t 782° and 1060° w ith
con gru en tly m eltin g compound KF:CaF2 a t 1068°.
Ternary system shows p e r it e c t ic p o in t a t 772
e u t e c t ic a t 682 .

0

and tern ary

R ecent rep ort (1951) by S ilb e r e t a l .

(Iii9)
CaF2-NaF-BaF2 , p a r t i a l d a ta g iven, lo w est p o in t approxim ately 750°,
used f o r magnesium p ro d u c tio n . (56)
CaF2-NaF-MgF2 , phase diagrams o f b in a r ie s and tern a ry .
CaF2-NaF system , sim ple e u t e c t i c , 8 1 0 °, a t about 50 mole p e r c e n t.
CaF2-MgF2 system , sim ple e u te c tic , 9 k5 °, a t about $0 mole p e r c e n t.
MgF2-NaF system , two e u te c tic s a t 8l5° and 985° w ith con g ru en t­

ly m eltin g compound NaMgF3 a t 1030 . (55)

167

CaF2-ThF4

( 177)

CaF 2-ZF3

(177)

CaF2-SnF3 , study o f s o l i d s o lu tio n s formed. (138)

CaFg-MgF,3, e u te c tic a t 970

and $2 mole p e rc e n t MgF3 . (160)

CaF2-BaF3 , e u te c tic a t 1277° and $0 mole p ercen t BaF2 . (160)
CaF3-MgF3-BaF3 , lo w est p o in t o f system , 817°.

Used fo r

magnesium p ro d u ctio n . ( 50)
MgF3-KF andMgF2-RbF, phase diagram s.
Two system s sim ila r in th a t compounds M^MgF3 and M2MgF’4 are
formed w ith congruent m eltin g p o in ts ; two e u t e c t ic s are
p r e s e n t.

( 36)

MgF2-KF-RbF, tern ary system d e sc r ib e d . ( 36)
MgF2-NaF-BaF2 , bin ary and ternary system s g iv en .
MgEr2-NaF, two e u t e c t ic s and in term ed iate compound.
BaF2-MgF3 , two e u t e c t ic s and in term ed ia te compound (u n c e r ta in ).
BaF3-NaF, sim ple e u t e c t ic . ( 8l ) (37)
MgCl2-NaCl, Mgdl3-KCl, MgCl2-RbCl.
T

These system s a l l show form ation

T

o f M MgCl3 and M ^gC ^ w ith congruent m eltin g p o in ts fo r
potassium and rubidium , but incongruent fo r sodium.
BaF3-SrF2 , minimum in. system a t 12 7 0 °. (7)
BaF3-NaF-MgF2 , binary and tern ary sy stem s.

o
BaF2-NaF system has sim ple e u te c tic a t 825 . (57)
B) Lithium F lu o rid e Systems and A n a lo g s.
See above under item A ).

A lso the fo llo w in g

(82)

168

LiF-MgF2 , s e v e r a l c o n f li c t i n g r e p o r ts .

Probably sim ple e u t e c t ic

(75) ( 16 I4) ( 67 )
LiF-MgF2 , NaF-MgF2 , KF-MgF2 , EbF-MgF2
Systems become in c r e a s in g ly com plex.

For LiF probably sim ple

e u t e c t i c ; fo r NaF, two e u t e c t ic s and compound NaMgF3 ; fo r
KF and RbF, two compounds M^MgF3 and M^MgF4 .
LiF~BeF2 , phase diagram.

( 8 l)

Two rep o rts th a t were somewhat d isc o rd a n t.

D escribed in co n n ectio n w ith d is c u s s io n o f r e s u lt s on t h i s
system .

Reported in 1959 and 1950.

(166) (135)

NaF-BeF2, phase diagram, s u b s t a n t ia lly same as th e LlF-BeF2 system

above.
.

Three r e p o r ts ( l 955, 1950, and 1951).

NaBeF3 and Na2BeF4 and o th er s cla im ed .

Compounds

( 120) ( 166) ( 13U)

--

CWaF>RbF, p a r t ia l stu d y . (135)
LiF-BaF2-MgF2 , rep ort o f lo w e st p o in t as 6145° . ( 83 )
o
LiCl-CaCl2 , sim ple e u t e c t ic rep orted a t 596 . (58)
C) B eryllium F lu o r id e Systems and A nalogs.
BeF2-L iF , BeF2-NaF, BeF2-RbF .

See under item B) .

BeF2-M ixtures w ith a l k a l i and a lk a lin e earth f lu o r id e s .
R e fer s to BeF2- r ic h p o r tio n o f system s where g la s s e s formed
which were th e o b je c t o f th e stu d y .

(68)

BeF2-MgF2 , d e sc r ib e s p a ten ted p r o c ess fo r Mg m etal red u ctio n o f
t h is m e lt to form Be m e ta l.

(8 0 J

BeF2-MgF2 , phase diagram stu d y , u n ce r ta in r e s u l t s .

(172)

BeCl2 - V arious m e ta llic c h lo r id e s (N a, L i , Ag, Cd, B a, Ca, Mg,
Pb, T l)

169

F or NaCl — shows e u te c tic and in c o n g ru e n tly m eltin g compound
Na2BeCl4 .
F or L iC l2 — shows e u te c tic and compound LigB eC ^ .

U n certain

w hether congruent o r in c o n g ru en t.
F or CaCl2 — e u te c tic o n ly .
F or BaCl3 — shows e u te c tic and co n g ru en tly m eltin g compound
BaBeCl4 .
F or Mg312 — u n c e rta in r e s u lts ^ pro b ab ly s o lid s o lu tio n s .
Very e x te n siv e a r t i c l e .

NOTEs

(li;2 )

R ussian in v e s tig a to r s have been p a r tic u la r ly a c tiv e in the study
o f fu sed s a l t sy ste m s.

T heir accumulated data have perm itted

c e r ta in g e n e r a liz a tio n s on the behaviour p a r t ic u la r ly w ith r efer en ce
to th e p o s s ib le ty p es o f compound form ation and t h e ir r e la t iv e
s ta b ilitie s .

R eference may be made to th e work o f Dergunov and

Bergmann (13) (37) in which fu rth e r r e fe r e n c e s may be found to an
e x te n s iv e s e r ie s o f papers .

Other p e r tin e n t r e fe r e n c e s are to

Gromakov (51) and Kryagova (87) .

APPENDIX 2

170
APPENDIX 2

PHYSICAL AND THERMAL PROPERTIES OF CALCIUM FLUORIDE
LITHIUM FLUORIDE AND BERYLLIUM FLUORIDE
Calcium F lu o r id e .
M eltin g p o in t — W hile g iv en as approxim ately 1325-1375° by most
handbooks, more r ec e n t v a lu e s would in d ic a te th a t the o ld e r
v a lu e s were ob tain ed u sin g impure m a te r ia ls .

Probably the

b e s t valu e i s 15.18° obtained by Naylor (116) u sin g a dropc a lo r im e tr ic method.
M o d ific a tio n s — A t r a n s it io n from ac to
accompanied by ab so rp tio n o f energy.

(3

form occurs a t 1151°
(116)

S p e c if ic Heat —K e lle y (79) g iv e s s l i g h t l y d if f e r e n t v a lu es than th ose
given by Naylor (116) which are shown below .
56.900
0
CaF3 ( « ) Cp = 1 5 .3 0 + .00728T *
T2
; Range 298-1525 K.
CaF2 ( @ ) Cp = 25.81 + .00250T ; Range

1525-1691° K.

CaF3 ( l i q . ) Cp = 2 3 .88; Range 1691-1800°

K.

Heat co n te n ts and Entropy V alues - Given by Naylor ( l l 6 ) , fo r the
above r a n g es, and by Todd ( I 69) fo r lower range 52-298 K.
H eats o f T r a n sitio n and F u sio n - As g iv en by K e lle y (79)
A Hl525°K = 1 >1^0 c a l - /mo]re ( t r a n s i t i o n ) .
^ Hl691°K = 7 j l ° 0 c a l./m o le ( f u s i o n ) .
C r y sta l System — Cubic ( f l u o r i t e s tr u c tu r e ) , a = 5.551 A .
R e fr a c tiv e Index — 1.5339 = ftp
Vapor P ressu re and V o l a t i l i t y (137) (97) (5 )
H yd rolysis Data (26)

171

D e n sity o f L iquid Calcium F lu o r id e .

In u n a v a ila b le r e fe r e n c e (108) .

Approximate valu e 3 .0 .
X-ray Data — The l a t t i c e sp a cin g s given by Hanawalt ( 63) fo r powder
d if f r a c t io n p a tte r n s are as fo llo w s :
3 .1 6 A0 *
1 .9 3
*
1 .6 5
*
1.370
1 .2 5 6

1.117
1.050
0 .9 6 8
0 .9 2 5
0 .8 6 6

0 .8 3 5
0 .7 8 9
0 .767
0 .7 3 1
0 .7 1 3
0 . 6I4J4

* S ig n if ie s str o n g e st l i n e s .
P rep aration o f Pure Calcium F lu o r id e .

Stockbarger has d escrib ed

p u r if ic a t io n methods n ecessa ry fo r th e p rep aration o f o p t ic a l
q u a lity m a te r ia l ( 155) (156) .

172

Lithium F lu o r id e .
M eltin g P o in t — Handbook v a lu es are approxim ately 81+0 to 81+5° *
but more r e c e n t d eterm in ation s w ith probably more c a r e f u lly
p u r if ie d m a te r ia l g iv e 81+5° (lir3) (l6 6 ) .
Thermodynamic p r o p e r tie s — K e lle y (7 9 ) g iv e s no data fo r lith iu m
flu o r id e .
C r y sta l System — Cubic (NaCl str u c tu r e) } a = 1+.0173 A° .
R e fr a c tiv e Index — 1.3915 =
B o ilin g P o in t — Approxim ately 1675°. ( 78)
D e n sity o f L iquid Lithium. F lu o r id e , approxim ately 1 .8 (78)
S tr u c tu r a l S tu d ie s — Numerous s tr u c tu r a l s tu d ie s have been made.
See c o lle c t e d r e fe r e n c e s under (1 5 9 ).
X-Ray Data — Powder d if f r a c t io n l a t t i c e spacings given by Hanawalt
(61+) are as fo llo w s :
2.32 A° -x2.0 0
*
1.1+19
*
1.211
1.160
* S i g n i f i e s stro n g er lin e s
V alues obtained in t h i s work fo r LiF are given in Appendix 1 1 .
P rep aration o f Pure Lithium F lu o rid e — The p rep a ra tio n o f pure
lith iu m flu o r id e by treatm ent o f a satu rated s o lu tio n o f lithium ,
carbonate w ith carbon d io x id e to y i e l d the more so lu b le b i ­
carbonate has been d e sc r ib e d .

(157)

The bicarbonate s o lu tio n

on treatm ent w ith h y d ro flu o ric a c id p r e c ip it a t e s the l e s s s o lu ­
b le lith iu m f lu o r i d e } thus affo rding a good sep a ra tio n from

173

the more so lu b le a l k a l i m etal f lu o r i d e s .

An analogous

procedure depending on form ation o f the bicarbonate was
d e scr ib e d by Schmitz-Dumont (lii3 ) .

17h

B eryllium F lu o r id e
M eltin g P o in t

Since t h i s substance i s undoubtedly polym eric i t

has no tru e m e ltin g p o i n t . The customary handbook value i s
o
approxim ately 800 3 bu t so fte n in g a c tu a lly b e g in s by about
750 , a t 800 the mass i s very v is c o u s , and co n sid era b le f l u i d i t y
o
i s not ob tain ed u n t i l about 900 according to the exp erien ce on
t h i s work.

Polymorphic changes are rep orted at tem peratures

below 800 ° } such as i|25° and 528° (120) .
M o d ific a tio n s — A r ep o rt o f a 11c r y s ta llin e " m o d ific a tio n resem bling

oc quartz has been given (120) .

Various oth er stu d ie s claim

g en era l resem blances to d if f e r e n t quartz m o d ific a tio n s.
Thermodynamic data — No v a lu e s are g iv en by K e lle y (79) fo r b erylliu m
flu o r id e .
o
C r y sta l S tru ctu re — Hexagonal oC quartz str u c tu r e w ith a = 1.72 A }
c = 5-18 A0 has been rep orted (120) .
R e fr a c tiv e Index — L ess than th a t o f w ater = 1.333
X -ray and S tr u c tu r a l I n fo r m a tio n — S tru ctu r a l s tu d ie s have been made
by Warren and H i l l (173) j Novoselova^ Zhasmin, e t a l . (120) .
Summarized in form ation on o ld e r work i s given in S tru k tu rb erich t
r e fe r e n c e s (158) .
The powder d if f r a c t io n p a tte rn g iv en by N ovoselova, Zhasmin,
e t a l . fo r th e ir " c r y s ta llin e " m o d ific a tio n i s g iv en below .

The

data la c k s con firm atio n and th ere may be in te r fe r e n c e f~om NaF
or Na2BeF4 a lso p r e s e n t.
a s c a le o f 1 0 .

R e la tiv e i n t e n s i t i e s are in d ic a te d on-

175

o
d ?A

I n t e n s it y

d jA°

3.81+
2.97
2.685
2.365
2.225
2.12
1 .98
1.92
1.83

10
h
6
5
8
7
5
5
6

1.7 7 8
1 .6 5 5
1 .6 2 3
1 .5 3 9
1.5 2 1
1.1+83
1.663
1.623
1.373

I n t e n s it y
7
1
6
7
1
1
2
3
1

d ,A°

I n t e n s it y

1.330
1.317
1.293
1.257
1.237
1.220

2
2
3
3
6
2

P rep aration o f Fure B eryllium F lu o rid e — D iscu ssed in Experim ental
s e c tio n on the p rep a ra tio n o f m a t e r ia ls .

APPENDIX 3

176

appendix 3
SUMMARIZED INFORMATION ON FLUOBERYLLATE'S
AND ANALOGOUS COMPOUNDS
W ith r e s p e c t to the system formed by calcium flu o r id e and lith iu m
f lu o r id e , no compound form ation was a n tic ip a te d .
In th e calcium flu o r id e -b e r y lliu m flu o r id e system , sev e r a l r e fe r e n c e s
in d ic a te d th a t compound form ation would probably occu r.

The compound

calcium, f lu o b e r y l la t e , CaBeF4 , had been m entioned in s e v e r a l fo r e ig n
p a ten ts ( 17) , but the p h y sic a l p r o p e r tie s were not d escrib ed and the methods
o f p rep a ra tio n in d ic a te d th a t the compound was not is o la t e d in pure con d i­
t io n .

S ev era l methods have been given by Rajr (129) fo r the p rep aration

o f the anhydrous compound, one o f which was claim ed to y ie ld a pure p rod u ct,
d e s p ite se v e r a l good reasons why an impure product m ight be ex p e cted .
The p rep a ra tio n s o f the compounds BaBeF4 and Ba2BeF6 have been
d escrib ed in an u n a v a ila b le p u b lic a tio n ( 7 5 ) .

Sarker and Ray claim ed the

p rep aration o f BaBeF4 in an a r t i c le (II4.0 ) which was the f i r s t o f a very
e x te n siv e s e r ie s o f r e p o r ts made by Ray in the German lit e r a t u r e on the
su b ject o f th e gen eral isom orphic r e la t io n between the BeF4
S04

io n .

io n and the

In summary, th e compound CaBeF4 was d e f i n i t e l y expected to

occur in th e calcium flu o r id e -b e r y lliu m flu o r id e m elts in view o f the
thermal s t a b i l i t y in d ic a te d by Ray (129) .

A compound o f the type C a^eFg

was not exp ected to o c c u r .
P r io r to th e appearance o f r ec e n t r e p o r ts on the lith iu m f lu o r id e b erylliu m flu o r id e system , one or two compounds were exp ected to occur
based on the rep o rted e x is te n c e o f compounds o f the ty p es M-^BeXa and M^BeX4 .

177

Numerous c h lo r o b e r y lla te s o f both ty p es have been rep orted as s o lid s and
s t i l l o th ers claim ed in io n ic form ( 36) .

Of the f lu o b e r y l la t e s , the

compound LiaBeF4 had been prepared by Ray (128) and two p a ten t r e fe r e n c e s
(38) (76) in d ic a te d the p rep a ra tio n o f NaJBeFa and K2BeF4 in somewhat
ad u ltera ted forms .

No thermal data was a v a ila b le from th e se so u rces, but

the g en eral ch aracter o f the lith iu m flu o r id e -b e r y lliu m flu o r id e system
could be p r e d ic te d from the a v a ila b le phase diagram due to N ovoselova,
Zhasmin e t a l . (120) on th e sodium flu o r id e -b ery lliu m flu o r id e system .
The l a t e r c o n f lic t in g rep o rts (166) ( I 3U) which were made s h o r tly b efo re
the s t a r t o f a c tu a l work on the system LiF~BeF2 confirm ed the g e n e r a lly
a n tic ip a te d b eh aviou r, except th a t the e x te n t o f the temperature low erin g
was g r e a te r than ex p e cted .

APPENDIX k

178

APPENDIX U
INFORMATION ON ALLOY SYSTIMS

The fo llo w in g r e fe r e n c e s cover th e a v a ila b le in form ation on a llo y
system s p e r tin e n t to the ch o ice o f a m e ta llic cathode fo r t h i s work.
content o f th e r efer en ce i s in d ic a te d by a - b r ie f d e s c r ip tio n .
Calcium A llo y Systems
Ca-Pb, phase diagram, h e a ts o f form ation (88) (U6) .
T h e o r e tic a l ( 1 I4I ) .

P reparation and behaviour (15U)

C1U3) ( 112) ( 13?)
C a-B i, phase diagram , h ea ts o f form ation (90) (88) .
Ca-Zn, phase diagram ( l 6 l ) .
Ca-Na, phase diagram (162) .
Ca-Mg, phase diagram ( 163) .
C a-L i-Pb, d e s c r ip tiv e (6 5 ) .
Ca A llo y Survey (72) .
Lithium Alloy Systems
Li-C a-Pbj d e s c r ip tiv e ( 6 5 ) .
Li-Pb (1 U 8 ).
Li-Pb-Na (1 U 8 ).
L i-P b , phase diagram, a lso numerous L i-m eta l system s ( 1 2 3 ) .
Li-Zn-Mg, phase diagrams (15) .
Potassium A llo y Systems
K-Pb, K -B i, K-Zn, K-Cd, K-Sn, p a r t ia l study (153) .
K-Pb, K-Sn, phase diagrams ( 8I4) .

The

179

K -B ia phase diagram (8 5 ) .
K-Cd, K-Zn, phase diagrams (86) (176) .
Barium Alloy Systems
Ba-Fb, B a-B ij Ba-Sn^ h e a ts o f form ation (89) .
General Survey ( 1 6 ) .

APPENDIX 5

180

appendix

5

GENERAL PHASE RELATIONS IN TERNARY SYSTEMS

S ev era l g e n e r a liz a tio n s used in the d is c u s sio n and in t e r p r e ta tio n
o f experim ental r e s u lt s w i l l be d e sc r ib e d .

The ternary system i s rep re­

sen ted u sin g th e c o n v en tio n a l Roozeboom method which employs an equi­
l a t e r a l tr ia n g u la r r ig h t prism to ex p ress the r e la tio n s h ip o f temperature
and the th re e com position param eters.

For convenience the l i n e s o f two­

f o ld s a tu r a tio n ( a t the in t e r s e c t io n o f two sp a c e-su r fa c e s o f s i n g le - f o l d
s a tu r a tio n ) are shown by p r o je c tio n onto the base o f the prism .

When r e -

qu ired j the b in a ry system s ( i . e . the s id e w a lls o f the prism) are shown
in "fold ed down" p r o je c tio n s along th e edges o f the tr ia n g u la r base o f
the prism .
From the manner o f use and the p r o p e r tie s o f an e q u ila te r a l tr ia n g le
i t fo llo w s th a t th e r a tio o f the components B and C i s a con stan t along
any l i n e con n ectin g component A w ith the o p p o site sid e (BC) o f th e t r i ­
angle j as shown in Figure 1 7 .
As a consequence o f t h is i t i s a lso tru e th a t i f but one s o l id com­
ponent B i s sep a ra tin g in a pure form from a m elt w ith the com position
the com position o f the m elt w i l l fo llo w the s tr a ig h t li n e mx m2 , e t c .
u n t il a d if f e r e n t s o lid phase appears, as a t p o in t p in Figure 1 8 .

Here

the occurrence o f a new s o lid phase C cau ses the system to become u n i­
v a r ia n t and th e com position o f the m elt fo llo w s the l in e o f tw o -fo ld
s a tu r a tio n e t w ith sim ultaneous d e p o s itio n o f B and C u n t il the ternary
e u t e c t ic p o in t t i s reach ed .

Here s o lid A commences to form and the

181

FIGURE 17

e

FIGURE 16

e

FIGURE 19

182

system i s in v a r ia n t u n t i l the liq u id phase v a n is h e s .

The arrow heads

denote th e d ir e c t io n o f lower tem p eratures, and the th re e bin ary system s
are in t h i s case assumed to be sim ple e u t e c t ic on es.
Any system c o n s is t in g o f two phases has a t o t a l com position rep re­
sented by a p o in t on the s tr a ig h t l i n e connecting th e com position s o f
the two phases ( e it h e r s o l i d or l i q u i d ) .
In the even t o f the occurrence o f a b in a ry congruently m eltin g com­
pound D in th e b in a ry system. AB o f Figure 1 9 , se v e r a l c a se s are p o s s ib le .
The o th e r two b in ary system s are considered to be sim ple e u te c tic o n e s.
1 .) S e c tio n CD i s q u a si-b in a r y .

I f a v e r t ic a l s e c tio n CD through

th e tern ary system e x h ib its a l l the c h a r a c t e r is t ic s o f a b in ary
system o f C and D the s e c t io n i s c a lle d q u a si-b in a ry and the
e n tir e system may be tr e a te d as two independent tern ary system s
ADC and BDC.

I t can be shown th a t the tern ary e u t e c t ic p o in ts

f and g must l i e on o p p o site s id e s o f the l i n e CD and th a t
p o in t h , r ep r e se n tin g the lo w est p o in t on l in e CD ( e u t e c t ic o f
the b in ary system o f C and D ), i s a p o in t from which the
temperature d ecreases In p a ssin g from h to e ith e r g or f .
A lso any m elt w ith an o r ig in a l com position ly in g on l i n e CD,
such as mx , w i l l s o l i d i f y to form C and D o n ly , in a normal
b in ary system o p e r a tio n , w ithout departure o f the m elt compo­
s i t i o n from the l i n e CD.
2 .) S e c tio n CD i s not q u a si -b in a r y .

(See Figure 20. )

I f th e s e c tio n CD does not have th e normal c h a r a c t e r is t ic s

N o te :

P a g e s 183 and I 8I4 a r e n o n - e x i s t e n t due to n u m b erin g e r r o r .

FIGURE 20

186

o f a b in ary system , th en i t may be shown th a t the p o in ts g and
f l i e on the same sid e o f li n e CD and th a t one o f them i s a
p e r i t e c t i c and th e oth er a tern a ry e u te c tic p o in t.

Again the

system s A and C, and B and C, are co n sid ered as sim ple e u te c tic
o n es.
In t h i s ca se the two p a r ts o f the system are not independent
and the behaviour o f d if fe r e n t m elt com positions may be d escrib ed
as f o ll o w s .
A m elt o f com position m^ w i l l fo llo w along lin e CD w ith
d e p o s itio n o f s o lid C u n t i l p o in t h i s reached.

S o lid A w i l l

then a ls o s t a r t to d e p o sit and th e m elt com position w i l l fo llo w
hg w ith th e l a s t o f th e m elt s o lid if y in g ju s t as p o in t g i s
reach ed .
A m elt w ith in the area CihC w i l l f i r s t d e p o sit C , then
fo llo w along ih g d e p o s itin g s o li d s C and A , and w i l l s o l i d i f y
b efore reach in g p o in t g , somewhere between h and g.
A m elt w ith in the area A jlhiA w i l l fo llo w a s im ila r p r o c e s s ,
f i r s t w ith d e p o s itio n o f A , and th en w ith d e p o s itio n o f A and,
C (o r A and D) along li n e ih g (o r j ig ) depending upon the ex a ct
r e la t io n o f the e u te c tic l i n e s .

Again s o l i d i f i c a t i o n w i l l

occur betw een h and g (o r 1 and g ) , p r io r to reach in g p o in t g.
A m elt w ith in area DljD behaves an alogou sly to one in the
area CihC .
A m elt w ith in th e area high w i l l f i r s t d e p o s it s o lid A,
and su b seq u en tly w i l l fo llo w along hg (o r lg ) d e p o s itin g C and A

187

(o r D and A) u n t i l p o in t g i s rea ch ed .

Then new s o lid phase D

(o r A) appears and the m elt flr e a c ts ,f to use up A a t constant
tem perature w ith d e p o s itio n o f D u n t i l A v a n is h e s .

The m elt

th en d e p o s its C and D alon g l in e g f u n t i l th e m elt com position
reach es f , the tern ary e u t e c t ic p o in t.

T here, com plete i s o ­

thermal s o l i d i f i c a t i o n occurs w ith d e p o sitio n o f C and D and
th e new s o lid phase B.

The amount o f B formed i s a fu n c tio n

o f th e lo c a t io n o f the o r ig in a l m elt w ith in the area h ig h ,
m e lts near th e li n e h i forming alm ost no B a t p o in t f .
M elts w ith in oth er areas req u ire no d e ta ile d ex p la n a tio n
ex ce p t to say th a t m elts in th e area ChgC (o r DlgD) w i l l undergo
a p e r i t e c t i c r e a c tio n a t g subsequent to the primary d e p o s itio n
o f C (o r D) and secondary d e p o sitio n o f C and A (o r D and A ),
w ith isoth erm al disappearance o f A at p o in t g .

T h erea fter the

m elt proceeds to p o in t f where B , C and D phases d e p o sit i s o th erm a lly w ith com plete s o l i d i f i c a t i o n .
Two g e n e r a liz a tio n s are i l l u s t r a t e d by the above behaviou r.
(F igure 20) .
1 .) A m elt

havin g an o r ig in a l com position lo c a te d w ith in the t r i ­

angular area ACDA s o l i d i f i e s w ith form ation o f A , C and D o n ly ,
th er e b ein g no p o s s i b i l i t y o f d e p o s itio n o f B , sin c e then the
com position o f the s o lid r esid u e would be e x p r e s s ib le o n ly by
a lo c a tio n w ith in the tr ia n g le BCDB.

T h is, however, i s impos­

s ib le s in c e the o r ig in a l m elt and f i n a l s o li d resid u e must
have th e same com position lo c a t io n .

188

S im ila r ly m e lts o f o r ig in a l com position w ith in the t r i ­
angle BCDB show f i n a l l y o n ly s o li d s B , C and D in the f in a l
s o l i d r esid u e .

2 .) From th e r e la t iv e behaviours o f F igu res 19 and 20 the g e n e r a liz a ­
tio n may be drawn th a t no l i n e o f two f o ld s a tu r a tio n may p a ss
a c ro ss a q u a si-b in a ry s e c t io n .

Thus in F igure 1 9 , the lin e s gh

and fh are two d if f e r e n t space curves w ith a common terminus
a t h , but in F igu re 2 0 , l i n e s ih g and j i g are continuous space,
curves which do c r o ss the q u a si-b in a ry s e c tio n CD.
I t may be asked whether I t would not be p o s s ib le fo r the binary
e u t e c t ic s on e ith e r s id e o f compound D to jo in a t a p e r i t e c t i c p o in t or
e u te c tic p o in t b efo re in t e r s e c t in g the l i n e s o f tw o -fo ld sa tu r a tio n from
the oth er two bin ary system s AC and BC, as shown in F igu res 21 and 22.
Such a c o n str u c tio n i s im p ossib le in ca ses where no s o lid s o lu tio n s are
formed fo r th e fo llo w in g rea so n s.
I t w i l l be observed fo r both fig u r e s th a t m e lts having the o r ig in a l
com position m1 w i l l in g en era l f i r s t d e p o s it s o lid D, then w i l l reach
li n e kg and p a ss down along th a t lin e w ith d e p o s itio n o f s o lid s D and B
u n t i l p o in t

g i s reached a t which the new s o lid phase A appears .

case o f Figure 22

In the

where a tern ary e u t e c t ic i s in d ic a te d at g , th e remain­

in g m elt o f com p osition g would s o l i d i f y to form s o lid s A, D and B having
a f i n a l com position equal to the o r ig in a l com position o f the m elt mx .
w i l l be noted th a t the s o lid mass o f com position mx does not l i e w ith in
a tr ia n g le whose v e r t ic e s are the s o l i d phases making up the s o l i d i f i e d

It

189

e

— IMPOSSIBLE
CONSTR UC T IO N S

e

FIGURE 22

190

m ass, i . e . w ith in a tr ia n g le ABD.
com position

in terms o f

Hence i t i s im p o ssib le to ex p ress the

B and D and the s it u a t io n o f F igure 22 i s

im p o ss ib le .
In F igu re 21, where a p e r i t e c t i c p o in t i s in d ic a te d a t g , i t w i l l
be noted th a t the shape o f the area gADBg (formed by con n ectin g the
in v a r ia n t p o in t g w ith th e s o lid phases p resen t a t the p o in t g) i s not
a quadrangle w ith no r e -e n tr a n t a n g le s .

I t may be shown however th a t

such a form i s n ecessa ry fo r a p e r it e c t ic r e a c tio n , and hence th e con­
s tr u c tio n o f F igu re 21 i s im p o ssib le .
In th e event th a t s o lid s o lu tio n s are formed, e ith e r a p e r it e c t ic
or e u t e c t ic r e a c tio n may occur depending on the e x te n t o f th e reg io n s o f
s t a b i l i t y o f the s o l i d s o lu tio n s oc , (3 and
and 2k.

& as shown in F igu res 23

P o in ts oC , (3 and 8 r ep resen t the com positions o f s o lid s o lu ­

tio n s o f A, B and D which are in equilibrium w ith the m elt o f com position
g. (These are not the same fo r m elts o f d if f e r e n t co m p o sitio n s.)

It is

seen th a t in the p e r i t e c t i c case o f F igure 2 3 , the r e a c tio n a t g i s one
in which s o lid s o lu tio n

& i s used up in th e eq u ilib r iu m ,

8 +melt g <—- oC +@ .

Masing (131) terms the r e a c tio n at p o in t g in F igure 2k a '‘e u t e c t ic ’1
one o f th e type 8 ~7~*~ m elt g + <*. +
s o lu tio n

, and fu rth er s t a t e s th a t s o lid

§ cannot be in eq u ilib riu m w ith th e m elt g a t tem peratures

below th a t p o in t:

The system

ot8(3goC

i s one o f a quadrangle c o n ta in in g

a r e -e n tr a n t a n g le , so th a t the r e a c tio n cannot be one o f the u su a l p e r i ­
te c tic ty p e .
The above j u s t i f i c a t i o n o f Figure 2k lea d s to no im p o ssib le ca ses
provided the fo llo w in g in te r p r e ta tio n i s a p p lie d .

M elts

having o r ig in a l

191

FIGURE 2 4

<*-

FIGURE 23

192

com position s to th e r ig h t o f l i n e

<x&$

s o l i d i f y b efo re reaching p o in t

g , or a t th a t p o in t any r e s id u a l s o lid D r e a c ts to form oc + ft + &
I f , however, th e o r ig in a l m elt l i e s w ith in the area

5(3gOt

th ere may

remain a t g ( a f t e r co n clu sio n o f the r e a c tio n ) one o f th e fo llo w in g com­
b in a tio n s j ( l )
m elt

g .

OC + S + m elt

g ,

(2 ) oc + p> + me l t g , (3 )

P

+ £

+

I t must be assumed, th a t i f any S rem ains, i t ta k es no fu rth e r

part in any r e a c tio n according to the Masing in t e r p r e ta tio n .
must then p a ss along the l i n e g f w ith form ation o f oc and P

The m elt g
, u n til i t

reaches the tern a ry e u te c tic p o in t f , where component C i s a lso d ep o sited
as such or as a s o lid s o lu tio n (n o t shown).

I t i s apparent from the

above d is c u s s io n th a t i f th ere i s a lim it in g boundary fo r s o lid s o lu tio n s
of D, th ere i s no p o s s i b i l i t y th a t p o in ts g and f may be ternary e u te c tic
and p e r i t e c t i c r e s p e c t iv e ly .

In stea d p o in t f n e c e s s a r ily must be the

tern ary e u t e c t i c .
The above arguments fo r a system e x h ib itin g an "open maximum1'
(co n g ru en tly m e ltin g compound D) are e q u a lly a p p lic a b le to th e case o f
the occurrence o f an in con gru en tly m eltin g compound in the system AB .
Again i t i s im p ossib le th a t th e two l i n e s o f tw o -fo ld sa tu r a tio n may meet
b efore f i r s t encountering some oth er l in e o f tw o -fo ld s a tu r a tio n , except
fo r th e case o f e x te n s iv e s o l id s o lu tio n form ation such as th a t d escrib ed
above.
M ention should be made o f a method o f study which perm its a quicker
d eterm ination o f th e type o f system p r e se n t, provided circum stances are
fa v o r a b le .
(1 0 6 ).

T his i s the "Klarkreuz"' method due to G uertler (59) (131)

F igure 25 shows a system in which two b in ary compounds D and S

FIGURE 25
CD OR BE MAY BE QUASIBINARY, BUT NOT BOTH.

19^

occu r, one in system AB and th e oth er in system AC.
s i s t s o n ly o f c o o lin g s o lid phases

I f the system con­

having the o v e r - a ll com position o f

the o r ig in a l m e lt, th e phase r u le fo r a condensed system s t a t e s th a t
on ly 3 s o lid p h ases may be p r e se n t fo r a 3 component system , i f the
temperature i s s t i l l a v a r ia b le .

That i s

F = n - P + 1 = 3 - p +1=1
and P = 3

fo r F = 1 = temperature degree o f freedom.

C onsequently w ith in th e s o lid m ixture o n ly th ree s o lid phases may
c o e x is t .

P o s s ib le com binations are obtained by c u ttin g the system in to

p a r t ia l system s by th e l i n e s ED and EB, or l i n e s ED and CD.

From th ese

d iv is io n s o f the system , th e fo llo w in g com binations are p o s s ib le fo r
the th r e e c o e x is t in g s o lid p h a s e s .

(l)

A+D+h
B +C + E
B+D+E

and

(2)

A+D+E
B +C + D
E+C+D

I f o n ly l i n e s CD, or BD, were drawn, p o s s ib le com binations would be
the fo llo w in g :
(3 )

A+C+D
B +C + D

/vv
^

A +B + E
C +B + E

I t w i l l be noted th a t according to the p a r tic u la r com bination chosen,
c e r ta in s o lid phases are or are not capable o f c o e x is t e n c e .

These are as

fo llo w s :
Combination (1 )

E + B can c o e x i s t ,

A + Ccannot.

Combination (2 )

C+ D can c o e x i s t ,

E + B cannot,

Combination (3 )

C+ D can c o e x is t ,

E + B + A can n ot.

Combination (U)

E + B can c o e x i s t ,

A + C+ D cannot.

195

The G uertler method c o n s is t s o f preparing a m elt w ith the com position
X (F igure 25) which i s common to th e v ariou s p o s s ib le ways o f d iv id in g
the system .

The m elt i s allow ed to s o l i d i f y and the c o e x is tin g s o lid

phases are determ ined.

According to the p o s s ib le com binations i t i s p os­

s ib le to e lim in a te unallow able methods o f d iv id in g the system in to sm aller
system s which may then be tr e a te d as d is c r e te system s.

T h is does not

n e c e s s a r ily mean th a t the q u a si-b in a ry s e c tio n s are determ ined, but rather
th a t the method d is c lo s e s where th ey are not p o s s ib le .

In t h is r e sp e c t

the method i s e n t ir e ly g e n e r a l.
For system s in which ADB and ABC bin ary system s show o n ly sim ple
e u t e c t i c s , the method c o n c lu s iv e ly d ecid es whether CD or BE i s a q u a si­
binary s e c t io n , s in c e in the two c a se s o n ly two s o lid phases w i l l be formed,
e ith e r C + D} or B + E.

I f one i s a q u a si-b in a ry system then the oth er

cannot be q u a si-b in a r y , sin c e two such system s can never cro ss each o th e r .
Here th e c o e x is te n c e o f C and D (o r B and E) s o lid phases means th a t CD
(or BE) i s a q u a si-b in a ry s e c t io n .
When the b in ary system s show any p e r it e c t ic r e a c tio n s , the method i s
not as c o n c lu siv e fo r two r e a so n s.

The f i r s t i s t h a t , p r a c t ic a ll y , p e r i­

t e c t i c r e a c tio n s may not go to com pletion and d e f in it e r e s u lt s may not be
o b ta in ed .

The second reason i s th a t n e ith e r CD, nor BE, n e c e s s a r ily has

to be a q u a si-b in a ry s e c t io n , so th a t the method can o n ly a t b e s t say
th a t one or the oth er i s not a q u a si-b in a ry system .
Other g e n e r a lit ie s which are a p p lic a b le to tern a ry systems are the
fo llo w in g .
For th o se system s, in which a l l the b in ary system s show one in v a r i­
ant p o in t , th ere e x i s t s but one tern a ry in v a r ia n t p o in t.

196

For each a d d itio n a l in v a r ia n t p o in t in any o f the binary system s
th ere e x i s t s another tern ary in v a r ia n t p o in t .

(Such an a d d itio n a l

in v a r ia n t p o in t in th e binary system a r is e s from the occurrence o f a
congruent or incongruent compound, so th a t in the case o f the form er, i t
should not be counted as an in v a r ia n t p o in t fo r t h i s purpose) .
An example would be th a t o f F igure 25 where as a consequence o f the
e u te c tic behaviour o f the binary system s ( in which f iv e in v a r ia n t minimum
p o in ts a p p ea r), th ree in v a r ia n t p o in ts must occur in the tern a ry system ,
o f which a t l e a s t one must be a tern ary e u t e c t i c .

The oth er two may both

be p e r i t e c t i c , but i f one o f them i s ternary e u t e c t ic , then the other i s
n e c e s s a r ily tern ary e u t e c t i c .
are:

Thus the on ly p o s s i b i l i t i e s in t h is case

(1 ) One tern ary e u t e c t ic , two ternary p e r i t e c t i c p o in t s ,

or

(2 ) Three tern ary e u te c tic p o in t s .
I f th ree tern a ry e u t e c t ic s are shown, then two q u a si-b in a ry s e c tio n s

are p r e s e n t.

I f a tern ary p e r i t e c t i c p o in t o c cu rs, no q u a si-b in a ry s e c tio n

i s p o s s ib le fo r t h i s

(F igure 25) ternary system , sin c e another p e r i t e c t i c

p o in t must a ls o be p r e s e n t.

As a consequence the lin e s o f tw o -fo ld sa tu r ­

a tio n (le a d in g to each p e r it e c t ic p o in t) cut across th e two s e c tio n s p r e ­
v en tin g them from bein g q u a si-b in a r y .
A s p e c ia l ca se may be mentioned in which the normal order o f s o l i d i ­
f i c a t i o n p r o c e s s e s w i l l not fo llo w the order o f:

( l ) primary c r y s t a l l i z a ­

t i o n s , (2 ) p e r i t e c t i c r e a c t io n , ( 3) secondary c r y s t a ll i z a t i o n and f i n a l l y
(1+) ternary e u t e c t ic s o l i d i f i c a t i o n .

With referen ce to F igure 2 6 , assume

th a t the bin ary system AB forms compound D w ith an incongruent m eltin g
p o in t .

I f th e two fo ld s a tu r a tio n l i n e s hk, ek, k t , tg and f t meet in

197

i___

FIGURE

26

198

the manner shown, th e p o in t k r ep r e se n ts a tern ary p e r i t e c t i c and p o in t
t a tern ary e u t e c t i c .

A ll m e lts in the tr ia n g u la r area ACD w i l l th e r e ­

fo re s o l i d i f y on reach in g p o in t k .

M elts in the reg io n jkd show the

normal primary c r y s t a l l i z a t i o n , then a p e r it e c t ic r e a c tio n at k , and
then secondary c r y s t a l l i z a t i o n along l i n e k t to f i n a l s o l i d i f i c a t i o n at
the tern a ry e u t e c t ic t .
M elts in the tr ia n g u la r area kde show a somewhat unusual order w ith
r esp e c t to ty p es o f c r y s t a l l i z a t i o n .

A m elt o r ig in a lly a t com position X

f i r s t d e p o s its s o lid A and the m elt com position fo llo w s l i n e H to an
in t e r s e c t io n w ith the l i n e o f "binary11 p e r i t e c t i c r e a c tio n ek .

A p e r i­

t e c t i c r e a c tio n occurs in which p r e v io u sly d ep o sited A r e a c ts w ith m elt
to y i e l d compound D w h ile th e m elt com position fo llo w s the p e r it e c t ic
tw o -fo ld sa tu r a tio n l i n e ek .

T his r e a c tio n i s not isoth erm al as o n ly two

s o lid s and the m elt are p r e s e n t.
the system i s u n iv a r ia n t.

S in c e , F = n - P + 1 = 3 _ 3 + 1 = 1^

Upon reaching p o in t z , however, the system

becomes b iv a r ia n t because the s o lid A has been exhausted by the p e r i t e c t i c
r e a c tio n and th e system now c o n s is t s o f s o l i d D and m elt z , the o r ig in a l
com position X ly in g on the s tr a ig h t lin e jo in in g z and D.

As the system

has become b iv a r ia n t, the com position o f th e m elt i s no longer r e s t r ic t e d
to fo llo w in g l in e e k , and the m elt d e p o s its s o lid D w h ile proceedin g to
i n t e r s e c t e ith e r l i n e k t or f t , where secondary d e p o sitio n o f two s o lid s
o c c u r s, and f i n a l l y the tern a ry e u te c tic p o in t t i s reached.
The p o in t il l u s t r a t e d i s th a t m e lts ly in g in e x c e p tio n a l a r e a s, such
as the one shown above, may e x h ib it somewhat unusual behaviou r.

Here the

s iz e o f area kDe was p u rp o sely ex a g g era ted , but i f i t were decreased so

199

th a t the circum stances resem bled th o se shown in Figure lU i t can be seen
th a t m e lts c l o s e l y r e la t e d in com position would d if f e r in thermal be­
haviour in an ap p aren tly anomalous manner.

200

C aFp
C C

BeF-

L iF

FIGURE 14
D ISTO R T ED VIEW
TERNARY SYSTEM
CaF~— L i F — B e F p

APPENDIX 6

201

APPENDIX 6
SAMPLE NUMBERING SISTEMS
The m e lts in v e s tig a te d in th e binary CaF2-LiF and CaF2-BeF2 system s
were id e n t if ie d in a g e n e r a lly sim ple fa s h io n .

In th e former ca se th e

m elts were I d e n t if ie d by e ith e r the approximate mole p ercen t CaF2 , or by
l e t t e r s A, B , C, e t c . s ta r tin g from near th e LiF s id e o f the system .
One s p e c ia l s e r ie s o f r e la te d m e lts near the e u t e c t ic o f t h is system was
id e n t if ie d by 1CL, 2CL, e t c .

S im ila r ly in the CaF2-BeF2 system th e m elts

were i d e n t i f i e d as A, B , C, e t c . s ta r tin g near the BeF2 sid e o f th e s y s ­
tem.

One s p e c ia l s e r ie s o f r e la t e d m e lts near the cen te r o f t h i s system

was id e n t if ie d by l a , 2 a , e t c .

Thus in th e s e two system s the gen eral

r e la t iv e m elt lo c a t io n i s in d ic a te d by the a lp h a b e tic a l seq uence, or by
the approximate percentage f i g u r e .
The same g en eral method was used on the LiF-BeF2
LiF s id e .

s y s te m ,s ta r tin g a t the

Due, however, to th e g rea t number o f m e lts stu d ied in t h is

system , o th er d e sig n a tio n s such as MM, LL, e t c . became n e c essa ry to in d i­
ca te m e lts in the v i c i n i t y o f m e lts M and L e t c .

In a d d itio n a la r g e

number o f s e r ie s o f r e la t e d m e lts were stu d ied in the L i2BeF4 reg io n o f
t h is sy stem .

These m e lts were d esig n a ted as I I I , 211, e t c . ; IKK, 2KK,

e t c .; IR, 2R, 3R, e t c . ; I d , 2d , e t c . ; l e , 2 e , e t c .

In t h i s same r eg io n

a la r g e number o f m e lts d esig n a ted by IQ, 2Q, 3Q, e t c . were stu d ied which
were independent (u n r ela ted ) m e lts .

Because o f th ese co m p lica tio n s i t i s

necessary to c o n su lt th e phase diagrams or ta b u la ted com p osition s in the
appendices to be c e r ta in o f the r e la t iv e com position lo c a t io n s .o f th e s e m e lts .

202

W ith in the tern ary CaF2-LiF-BeF2 system , a l l m e lts are d esig n a ted
by id e n t if y in g symbols which term inate in a num erical s u f f i x .

The f i r s t

p o r tio n o f the symbols i d e n t i f i e s the binary parent m elt from which the
tern a ry m elt was d e r iv e d .

The num erical s u f f i x in d ic a te s the r e l a t i v e

order o f th e tern a ry m elt in th e s e r i e s d erived from th e bin ary m e lt.
Thus, fo r exam ple, A -l i s th e f i r s t tern a ry m elt d eriv ed from m elt A ( o f
the LiF-BeF2 system ) by th e a d d itio n o f CaF2 .

A-2 would be the second

tern ary m elt o b tain ed by fu r th e r a d d itio n o f CaF2 .

As a more com plicated

example; 12Q-5 would s ig n if y th e f i f t h tern ary m elt ob tain ed from bin ary
m elt 12Q o f the LiF-BeF2 system .
A ll tern a ry m e lts were d eriv ed from bin ary LiF-BeF2 m e lts , ex cep t
fo r two s p e c ia l s e r ie s d esign a ted t - 1 , b-2 e t c . and g - 1 , g -2 , e t c .

A ll

ternary m e lts w ith the ex cep tio n o f the b , g , and Q s e r ie s were s ta r te d
in to th e tern ary system by th e a d d itio n o f CaF2 to LiF-BeF2 m ix tu res.

A ll

ternary system m elts d esig n a ted by upper case l e t t e r s such as A - l , MM-5,
e t c . were c a rr ie d a c ro ss the system by the a d d itio n o f CaF2 o n ly , w ith
the e x c e p tio n s o f th e Q and S s e r i e s .

The l a t t e r and a number o f th o se

d esign ated by lower ca se symbols such as p , t , e t c . were s h if t e d along
s e c tio n s by th e a d d itio n o f m ixtures or o f oth er components, as more p a r t i­
c u la r ly d escrib ed in th e d is c u s s io n o f procedure and r e s u l t s .

APPENDIX 7

203
!k>

©

rH

co

73
• •H
a © o
o o ©
-p Pr
CO © CO
p p>
XJ
©
p p © ©
© > H
PI -p O P .
-p co d
o © © c3
m d © co

-p
a
©

O
O

-P

o
o

X3

©
I—I
O
o
o

Oo

•H O

(Note

RESULTS OF THERMAL ANALYSES OF CaF2-LlF SYSTEM
- See footnotes for sig n ifica n ce of a b b r e v ia tio n s.

-P

XJ
©
!>
O
d
©
I©

CM
I
CM(A I OK I CM
c—r— c—p*- i p— i c—

CM

I

*

o •
0) Oh]
13
£
0 (D
(xj

73
•H
o
©
Ph
W

I CK C CM CA !—1 I CM
imo o t '- k - I c—

c— c—

c—p— I c

-p
©
©
Ph
©

rH

P— OK

ir—c-^— c— e—c— I c—

©
H
p,
d
©
co

I

I CM

I CM CM CO CM CM
I P - CK- O - K - C—
I P— C— C— C— P—

I C—
I r—

I c—

©©
o

-p
ra
>5 *
u O
o o
•V
Ik , •
P &
© d
s ©
•H EH
P
P-t

page

CO

next

o

rH

CO

<A
&
CK
p.. I
(>.
I I
I
CMCMH O\r\co © \A O
CA © \0 p- ia cm P 'O n c o
OOC
—C—QOCOCKCKp—
'L
A'LA-©' rA CMo rH CKCO C~- IK- □a
co c—r—r—c—c—c—c—
C O C O C O c O O D f f i C D lS - [ ^ C ^ - lS

O O
1A 1A

H
s

§

o

po CMAO

<3
CM

N H H O
CO CO QO CD

rt
o

CO

©
O

ph a
o o
Xi -P

2 °©
Pi
- P -P
© ©
a o
Ph

©

O

C

O

C

O

C

CO
C O

O

>

CO
C O

>

CO
> >

C O

>

C O

CO
! >

oj

S In
O
ro
OO
X! 'feR
o ©
I—I 1—I
© o
Os

rH

C

'

ds
CO
©
•p
©
Q

O• O* O• Os'O
• • H
• K
• \0
• •o • •
O o cpix\oo rH fAU\co co CK
r l H i—I I—I i—I rH

O O O C O V N O O O

« « » • » • • •

PA m

O O O O O O

O O O O i—t CM CM CM
CM CM CM CM CM CM CM CM

CM m
CM CM

-© ©•-© 'lA'LA'LA

O

rH
Pi PI
I I <; PQO H Q U \ c b £3 CN
I I
f—I i—Ii—I i—I

=

c
-

P I PI fH PI
O Q O O =

P O O O P -

mO

h i a c m

CK CK
C K -© -©
—© W
i— I i— I i H

...........................................

' v0

C~— (—t CO GO CK i—I £ K j H (—1
"

-

■ ~ M3

i—
I rHM3 M3

PJ CA
O •
CM rA

^

rH. rH rH

CM

CK CK CK CK CK

CK
-©

p—UK1A"LT\UKC
—
« x 'v0

CM CM

-© - © , © - ©

i— |

CK rH rH " \ ^ \ " ' ' ' O K - xH
'0

_

C A - © © ■ P t, ©J TAv

CM

CK
CK CKCK
C K _ p CK CK C K -© - O n _©-_©*
—
©"k ^_©—
©©
i—| i—I i—I
\ n \ \ \ o \ ^ o k- u \ xafA
rH i—I i—I

• » » • • •

CM CM CM CM CM CM

©

©©
o o

i—I

• •

MO MO M3

r—I <—I i—I MO MO MO MO i—I

i

A

K

K

O

H

K

'O'1—^
''K»l—II—| I—| I—I I—I
MO

| rH rH r l H

Continued

|S

2Oil
to
©
rH
Ph
E
cd
to

co
-P
P
CD

.

i—1 d
cd ©
•H >
o Q
© E
Ph©
co P

O
O

•
rH
AJ
o
-p g
■H E
©
E d
©
-p P
ra p
P "H
•H -P
Cp ©

d
<D
-P
•H
K>
O
Ph
■H CD
iH d
rH
cd PI
-p o
m co

C\!
%

o

-P
©
©
Ph
©
ft

CM

CD
p
p
Ph

£? ©
o !>

t—
o o

•HO
-P .
O '

CD PJ

p t)

cd

On

I

ipco
i \o c - n c - H
i o- c-~ o- i>-1>- [ > - 0 i ©- c—i>- i>- r— [ > - IN -

I CM 1
I p- I
I p- I

I

I

I

I I I
I I V

cd
o

e- i

00 4
>s '

PJ

OF

THERMAL

A N A L Y SE S

OF

C a F 2 -L lF

SYSTEM

-

C o n tin u e d

1 S

RESULTS

P
i—! rH O
cd - P
•H
■P
-P cd
co
cfd
•H
©
to
o
£?g
-p
ft
cd
E
rH O
o
CM ©
d
P
cd to ©
£ *H P •a <H
•H W
o
o

i—I

•P
to

P

E to
•H -P
P rH
Ph©
S
P PH
O O
•H
-P -P t-H
cd P
N •H 8
•H O
rH P h ©

ON

op
UN CM

O

O

I I
"LA H Q
T-P'-O m5

r—

CO
on

CO CO CD OO QO CO

CO
ON
co

I
CM C p CP CO c p i p i p H r p
f ip P H -J fP H C O H
On On On O O O H iH - P

cd
o

•S ^

ft

I—I I—I I—II—II—I I—I
cd CCS
ctf

cd
o

o

o

o

O d
cd o ©
-p
o
•ri
O
p ©
-H
o O
•H p .
P
-p ©
•H
cd d
O
P h© N
. . ‘H -P
P h -P rH •H
o
rH
cd ts
■P o
P
co i—l
o
• H •p f>» ©
-P P P £>
cd •H o
>
©
P P h rH co
©
Q
CO P cd X!
P • , ■P
o •H *H
-p p
I—I o p
cd ©
P H «H
co
O £
•H
•H ©
rH
co CO
© <D -P cd
■H
•H
«H Ch fi -p
•H
P
t
bo S )-p •H
■H *H Ph
CO CO *H (xh
P h rH

Ph P
o o
•H
d -p
o o
,P CD
-P -P
CD CD

&
03
e it

to

i> !> f> f=* CO !>

O CD

On O

d

'feA

UNMOc—On rp op
CMCMCMCMCp op

o
H
CIS
O

CD
rH
O
5!

q o

©
rH
Pi O
g IS
CO

CD
Id
n

O

O

cpcp

CO t o

op cp
CP c p

cp op

t

o

c

CO CO
o
>
>

c o
>

nO

F - On P3 CO
CM CM CM

On On On On On
Ip M D c O fO
i—1 i—I

CM
rH

rH rH i—I

MD!i—I i—I i—I i—Ii I

H Hd

O n On

>

>

o

O O O O O O O O O
• • • • • • « • •
c o o o o p p -o O Q O O

n c p j - j - i p i p \ D p- o

i—i

cd
o

^
3

>

CO
c

cd
o

©
coP P O O O O I
op cp_p - A t ip i p N O P - I

>

C O

H

CM r p

O nO nO N O nO nO nO nO n
1
n A n t- H f On
CM - d —P

i—I i—I
i—I

_ d " L p ip ip "

,

i—Ii—I r H i —I H i—I i—I
i—Ii—I r H i —I i—I i—I rH

co

©

-P
o

CO

!>s
-p
P
-H
Cd
-P
p
©

o
§

i—I
cd
S
P
o
P
P
cd
^
-P

©
P

I
CO

PH
fciD
•H
co

cd

APPENDIX 8

205

APPENDIX 8
EESUITS OF THERMAL ANALYSES OF CaF2~BeF2 SYSTEM

In l i s t i n g the r e s u lt s o f th e se thermal a n a ly ses the com positions
given in column 2 r ep r e se n t c a lc u la te d v a lu e s based on in p u t to the m e lt s .
The m e lts analyzed subsequent to th e t e s t s are in d ic a te d by a sample
l e t t e r in column 3 and an accompanying a n a ly t ic a l r e s u lt i s given in
column k .

The oth er com position v a lu e s given in column 5 in square

brack ets r ep re sen t in te r p o la te d v a lu e s based on s ta r tin g co m p o sitio n s,
a n a ly t ic a l r e s u lt s and observed w eight lo s s e s during the t e s t s .

I f an

a s te r is k i s shown in column II, the com position value i s th a t o f the main
p art o f th e m e lt, a d d itio n a l a n a ly se s having been made on oth er samples
o f m a te r ia l on the w a lls o f the c r u c ib le , e t c .

(See A n a ly tic a l Appendix

15, P art H) .
The d e s c r ip tio n o f the thermal behaviour shows the temperature o f
i n i t i a l d e p o s itio n o f CaBeF4 or CaF2 in columns 5 and 6 , a c tu a l o b serv a tio n
o f d e p o s itio n b ein g in d ic a te d by an X in column 7.
In column 8 the i n i t i a l tem perature o f p la te a u s i s shown,with a
d e s c r ip tio n o f th e type in column 9 , i . e . sh o rt f l a t , sh ort s lo p in g , lon g
f l a t , lo n g gra d u a lly s lo p in g .
In column 10 the tem perature o f disappearance or appearance o f
apparent liq u id i s given accompanied by C or H to in d ic a te c o o lin g or
h e a tin g .

In c a se s where the i n i t i a l c r y s t a ll i z a t i o n was not o b served ,

therm al i n f l e c t i o n s were ob served .

206

The i n i t i a l s tu d ie s were made in 1950.

The second t r i a l s which were

co n sid era b ly b e tt e r in g en era l accuracy are rep resen ted by the un d erlin ed
s e r ie s o f "a” samples o f th e p e r io d February 13 to 1 5 , 1951.
A number o f th e e a r lie r runs on m e lts contam inated by BeO are tabu­
la te d even though t h e ir low v a lu e s are o b v io u sly d isc o r d a n t.

These

samples are in d ic a te d by t r i p l e q u estio n marks (???) in column I;.

A s in g le

q u e stio n mark in d ic a t e s a la r g e r than normal u n c e r ta in ty o f com position^
p o s s ib ly t 1 mole p e r c e n t.
Above 50 mole p ercen t CaF2 i t w i l l be noted th a t on many o f th e
samples th e i n i t i a l CaF2 d e p o s itio n was not d e te c te d by thermal e f f e c t s
and a t th a t time v is u a l d e te c tio n was not b ein g u sed .

The observed

tem peratures th e r e fo r e rep resen t the beginning o f the e u t e c t ic r e a c tio n
in th e se c a s e s .

207

0

l

cr*

I

l

CVH
P rft
<d

l I

I

I

Mv
o *■

uo

CO

0 0
O CD =
ftCO pft

p
0

cd

(D
■fO

cd

-—■%f t
ON f t
<w/
CD

O

■rH

f t p

p

P

0 o 0
o .rH.

O

CO

co

f t CO

t P

p cd r
rH
bOft

ftCO

bO

0

P

O P

Sh
O

•H P

p

0

p

■H P

>3 0 >3 U f t Sh Sh Sh *£ f t Sh
0 O 0 Q o o O O h
Sh Q O O =
CD f t CD
I—I ft ft ft CD i—I ft
CO co > CO > s o c o c o c a c o > c o c o c o

pH
X
ft
ft

fcuO
0

ft
o
CO
<D p
ft 0
^
2
CD ft

-

o

ft
o o ft

o

O "LA I
H O
I s

c—c—c o

s

co

ft o
1 1A A 1
1 1 i—1 f t 1

oo

CO CO

V

ft
o
rH
bO
W bQ
0
^ 0
0 -H
•H
© ft =
-ft
O bDffl O
i—1 0 P rH
CO •H
" ©
P ' i0f ^
ft O
0 s
© 1—1 1—1 ©
to f t f t
P
p
•'-P
*»
P bO 0 f t
■cd 0 o cd s
r—1 O f t H
f t f t s/3 f t

C
— OS
C"- ft
p

o

O
On cOo

r — i> - o - c o a o - c—
•LA f t

o
ftw

o

■C
—1>—C-—0 °° nD ®

1 1 1 1

l

V

u

ffl ffl

O O O O KO o
I I I I O lA O O A Q o

P! t J

bO
0
•H
ft
O
1—f
Oi
0
ffl
b O ft
0 -P
•H
"
f t H>
o 0 =
1—1 1—I
CO f t

g “

0

0
*—
n 4s ©
CO Sh p
■
p cd cd
p i—1
CO ft

oo

cm oj f t

o - d o O <ft f t o O O

ca c o

On

co

n A n n c v i'O p m w 'O f t jo

S?

01

■ l

I

o

i

I i
I

l

I I l i l l i
i i
i
l l I 1 I X I IX

X I

i

-P 03
ft ft
■o Sh O

I
I

I
I

I
I

I

I

I

I
I

■ I I

ti

i

I X X I X X X X X I I

i i —^ i i
I X I

i
I

I
I

I
I

O

S**- !
£ cd

I
I

■ElS

I
l

I
I

I
I

I
l

I
I

On CO

A A
I CA CA

I i—I OJ I

I
I

I co

oo

NO Pd" A
OO NO
I CA CA <A CA CA

I oo co

i

i co t o co co co

I

i

I

i

O

n»

P hIx.

£ ro

oo*

O O

ft° .^
rH
0 CD
0

H

O'
CA O

co

C A NQ n o

I rH On rH oo 00 CO CO

Io

H CA L A ON O
r H i —I i—I
rH
L :H , HuPijC

=aj O

s

i—I

(A 0 #

LA

s s

oo oo

*

CD

i—I
ft

ft

P < d l

I w

ffl

l

I

l O

CO

I

I

I

I
I

I
I

I
I

o

l

I

C
—1
H <->pf rH ,—N0Q• O•J p• t f«t *O-! nO»OO• C
—
• • £>., • •
•
00 co O' APT O' LA la NO 0 -0 - I COOO ON

p j - p f c '. j - j c '. p t _ j p t p J - 4

cd*

I. -ij!<

' 1--- 1 1_O'---"UoJ LiO !---HU P u p

k

o

•

lifted

Hr

I

cd

I cd

0

CO A

o*
o

CMft A
O CD

LA oo
<A A

I—I
I C—

I

ftcd H
o
O

SjC

0 0 0
ft A A

O 0" -LAO o
O LAOn C
—On O O
OJ OO00 A A Pf Pt

1 0
O
r—1

rH L A A

A
CM A C
— IN-

-ft
ft p f-d

ft* O- CO CO A o
Pf f t f t f t f t A

p

Tp i -

o o

X I

O. -ft
P P
A
A

g

CD

-P

r—I

I—I
I—I

p f P S Pi "

o

CO

•—«,

&

^

*Eh ft 0)

•

I

co

CD •H ptr

H

f t co

D
—ft O ft "LAfOft LAA "ACO C
—O H rl H CO00 COft A OJ c^ A A A c"i A A AOJ ft
c-—c— o—t>—o—■c~—c— o -r—i>-[>-i>-cocococo a o c o c o c o c o c o c o c o c o c o c o c o c o c o c o a o

xi

A

bO
0
ft
ft
o
I—1
to
(ft
y-\
©p
ft 0
P ft
ft
bO bO
0 0
o o
ft ft

bO
0
•H
ft
O
1—1
w
p
X
r—1 0
i—1 f t
cd ft
0
ft p X
0 0 0
0 1—i 0)
a ft >

O O O O O O O P

A

LA

la l a

a

a

p

p

O O O O O H

a

a

a

cd

ft

OnnO''a 'o '
H rH

o

' C
— H H r l H O D A A W a 'a
oo^
O J ^ A 'A

C\T o T C O O O C O O O O O O O O J O O O O A A - f t A O J

CMPt LA
CMCMH
CM A

OJ

OJ A

i—1 t—11
—1r—1 ft
A A A A O A
A
A A \
A
''A
Pt f t f t A A A
f t ft^ f t i—1 O- ft
A
X X
\
CM OJ OJ OJ CM A OJ

P
o
tA A
co o

CM rH

208
-p
P tS
© -H
b cr
*
©
P -H
PP

O

1 I I

I

I

o

! o

I

o

oo oo

oo

r->
ON

P
o ©
—. p ©
00 p P
©©
p H
co P
#
"d
--V !>
to
bP
o
p
co
M
O«H !>>
s-r O P
.o
P rH
©
SD•H
C
r-v P
m/ •H
—
P
H
• 01
p%
o£
r—
.O o
m •
=—*rH
cti ©
p 1—I
■=© o
S
m

©
i—l
P

CO
•y
P ©
E
o
o
r-N o
CM•"
TJ
s_/
O©
rH 1—1
© o
O S
r".
rH

©
p
©
Q

OO

©

©
o

■=©

P
rt
CD
P
cd
i—1
P
©
P-i
EH

o

o

1 I
II oCM II 1rH
-nm
o 'I II rH
o'! I I

bO
p
*H
ft
O
rH
CO
P

P
P

bOp
*P

2

-H *H *H

P P P
o o o
■ rH

CO CO

P
CD

CD
P

bO bO bO
P P .P

tuO bO

P P
•H -H
P Pi
o o
<—
toI rH
to

P

bO
p
op
I—
toI

i—I

•H

CO

p
CD

bO bO bJ3
P P P bO
_
P
P P P
•v ** •'•H
p p p p •H *H -H
P P bOp p p p p p P P P
©© P O©o o o o o o o
H rl O rl
■p PI ■

p

-H
P h b0P

o

o

P ©
O r—I

PI P

f'lHONOJW
HHrbOM
H POJrlOsH
(M rbnnnfM CbcbCM
WJM
cyrbbinr'lcM
fb

oooooooooocococoaocooocooooooococo

111
I I 1 I I I
I I I MMI • I I I
O

0\ p

C
CO
O

b -

I

O

II1A
b- O
on1I II II •I II
0O(D

O

m

I

Q

II0®\ 0Oi 0N II V
rlJ II
rH rH

I I I I I I I I I I I I I I I I I
I I I I I I I I I I I I I I I I I
b*
bb*
in
o
m

bb- b* bP-*
•md m
b- b* m
i n m n i mo r I CK i
i lA i r n n i i-n t i -on i
bb«
b*

i
i

^

*!<

E-h I I ©

I§

H

b«
b<
b_JM0

mo

—P

hd

I

I in b-

i I b- b* *

•=3|
I «aj

1AO O b;
o r H CM CM
lA m m m

mro o
no mo o o
CAMD c—b—On CNrH-PHD n H o o
la m m in in m m \o hd b- b- b~-

O
H
in o o in
^ ir \in

OOOO
OOO
o
om
mm
o ><m
mo
l' \ *mmmWmmo W
^ >m. m
. cmo m on'^.H m h V hN
PCMHCMCMCAHWHb^N
cncM n .

•m m m m m .mm

APPENDIX 9

209

APPENDIX 9
RESULTS OF THERMAL ANALYSES
OF LiF~BeF3 SYSTEM

E xplanatory N otes
Table A and B g iv e th e data fo r th e o v e r - a ll phase diagram o f
Figure $ .
T ables C , D , E and F g iv e the data fo r th e more d e ta ile d study o f
the r e g io n near 33 mole p e r ce n t BeF3 , p lo t te d in F ig u res 6 a , 6b , 7a and 7b.
A b b reviation s
Under C om position, one or more o f th ree headings appear, s ig n if y in g
the fo llo w in g :
ORIG. s i g n i f i e s com position c a lc u la te d from inp ut o f LiF and apparent
w eigh t o f BeF3 a ft e r d r iv in g o f f NH4F from (NH4) 2BeF4 , or based
on input o f BeF3 and L iF .
RECALC. s i g n i f i e s com p osition c a lc u la te d from in p u t based on subse­
quent a n a ly s is o f (NH4) 2BeF4 u s e d .
ANAL, s i g n i f i e s an analysed m e lt.

I f r e s u l t i s g iv en in p a ren th eses

i t in d ic a te s a value ob tain ed by back e x tr a p o la tio n from a
tern a ry system sample a n a ly s is .
Under F i r s t I n f l e c t i o n , th e a b b rev ia tio n s s ig n if y :

S , sharp; G,

good or d e f in it e ; F , f a i r or l e s s d e f in it e ; P , poor or in d e f in it e ;
VP, very poor or b a r e ly s u f f i c i e n t fo r r e c o g n itio n .
Under P la te a u , th e ty p es shown have the fo llo w in g s ig n if ic a n c e .
Long, f l a t — f l a t or very s l i g h t l y slo p in g w ith in tem perature range
shown, fo llo w e d by abrupt drop in tem perature.

Length about

210

1 0-20 m inutes app roxim ately.
F l a t , s lo p in g — d en otin g behaviour in which q u ite f l a t p le a te a u
o c c u r s , fo llo w ed by a sm oothly slo p in g t r a n s it io n to a r e ­
sumption o f o r ig in a l g en era l s lo p e .

Length o f f l a t p o r tio n

v a r ia b le from about 10 to 15 m in u te s.
S lop in g - d en otin g behaviour in w hich p la tea u i s never very f l a t ,
b u t in s te a d shows gradual drop o f about l° /m in . fo llo w e d la t e r
by a more r a p id ly c o o lin g s e c t io n , w ith ou t any abrupt t r a n s i­
tio n .
G radually slo p in g - same as above, but w ith even more s lo p e .
S h o r t, slo p in g — denoting behaviour where a sh o rt p la tea u o f about
5 m inutes or l e s s du ration o c c u r s , fo llo w ed by a non-abrupt
t r a n s it io n .
Short — denoting a sh ort p la te a u o f l e s s than 5 m inutes duration
fo llo w e d by an abrupt type o f t r a n s it io n to more rap id c o o lin g .
In th e columns headed ’'Binary E u te c tic " or "Apparent Liquid" the
l e t t e r s H and C In d ic a te o b serv a tio n s made during h e a tin g and c o o lin g
r e s p e c t iv e ly .

"Apparent Liquid" r e s u lt s are e n t ir e ly based on o b serv a tio n

o f the disappearance (o r reappearance) o f the l a s t (o r f i r s t ) liq u id during
c o o lin g or h e a tin g o f the m e lt s .

Temp.
Apparent
L iqu id

211
o o o o

o o o

O Q

O n O 1A
-P T X A X A

rH X A

O nD-la in

nD

V v

M
C

W
>
C

*H
IP
o

•H
O,
O
rH
CO

P P
P
cd cd p
cd
i—I rH
i—I i—|
H
i—I th i W
<i-l

First C ryst.
Observed,
Temp.
Type

First I n f l e c t . ,
P la tea u ,
Temp .
Type
Temp.
Type

*\

I

|

|

|

I p

*>

p

-p

*\

hO bO t>3 t O P

I I I I i t i i d t f l f l Cj HCn )
( H r H r - I O O C D O t —I

(X 4fpfX (|-3l-31>PJ!i|
O - O -

XA
X A —V

J - O
J ;
- j- ^ A -d "

CM X A - V

I I I

I
I

I
I

I
I

I I I

I I f A i—I X A O O n —V CM O
I I X A X A X A XA P t X r\ v O XA
l- .p tp d - .p j-

n o p q O h p L .^ fx .p m ^ p q

On

On

I

I

( A CO X A NO X A C X A
(A n n a h c o i a

O O
j q

CD 0)

i—I £ £

NO
o n w

o

o

tt)a5(H-C'-C'-NO\ONOND_d-XAP! £
! ! ! ! ! ! ! ! ! S ^
ip i-H

§
(P

o
<A
I [—
XA rl
...........................I

I

l> - [> - l> -

X A O O O H nO
nQ P t Q
O CNJ CO
I NO NO NO X A X A - p - .
cd

Melt

C om position,
Mole % BeF2
P rig. Kecalc . Anal.

o

.X A

■i—I

• CM X A ON CM ( A X A CO O
. P t CO i—I i—I i—I CM CM CM CM < A

CM
H O
« » •

•

(A O

n

I
I

O• (A
O ® CM—
• • • • • !

GOCOCO O

O

<A
CA

XAO« •

P t CO CM X A O n CM < A CA NO OO
i—I H ' —IC M C N IC N IC M C M

I

( M r —I
( A CA

CA X A X A P t CM X A

.

C A O - H A CO CNI C A - V O O n O CNJ CNJ
i—I i—I (—I CM O J CM CM CM ( A ( A ( A

4 | F q o n W ^ l l ^ K H H ®^
o o o o o o

X A X A X A X A "LA X A O

Date

I
I

o o o o

XA XA X A ^ A H

O

C A P t ON CM A L A N O
O n CM
rH H rH CM CM CNI C— O J CNJ CNJ i—I (SO H
NO

n O n O nO nO nO

C—

n O nO nO

O N X PvO -

212
p

•>

Sd X •
(D-H P
d

E

G

«i c r ffl

ft-H Eh
ftid

O WO

O EC O m
O
QM
Q3
CNO
lb-®

l o o p

^AO^A
-=t_d- A

<0

I IOm
1 ' oA pA 1

rn co rftro

2Sd Q°> (D
s
■rl -P Eh

mo

P O n o O pco

m m m iA iA m in
A A A f*^ (V^
or,

a p t

A

A

I
0
X (ft
sd 0
G P3
©

O

j o m o m —t o m

M3 A

-d

X

a a

o
£?-p

CO
0 A

i i
i i

XA

o

o

o

XA
XA

O CO b - CO M3
CA XA XA XA XA
CA CA CA CA (A

CA

xa

E

o mo

sd

©

o
ft
©

ft

©

■
—i
bO

-p
cD

rH

ft

■CD

t?
CD

*\
G ft

0
©
p
0
H

•

ft

ft

H rH CO
ft ft

tuO

sd
o

U

0
>

XA
I

O
i—
II "LIA

0

ft

bD bO

bO

sd

sd sd
•H H

sd
•H

ft

ft f t
O O

•rH
o

I— 1
CO

r—1 r—1
CO CO

r *\

•\

>

CO

E

sd
•H
ft
-P -P o
cD CD 1
—1

bD

m

bO

sd
o
ft

bo be

sd sd
o o
id d

ft

o
1
—1

bO co
0

•H

^

ft-P
O 0
H H
CO f t

bfl bD

CO 10

•
D
0
Sd
X

sd
•H
ft

P

CO

p
G

sd

•H P
ft G
2
H° 4d

co co

H P

bD id
G O
O 4d
Id CO
M3

M3

CM H XA
XA X A Pf

(A A

p j

CM CM

o o O n ce b
p t P f CA CA <A

I

CO

CO CO t o

CO

CO co

o o

ooom <E 5

ft

ft
•rH
ft

.
p)

CO

E
©

-X
Pt

ft

CM H XA
A A P t
p f p f p )-

XA
pf
Id

C— r—I
tA _ d
_d -d

CD

W *> Q
tO>X S

G © ft
O >
-P

CO

CD
CO

ft

CA CA CM CM A
o O CNCO b _ d _ d CA CA (A

0
wCM
.
•H

•

ft

CA H XA
XA XA Pf
p j-p fp i-

h£ S
CD

•H O

ft

b- o
AA
pf pj
cD
O

sjt

-P

ft

b-

Id

A H
3 3

ON o
A A
p f -d

b - <A A A A
O O ON CO b d d c A © 0

1 On

0

I

■ H O C

P

0 A
-P
CO f t

©

H O
E

©
id
©

>* 2
42 r—j
0

©

>

O
r—I

X

CO

’rH

o
P-i

0
X

©

ft

M3 M3

CA CA

co

M3

o o o

O CM O
b - b - co

qq <D

f t CD M

E H H
O O d

OS O

O

XA

to ^

o

•
I b—8 cD* cD cD cD
d
o o • CD -d XA O o o o
CD 0
O -P CA |3 Is d £ Is IS
Oo o o o o
5 1—1 2 H I—I I—I H H H
cd f t cD 0 0 0 0 0 0
0 d x x X 43 43
o
-p -p -P
c
D
c
D
C
DX X O X X X
p f -d
M3 M3 H H r
—I 0 0 0 0 0 ©
f
t
f
t
CA CA
f t Sd sd c
m om
0 © 0 © © ©
bO bO bOpn
m
x ft
Sd Sd G O o © o o o
o o O -H •H *H •H vH *H
i—I t—I
-H -p -p -p -p -p
tA iA
H !>a r~a Ja tA rN rN
id cd Cd i—I rH rH rH H H
o o O H r
—Ii—1r—I r—I r—I
cD 0 0 0 0 0
co co to G G d G G d
£ |3 3 'S x x x x x
O cD 0 0 0 0 0
X A -d O O 1
M3 M3 4d 4d 4d G
X d d
CA CA co

co CO o

O O o

cD
0

Sd 011—I

O ft

ft A

Q 'A
M3 CA o
CA
A A

P 0
o
0
H ft

CA

O

(A

pj- xamd

CA

ON o

< A CA CA

O

o O

co

A

A

I H I I CA I
I -II
*1

i
i

i
i

i
i

i
i

i
i

O N Q \ ' O p q o o q o q o

A

o O CA M3 Q Q XA O O O XA
XA XA XA XA M3 MJ M3 b -C O On On

A M3 CO O
" A P fP fA
<b«

a

id

id

s

CD
-P
CD

o

o

XA O XA

ft

s CA
ft
XA i—I

ft

A

O H
XA XA

a m f t o ft
o

o

XA i—I i—1 XA i—1
\
a ;a \
xt\

f t - c jd .q
A A H

a >

!xj >H b l iSI

O s ON CO

A CO XA A XA

XA

CO M3

ON 'M3

CO A

M3 CO M3

0

ft

o o
'H
X X

sd
H

0
0

-p

H ra
H sd
=D -H

sd H
■H
0 P

0 O
O ft
c ©
G ft
m©
H i—I
O

m

e

P A
H Ai
©
CM
a
(ft

ft

0

sd
o
■H
P
•H

p
sd

co

0

ft
ft

O P
sd
© ©

O

ft o

E
o
o

H ft

©

I—I

I—I 1—I I—I

O r —I '—I 1—I *—' 1—' X A i—I X A X A X A

X A X r \X f\iA X \x r \\X A v s W
i—I "

A

Q

sd 0
o r^l
H -P
-P

ft © ft
©

-P

H
CD

CO

o pq

XA

o < A _ d H O

CA

o
e

CO

pt pt

*>

H
P
CO

©

o

o
1—1

bO

p p
Sh S
o o
4d m
CO CO

.
o

bO

sd
•H *H
f t ft
o o
1—1 1— 1
G

v i—I O n O

T -d M 3 H A A

A ft

ft
*

©

p
o
a

O
E

A

©

Apparent
L iq u id ,
Temp.

213
O

i i
i i

Ooi SOK

Q

I l>~ Q

MD
A

CM O

I V \vD J O

'

O

CM

M
fi
•H
Oh
O

(—
I
CO
a

i

A
A CD
■P -P tc -P cd P i
cd cd H cd H A

l—1 C
h
Al 'H «H bO bO
*'
tO)
bOo C
!>
>
h>
C
o O o POO
0) CD PI O PI PI
> > D—^ nO nO
TA

1A 1A

8 S IS
|-A A

II
I

I

I

I

I

I

II

I CO CO I

l

I I Q «H On c o Q

I I I I I NOMO"LAUNVO

A A A}A'A

•H = e e s

S

PI

e e

CD =

cq

0
“

° O O C \ I O O on* r l O ' \ Q Q
C— T A O r l O
V A M J m5

1A T A T A TAT

rco

11
1

CNJ

i TA» |
1 rH 1
CA

1|
1

TA P• 1 O•
i■
i CM AT i1 NO
CA

CA CA

c— NO NO T A CA o~\ O n CO TA CA O
rH CNJ CVI CM CM A T A N O
CNJ CNJ CA CA CA r a CA CA CA CA CA

o o ON O

o

O

0 0 GO A

O

o

•
TA o
cA ■rH
A
A o
0 CD
o A

Acd C£D

•H
PI

CO CM CA O n C—

A

O N O O H CM CVI CA CA A T A nQ
CM C A < A C A C A C A C A C A C A C A C A

'p

Melt

'r

TdrO rO'TdrO rCSrO rO rOTdrO
o
0 \C O p
t
H IA
cm c a A
I—I

Date

C om position,
Mole % BeF2
P rig. R ecalc. Anal.

First C ryst.
O bserved,
Temp.
Type

First I n f l e c t . , P la tea u ,
Temp.
Type
Temp. Type

rH i—I A)

TATAlAIAlAlAlA'ATATAlA
nO';^nsU;A;A.A;^0^;A;a TA

r l H r l H H r l r l H H r l r l

X A lA 'lA 'L A X A lA 'lA 'lA 'A lA 'A

A U
cd
o
A
A
O
CD
A

0
£
O
i
—1
A
a
o

0CD 0

fd
Oh
O Oh
CD cd
£
O
0

A o
A A
CD

0
is i
s
O or^c
~] i
co 'm
TJ "d
A
CA

I

I
CO
CD
-P

o
S
*

21 b
-P

o

O
I
I

I IA I
I XA I
P f

o m o o o K o

H ) |p |p t P A O
O O v©
PA H A U M f t O lA ^
P f p f p tP tP t A A P f

©

© ©

-p
©
R
R

©:

ch A

h ^
h

G G

G

©

© ©

O !

CD
-P
cd

"

Ah P h|
P
©

Eh

PA
m i A I A - j \O vO
PA
PA PA IT \IA PA LA
I
I I I I I I
t — A- O- It—t — A- CN APA PA PA PA PA PA PA PA

p fp fp t.^

G ©

At

©

’

pH

Eh

Ph

Ph

l A P - 'O A - P t
PAPAPAPAPA

PA A
TA LA

CO CO On OO CO CO APA PA PA PA PA PA PA
" P T P tP t.

TATA

I I I I I

■ ■ • -•

P>
©
R

£

II

onoo

i
I

P—C— C— On
i tatatata i
I P tp rp l-p f I

i
I

i
I

I
i

CO CO cO AI PAPAPAPA
i p fp t.

R o

Ph R

NO
PA

I

nO a o
PAPA
P t P f

CO CO

NO A PAPA
P t P t

I

©

©

%
R

e

PR

d<
Ah

Ah
0Q

cdr =

•H
R

ft, e e = =.

•H
R

•H
R

©
o

*> i—©1 o• 1
G cdp C M l

O Ph «x!
*H ©

1
I
1
I -P
cd
H

o
O O c— r—m o c o c o q o p a Q r— c o c o a p - co
O \ 0O P-PAPAPAPACO GO P-MD NO PAPAPAPA PAPA
p t p f p f P f p f - .......................................................................................

i

©

tiP E
- HO©

11
1 to
G
O
R

NO o o
PAPA

Ah <GH ©
W
I—IE-)

°
t,
-P ffl
co to f t

1
1 to
G
o
R

OncO
papa

ft

43
CO ^« 94
P>5
1>J'0 E-I
G©

I1
1

•P
cd
R
R

I l l
I I I I I
I I I COCOCOCO I I I I I COCOCOCO coco

•s gj
Pm Po Eh
•H H

1
1

to
G
O
R

> > > A

3O^
T

rH

AR
M

-P
cd
■—l
R

to
G
■H
Ph
O -P
i—1 ©
co H
R
«\
-P t o
© G

1 1
l l

<A

I I I 'A

1

1

i i I
I l l

1 CM
A

i
l

I
I

i
l

i
I

I
I

I
I

I

I

P t

CA
CA

-p oq
.rl
© T^
O

A©

g rH
O
O

-P
rH
©

s
©
-p

©

R

A - OO C> CM CA P A

nO i—
I

CA

CM CM CA A P f P f P t H CM CM CM CM CA A CA CA
A < A CA CA CA CA CA CA CA CA CA CA CA CA CA CA

CA PA
CA CA

PA
CA

ad ci fr? fR eg. gTr R fR P S cS R R a d rtR

£O O&n

iR w

P tO N A A -rH P A O \C O

AMO

OO P-MD PA PT CA CM ONCO P - nO P A P CA CM R
i—I r l R i —I i—I i—I i—I i—I i I
O
tA

Pt

.CM S

O

PA

&

CA

NO

OJ C

cm

CM

CM

I—I

o

PA
PA
e

CM
CM

i—I

I
I

o

P A i—I

\P A

CM \
CM P A
\
CM
CM \
i—I i— I

Compositions of IE and 2R calculated on the basis of e ffe c tiv e
,in the starting material^ L i3BeF4 product of 1 2 /1 9 /5 0 .
^'Starting material for S was L i3BeF4 product of 1 /1 9 /5 0 .

-P -P -P -P
r l r l r- | r l

bO
G
•H
Ph
O
i—1
ra

—

R

35.3 mole percent

BeF

P r-H
P .R

O ( o o o o o o
■LA I papapap TnOnO
p) ^ ^ ^ 5 ? P t P ?

Notes

©
IH

a tj
(D -H
h
^
© tT

215
*1

-p
p

T3

P

O' fe
—
a
, -H 0)
ft
PIE-i

(13

o

W OOiS o

OJ
A

A oXnco A
3 ojoj3 oj
3 3 3 3 - 3

3

0
T' f t

3 ^

o

A A

O O OO

3 3

QUA

Sr^

(0
O

<*!

OW O W Offi w o

I0 0 - A

ft
0
1—1

4^ f t 43 43
0 0 0 a)
1
—1 rH 1—1 rH

43 43 43

PH

Ph Ph Ph Ph

P h Ph P h

0
1
—1 n

0

cti

h

43 43
0 0
1
—l 1
—I
Ph Ph

43
0
1—1

43

PH

Pi

0

t—1

hO ho hO hp
f>» f t >5
P 0 P
P P
P P P
0 rH 0 0 O O
O O 0
> Pt, > > PI PI PI P I >

!>> 1>3
P P
t?
0
0
>
> >

0
>

OJ CM (A A CA A
A D A A A A A

A 3 A
A A A

OA OJ
A A

H
A

XA

3 A

3 A

A

A

A A 3

A 3

CA

3

3

3

3

_03 3

3

3

0

!>>
P

0

-P

0

rH

I I I I I I

ft f t
e

<D

I II

II

I

PA XAXAXAXAPA PAPAPA PAPA PA

&H

3

"0
-P’ f t
O &H

I
I

ft

3

I
I

I
I

I
I

I

3

I
I CO

I CO I

co

I

CM

A

co

C
D

i—I

hQ
P

*2

M

•rH

■P

0

-P

w

1-1

CO
P
*
•H p,
* §

PA
PA

I—I
OJ

PA

• I
rt I

EH

i
i

o

J
X
a

3

i i
i i

-3-

A
PA

3

CM

A
3

3r-I
0
0

P

P

o

'i'

-P
S
CO TCi f t
>3 03
P >
P
o
CD
-p CO
CO £>
P O
•H
PL
ft
E
C
D

Eh

PI

•rH

0
m

4=
O

•H

4J
P

cd

V
■Hft

ft

o

ft

fto

CD

K
—
1 PI

0

pi

0

H (A CO XA rH PA AJPACA X A J CA
C
MH P P P ' O
M
DA C
— M3 PA XA
PA PA3" 3" u j 3" 3 - ....................................

A
A

M3

3

A
A

0

0

43

0

o

O

Pd

0

P NH
O ft

•rl 0

I

P

I

I • » I I

I

I I I

fl

I —
0 CM
A A

4 3 CQ <C|

I

I

43

'Si

w ^
o

ft® M
E rl -H
OOP
OS O

0
P
o

rH A

CA O

CO O

rH

A M3 1A

CA A

PA

O O C M O JA A
A A A A A A

A CA OJ
A A A

OO A
A A

A
A

•H
rH

3
A

0

a3 0
o o
rH
0
s

0
43

0

n

co
ft

h

h

h

h

ft
ft
3

h

H H H H H H
H rH CM A 3 A

O
A
\

oo s

"3.

CA

O
3

A
s

—
j* — c

"3.
OA

£
o

o
M3
CA

ft
ft

%
ft

o XoA

O
A

ft
ft
A

CM

XA

e

3

co

CA

CA

rH

CA

rP
C
O

-p

K
o o

O o

*

G ffl!

OOOO

® -H ft T A -ffl m p f t A :
1ATA TaTA
ht c
*rȣffl -ffl"-P
r
_ffl_ffl
Ph-H E~*

5= O

O

O

O

i—I T A T A O

-ffl-ffl-ffl-ffl
V V V

P ji-H

M W) tiO
PPG

*H «H *H
Pj Pj P ,

0 0 ,0
ra w w bO
G

-p

ffl s

rl rl H

rH

Ph

■H

txO

-P

H
ft,

TA
TA

oj
-P

O
ta

ffl

o p,
fflI fH
K
i—
'S
M

WCOCQCOCOCOCQCO^COCO

.
-p
ra &
P e
•H ©

fflT

_ffl_ffl HD HD

!> -C ~ -sO -ffl TA CM

"U~\ TA "IA TA TA TA TA "LA TA TA TA
-ffl-ffl-ffl-ffl-ffl-ffl-ffl
_ffl _ffl-ffl-ffl

©
ft

*
-p
ra ffl!
|>a CD Ph
>
p
P
©
W
ra
*
P O
•H
Pe
©

o
-P

*}•
©
m

ft
(ffl

■H

(ffl

_ffl O

HD MD OO 0 - C--_ffl_ffl HD CM
OO C N C O H D T A T a T A T a T a T a T a

' -ffl-ffl-ffl-ffl-ffl-ffl-ffl-ffl

CtJ

E-t

G

fflllM r v lA ffl
i_r\ i a "lts t r \ "l a

I
I I I I I
I
HD -ffl -ffl HD
O— C— \ 0 —
ffl TA OJ
TATATATATATaTATATATaTA
-ffl-ffl-ffl-ffl-ffl_ffl-ffl_ffl-ffl_ffl_ffl

p,

o

oi

O Pe
ffl
•H
■P pq
•H
ffl -H
O

P

P, ffl O
5O
3H
O
o S

<d
S

»

«

•

ffl
ffl
Q

»

o

*

CM O

•

•

©

»

C -

*

rH rH rH CM CM C A - f f l - f f l T A T A HD

cACAfACACAfArAfACACACA

(yorarararcycyar&cyar
A U M ^ O D O cm A O n H D j—| —
ffl
H H

o

-p

V'*
• •

CM n t ^ f f l O X ®

H

o

O TA

O O O O O I A O O

■\cos
HD H

H

~

— All melts prepared from Brush BeFa and L iF.
Composition of 8Q uncertain due to probable weighing e r r o r .
On 3^ and 5>Q the in it ia l observations of surface c r y s t a l l i z a t i o n

P G ffl O
O O H rH
PI ffl) tx, CO

C M C ^ -ffllA C A H O O C O
. . CM CM H O
H H H H H

H O
H H

Notes

rso 5-1
ffl
a

P,

U ) bO - p

Gs
o
rH

APPENDIX 10

217

APPENDIX 10
CALCIUM FLUOBERILLATE; RESULTS OF WET AND DRY SYNTHESES,
OPTICAL AND X-RAY DATA
R eferen ces fo r the e x iste n c e o f the compound CaBeF4 were given in
Appendix 3 .

The method o f wet s y n th e s is o f BaBeF4 b r i e f l y d escrib ed by

Sarker and Ray (II4O) was t r ie d in se v e r a l m od ified forms in an attem pt
to prepare pure CaBeF4 .

X -ray and a n a ly t ic a l data both in d ic a te th a t

a l l th e products were contam inated w ith calcium f lu o r i d e .
E s s e n t ia lly the Ray method com prises m ixing a c o ld sa tu ra ted s o lu tio n
o f a so lu b le a lk a lin e earth m etal s a l t w ith a c o ld sa tu ra ted s o lu tio n o f
(NH4) 2BeF4 .

The somewhat m ic r o c r y s ta llin e product i s washed by decanta-

tio n w ith s e v e r a l s o lv e n ts to remove so lu b le s a l t s and f i n a l l y w a te r .
The p r in c ip a l m o d ific a tio n made was to use c e n tr ifu g a l d e c a n ta tio n , but
other methods o f combining the s o lu tio n s and d if f e r e n t d ig e s tio n methods
were a ls o t r ie d .
The f i r s t p rod u ct, "A", was prepared by r a p id ly mixing c o ld saturated
s o lu tio n s o f (NH4) 2BeF4 and Ca(N03) 2 fo llo w e d by s u c c e ssiv e c e n tr ifu g a l
d eca n ta tio n and washing w ith fou r p o r tio n s o f w a ter, th ree p o r tio n s o f
a b so lu te e th a n o l, two p o r tio n s o f hot g l a c i a l a c e tic a c id , and th ree
p o r tio n s o f a b so lu te e th a n o l.
The a lc o h o l su sp en sion was then evaporated to dryness on the steam
b ath , y ie ld in g a w h ite im palpable powder w ith an average s iz e o f about
1 to 2 m icron s.

As the (NH4) 2BeF4 was known to con tain NH4F as im p u rity ,

a $0 % e x c e ss was u se d .

Calcium flu o r id e would be expected to be p r e se n t

218

in th e product under such c o n d itio n s .

Upon a n a ly s is , th e m olal r a tio o f

CaF2/B eF 2 was found to be approxim ately 1 .9^ but the a n a ly t ic a l r e s u lt s
did n ot t o t a l 100$ by w e ig h t.

X-ray exam ination in d ic a te d th e p resen ce

o f f i v e o f the stro n g er l i n e s o f calcium f lu o r id e , in a d d itio n to about
t h ir t e e n oth er l i n e s id e n t ic a l to th ose observed fo r CaBeF4 occu rring
in th e samples from the CaF2-BeF2 system m e lts .
In an attem pt to prepare purer (NH4) 2BeF4 th e same s ta r tin g m a teria l
was f r a c t io n a ll y r e c r y s t a lliz e d to y i e l d a product which turned out to
be about 95$ NH4F on subsequent a n a ly s is .

N eed less to say the CaBeF4

p rep aration s f,BM and “C*1 ob tain ed w ith t h is “p u r ifie d 11 product were p o o r.
In another attem pt u sin g about $0% p u r ity (NH4) 2BeF4 , a d ilu t e
s o lu tio n o f one reagen t was s lo w ly added to a d ilu te s o lu tio n o f the oth er
w ith a g it a t io n .

W hile co n sid era b le calcium flu o r id e im purity would be

e x p e cted , a n a ly s is o f the p rod u ct, "D", showed a r a tio o f CaF2/B eF2 o f about
1 .2 .
F r a c tio n a l p r e c ip it a t io n was a lso tr ie d in the hope o f removing most
o f th e contam inating flu o r id e io n in the f i r s t h a lf o f th e p r e c ip ita te d
p rodu ct.

The product o f the second p a rt o f th e p r e c ip it a t io n , "F",

showed a m olal r a tio o f 1 .1 2 .
X-ray exam ination o f th e se p ro d u cts, HD" and HF", showed s l i g h t
amounts o f calcium flu o r id e contam ination .

None o f the a n a ly t ic a l r e s u lt s

showed 100$ r e c o v e r y , the t o t a l s g e n e r a lly being in th e range o f 89 to 93$.
T his recovery could be accounted fo r i f the oven -d ried m a ter ia l had an
em p irical com p osition CaBeF4 :l/2 H 20 , or i f i t were oth erw ise so lv a te d .

219

A lthough a l l o f th e above r e s u lt s were obtained u sin g (NH4) 2BeF4
contam inated by NH4F , i t i s s t i l l probable th a t d i f f i c u l t y would be
encountered in o b ta in in g com plete freedom from calcium flu o r id e even
though pure (NH4) 2BeF4 were u se d .

Since the l a t t e r m a te r ia l probably

d is s o c ia t e s in s o lu tio n to form some flu o r id e io n s , s l i g h t p r e c ip it a t io n
o f calcium flu o r id e would be ex p ected a t the l e a s t , and i f the m ixing
o p era tio n was c a r r ie d out very g r a d u a lly , s u b s ta n tia l calcium flu o r id e
form ation might be p o s s ib le .

In a d d itio n h y d r o ly sis o f the produ ct,

CaBeF4 , may occur w ith p o s s ib le o x y flu o rid e and calcium flu o r id e forma­
t io n .

The hot g la c ia l a c e t ic a c id used in th e wash treatm ent i s supposed

to remove h y d r o ly sis p ro d u cts, but i t does not have a la r g e so lv en t
e f f e c t on calcium f lu o r id e .
Because o f the apparent p r a c tic a l and th e o r e t ic a l d i f f i c u l t i e s o f
preparing CaBeF4 in pure form by the above method, some a tt e n tio n was
given to another p o s s ib le p roced u re.

The complexing agent n i t r i l o t r i -

a c e t ic a c id , N(CH2COOH)3 , shows co n sid era b le s p e c i f i c i t y as to the typ es
o f calcium compounds w ith which i t forms so lu b le com plexes.

C onsiderable

work was done on the p rep a ra tio n o f t h i s reagent and t r i a l s were made o f
i t s p o s s ib le valu e in d is s o lv in g CaBeF4 to lea v e the in s o lu b le CaF2 as
a r e sid u e but no s a t is f a c t o r y r e s u lt s were ob tain ed .
The formula o f the compound formed by CaF2 and BeF2 i s probably
CaBeF4 fo r s e v e r a l r e a so n s.

The o r ig in a l d r y -sy n th e tic product rep orted

by Hay (129) showed th e proper a n a ly s is .

In a d d itio n i t m aybe s e e n t h a t

as the produ cts o f t h i s work ware ob tain ed in purer form the a n a ly t ic a l
m olal r a tio o f CaF2/BeF2 approached the valu e o f 1 from the h igh s id e .

220

C onversely the sample J 1 c o n s is t in g o f s in g le c r y s t a ls o f CaBeF4
(removed from, m elt J o f th e CaF2-BeF2 system) gave a m olal r a tio of
about 0 . 9 , y e t both products by X-ray methods were i d e n t ic a l. (See Appen­
d ix 1$3 P art H) .

I t i s th e r e fo r e concluded th a t the formula o f Ray co r­

r e c t ly e x p r esses the com p ositio n .
S in ce no X-ray str u c tu r e or powder p a tte r n data had ever been pub­
lis h e d fo r t h is compound, an attem pt was made to determine

the c r y s t a l­

lin e system and u n it c e l l dim ensions o f CaBeF4 .
O p tic a l exam ination shows p r a c t ic a lly e x c lu s iv e ly one c r y s t a llin e
h a b it, th a t o f a th in p rism a tic c r y s ta l ly in g on i t s broad fa c e and
p r e se n tin g the appearance o f very elo n g a ted p arallelogram s w ith
an gles a t th e acute c o r n e r s.

The c r y s t a ls when viewed in t h i s p o s itio n

are a n is o tr o p ic and show p a r a lle l e x tin c tio n .
The in d ic e s o f r e f r a c tio n were determined u sin g c a lib r a te d w aterg ly c e r in e m ixtures (fo r both tu n gsten and sodium lamp illu m in a tio n ) by the
Becke l i n e method, u sin g d is p e r sio n e f f e c t s fo r f i n a l m atching.

In the

crushed samples on the m icroscop ic s lid e some n e a r ly is o tr o p ic s e c tio n s
were r a r e ly ob servab le (n o t always p e r f e c t l y o r ien te d ) which had an in d ex
o f r e f r a c tio n o f 1 ^ ^ = 1 .3 8 6 .

I f the c r y s ta ls were u n ia x ia l t h is would

determine the index o f th e ordinary ray as b ein g

1 .3 8 6 .

For the a n iso tr o p ic p rism a tic c r y s t a ls (ly in g on th e ir broad fa c e s)
two in d ic e s o f r e f r a c tio n were obtained fo r the ordinary and e x tr a ­
ordinary r a y s ,

6 4 /NaD= 1 .3 8 6 and

£ Na£= 1.1+01, assuming again th a t the

c r y s t a ls are u n ia x ia l and th a t in t h i s o r ie n ta tio n the o p tic a x is l i e s
p erpendicular to th e a x is o f the m icroscop e.
are probably accu rate to w ith in 2 0 .0 0 5 .

The r e f r a c tiv e in d ex v a lu es

221

In te r fe r e n c e fig u r e o b serv a tio n s made on the is o tr o p ic s e c tio n s
always showed a t b e s t incom plete u n ia x ia l c r o s s e s , in which the cen ter
was s l i g h t l y o u tsid e the f i e l d , but the arms o f the c r o ss were ob serv­
able on r o t a t io n .

T e sts w ith q u a rtz , mica and s e le n i t e r e ta rd a tio n t e s t

p la t e s showed p o s it iv e o p t ic a l s ig n .

T e sts w ith the s e le n i t e t e s t p la te

on a n is o tr o p ic s e c tio n s (w ith th e lo n g a x is o f the c r y s ta l p a r a lle l to
the slow ray v ib r a tio n a l d ir e c tio n o f the t e s t p la te ) showed a blu e
c o lo r a t io n , in d ic a tiv e o f p o s it iv e e lo n g a tio n .
I n te r fe r e n c e fig u r e s observed on the a n iso tr o p ic s e c tio n s gave what
appeared to be " fla sh fig u re s" ( c o n s is t in g o f two hyperbolae which en ter
from o p p o site quadrants and lea v e by the other two q u ad ran ts), which
m ight be in d ic a t iv e o f e ith e r a u n ia x ia l s tr u c tu r e , or o f b ia x ia l "op tic
normal" and "obtuse b is e c tr ix " f ig u r e s .
The r e s u lt s o f th e above o b serv a tio n s s tr o n g ly in d ic a te d a u n ia x ia l
s tr u c tu r e fo r CaBeF4 , but were n o t a b s o lu te ly d e f in it e in exclu d in g a
b ia x ia l s t r u c tu r e .

The o r ie n ta tio n o f the few is o tr o p ic s e c tio n s ob­

serv a b le was not s u f f i c i e n t l y p e r fe c t to perm it c e r ta in ty th a t the observed

w value was id e n t ic a l w ith the OJ v a lu e shown by the a n iso tr o p ic s e c t io n s .
The f a ilu r e to ob tain cen tered u n ia x ia l cro ss in te r fe r e n c e fig u r e s on
the is o tr o p ic s e c tio n s a ls o in d ic a te s a s l i g h t d e fe c t in the o r ie n ta tio n .
The f a c t th a t o n ly two in d ic e s o f r e fr a c tio n were observed fo r the a n iso ­
tr o p ic s e c tio n s does not u n eq u iv o c a lly mean th a t th ese are the exact v a lu es
o f oo and £ s in c e a very predominant c r y s t a llin e h a b it might cause the
c r y s t a ls to always l i e in the same o r ie n ta tio n , w ith the o p tic a x is non­
p a r a lle l to the sta g e in a c o n s is t e n t manner.

The a n is o tr o p ic in te r fe r e n c e

222

fig u r e s co u ld be in te r p r e te d as b e in g due to e ith e r u n ia x ia l or b ia x ia l
stru ctu re " fla s h fig u r e s" .

The o n ly p o s it iv e in d ic a tio n o f u n ia x ia l

character was the o c c a s io n a l o b se rv a tio n o f a poor u n ia x ia l c r o s s .

The

agreement o f the to v a lu e s fo r the is o tr o p ic and a n iso tr o p ic s e c tio n s
may be taken as con firm atory.
Larger m a c r o -cr y sta ls o f CaBeF4 have a somewhat rhombohedral appear­
ance excep t th a t th e elon gated fa c e s meet a t approxim ately 90° a n g le s ,
The ends o f the c r y s t a l s , because o f c h a r a c t e r is tic h a b it, or prominent
c le a v a g e , have a rough resem blance to' rhombs as in d ic a te d in th e fo llo w ­
in g sk e tc h .

I r r e g u la r ity o f th e fa c e s preven ted any s a t is f a c t o r y measurement o f the
a n g le s .
By means o f X-ray exam ination an attem pt was made to determ ine the
c r y s t a l system, and u n it c e l l dim en sion s.

The powder d if f r a c t io n p a tte rn

fu rn ish ed accu rate in te r p la n a r spacing in fo rm a tio n , but the in d exin g o f
the observed l i n e s was not obvious from the p a tte r n as would be the case
fo r the cubic system .

Attem pts were made to in d ex the l i n e s , u sin g a lig n ­

ment ch a rts fo r the te tr a g o n a l, sim ple hexagonal and hexagonal rhombohedral
c r y s ta l system s in th e manner .o u tlin e d by Bunn ( Ilf.).

For a l l th ree system s

223

i t was found th a t no a x ia l r a tio valu e gave a s a t is f a c t o r y f i t between the
observed and t h e o r e t ic a l l a t t i c e plane se p a r a tio n s , in d ic a tin g th a t the
CaBeF4 c r y s t a ls did not have u n ia x ia l c h a ra cter.
Crude attem pts were made, u sin g the powder cameras, to o b ta in r o t a t ­
in g s in g le c r y s t a l p a t t e r n s .

Due to the la c k o f s u ita b le o r ie n ta tio n

c o n tr o ls , and because o f the r e s t r ic t e d a x ia l le n g th o f th e se cameras,
on ly a few d is to r te d la y e r l i n e s were o b ta in ed .

Use o f the approximate

u n it tr a n s la t io n v a lu e s thus obtained (fo r th ree d if fe r e n t o r ie n ta tio n s
o f the c r y s t a ls ) was found to fu r n ish no help in determ ining the true
a x ia l r a t i o .

Attempts were made u sin g th ese v a lu e s , and to a lim ite d ex­

te n t by t r i a l and e r r o r , to index th e c r y s ta l in the orthorhombic and
m onoclinic system s w ith out su c c e s s.
U sing analogous s tr u c tu r e s , such as CaS04 (w ith which CaBeF4 might
be expected to be isomorphous) as a p o s s ib le c lu e to a x ia l r a t i o s ,
a d d itio n a l t r i a l s were made but again no com p letely s a t is f a c t o r y ind exing
was p o s s ib le .

I t was d ecid ed th a t proper in d exin g would o n ly be p o s s ib le

by the use o f s u ita b le s in g le c r y s t a l cameras which were not a v a ila b le .
The X-ray exam ination i s th us a t odds w ith the o p t ic a l m icroscopic
r e s u lt s in th a t i t i s in d ic a te d th a t th e CaBeF4 stru ctu re i s not u n ia x ia l.
I t i s worthy o f note th a t th e stru ctu re o f CaS34 i s a lso in some doubt,
as i t does not e n t ir e ly f i t th e rhombohedral system according to some
in v e s t ig a t o r s .

I t would be ex p ected from th e s tu d ie s o f Ray on th e i s o ­

morphism o f BeF4 T and S04 ~ th a t th ese compounds should be q u ite s im ila r .
The in te r p la n a r sp a cin g s from powder d if f r a c t io n p a tte rn s are not very
id e n t i c a l , but here s c a tte r in g a b i l i t y p la y s a la r g e p a rt so th a t com­
parable s i z e s would not n e c e s s a r ily mean id e n t ic a l X-ray behaviou r.

22k

The a n a ly t ic a l r e s u l t s fo r s e v e r a l o f the wet p rep a ra tio n s are
ta b u la ted b r i e f l y b elow .

For o th er d e t a il s see Appendix 1 5 , Part H.

Date o f
A n a ly sis

W t. B eO ,¥t. % V ol. KMnO^ Wt. % M olal R a tio ,
g.
BeO
m l.
CaF2 CaF2/BeF2

5 /2 9 /5 0
ti
n

CaBeF4 Sample
P rep .
W t .,g .
"A"
"D"
"F»

0.1*000 0.01*55
0.3950 0.06k 7
0.1*000 0 .0 692

21 .it
3 0 .8
3 2 .5

133.3
123.0
119.25

67.3
62.9
60.2

1 .89
1 .2 3
1.12

Data on the X-ray powder d if f r a c t io n p a tte r n s o f CaBeF4 and CaS04
are given below .

The m a ter ia l used fo r the farmer was a crushed sample

o f m a te r ia l (c o n ta in in g a l i t t l e CaF2) removed from M elt J o f the CaF2BeF^ system .

The l a t t e r was ob tain ed from a resid u e o f one o f th e salvage

o p era tio n s which was found to be CaS04 .
fo r CaF2 and CaS04 are a lso ta b u la te d .

The Hanawalt ( 63) v a lu e s given
R e la tiv e i n t e n s i t i e s are given byi

VS, v e ry strong; S , strong; M, medium; W, weak; VW, very weak; and D,
d if f u s e .

The l i n e s marked w ith an a s t e r is k in the CaBeF4 column may be

due in whole or in p a r t to the presence o f CaF2 which in t e r f e r e s a t th ese
p o in t s , but i t i s b e lie v e d th a t th ey are due to CaBeF4 p r im a r ily .

Addi­

t io n a l l i n e s were p resen t in th e back r e f l e c t i o n r e g io n , but th ese were
not measured in t h i s c a s e .

The f a in t l i n e s , known to be due to CaF2 , in

the CaBeF4 p a tte r n are not ta b u la te d .
The p o s s i b i l i t y o f in te r fe r e n c e by CuK^ r a d ia tio n was checked fo r
the CaBeF4 p a tte r n and no in d ic a tio n s are p r e se n t th a t any o f the l i s t e d
lin e s are due to anything except CuK^ r a d ia tio n .
W hile some apparent s i m i l a r i t i e s may be noted from the c o in cid en ces
o f CaS04 and CaBeF4 l i n e s , i t i s b e lie v e d th a t t h i s i s p u rely c o in c id e n c e .

225

COMPARISON OF CaBeF4 AND CaS04 X-RAI PATTERNS

Hanawalt
CaP* V alu es 9
k°

CaBeF4 Sample J ,
F ilm # 9 ,
A0
Ca. It.55
3 .I4I

3 .1 6

M

3 .7 2
3 .5 7
3 .1 5
2.8U
2 .7 9
2 .5 6

s

2.32
2.27

MD
VSD

S
2 .5 6

2.27

2 .II4 S
-H.
1 .9 3

S

1 .6 5

5

1 .9 3
1 .8 2
1 .7 6
1 .7 2
1 .6 8

CaS04 Sample ,
Film #17 5
A°

M
S
S
M
VW

1 .3 7

1 .2 5 6

1.117

1.50

w

1.143
1 .3 9

M
W

VW
VS
VW
S
W
W
S

3 .89
3 .5 9
3 .1 1
2 .8 5

2.32

S

vw

2 .2 6
2.20

S

2.20
2 .1 7
2 .0 8
1 .9 9
1.93
1 .8 6
1 .7 5

S
VW
W
W
W
M
W

1 .6 5

M

1.59

vw

1 .5 6

VW
VW
w

1.52
1 .5 9
1.53
1.50

Hanawalt
CaS04 V a lu e s,
A°

w
W

S

2 .5 6

2.08
1 .9 9
1.93
1 .8 6
1 .7 5
1 .6 5
1 .5 9
1.52
1.587
1.520
1.3 9 5

1.360

1 .3 5 M
1 .3 0 M
1 .2 6 * VW

1.32

W

1.30

w

1.29

w

1.25*?WJ

1.23

vw

1.318
1 .2 9 6
1.2 7 5
1.215

1 .1 6

M

1.15

w

1 .2 2
1 .18
1.12

VW
VW
M

1.197
1.103

1.09

w

l . l l i vw
1 .1 2 * VW
1 .1 0 VW

S

226

The g en eral accuracy o f measurement i s to w ith in 0.0lA.0 ^ excep t fo r the
f i r s t two l i n e s fo r CaBeF4 which are so d if f u s e th a t0 .0 2 to O.03A-0 error
might be q u ite p o s s i b l e .
Exam inations o f oth er m elt samples and the w e t-s y n th e s is products
gave e s s e n t i a l l y th e same v a lu es fo r th e CaBeF4 l i n e s as th o se shown
above.

APPENDIX 11

227

appendix

11

LITHIUM FLUOBERYLLATE; RESULTS OF SYNTHESES OF MONOHYDRATED
AND ANHYDROUS COMPOUNDS; X-RAX DATA
In p art E

o f Appendix 1 5 , the d e t a il s o f a n a ly s is and c a lc u la tio n

are given fo r p rep aration s o f L i2BeF4 :lH 20 and LigBeF^. obtained by

a

com bination o f

p re­

wet and dry s y n th e s e s .

The f i r s t product^ L i^ e F ^

pared from analyzed (NH4) 2BeF4 and LiF by v o l a t i l i z a t i o n o f NH4F had an
a n a ly s is o f 3 3 *b mole p ercen t BeF2 .

T h is checked e x c e lle n t l y w.Lth the

c a lc u la te d inp ut o f the rea g en ts and w ith the w eight o f product obtained
fo llo w in g v o l a t i l i z a t i o n .
T his product was converted to th e monohydrate by s o lu tio n in water
and vacuum ev a p o ra tio n .

In the recovery o p era tio n s the r e s u lt in g product

probably l o s t a l i t t l e o f the th e o r e t ic a l water o f c r y s t a lli z a t i o n .

The

hydrated product had the fo llo w in g a n a ly s is .
Wt. P ercent
H20
LiF
BeF2
1 3 .9

U6.1

3 9.7

T o ta l
Wt. P ercent
9 9 .7

Mole P ercen t
H20
LiF
BeF2
23.0

5 2 .2

2^.8

D esp ite th e good a n a ly t ic a l r e c o v e r y } the v a lu e s show more d e v ia tio n from
the t h e o r e t ic a l v a lu e s than would be expected from the probable error o f
about - 0 .5 mole p e r c e n t.

S lig h t dehydration o f the product would e x p la in

most o f th e d e v ia tio n .
A fter thorough dehydration (r e q u ir in g more than 1 1 0 ° ), th e same
sample had th e fo llo w in g a n a ly s is c a lc u la te d on th e dry b a s is .

228

W t. P ercent
LiF
BeF2
5 3 .5

T o ta l
w t . Percent

5 6 .1

Mole Percent
LiF
BeF2

9 9 .6

67.8

32.2

This r e s u lt d is a g r e e s w ith the t h e o r e t ic a l value by about 1 .1 mole p er­
cent .
A s im ila r r e s u lt was obtained on a n a ly s is o f M elt "d" o f the LiF-BeF2
system , which had been prepared from the same monohydrated product.

This

m a te r ia l, a f t e r u se in one thermal a n a ly s is , had the com position 67.5
mole p ercen t LiF and 3 2 .5 mole p ercen t BeF2 .
The s o l u b i l i t y determ ined fo r the compound (See Appendix 1 5 , Part E)
was approxim ately 23 grams o f L i2BeF4 per l i t e r o f water as determ ined by
a n a ly s is o f th e mother liq u o r .

This checked f a i r l y w e ll w ith an estim ated

s o l u b i l i t y o f 26 g . / l i t e r based on f i r s t appearance o f permanent c r y s ta ls
during ev a p o ra tio n .

These r e s u lt s are about double the value o f 13.2

g . / l i t e r given by Ray ( 1 2 8 ) .
In connection w ith the ta b u la tio n o f the -X-ray powder d if f r a c t io n
p a tte rn fo r L i2BeF4 obtained in t h is work
fo r r e s u lt s due to T h ilo and Lehmann ( 1 6 6 ) .

a d d itio n a l p a tte rn s are shown
These authors have p resen ted

evidence fo r the e x is te n c e o f th e compound LiBeF3 .

In th e s e c tio n d is c u s s ­

in g the r e s u lt s fo r the LiF-BeF2 system t h i s X-ray evidence was q u estio n ed .
Their data i s reproduced in co n ju n ctio n w ith measurements made on L i2BeF4
(a s ob tain ed from m e lts during t h i s work) and a sample o f pure L iF.

The

la t t e r i s inclu ded fo r comparison s in c e , in t h i s work, a t l e a s t tr a c e s of
LiF tended to be p r e se n t in the s o lid products although forbidden by a
t h e o r e t ic a lly com plete phase r u le " r ea c tio n ” p r o c e s s .

For both LIF and

229

BeF2 c e r ta in l i n e s were found by experience to be p r e se n t in m elts o f
w id ely d if f e r in g co m p o sitio n s, and th ese l i n e s are id e n t if ie d a c co r d in g ly .
The T h ilo and Lehmann data co v ers three d if f e r e n t sam ples, d esign ated
’’a", "b11 and "c” .

T heir "a” sample was b e lie v e d to be L i2BeF4 .

The nb"

sample rep resen ted m a ter ia l from a norm ally cooled m elt having the o v e r - a ll
com position o f LiBeF3 .

According to them such a m elt would form L i2BeF4

in s te a d o f LiBeF3 because s p e c ia l c o o lin g technique was not u se d .

The

"c" sample was b e lie v e d to be LiBeF3 , obtained by s p e c ia l c o o lin g o f a
m elt c o n ta in in g a s lig h t e x c e s s o f BeF2 over the sto ic h io m e tr ic r eq u ir e ­
ments .
The ta b u la ted v a lu e s are in terms o f
Lehmann) where G i s the Bragg a n g le .

& ( c a lle d 9 /2 by T hilo and

For convenience th e v a lu es given

o
for F ilm #33 o f t h is work have been shown a lso in A u n it s .
The i n t e n s i t i e s are shown by th e fo llo w in g a b b r e v ia tio n s.
VS

- very strong

S

- stron g

M

- medium

W -

W - very weak
VVW - v e r y , very weak

weak

D - d iffu s e

(3

-

Cu

lin e

The samples o f m e lts o f the LiF-BeF2 system , whose p a tte r n s are
shown, had the fo llo w in g com p o sitio n s.
Film #33; M elt B, c o n ta in in g approxim ately 8 mole p ercen t BeF2 .
Film #76; M elt 3Q, c o n ta in in g approxim ately 32 mole p ercent BeF2 .
Film #52; Sample c o n s is te d o f crushed L i2BeF4 s in g le c r y s t a ls
dipped from. M elt 0 (w h ile c o o lin g ) , com position o f the
m elt approxim ately U8. mole p ercen t BeF2 .
Film #32; L iF , reagen t grade, powder.

230
m
0

•H A

^ -ir

fa
O

CO

O

- P

p

- P

0
d
p
0
o n
Z M

O
O
d
•H

41 f

0~
"f

a
41

fa fa fa fa fa fa
0 0 0-0 0 0

fa
0 fa
0
N n n on
fti Tl oi
0 P P
ffl P P

•H i—I
to

A
41 -f

Tf

CN

d

^

^

^

fa
fa fa
0 0 0
H

A

^ f a

^

i-P t-Ph—
I h-q i-Pi-P

•H P

p

*r*l P

P

P

p

< P CM
CP r l
•
<P fP

A
XA
•
A

•rl

P P

P

P

P

P

A
^

^

fa
0

^i» M
1M
1M
*M
* ■sf
fa
0 fa
0 fa
0 fa
0 fa
0fa
0 0fa
^
mc,D3>‘V X D3.fc, £

^
mo A f a
•rH *H *H *in

•H P
P p

1-l^hlhl

P

P
P

P
P

P *H iH *H * n
p p P P P

<

#

0

e

o

a

p PP

CO

r l M3
P H ®
M3

.

•

op

CM
<D

A

Cp

A

c—
•

xrC'op
cp cp
•
•

A
A

A

A

A

H *P
^

HP
P

A
O

•

1—1

fa

CO

co

CM

CD

A

op

CP

!>

o

0

o

A -sf

CM

CM C p
CM CM

P f

a

CN
CM

to
P

O
CM 1
XA O
E

P

p

CO
CO

g

i> £

>

g

>

; s ^ >

[S

Is

ACN
•
M3
CM

!> r> g

!>

0

A 0)

rl rl
fa
£j

C—1—1

vO \ o

r 1

d

P
IS

CM VO o
CO h H

1—I t —I

1—li—I1—1

«

( « «
c r\ r o 4 4

vO CD 1—!

CO

0
0

M3 O I P On O XA
cp On cp c— cm p f
• «
'LO\ ' O

rH 1
—I rH rH

a

-

co

a

00

O

O

CO

r-

c p

A -

O

0 \

O r l CM
CM CM CM

Cp 1A
M3 M3 M3 c p
•
• • •
r p _ d - X A M3
CM CM CM A

CO
>

3

I—I

H CO
CO P f
• •
A -C O
CM CM

fa
M
3 ry
A- fp

OF

s1 r-Pl
1
—

A

CD

•H 0

fa g

COMPARISON

p
co

p

co

CO p r

co
>

1
—1
co

1
—1
1
—1

£

C\J

CN
1—1

CN
1-- 1

0

1
—1

A1
—1

g

Is

g

P l ’ XA
A OO

C— M3 P t

s

M3 0

A
1
—1

1
—1
—
11
11

g

£

IS ^

M3
rH 1
—1

Q

*P-73. I s I S

A
A cq

pf CD
■H 0
fa g

0

O

Pi

g*S
o P

rH

ctf

co
IP A -P C

A

M3

A
1
—1

3

o rcs
: R ffl
P fa

g

P
1
—1

0

P j-p jr—1 i—
1

>

CO

0

A CD

rpXA

—»
1O

1

•

CM

>

CO

A
cp

00

co

A

A
P

M3
1
—1

AP

CN CN
1—1 1
—1

A
P f

xpco o

C— CM CO CO X P

Cp

CM
CM

A P f X A M3 M3 [— CO
CM CM CM CM CM CM CM

CN
CM

A

CM
CM
CM

O
CM

CO
>

co
>

CO
>

r-.

A

A

•A•

00

A

A

O

o

>

O

•

A

P

•

OO
P f

ro

CO A

A

A

•

Is*

•

CM

Ox J

•

A
A

•

A
A

1—1

•

•

0CM

CO
CM

-P c o l >

P
P

O
PI
"S

cti

fa

O
■H

g

s p

P
<0
P
0 C
O <3
-4 O
& &• a

CO

M3 M3 M3 r l 1
—
I

CM X A

CN r l A

0•

co

O H CM A P f

A M3

A- OO CN

0
A

g

CM CD

l c

q

!—I I—I

1—11—I1—I 1—I 1—I

£3 o
CO o

CO

>

■~=\ IS

0 0
1

co co

rO

O CN M3

•
•
•
— CM
O 1
CM CM CM

A P T M3
CM CM CM

A - O '- CO
A CM CM

CO CO P

CO

o

I—I
•H
fa rt
Eh

H

s
0
rl

CO

m
C
M
P
p

d
1
rl
O
O

CD

Q

IS > «
A A P f
H

CM c p

1—I1—I1—I

CO

>

CO

vO
*
—J
1
—1

'LA
•
vO
1
—1

CN
•
f> 1
—1

CM
•
CN
1—1

co 2 ?

P

> E>
-=s 0
O H
CM CM

•
CM
CM

C"— C— vQ
•
•
•
CM CM CM

H

CNMD
•
•
C— CO
CM CM CM
•

0
1
—1

A

CO

1-- 1
CM

O
A

CM CM A
vQ vO M3
CM CM A

!>

P P

O xp H

IS

CO >

OO 0 \
p p

A

co
>

Is co
■A •

IS ^

P

P
CO

p f

XAM3
P P

rp p f

S3

co

0

C pco

co

VS

POWDER

PATTERNS

5

X-RAX

>
SQ.

rl P
fa •H
S
h

co • H
-p f a

*

p

co
>

A

O

1

rd °
P

a

fa

A
CO
M3 T A
•
1
— rH

• *

A

p

O
IS

*

A X A CO
CN CO A •
P i —I P

Continued

a

next

cn W o

page

I
o
cd
Pi

231

M
1M
1
© m3 © <d

plH pCj

nO pa On
la L A d

CA
d

H

•H

d d d d d d d

iH

H

*H *H *H 'H -rl *H *H

>!<
co
1
—1

rH rH

rH

g

•H

CM
CA

hp*

s

On
CA

d

d

iH

iH
3: g g

g

CM

d

O

"LA

LA "LA"LA
CM LA CM

CN

nO

OO On
LA LA M3

-d

d>

©

CO

NO

d

d

'S *«:

£

ft f t f t

ft

d

CO

00

NO

£

C"H
CD

d

LA CA
CM CM

VSD

H

t—
1

1 .2 9

CO

%

%

1 .3 2

*H *rl *H *H

hlhlhlK l

^ ^ vj* ^
Tjl
‘S ' S ' S ’S 'S 'S

1 .3 6

Os

0

H nQ "LA
LA CM H

c—no no la
L A d rH ON

d " 1—1
LA CA

"LA
NO

d

CN O rH
CM A A

d LA nO nO
CA CA CA CA

CO CN
CA <A

d

NO
d

£h

© -p

d Co m
-P -H S3 CvS
o i-q ©

co

g

h^
J>

■LA

U \N -H lA
no CN
J C\J CN
* • • •
ON O rH 1—I
CM CA CA CA

1
^.

K=* ^

g
d
LA

d d 0 a - cm
1
—1 CN S3 LA CA

.CM
CA

d
CA

no no

- Continued

o

A-

-P -H ! 3

^

g

O d

d

a?
©
g

QQ

g

S g

Q CM LA
MJ CM H

A~
NO

d
LA

CM O
tH CN

LA ALA CM

d
A

CO

C A N O CA

ON o H
CM CA CA

CM
CA

d
CA

NO NO
CA ca

OO CN
CA cA

H
d

CN
d

NO CO CN
LA LA NO

to

CO^ £>

CO

NO
•
ON
CM

CM

LA

NO LA L A d A

A A

CM

A d

CM CN

CM

NO

rH
A

CM
A

A d land aA A A A A

ON O
A d

A
d

d A
d d

C— Ad d

CN
d

ft

•

•

Powder

Patterns

N
=- 1>
1
---

U CO
CD Q) -p
CO
d fl O &

Q

to

CO

CM

N O L A C A

ON o iH
CM CA CA

l a

^

^

g

A-

O CN A-LA

CA

LA nO A—
—
j —
—
-1 —
■
—
j

ON
d

0 H CM A
LA "LALA LA

to CO

CO

^

3s

CM H
•
NO AA A

NO
•
OO
A

CM

1— I

nO

"LA

cm

l a

CM

CA_d"LANO A
CA CA CA CA CA

CN

CO

NO

CA

c—

^

rH CM

L A nQ

C om parison

of

X-ray

■P
©

r—co cn
CA CA CA CA CA

>

CA

P-

LA
A-

CO

d

g

s IS
" L A - d CA
H

o
O

*

*

*

*

CN o
H CM
CM CA CA CA

ON
•
CM
A

•

d
A

g

CO O H

NO LA

LA

LA

O CN LA A

H A d
d d d

A

CO

CN
d

H rH CM A
"LA LA LA LA

NO

d d

d

232

In th e ta b u la tio n j th e h o r iz o n ta l r e la tio n s h ip given by T h ilo and
Lehmann fo r th e variou s l i n e s has been p r e se r v ed , although o b v io u sly un­
j u s t i f i e d in s e v e r a l c a s e s .

In s t i l l more ca ses a r e a l i s t i c a p p ra isa l o f

t h e ir measuring accu racy, as judged from the in te r n a l c o n siste n c y o f the
d a ta , would in d ic a te no s ig n if i c a n t d iffe r e n c e s between some v a lu es shown
at d if f e r e n t l e v e l s in Columns 1 , 2 and 3 .

I t i s assumed th a t by so

d if f e r e n t ia t in g the v a lu e s th ese workers claim ed th a t they belonged to
d if fe r e n t compounds.
The r e s u lt s o f t h is work, shown in Columns k, 5 , 6 and 7 appear to be
in t e r n a lly c o n s is t e n t to a t l e a s t 1 0 .03° , except fo r very weak or d iffu s e
l i n e s where measurement becomes more d i f f i c u l t .

From oth er evidence ob­

ta in e d during camera c a lib r a t io n , a gen eral accuracy o f t 1 p a rt per 1000
was found under fa v o ra b le c o n d itio n s .

The camera s iz e (tw ice th e diam eter

used by the oth er workers) made more accurate measurement p o s s ib le , but
corresp o n d in gly low ered l in e i n t e n s i t i e s .

As a consequence some li n e s were

ob tain ed by T h ilo and Lehmann which were not observed in t h is work on 3
hour exposures a t 35KV and l^ma w ith n ic k e l f i l t e r e d Cu

r a d ia tio n .

A thorough in s p e c tio n o f the comparative data w i l l in d ic a te a t most
o n ly two or th re e l i n e s p r e se n t in th e "LiBeF3" p a tte rn which may not be
reason ab ly a ttr ib u te d to L igB eF ^

The f a c t th a t the balance o f the lin e s

show good or f a i r l y good matches w ith L i2BeF4 l i n e s i s strong evidence fo r
concluding th a t th e p a tte r n s were formed by the same compound.
In Column ( 6 ) , the v a lu es o f l a t t i c e plane sp acin gs fo r Film 33 have
been g iv en in A0 u n i t s .

Those v a lu e s b e lie v e d to be due to L iF , or BeF2 ,

rath er than L i2BeF4 have been marked w ith an a s t e r is k .

233

The v a lu e s giv en fo r the r e s u lt s o f t h is work are un corrected v a lu e s
sin c e c a lib r a tio n in d ic a te d no s ig n if ic a n t c o r r e c tio n ( f o r film shrinkage
and camera diam eter) was n e c e ssa r y .

The c e n te r -to -c e n te r measurements }

made in d ep en d en tly on p a ttern s o f m elt samples show' no s ig n if ic a n t d i f f e r ­
ence in th e L i2BeF4 product produced under a wide range o f c o n d itio n s .
This range extended over m elts from 8 to I4.8 mole p ercen t BeF2 .

None o f

th ese m e lts should have produced "LuBeF^ 1} according to the claim o f T hilo
and iehmann, y e t the s im ila r it y to t h e ir r e s u lt s (Column 3) makes i t un­
l i k e l y th a t th ere i s any r e a l d iffe r e n c e in the sam ples.

APPENDIX 12

23b

APPENDIX 12
RESUITS OF THERMAL ANAIISES
OF CaF2-LiF-BeF2 SISTM
The order o f arrangement o f the ta b u la tio n i s an a lp h a b e tic a l sequence
in which a l l sam ples having a c a p it a l l e t t e r in t h e ir p r e fix have been
placed ahead o f a l l th o se i d e n t if i e d by a lower case l e t t e r p r e f ix .
The ta b u la tio n l i s t s com position v a lu e s and therm al r e s u l t s .

Almost

a l l com position v a lu e s are based on the c a lc u la te d s ta r tin g com positions
o f the parent member o f each s e r i e s .

T his value was e ith e r based on the

w eigh ts o f BeF2 , L iF , and CaF2 u sed , or on the w eight o f th e m elt a ft e r th e
e v o lu tio n o f NH4F was accom plished in c a se s where (NH4) 2BeF4 was u sed .

The

A to K m e lts were corrected in com position by a compromise between the
a n a ly se s and c a lc u la te d com positions as the l a t t e r were not a c cu r a te ly
known.
In order to perm it con d en sa tio n , the mole percent o f LiF (o b ta in a b le
by d iffe r e n c e from 100) has been o m itted .

Except fo r the Q s e r ie s and

the e , g and t s e r i e s , CaF2 was th e a d d itiv e substance used to s h i f t the
com position in a l l oth er c a s e s .
In ca ses where more than two in f le c t i o n s occurred the th ir d in f le c t i o n
has u s u a lly been p laced in th e fo llo w in g lin e under the second i n f l e c t i o n ,
or has been oth erw ise n o ted .
Where the column shows a dash i t i s in d ic a te d th a t the m elt was not
cooled to the l e v e l a t which the expected p ro cess would have occu rred .

235

A b b reviation s
In column k, the a b b rev ia tio n s used

to s ig n if y d e p o s itio n

o f CaBeF4 and

L i3BeF4 are GBand LB r e s p e c t iv e ly .
In columns 5 and 9 , X s i g n i f i e s v is u a l d e te c tio n o f f i r s t or second
c r y s ta lliz a tio n .

T s i g n i f i e s thermal d e te c tio n .

o f in f le c t i o n s th e fo llo w in g were used:

With r e sp e c t to the type

S , sharp; G, good or d e f in it e ;

F , f a i r or l e s s d e f i n i t e , P } poor or in d e f in it e ; VP, very poor or b a r e ly
s u f f i c ie n t fo r r e c o g n itio n .
m e lt.

D s i g n i f i e s d e te c tio n by dipping sample from

M s i g n i f i e s th a t m elt o n ly th ick en ed w ith no d e f in it e liq u id u s p o in t.
In column 8 , th e shapes o f p la tea u s are in d ic a te d by:
Long f l a t — f l a t or very s l i g h t l y s lo p in g , w ith in the temperature
range shown, fo r about 20-25 m in u tes, fo llo w ed by abrupt drop
in tem perature.
F la t — same

as above, excep t about 10-20 m in utes.

Long slo p in g

—same as above, but w ith a decrease o f about 5 to 7

degrees during a p erio d o f about 20-25 m in u tes, about two
th ir d s o f drop occu rrin g in l a s t 10 m inutes.
F la t slo p in g
by smooth;

—f l a t p la tea u o f about 10-15 minute du ration follow ed
t r a n s it io n to a resumption o f the o r ig in a l slope .

S lop in g — p la te a u w ith gradual slo p e o f about l° /m in . fo llo w e d by
smooth t r a n s it io n to more r a p id ly c o o lin g s e c t io n .
Short slo p in g — sh ort p la te a u o f about 5 minutes or l e s s d u ration
fo llo w ed by smooth t r a n s it io n .
Short — same as above except w ith more abrupt t r a n s it io n to more
r a p id ly c o o lin g s e c t io n .

236

In column 9 3 nc r u s ttt s i g n i f i e s unusual type o f s o l i d i f i c a t i o n mentioned
in t e x t .

"Abrupt11 s i g n i f i e s unusually- sharp drop in tem perature a t end o f

a p la te a u .
Where an a s t e r is k i s used the value may be e ith e r approximate or un­
c e r ta in .
In column 1 0 } th e temperatures o f disappearance o f l a s t apparent liq u id
or appearance o f f i r s t liq u id (d u rin g c o o lin g or h eatin g) are in d ic a te d by
C or H.

CH i s used i f the two tem peratures were n ea rly the same.

These

tem peratures are v ery approxim ate, although th ey should be comparable w ith
one another.
Large i n f le c t i o n s a fte r h e a tin g through an in v a r ia n t p o in t are in d i­
cated by "Jump-".

-p

C o m m en ts

i—I
ft
O

ro

CO

ra

Is
o

£
o

£
o

rft

Shape

h CO O
Eh E-i Eh S

£ h EH £h

CO
»■ Eh

O
o

CO

CO

CO

ft ft co

O CO O CO
Eh Eh Eh E h B

B

Eh Eh E h

UN

I

CM f t U N U N MO MO GO CM "UN 1 A U N

o

OO C M - ft U N U N U N U N U \u \ O - H

ON-

f t

UN UN UN UN

I O CM 1
—I CN O M 0
I CO H CM CM ON ON
MO O O O O O

p

I o

—f t

ON 1
—I 1

[>— C— C— i f t C— [— I f t ( f t MO 0 - 0 - 0 - 0 0 - 0 0

I

CM ON - f t ’ NO ON U N
CN f t CM CM CM ON

1 'O N N N N N

Type
T em p.

CN U N f t t C Q
M3 M 3 0 C M C N Q U N C M Q C N - f t OO CM O U N ON f t UN on O O Q O
U N CO UN ON ON U N U N 5 5 o
C N O O M3 - f t CO H f t M J MO U N —
— C— C— O - o o o o o a o c o c o O C— O C— C— O O O - O - f t - C— O - N - CO
MO CO O - O - C — C
>r
»!< i|<
>!<

O b sv ,

0
1

C— %
o

UN

S

%
o

S '

-ft

Type

B

= r

^
P

b

ft

b

O ft

O CO
E

B

S

CO

(3
4 p-l
Eh Eh S

o

£ 3
P

cd <3 s

p o

CN CO O

•

•

•

&
p

o

( f t M3 CM 0 - 0

•

•

*

•

0} 01

r

b

o c o o c o

•

& =

=

ft

CM 1
—I 0 \ 0 0

I f t ( f t M3 M3 ' 0 ( M

H

I

CM P N - C t l f t ' f O - C O

I

I

I

I

I

I

I

c o o q m O 3 0 .MO CM U N —f t o

O \'0 -J o o m
m3

- f t

I

I

I

I

I

ft

I

I
o

=

ft

% ‘a

OO

OO 3 0 l>— CO

- f t

U N -ft

O

- f t

O

CN CN O n ; ON CM CM f t

I

CN

ft ft

O

1—I 1—I ft 1—Ift 1—I

CM O N - f t U N M 0 O

I

—f t

O - O n o f t CO f t U N C O f t CM
f t f t f t CM CM
1
—I 1
—I 1
—I r—I CM
f t

CN U N rH c o VO CA f t MO ON CO - f t O

M3 U N f t O

I

^ b

B

00

O

1
—I CM O N - f t U N M3 MO O

I

^

o

C— M3 U 3 MO

PQ

=

o

—| U N O - CN O CM O N O C M U N O ON ON M3 CA O
ON ON O - 1
rH I—I CM CM CM CM
H r l H H CM (M

CM ON ON CM CM CM CM CM CM o

ft

e U

01 0!

oj 01
f*T. pEj Cm

- f t

f t O ft

0EH ^ =

=

Eh S

b

1 ^ = ft* = =
1o
ft

ft ft

ft —
ft -ft

H CKOO 0^0 © O\03

o~

C aFP

I n f le c t io n ) ,

T em p.

fo fo to

ts
o

ft

r^-l

*

B eF a

P la te a u

2nd.
(o r

or
C r y s t.
1 st.

I n fle c tio n ,
P ercent

C o m p o s itio n ,

M o le
M e lt

CN

ra

- f t

- f t

CO

PH
o
o

ON

ON

-ft

ft
ft

o
o

O
o

- f t

0

-P

ft
ft
ft

ft
ft
ft

a3

ft
ft

I
0
-p

0
-P

I

I

I

I

ft

I
Q

CM O N - f t U N M 0 [■—

I

I

I

I

I

next

ft

u
0

p age

rH

n tin u e d

L iq u id

A pparent

237

I

C
O
•H
-P
O
CD
I—I

o

CO S
I—I
ctf
nQ

G
■H

s

<0
G
O

-p

IN—
NO

bQ

>

cti

o

CD

G e

G

o
o

ft

O

Continued

ft

next page

238

Eh

o

LA
-G

ElO
G
•H

I1

r~ 1

CL

-z£

-ft-

o

On
CO CO

CO CO CO

H Eh £h

Eh S

o o
30 C C
no n o

(A G5

EH S

CM
(N. CV.
I
O O c o O On
< D -p o \n
I CM LA C— CO CA
G On O H
o n o in- r— | NO NO NO NO kG
S

CM

ft =

CH
Fa
•H C p q

ft

G

O

G

G

OOG

I no r - I
I NO NO I

c

cO G

PL,
Eh

P-.

I

i

ft s

C-

-G

G

(H O

H

o o G

g

s[i

*

-G

I

if t

cn f t

pq

ON ON o

I co

o O

la

c— r— c o

GGLTv I c o G q O n O n G l a
NO IN -C O L A L A NO nO
nO n O nO
Lit vV
*
H
< *r«
:«
Ot

PG PL, Fa * G3 G LO
& * £< ' & ^ ft Xi XJ
=

l-ft r
I O

l<! X!
0CN. CM

CM

ft ft ft ft
G O O G

=

f t B f t ft
G

G O

H

G

•

LA ca
On p a
NO P-

t;c at
Ia

Eh S

CM

ft,
O

_G O
G -G

On

On

cm

•

rH rH H

nO

CM CM CM i—I i—I CM CM CM CM ,—1 CM CM CM CM CM CM CM CM CM CM CM

1C>1<1'fi. * *

rH CM p a - G - G LA rH H CM CA-G G CM C A - G L A
I
I
I
I
i
I I
I
I
I
t

G

ft

at
p a nO

•

f t

cO

L A -G p a

•

G

nQ n O

G G G G G

I I IG

IN—c— nO nO nO nO O O O On CO CM G G O On CA CM CM G G O O

•

Eh

On On co

no

ON O CO LA LA CM O n O n G L A O n - G O n H - G QO LA CO CMLA l a On CA G
•

c5

CA Q CA G

O X - CA rH vO• H • -G• L A• O • G O CO L A L A O -G G • O • co• -G•
• • • • • • •
•
•
•
I
»
CM CM CA L A "LA c o H - G CMLA CO H -G C A N O CO G G -G NO P A NO CO G
G i—t
!< >!C
—I i—I
I—I

i—
I i—I i—II—I

rH

On no
_G

"LA O n CM CM MO O

PA L A ON O

CM

o

c- 0J 03

ft s ft ft = = f t * = ft ft f t s

co

O CM _ G
I L A l a La l a
I nO NO NO nO

c*»

01

CM

I

I l G

PH C
a
U*

c^. O"

ft i—1
M CO

P.
O
t—1

at

CM L A -G O O CM LA CA C O O l a g c o l a cm
o o n On 1 A J CA ON A G 0 \ J nO nO l a o l a
NO cn -n o n o p— in- in- sO nO nO nQ in- vQ NO NO IH- r
;
&
;jS ?JC
cO
■LA Eh S
Xj

I
I

“rl

o

cO

EH EH

bO
C3

-P bO
o C
<1) •H
i—1 f t s

I I I I I

f>
h

Fh
Fa

G cm ( A - G G L a n O G CM C A -G

I I I I I I I

o

I

EC

I

I

I

CA PA
CM CM

-G

i l a■

239

eg

0

O

aCA

1

o
CM
ft
V/

V

o
ft
ft

A
CM

ft
V

page

eg

9

-g
TO
CD

W
H
G
to
CD
G

Ed
CM
XA

Ed
O
XA

ft
G.

ft
&

A)

As

Ed

o
ft
|
A>

tH
r—I
to

I—I

bQ

bQ

g

g

O

ft

ft

K

o

ft

—
l
to

bO

bfl-p

ft

o

ft

Q

Q

CO

PI f t

PQ pq pq pq

J

O 'O O C M 'IA
On On rH ft" CM
f t f t A A NO f t f t

O

8

XA

O

s

m s

M X M M lx| M X
<g
Pi CM
ft
ft r
^ = f t ft

%

%
o o

ft

ft O

^ W H O J
A

CO X A

iH A M D ON rH CM X A

i—I

ft O

CM "LTV On A

H H O Cr\°Q
A

A

• N O«
ft

CM
CA CM C\J A A

H CM A f t A H CM
I I I I I I I

H

A)

CM

ft

CM

ft

o

H

X A O n ON

nO

*

A

cA
I

G
o

o=

f t

ft

•

P
G:

e

l-l

P
G

ft
ft
bO

p

p

G

ft
ft

bQ

G
ft

G
O:

o

ft

CO f t

O nOXAXAXAnOXAXAC—COCO
ON CO
■f t f t f t f t f t f t f t f t f t f t I ff tt- f -t+
■ f t f t f t f t f t f t I —j f t

pq

O CO OO X A
G O ® W

XA 'O NO C

pq

J pq pq pq

s

O

CM X A CM CM o
CO A - i—I [— O
f t f t X T \ f t XA

o

co

a

- co

c— c—

f t A - f t f t f t f t f t
f t f t f t - f t f t f t f t

a

<■ f t f t o

C“

-

- f t
f t

A

ft ft
^ ^

lx!

Cg

’

f t

O = = = = ■= e = f t r

COft COco
u

p q

Q

Q

p q

&

}>^

pi

cm
cv.

^
o

.

s

= = PQ

pq PQ pq pq ^

ft

I O

o o o o o

A ® G
CM
CM
CO♦ ON• a-• O O G l S A C - H H v O
•
*
O CM CM CANO ON f t XA ( A t - O (A XA O O O
»

o

^

o ft

O

CO

S

o

to f t
to

A

e~

P

ft

G

P3 e =
ft

e

pH

CO
!x! >x! E—
i

X

ft

xa a- co O n O
XAXan O f t A
*!<

A

O

r~o R S S
f t AMO

CO

f t

bO
C

p

'O
f t fOt fftt fftt I
ftftft I

O O ft ft = 1

O ft

e

to

bO
G

O O ft
ft ft

J- f t f t A f t f t f t

PG

o

H

bo bOP
G G G

O O
CA ca
I I
4C O Q O Q O C
M
G A 0 \0 -G -d -J

ft

G

O

bQ
G bO
•H G
ft-H

bO
G
■H
ft

H
ft
to f t

G G
O H s

s

£
CD
>

ft

bO
G

•H
ftS

G O
i—I i—I

d

■H
G
G
G

ft

A>

G

-P

P
G

o

|

bO

G

!

CM

A)

EG

G
■H
f te
Q

-P
G

O

EG

ft
ft

ft
ft

bQ

bO

g

ON

EG

P

to
CD
ft
Tj
©
©

A

Continued

ft
ft

•H
TJ

next

-P

A H

f t®

X A H ® G G ®

H ft

•

i—II—I i—I

i—Ii—! i—I

ft ft

O H H H W

c A A f t ft ft A

—
\

A- O
CO
•
•
A O n ON
CM CM CM

CM CM i—! O i XA CM O f t A O n c O t— nO NO X A CA O CO X A c A CA ft O
CM f t O ON CO f t (A CM f t o A - A - l f t A - C ^ l f t C - — A - MO NQ MO MO NO MO
CA CA CA CM CM (A <A CA CA CA A CA CA CA CA ( A CA CA ( A CA CA CA CA CA

ft A A
I I I

f t CM C A f t XA
I I I I t

f t

CM C A f t X A f t

ft

ft
ft

I

I

I

I

I

I

CM A

I

O i—I i—I OmJ
f t XANO t - C O O N H H H H H

I I

I

I

I

I

I

I

I

I

1 I

A

CO
XA
A

H

I

2k 0

-O'

W
O

O

8 33 3
*

p t -

o

IX!

a

UN

3

-

iC »M

§

0
+3

CM
-O

CO

o

COQ
CO CN
CO CO

K

c c

- G
- G

- U

CO

UN
CO

3

ffi
O
-U
-= t

-P

3

-G

G

fa

ctJ CM
G
G O
s

G

—I
PQ

PC

cUN
o
CO -U

o

o

o

U N

o
c o
- G

O
•H X

t

c o

o

CO

td
o—
c
CO

o

next page

o

CO

PC
c—

UN

CO

&

Continued

ffi

B

&

fa

bO

G
•rl

•H

fa s

-p

to

fa

fa

CO

M

G
r—I :

I
I Q

oo—
oU
N
-O
'
C
O
-O
' o -O
'
- O _ 0 N iN -O

to

-p
u

-P

O
O
P

H s

l—1

to

bO

-p
G

BOS

I—I

rH

fa

o
I

o

I 1 -0

s-

o
G s

to

PQ

5 O P

fa

o
1—1

to

fa
O

•H

ho
a
•H

-p -p -p -p i— I
G G G as to
I—I i—I rH r—I
fa fa fa fa -p
u
o
C O CO CO CO

f a - p -p
o g g
H H H
W

fn

pH

C O

C /3

CO

&&&

G

•H

O

fa -p

i— 1

O oi
)— 1 i— 1

S

to

to «H

-p
u
o

bO b O

CO

CN

O o

CM

CM r o

I -U

I - G CM

I -G

I NO -G

G

bO

G

C—

CO NO rH

(M
(t |

(M
|V |

-p

G CO r H

C— _ G

U N

CM
pr |

CM
f-r j

J-COPCNPO-C
c o c o c o c o CO -G c o

CD c a

J PQ

I PQ
I PI

= 9 = = !• 9 .

-p
i—1
fa

bO

G

CO

-G -G

P>
G

H
fa

CO
G c5
o
° Pi
Pi N
PQ PI f a

rG

QQ B

O

bfl

G

•H
fa

•H

(QNONPOOWPPO
! - ^ ^ . cm^ 3 3 3 3 3 3 I
CM
f a O* O*
I 0 PQ PQ PQ
I PQ O O O

CM

G

bO

bO

G

O

i—
I
CO

bO

fa

fa!
O

G

o

I—I

On

bO
. G
•H

bO

bO
G

g
fa

•H

IO

(D CQ

<D CQ

O

o o

0 1-0

C
U N UN

NO

o o
U N

CM

UN U N

/\ A

A

rH CO CM CO CM UN CN
UN O O O Q O ( ff u N U N O UN Q O
CO
CNJ nQ CN P H Q r l C J M N CN I
c o CO r
r oo CO CO CO CO I H O O
^
, U N lA U N U U A U N U \v O NO nO NO NO NO CO I

-U -U -P ■
'- U

V

cn

Pn
UN

n

Q

-Zf
CM
Izt

^

fa

o q c Q p p ^ pfa

x

UN
CM

-G
A

I
I

c~
CM CM
O -

! mm

-O '

PQ

O

i o o

CO
CO CM CN
•
r o

•

CM nO

CN no

no

H

H CM CO
I I I
S

PQ PQ

NO - U CO

ls *a

PQ
O

o

_

C
Oo

O '

O O O rlH H NW torO O O PP

O

nO nO

CN

OO N—nO UN —G" CM H CN C— b—nO UN CM i—|

J nOnO H H I A

•

nO

UN

O O

c o

H

P O - O rH
I

I

I

O

l>- H - U UN UN UN nO nO UN

C O U N C - O -P 'C O H U N O

•

r-l

CM c o r ~ - N O
CO CO CO

I PQ

I O s O O s s

I—I

O N CO

•

CO

(N .

co

3 3 3 3 3 3 3 3 3 3 3 3 3 3

N

N

H H

O

O

Q Q co

-G -G -G -G -G

O O

P

I—I I—I I—I

c o

H CM co-uT
CM C O -U UN n O c — C O C K H r l H H r l H rH CM CM r o C O -G
I I I I I I I I I I I I I I I I I I I I
S

Q)

PQ PI pq PI pq p i pq

o

PQ

PQ

lfa -G

c o

IN-

O

CM

CO

a

i

H

O

CO - G

CN

IN- C—

3

k G -G

rH CM
I

o

I

O

NO

_G

-G

CO

-G

I

I

2 111

S

CO to
O O
K

CO
O

o Q ^0

pf
pf

CO
CM

p f p fp f
pf pf pf

UN

m
o

pf

3p f

pf
Pf

8

CO

£

ft

cm

P f to p f

p f CO
CO
O CO

o sOtN'O
o

8 8 ^ 8 3 5 -^
© O

m tK
CM CM
U N P f " U N P f CM U N U N C M P f P f P f
CM Pf P f P f P f
P f P f P f p f P f P f p f p f p f p f p f p f p f
p fp fp fp f-

pf

V

ft

©

o

bO

©

ft

ft
o

CD

Pf
I

Xj

H

■H

o

ft

CM

ift

ft

CM

d

d

s

©

ft

CD

a5
H

ft

d

«

ft

P
ai
rH

•H
O

i—
I
to

•»

ft P
a3
bO CD

bO

d

d

o s

s

e

i—
I
ca

ft

p

ft

© E
1—1

s

ft
ft bO
©d
ft o =

P

cd

l

o
Pf

t

CM IN - U N H
U N Pf P f P f
I
I
I
I
U N f t U N P f CM
U N UN UN UN UN
p f P f p f P f P f

I

(Mpf Q\ CNO
H cm CM UNUNP f CM CM f t On IPCO
ir\M 3pfNO ir\ I ao Un On UNTMUN UN UN UNp f "LAp f
CM CM cm (M cm I cm (M (M - ‘- p f p f p f p f p f p f - p f - p r

0}

w

f t f t f t
cd co

CO f t

C
M

CO

ba
d

o bO
i—1 1—1 d
ft * •H
* 'f t ft
E © o
f—1 1—1

ft

CO

d

E

5

.

o

-p
©
i—i

ft

bO
d
O E
ft

d ©
© 1—1
> • ft

d
o

d
©
f t faO
ft d
© O
i—1 i—i

ft
a5 a
i—1
ft

h, ft

O E
ft

8
ft
ft

E-

f?
E■
' ©
>

CM
P f
I

o

U N U N p f CM ON f t
U N UN U N U N U N P f p f
P f P f P f P f P f P f P f1

CM

ft

te m © i

CO f t

I)

CO

i

E- B

9
1 d
M O

©

E

£

d

bO

It

C—On mO p f
p f CM(M-Pf
l

©

ft

ft ft

II

1 ft

ft
it

I H s s

e

•H
~H
ft
O S E S ' © S
i—
1
r—1
ca
ft

I

bO
d w d co d ftco aj CO d
O Em O f t O CD f t rH f t O :
ft Pi f t Q f t Pf Q ft Q ft
bO

ft

ft
©

bO

bO
•H
ft

Xf

UN f t
P f

© =
d
d

ft O

4

o

ft
ft
f tf t
©
d
f t

©
d

p=

SC d

~

©
ft

d

bO

d
d

P

^3

ft

d
O

?

P

©
i—
1

P f

d
X3

ft

•H

ft

UN

Ph

M)

<n

P

O

-P
©
r—
I

ft

ft

(13

£c o3

f t
U N Pf
P f #

Pf

ft

to

o

l> UN

X3

d

P
UN

CO

P

&

xj
©

CO

d

u \
pf

to

UN
Xf
d

o \

ft
o

d
m
CO

TJ
ft © _

%

CD CO
CO f t

I

ftftft

ft
ft

I

I

I

I PQ

I I I I ft

a?

vO

(M O
ft CM
CMPf

Q

Pf
UN

A

UN

ti-o.

S M

n

c-

p f PQ PQ
ft O

PQ

a

O
1 CM
1 UN

O U N P f CM CO ON nO c m H
OOCOCOVD r-— C— CO CO
8
pf pf pf pf

V

A

i
1 PI

I pq
1 O

f t CM O CM f t ON O CM O n ' f t i—I n O O n CM CM <M CM cm On c o r —
On c o UN U N U N U N U N f t c o
UN f t C
— CO ON ON rH O n O n
' ■p f P f p f P f P f U N p f p f U N U N P f P f p f P f P f P f p f p f p f o

o O

ft

T3
©

ft

CO CO CO w
p>

X f x J I x iX X M

Ix if x i

pq f t
o ft

pq

^

ft ft s s.

@B

.

B . e

d

^

><j fKj jx j Ix j fx j ix*

fcs i f t

©
>

o- d
at o
o

%
= ft

PQ

ft =

X

%
o

co
©

ft

UN

M

f t CO
• •
O ft

NO
«
CM

UN
*
r^\

l>•

CM C M f ) C O

pf

O

O

O

O

O-f

f t H

H

b - - f
C

M

rH CMP f C— O CMO- CM U N ON
O O O O H H ft O o o

ft ft ft nO I—i d©

c m i—1 c o

CMUN

f t

O O f t f t CNJ CN! CM

CO

CNJ CM

Of

UN

CM

ft o
• •
o o

UN UN

NO
•
On
Pf

—1
•
On
Pf

NO
•
CO
pf

no

c—

co

On O

O

cm

UN

M cmcM
CM CM ( M CM CM C O c m c m
c m c m c m c n jc m c

l A C O CO P f UN nO CO O CM o
UNCMi—Ift I—I C—CO3 0 ON o
........................S
CMCMCMCMCMft ft O O i—I i—I i—I i— I UN U N P f CM CM CM f t
CMCMCMCMCMCMCMCM CM CM CM CM CM CM CM (M CM (M CM CM I
cm

©
rH CM
I

ft

I

CM
I

-P f

I

O IN- 0 3
T-TV rH CM C M P f UN M
I
I
I
I
I
I
I
I
I
ic y

ft
l

y
CM

cm

l

C M P f u n f t r—f t CM C M P f U N f t
I
I
I
I
I
I
I
l
i
I
i
a?
CM

ft
O

f t CM C M P f U N f t
I
cy

pf

I

I

I

I

I

IS

I

2 1 4 -2

WWW
_ p CA CA
-= E -= t-c t

J

mw

ec

O

1 A U M A CM CM O n
- X _ p -_ p -d
_d -d _ d

<X c o

c a

f f iS T T

- UN c a c o f t f t
C A ft ca P A f tf tf t f t f t f t - u f t

ps ps
O r
ca
ft ft

•
• p OO
o -d
0 ft­
1—I

X

PS

[> -

W

NO O F- o O ON

F - ca

ft
ft

- f t

CA CA C A C A f t C A f t f t

ca

ft ft f tf tf tf t f tf tf t

x

X

(D
EC
Eh UN
0 UN
£ -d
o

H ft

s

-p

e

ra

■s

o

ft

ft

-P

-P

m
P

B

P
Ph
o
m

CO

£

X

-P

- p -p

ft

-P

aj s

i—I
ft

II
II

ON d S
o

ft

f t f t

f t f t Pu

ft

CM H

UN

On O

is
o

ftft
co

ft _
I—I
ft
=

f

t

f

t

f

t

f

t f

t f

t '

f

t -

f

t f

t f

t f

t f

-P

ro
P
Ph

O

t

C
f

aj

aS

I—I I—I

f

t f

—

ft ro

ft

ft ft

to to
a PS C
o O

P?Sf =
0

-P
ai Si
i—1

ft

to
s: ■ P s
O

=

ft ft

>

> ft

t

-p f t __
a3
f t ft

=

ft

CD O

On co
t

•P
ai s

-P -P

fr r

>

ONCO
t f

Ph
O
ft

P
•rH

bO
p

CD

ft ft

f t CACM o

ro

U N U - \ X A f t f t T-A U N m U N U N U N f t f t f t - f t f t

CO

X
0
£

ft-P
0
o
9"
f t f t Ph Pi
O o X
Ph
f t ft
X
ra ra -p a3
ft
0 0 pj X
Ph Ph Ph Ph
o
O JD 0
5E f t >
0

to

II
U

bO

u

bO
to to
P
Pd P
•H f t
f t- P
ft ft
o aS
= = o o =
f t i—
1 Ph
ft
ft
ra f t 0 ft f t
ro ro
to bO f t Eft to
tO a0
a c aj Ph fl r
O O ft CD o
o o
•H

£
O
i—1
fi

ro
P e
Pi Ph
o o

to
a
o
i—
i
*
-p -P
03 aj s
i—
1 i—
1

Ph
0

o

X

ro
P

Ph
O

CD

o

p* & P P

ft
ro
P Si

w

Ph
O

O
S A)

CD
P
Ph s:
O

0)

E

*H

3 ^ 3
UN UN UN
~ d —d —d

X

Continued

rH

next page

ft
£
ft

+3
o
a)

ft

U n U N c a c a f t O n c o a o s O U N f t C A CM f t O n
U N U N U N U N U N f t f t f t f t UN U N U N U N UN UN f t
f t f t f t f t f t f t f t
nO

ftftftftftftftftft

m

OOCO OQOUNO COCMCM
O O 03 CO
o

f t UN
00
N
O OCOUNCAO'ftCM
t- nQmd nQvO U\ 03OO
On UNUN

> < j ! X X X X X X X X X

X X X X X S ^ O ' ^ X

n
n
UN f t CA Q f t MO r— U N NO
f t- lf tf t- f tc o
UN UN UN nO O -C O ON O CA n n
vO f t f t f t f t f t f t f t U N U N f t f t - f t f t f t ' f t f t f t f t U N

■LA

to

co co

X

X

ft Eh Eh
X

X

iX

X X

X

X

CA CA O O f t
f t CO O n

ft

nO

■ f t f t f t f t f t f t f t f t f t f t f t f t f t f t UN

co co
£ ^

x

X

X

X

X

CJ

ft

&

■ ft

ft

ift Si ift CO

CM U N O n CM N O O N f t
*

O

•

•

O

o

•
H

•
iH

•
H

•

O • C—CM

ft

ca ca cm

c m c a c a

iH O

H

CM c A - f t U \ n O

NO O n CM I f t H

f t

f t -

C— C A CU

•

O O O H H C M C M C M

f t UN

CM C— CM f t - r H U N OO OO CO
CM

ft

ftft

•

•

•

CO O

•

H

•

•

•

C A f t

CM f t " I f t

«

*

n r l r l r l r l W C M W C M l M
CA CA CA CA CA CA CA CM CM C A C A C A C A C A C A C A C A C A C A

I
fcy

NO

I

I

I

CM

I

I

O nco

ift c o

I

I

On

I

ft CMCAftUNMO Ift OOONft

I
O'

I

I

I

I

I

I

I

I

I

O= =

-

CM f t

f t

ft

ft s

f t CM OO U N CM ON NO U N C O o

O O Oft
o i—I

.p q

ft

CMCA c A f t O

CMCA UN f t O

O

CM f t CO U N

f t i—I i—I i—I cm

CM —d f t f t C A O n nO O o n

CM CM CMCM CM CMCA CA CA
CACACACACACACACACA

ca
ca

CM CM i—t i—I i—1 O n
c a c a c a C A C A CM

ft CMCAftUNNO ftCO ONft CMCAftUNNO f t
I

■ry

co

I

I

I

I

I

I

I

I

I
cy

On

I

I

I

I

I

I

2 hi

m m
ftQ
O

S K K
Q
®c a O4
4

4

4

4

4

4

m

m m
o
cm
4
ca
4 4

la

ca
4

4

m m
CA o
C A -4 "
4

m m
O LA
—4 —4
— i -4 "

m m
f t xa
—4 —4
—4 —4

m
1

m
m m ^.m
o c o <55 o o

m

pc—O
CA CA CA
4

4

4

ca

4

f t

—4
4

14 O- O

V

P
u

0

fed)
•H

•H

to
£

LA X A f A < A f t

ft

ft

&
G a
G

ft
■=*!

W
) to
G £
*H
P.
lftt?
o cd 0 a
1
—1
—
I
ft
CO
co f t
to
£ a
O
ft

>»
U

©

0

0

X

G
ft

>

Ph

ftO
ftro

XA
ft 4

a
*
!>aa Q
G p
G
g £
O
ft p
«sj 4)
>

a

e

bO
G
O
ft

PQ a

ft

a

a

a

a

a

CA

c o c4- L A nO L A OA
LA L A L A L A X A L A
nO

ft

X X X X X X I X!
d

ft

=

a

=

a

CA

CO
On
4

n

O 4

go

O

o j

4

X
01

0

0 a

1

a

—1 1
—
1—1 1
f t ra ro

bQ
£ a
O
ft

bO bQ ©
£ u a

0

0

ft ft

ON CO OO

0

a

S

ON ON ON ON O

4 4 4 4 4 4 - 4 XA
4 4 4 4 4 :

ft “

ft O

4

PQ PQ

r- o

CA CA CA CA CM CA X A X A 4 4 n o x a cm o n n o
O n i—I L A X A X A f t 4 CO ON O L A l a l a l a l a 4 f n o
’ 4 4 4 4 A
_ 4 XA 4 4 4 4 4 4 4 X

C
OC
QC
OC
OC
OC
O

XX XXX X
d

X ^ ^ XX X XX X
d
a

a

la

%

%

0

O

O

4

44

ft 1

o

44

LA

i—[ i—I i—I i—IC M C M C M C M

O

CM '

c—

4

CA

44 •
CA

C A CA CA CA c A 4 f 4
4 f 4 t XAXA
< A CA CM CM CM i—I i—1
CACACACACACACACA CA CA OACACACA CA CA CACACA

On

O
1 1

1CA

Xa
C\J

O ft

•H ©
Ph Ph

L A XA 4 4 - 4 - 4
—4 4 4 - 4 4 4 4 4

CO CO w

LA

4

PQ PQ a

PQ PQ

ft O

p
ft

G

C~

4

a ■' cpNj

bO

•H
Ph
O a

1—
1

bO
£
O
f t

a

P

c4 0 G
ft ft 1
—1
f t CO f t CO
r-j
G
CD
>

m

O

£

ft PGft

a

CA
XA
44

G

bfl

bO
£
•H

CM o

O N ftL A O M A O O o o

00 L A X A L A L A X A 4 X A L A 4 X A 4 4 j - —
.—
J—J4 4 4 4 .

£
5a

B

co

ft

©

O
ij
m
ft CM

II

1—1

bfl M
£ £
O
ft f t X

&0

On p
a
O a
O '.
ft ft

ft
cti 2
ft
ft

Ph
O

p.
0

i— 1
CO

&
£

h

II

ft
CO

G

bQ
£

£
ft
cti
1—1
ft

O ft
ft
_p
ft P.
co £
G G
Ph f t
O -4

ft
P

n

ft
CO

ft a

a s

0

It

ft
CO
G
&
O

h

O

ft

a

E

Ph G

II

txO

«%

CD P

u

ft
cti a
i—i
ft

M
£
•H
ft
O !

ft
c4

V

ft

It

ft
9*
PC
ft

s

G

a

-f t
(D
s
Ql
O
f t

II

ft
&
£

ft

o

CA

4 4 CA C A 4 4

i4>

-

, CA
1 CA

ao

1 I I I I I i I
K?
O

ft CM X A 4

ca on x a

O

o

O

CM X A CO CM 4 4 CO r - l O O C— ON X A CM O n n O

4

cm

O

co 4

O

4

O

ft

ft

CM C A 4 L A X A N O

nO

C— O n CM X A 4

O n i—I CA

O

O ft

t CM 4 4 XA nO 4 0 0 O N f t f t
—i l f It CMI C1A - 4IX AIN OI C—I COI f I
1 II I It II I I
J3“

cy
,t—
I

CM
rH

next page

m

Continued

44444

2hh
3

CD
3 i

3

3

O

H

H !—
Qi O

33

Os Os o^

O O
-3 -3
wq
-3 -Hr -3 _3 -3 -3
v V v V

3 3
8 8
SB
V v

%

B
3
3

S
v

O

(D

O

O
3
p

R =
Fh

o

=

bfl
R
•H ©
P
Ph Ph
o o s
R e
F
h
i—l
1—1 1—1
m w
©
3
P
bO 0
bO
P
R Fh S = R s R s
O o
3
o
&4
3
3 S

bO
-p

•H

3

bO

R

Ph
O

Oh O :

W

bO
R

iaO|
R

p
Fh

p

P

rH

3

3 C
OC
O
C
O 9-i 3 P

O

o

R H P

3

R
R
3
P

O3
H ra

o

rH

rH

H

bO
R
•H
3

ra

ra

3

O

•H
P
R
o

3

o
-H
3
P

o

Cxj R
CD R
® VR
m Fh
ra bo
o
i—I R

o X

O:

©

g

013

Fv

(xr CO

PI R i-l

R
3

2m Hcd

13

PM
i—
rai

P h

o

R

bO
bO R
R 3

bG
r

R
*H

£0

"cT bO

O

O

P h

•P

*■rH2

Fh

S5

CD

3
p
R

CD

w
R

R

Fh

3

P

Fh

o

Os

O
3

CD

ra

r—1

o
•H
p
o

CD

PM

O

0
bO

a

3

feS .

>s

O H
i—I R
o
«

lr \lA V M fW O

CO

MDO O 3
33 3O 3<3 I

r -N O M S r O lH P I H

3 3 3

I

3 B 3

£

o
Fh
ft
Ph
R

o
3
ra

0

3

ra
o

3

£
CD
P
a-

W PR PR PR CQ CP
O

R

O

H

O

h l

ra
!>»
ra

j eg pr
O

=

=

R O

01

Ph

•iH P
3
R
R P 3 0
P
R Fh
Fh © 0r —l R
© O 3-i O
A P
•H
P
Fj
3
R
3
R
•rH r a O 3
V_ /
•H
ra
ra
© 3
p
•
3
•H
R 3 © R
© o £
o
ra
•H
1 P
R i—
3 o R O
•H •H ©
o P P i— 1
R ■H • H
ra R
ra o • H • H
R
3
©
£
e
o r H cti
© o P

CD

A- MO A rA R A -3 A- CO RA OO _ 3 O A) vO
L A 'L A 'L A X A -3 _ 3 -3 'LA'LAT-AvO a - A-vO
3 ' - 3 - 3 - 3 - 3 3 ‘- 3 - 3 - 3 - 3 - '

CA

c—

-3

-3
co
-3

?
•H
3

CM A 3 M D A O
O co O A- Q 3 I
3 3 A- A- CO CO I

R
•rl

CO CO CO CO CO 00
lA

rH

|^ |< ^ ^ ^ ^ x x x x x x x x

CD
X

X

Ri

*H

X X X X X X

I

P

R
CD
■3

W
3

%
o

13

CA C—3 r l RA Os. <A [— i—I 3 C— rH vO CA
<D
O H CM C\J <M CA C A 3 3 3 P A 1 A V O
A

0_
A-

AA-

ra

CD
•H
Fh

CD

ra

O

A - OS CM A - 3 MD A - OO CN 3

CM^ C R \Q

CM CA CA < A 3 3 - 3 - 3 3 3 3 TAvbAlAbA

A A A A A A A A A A A A A A

o fH C\j A 3
i—I CM A 3 " 3 MOt— CO CN 3 3 H 3 3
1 I I I I I I I I I I I I I
IQ*
A

CO

Q#

3
A

3
A

RA
rH
I

3
3
I

3
i
i
CD
P
O
3

0

>» rar
6 CHJ
->’tM
R

<A

I 3

Fh

A

3

CO O

Q \ A

A

A

'OOJCOlAAOOO
3
3
CM CM A
A A

Fh

© ©
HH 3H
3
3

CM CM O O CN A- A- C—,
3C N A -3A -Q O O
acmcmcmcmaaa

3

3

CM A

3

3

3

sO 3

I I I I I I I I

C
O

a

1--- 1 cd
1-- 1 1-3
•R p

2hS
CO

O
XA
J
-P

cD
0

atxO>
cD

d

§

ft

ill

£

f t ca ra

d

H

J

by M ).

d

ft
o

o

co
0

f t!
-p

d

o

ra

-P

i—I
W
) ft

-p
o
CD

txO

O ft

d O
O rH
P I CO

^ f t

UArl C
MC
M

OO

CO

OO

1A H

H

pq

ftJ ft\D
ft O
O

IA -P A X A

0— 0— OO

I vO VO c— CO

Q ^ !*! I*! !H S ft ft ^ ^

I

I

I

I

I

©

pq

PQ c q

%

ctJ

rl O
•

W
*

CM

•

•

0 -4 H ® l A CA f t - rH X A rH
H r I C M
H rH CM CM r a

ft C
MJ

O - 4 cm J

- 4 c\i ,—| o \ f t - c a o \ X A f t x a c a o
X A P A X A -d ' - d ' - d r a c a X A X A X A X A

2

I J

J

O
Q

c
H

I

I

I

I

o O
a j O

i vD\0\oc^-o-f~r^iA-®co

-O

CO CO

rH 0 \ J

ft
ft

f t

O

ft

' <B

X A I A 4 4
J * X A X A XA
J
f t f t f t

Wft
o

0

CQ p q

cd £

ft O
f t j
o
-d x a q

f t

f t

£0 rH
CtJ
ft Cha
tJ *rH
ao o
o Pa

f t <x > c o c o

ra £ f t
ra -—

ra

ra
d
CtJ
d
d
0

O O O O o ft

r.
d
-rl
CtJ
ft
d

-p
d
•rH

01 01

ft

I CQ pq

CM CM o

f t O

■s
= s;oC =

°a I

O

PA

s OO

=

ra ca f t O

XA O

O

C— C— CO C A f t CM r H 0 - \ O

L A O -O M A O

OO J

r a n

j
j

C O M 3 MO O

J

ra CMA cm o n o - J O
o o ^ O j
- J i—i f t
J ra ra O l i A D A l i A P i - c J P f - a ^ t - C r a

\0

m j a

cm

ca ra j

o o ®
x a - 4

:

XA

O

j

j

J

h

XA

h o o

ca

C\J c a - j - l A v O

r— CO H

cm

c a j x a

vO CO I-H CM CM r a J P A aO f t c o C K H
I I I I I I I I I I I I I I

rH

I

X

CM r a J P A M D

I

I

I

I

I

C— A - CO

I

I

I

C
-rH
0
rd
ft

O rH
d rH
0 « i

ra

•rl

•
v—-V
d

is
o
0ra fto ,d
ra
d
ft •or l CdtJ
o -p ,d
d • r l ft

ra
ra o cm
0 Oft
is £ CD
0 opq
o
ra
d

0

© ra
xi ft ©
f t f t i—i

I
I
H

0

0 ft
•rl
d

o

O A C a c O X A rlO
r l

ra

ro co

c^cao
raj* Jaoc\JAOojPAa3HHlA-c\i
rH rH CM CM CM CA r a CA J

H

o

d
f t ra
f tf t
CtJ r H

ra x a o
m) o J

cxq O n c \j r a P A O
CM r a J N O q o o

O

jq c0 O
CO E-I f t

I I
I I

I pq

ESI

d

I I
I I

ra ra J

ESI

ft

&1
CO

I I -H I I o

a = = Io o = sr

I

2

03 cd

X X X

o o o o

P

O

I

I I

ft

w
m

i—I
CO

d

I I ft C
—ft J

CO CO

xia5£ H

o

d
H

d

rH PAJ

i—I rH i—I

O JONVO 03

ft

C\J

O= B
CM Q '\ O
- w- - o

CD

0ra

1—ow1

rH
O
O
O
d
0

ft
ft ft

o

1—I
‘a

0

ctJ
i—1

-p

bfl
d
*H
a

ft

ft

■§

•rl
d

0

ft

O
13

(Indicated

0

-rl

an i n f le c t io n .

o0
i—I

than

-p

H d ft
d
d
'a
h h h

p
J

M

ft

o
o
CD CD

rather

-p

o
0

thickening

-p

-P

d

xs
0

a gradual

hO

show only

M

2k6
o
o

o

XA A

©

G CA

o

O

G

3

O
O

A t

co i_r\

CA

CA
0)

P P
■H O

P

tH i—I

i—I
'a

£

CD

T

XA

H

Q H

-p p

p

At

o

o

o

p

p

o

o

A>

©
1—1
'a
H

H H M M

bO
G
d h
© Ph
H O
O i—I

On

o
o

d
©

H
o
o
o s

©

u

A P
© A

I
I

r^-

I I I -=t c—
I I I A fC O
At t—

I
I

I
l

I

I
I

I
I

I
I

I
I

PQ p q

I AI o

O
I o

o CM P— C\J NO IN- A
On o -3 " F— CO CO n o

i

A P^- p— 0 - C
— o - go

I
I

I
I

I
I

I
I

CM

c j ^ 0 4 p- 0 n O
po-p-c^r~-c^^-®e-NOO

%
o

-A pq pq
o

o

H CO A

I £

pq

i

o

I
i

O

J pq

I
I

I o

in-

rH CNJ CM CA-=t-=t

IP- no A t On
«

*

o

«

ca
•

CM CM NO H NO H

P~- On
•

a

c

H

A

O

o

c

i—I

H

H

H

O

o

On A G O a o o
CM CM A c A A t A t
a

O

c o i a p

- O nco

•

CM ( A ®

p—no nOA A J ' - J

A CACO CA ON CA o 'd)
30 P— 0 - P—® O \ A U M A _ J

l ^

f t

OO O n i—I o

= =

T A 1 A H NO U \

CM C A cO q H H i—I i—I C A On
o - p— r — e o c o c o a o a o a o c o

C
OC
OC
O

i o

X X

!xi

01 01
%%
SCOO:

i m a:
o

ft

S fi ^ X

co

ON O n n O A t rH CAO OCOIA H ON H XA O

CA On no <ACO UN rH nO

•

o

©

i—1
A
•H
A O
W
id)H
•H CO CO c o ra
p

ft

CM Cl
O'

ft t

I PL, CO CO
ft ft ft
I f t E-i E—
i £—
i 1^

E-i Eh &-i E-i E-h &-i £h O.

Eh H Eh Eh E-i Eh X

•P
P ©
O *
© P
i—I CD
ft ©
G H
•H ft
G
O -H
S
ft
A
bfl
•H
rH
W

I H O On
I H H Q
a o CO <25

I
I

pH Ph O C
OC
OC
OC
O

OC
OC
OC
O
IA ft O C

G bO
ra G
»H

I O On
! O On
<25 p -

O NO CM CA CM

O

A
©

o

G
CO

I
I

u
<D

G
ra

©
9*

ft o
G -A CO
co co E-i

o

o s
o
S

© A t

©
>
G
bO OJ
G W
•rl A
H O
O
o
o

bO
G
•rH
H

&

p

d

P
o

© 000

0

CD

<H 'H
<D C

O c o 'O
On O n O
P~- c— c— c o

$

o

©
A
P
O
A

PC
o
XA
j#A

o

On I A O n On CNNO rH O

C A CM

O C O X A CM T A C O H A t X A [ f t O n
i—I i—I CM A A C A —Cd A J A t A t A t

i—I A

A

H

CM nO

CM l A

O

A

XA

AlHQAfHOONOACMOCO
CO f t

f t nO n o A

A

l A

l A

l A

d

(

H CM AAtXT\NO C'- H

CM A

A

A t C A N O A - CO ON

H

CM A

A t X A vO

A

A

On H

H

j

I I I I I I I I I I I I I I I I I I I I I I I I I i I 1
Cd

M

INJ

est

p a g e

ffi
O

m k m ng

ffi

next

O

O G
At
XA ©

p

CO

d

•H

d

o

o
co.

£

C o n tin u e d

p

O

•

.

2b7

X
CD

X

CM

PQ

CM

S t=

•
XI
o
-p
cd
X

o

o

mCM
v

= s ft

H-l

‘H S

X

o

PC K Ed

o Q®

CA MJ CM
-X A - X
-P -P -P

X
ra
P
o

•H

P hS

ra

£
O
0
X
X
0
0
0
Ph
0
0
Ph

0

-p

cd

0

i—I

X "s'
0 0
0 X
Is ra
.
o IH
X ra

«H

fctO
id s
o
i-H

O
X

A _=t O

On

M

0

P H P H P H -X

I

I

1

1

I
I

I t

I I I I I I

I
I

I
I

i—I

•

P.

txO

0

•H

Ph
co co O
CO CO CO Eh E-c rH
E-i E-< &-* M M CO

Eh Eh E-i

O HD _X
I A A CM
I -X A j

H ft

Ph X
0
0

ro
0
•H
0
0

I I I
I I I

ra

o
|2
-p

I
I

I
I

I
I

I
I

I
I

I
I

f t H-h
O X
0 0
X •H
X —-

CM
pr ^

CQ
iI O
= = e

s

CQ

I

I CQ

3

I

I X

0

CQ

PQ f t

0

-rfPH APH-X- (ACO^-nO-J- rH ONCM CD -X O -X CM nQ CM
3D 00 - X A Q Q
rH < g ON O O H J P Q O Q
Q On On
P\'LT\\AITivQ NO 1 A IA irilAAlQ'ONj'SNJirvXA

NO 0

n

0
0
IS
!>>
i—
1
H
P\ 0
0
X
o
0

A

e

t
0 £
s
o P IS
rH
X 0
0 X

0-«

03

CM

o*

0 £h
PH HD PH Q A
0 0
0 CQ CM H A CM NO CM

On On co CO A A

£ x
o 0
0 0
<HH o
0
X 0
0 P hHi |xj (X) |x) X
•rl
0 0
0 I—
1
0 o
o £

ft

ft %

ft

ft

♦

*

*

*

•

*

’

» * *

1

NO NO A

CM -P 0 H O O n C O C ' - nQ I A nQ nO C —; rH rH o ON CN-NO PH HD A A 1
(ACAIACM OJ CM CM CM CM CM
0
0
0 0
X
-P O'
o
0
O
K
CO

HI CQP"LA^O
I I I I
CD CD

(D

CD

CD

O PH HD IN-CO O
J ON
— CO ON i—I i—I i—I rH i—I CM
r H r H r H r H CM C
CM r o

I I I I I I I I I I I I I <

CD CD CD CD O 0 0 0 0 0 0 0 0 0

ft

CD
S

EH

=

S

A
A

nO

Ph
EH

CM

ftft

—
X C' i—I

O

CO CM A
CO A PH
NO HD NO

CO
ixj

ro On
0 —
0 ft
O S S
s s o s X X £
s s s p o =• s s H“1 C S S
X
ra 0
0 o
•H X
X
0 0
0 X
liN Q C A „
0
0 0
- A HD H5 PH HD 1 /\ CO CANO ra X
A CM O C A P
£ X PH -X (ACM rH I—I PH CO O ^
4 • » I « • » * • •
« * * • • • • • ♦ *
0
X
0
I i—
I i—
I i—
I i—
I PH C nO nO ph
CMNO O
ro o A A A A A NO On QO CO A i—
M X PH -X -X ACM CM
-p •H
0
H X
0
•H
0 0
O
£ •H
X Ph
PH On CM
0 0 CA IA c o r — —X IN— IA CA CO CM On ca CA i—I HD PH On PH CM nO

J

X

Oh Ph fr-,

C
M
X
0 Oh
•H 0
PH PH A C
— PH CM
o PQ
Ph 1 I ON nQ P f rH O PH I
i a c— t—c— ex -X i

■
I

0

i—ii—ii—i

P
n3

X

0

O
rH

X
X

0

"a‘a ■a

O
Ph
£
o
o

0

o\ 0

0

i—I i—I rH

•H
ra

M

ro

OOO

0
o
*H
X

0 =

0
£
cd

a

^
P^i

S

X

0
0
-P
ro
cd
£

X)

e- »

CD

ft

•H s
HI

=

s

x0

prT
S
CD

cfl

page

X

next

0

X

rH CM A -P-PHN Q

I I I I 1 I

HO

PQ
O

rH A A
co _ x h

On

A

ON

i—I rH i—I

nO

CO

A A A

- X .X - X

A CO ON
I I I

A
-X

o

Continued

S~.
-X x

2i|8
ft

©
ft =
ft
©

N

g

i- 1

j-q

g g s
\OOn
•

©

n

O

•

ft

ft

© ©
H H<H

p

P

P

o o ©

CO

H M

ft

©
©
P & s
©
H
ft
bfl
G
•rl
ft B r
O
rH
CO

ft

©
G

ft

©
E-t

i—I —G" C*—CO
O

O

ca

ca

03

ft

fO

pq

ft

p

o

O
•H
P
O
©
P
G
©
i>a

©

m

co
G
G
o

G=
g

's'a S

H

ft

©

O

P

ON

O

ft

C A -= t

*cc!

tuO

M
G
*H

G
'H
ftB
o

£
o

I—I
©

£>
'O
©
P
O
3

co
©

rH

co

ha
G
o r

b0|
G
O

ft

p

<a
i—
I
LATA I I I "LA
-ft-ft I I I -ft
lr \

i—
I
©
•H
o
©
a

eg

o

i—
I
co

I I I fq

s

3

I

I

I ft

CO

'•©
©
O

On O

H

I—I

\a

•g

NO H CO "LA CNJ
NO \

\

a

a

l a o P i - <H cm CN
0 - 0
f A n O O n i—I
t A 1 A U N "CA NO

©

©
G
©
u \

ft ft ft ft
£-J f t Jft f t

Is
CO

p
rH

©
g

-©•

CQ
Os s B

ft

o
©

o»

%

O

r

=

cv

O

=

O

O n c o 1A

cm

On H

O

NO 1A CO

©

•H
CO i—| NO U"\
ca

o— n o p j -

ca

O f t CM CM
•

»

•

*

CNJ 0O CO CO CO

G
©
co
i
p

nO nO

©

ft

E-J

P t_ © - p } p J

On CM H CO NO O
* * * • • •
O O on p - NO NO
CA CA CM CM CM CM

©
P

O

H

Cnj < A —d

i—I i—I i—I i—I

t i l l

bO

ft

iH CM CA- f t XA NO

I I I I I I

P

APPENDIX 13

2h9

APPENDIX 13
QUALITATIVE X-RAY DATA FOR TERNARY SYSTEM SAMPLES
X-ray d if f r a c t io n p a tte r n s were obtained fo r numerous samples as a
means o f d e te c tin g the compounds p r e se n t as sep arate phases in the
s o l i d i f i e d m e lts .

The r e s u lt s obtained on binary system samples are not

d isc u sse d here as the o n ly noteworthy r e s u l t , an in d ic a tio n o f incom­
p le te n e s s o f the L i2BeF4 p e r it e c t ic r e a c tio n , i s ill u s t r a t e d by the data
in Appendix 1 1 .

T his same tendency fo r some incom pleteness o f p e r it e c t ic

r e a c tio n s extended in to the tern ary system and somewhat com plicated the
in te r p r e ta tio n o f r e s u l t s .

Thus o c c a s io n a lly four s o lid phases were de­

te c te d in the s o l i d i f i e d m elt where the Phase Rule fo r b id s more than
th ree c o e x is te n t s o lid phases under non-isotherm al c o n d itio n s .
An a d d itio n a l d i f f i c u l t y i s th a t BeF2 i s d e te c ta b le by on ly one e f f e c t ,
a d if f u s e halo a t a low value o f the Bragg a n g le , no l i n e s being a t t r ib u t ­
able to t h i s amorphous form.

O c ca sio n a lly t h is halo was too i n d is t in c t

fo r p o s it iv e d e te c tio n and i t s presence had to be in fe r r e d on the b a s is
o f sto ic h io m e tr ic n e c e s s it y .

Thus a m elt co n ta in in g o n ly a very sm all

percentage o f CaF2 might not show a p a ttern fo r e ith e r CaF2 or CaBeF4 and
the presence o f one or the oth er might have to be in fe r r e d u n t il on succeed­
in g m e lts th e co n cen tra tio n s had been in crea sed to d e te c ta b le l e v e l s .
Another example o f such a s it u a t io n might b e.on e in which the added LiF
and CaF2 were in s u f f ic ie n t to combine w ith the BeF2 p r e s e n t, so th a t i t s
presence would be n e c e ssa r y , even though not d e te c ta b le .

250

Both CaF3 and LiF e x h ib it o n ly a few X-ray l i n e s , due to th e ir cubic
str u c tu r e , and th ese l i n e s do not in te r fe r e s e r io u s ly w ith each o th e r .
On the o th er hand, CaBeF4 and L i2BeF4 have q u ite complex X-ray p a tte rn s
and much overlap pin g o f the l i n e s o ccu r.

F o r tu n a te ly , however, a t l e a s t

one f a i r l y in te n se l i n e e x is t s fo r each o f the fou r compounds a t lo c a ­
tio n s fr e e o f in t e r f e r e n c e ,

I t was th er e fo re f a i r l y sim ple to d e te c t the

p resen ce, or apparent ab sen ce, o f p a r tic u la r phases by m erely matching
the p a tte r n s v is u a lly w ith other X-ray p a tte r n s .

N a tu ra lly th e method

cannot prove the absence o f very sm all amounts o f a p a r tic u la r phase in
such a r o u tin e way, but the s e n s i t i v i t y in t h is case i s probably b e tte r
than $% and p o s s ib ly as good as 2%,

T his u n u su a lly high s e n s i t i v i t y i s

apparently due to the low atomic numbers o f th e elem ents in v o lv e d .

This

s e n s i t i v i t y may be il l u s t r a t e d by the la r g e number o f l in e s given fo r
L i3BeF4 , F ilm 33 in Appendix 11, where the a c tu a l Li^eF.^ co n ten t was
approxim ately 25$ o f the w hole.

P r a c t ic a lly the same r e s u lt s were o b ta in ­

able when t h i s value was reduced to 1 0 $ , i . e . fo r m elts c o n ta in in g about

97% LiF and 3$ BeF3 .
The r e s u lt s have been ta b u la te d o n ly fo r a reg io n o f p a r tic u la r in t e r ­
e s t , i . e . in the L ijjB eF ^ rich p o r tio n o f the ternary system .

I t i s under­

stood th a t o u tsid e o f t h i s r e g io n the X-ray exam inations showed o n ly those
phases which would be expected from the nature o f th e system , except fo r
f a in t tr a c e s o f components p resen t because o f incom plete phase r e a c tio n s .
The f i r s t ta b u la tio n shows the rough e stim a te s o f the amounts o f
components p r e s e n t, in a l l c a se s L i3BeF4 b ein g the most prominent phase.
A ll th e se samples showed a t l e a s t tr a c e s o f BeF2 , an e f f e c t which was

251

noted a f t e r th e a d d itio n o f o n ly a few p ercent o f BeF2 to any m e lt.
Consequently the p resen ce o f BeF2 i s o n ly mentioned where the in crea sed
in t e n s it y warranted the n o ta tio n .
The second ta b u la tio n groups the r e s u lt s o f the f i r s t in to one o f
three major com bin ations.

I t w i l l be noted th a t f i v e r e s u lt s show the

presence o f a ••forbidden" component.

The choice o f which component was

the ••forbidden" one was based on r e la t iv e i n t e n s i t i e s o f the X-ray p a t­
te r n s .

These com binations o f phases are as fo llo w s:
(1 )

IdF + L i2BeF4 + CaF2 .

(2 )

L i2BeF4 + CaF2 + CaBeF4 .

( 3)

L i2BeF4 + CaBeF4 + BeF2 .

252
TABLE 1

ESTIMATED RELATIVE AMOUNTS OF PHASES
Note — A H samples show the p resen ce o f some BeF2

Sample______________
H-3
K-2
K-5
L-2
L-5
LL-lIia.
LL-lI|.e
M-l
M-2b
M-3b
M-5
MM-12a
MM-12c
MM-12d
MM-12f
N-3
N-ljf
0-2b
IQ-J4
IQ-6
IQ-8
2Q-3
2Q-1;
2Q-5
2Q-7
3Q-2
3Q-I4.
3Q-5
3Q-6
kQ-2
l+Q-3
lj.Q-5
l$ - 6
i|Q-7

Phases P resen t

L iF , L i2BeF4 , CaF2i m oderate.
L iF , s l i g h t .
L i ^ e F ^ CaF2 , str o n g .
L iF , sm a ll.
L±2BeF4;} s tr o n g . CaF2 , m oderate.
L i2BeF4 , str o n g . CaF2 , sm a ll. (CaBeF4 in f e r r e d ). No LiF
L iF , s m a ll. L i2BeF4 str o n g . CaF2 m oderate.
L i2BeF4 , str o n g . CaBeF4 , m oderate. BeF2 , m oderate.
L iaBeF4j str o n g . CaBeF4 , m oderate. BeF2 , m oderate.
L i2BeF4 j str o n g . CaBeF4 , m oderate. BeF2 , m oderate.
L i2BeF4 , str o n g . CaBeF4 , m oderate. BeF2 , m oderate.
L i^ eF .^ , CaBeF4 , str o n g . CaF2 , tr a c e .
L i2BeF4 , CaBeF4 ,
CaF2 , m oderate.
L i2BeF4 , str o n g . CaBeF4 , m oderate. BeF2 m oderate.
L i2BeF4 , s tr o n g . CaBeF4 , m oderate. BeF2 m oderate.
Li^eF.^., str o n g . CaBeF4 , m oderate. BeF2 m oderate.
L i2BeF4 , s tr o n g . CaBeF4 , m oderate. BeF2 m oderate.
L irgBeF4 , str o n g . CaBeF4 , m oderate. BeF2 , moderate.
L i ^ e F ^ CaBeF4 str o n g . CaF2, p r e se n t.
L i2BeF4 m oderate. CaBeF4 , t r a c e . BeF2 , m oderate.
L iF , s l i g h t .
L i2BeF4 , str o n g . CaF2, sm a ll.
L iF , s l i g h t .
L i3BeF4 , str o n g . CaF2 , sm a ll.
L iF , t r a c e . L i2BeF4 , str o n g . CaBeF4 , s l i g h t . CaF2 , sm a ll.
L iF , s l i g h t .
L i2BeF4 , str o n g . CaF2 , sm a ll.
L iF , s l i g h t , L i2BeF4 , str o n g , CaF3 , sm a ll, more than 2Q-3.
L iF , s l i g h t .
L i2BeF4 , s tr o n g . CaF2 , sm a ll, more than 2Q-U.
L iF , s l i g h t .
LiaBeF4 , str o n g . CaF2 sm a ll, more than 2Q-5.
L iF , s l i g h t .
L i2BeF4 , str o n g . CaF2 , m od erate.
L iF , s l i g h t .
L i2BeF4 , str o n g . CaF2, moderate.
L iF , s l i g h t .
L i2BeF4 , str o n g . CaBeF4 , CaF2 , sm a ll.
L iF , s l i g h t .
L i2BeF4 , str o n g . CaBeF4 , CaF2 , sm a ll.
L i2BeF4 , s tr o n g . (GaBeF4 , in f e r r e d ) . No CaF2 or L iF . BeF2 , sm a ll.
L i^ eF .^ , str o n g . CaBeF4 , s l i g h t . BeF2 , sm a ll.
L i2BeF4 , str o n g . CaBeF4 , CaF2 , sm a ll.
L i2BeF4 , str o n g . CaBeF4 , s l i g h t . CaF2 , m oderate.
L iF , s l i g h t .
L i2BeF4 , s tr o n g . CaBeF4 , s l i g h t . CaF2 , m oderate.
L iF , s l i g h t .
L i2BeF4 , str o n g . CaBeF4 , s l i g h t . CaF2 , m oderate.

253

TABLE 2

Combination ( l )
LiF + L i3BeF4 + CaF2

Combination (2)
Li^BeF,^ + CaF2 + CaBeF4

H-3
K-2
K-5
L-5
1Q-U
IQ-6
2Q-3
2Q-1|
2Q-5
2Q-7
3Q-2
3Q-I4
LlQ-6 + CaBeF4 tr a c e
5Q-7 + CaBeF4 tr a c e

—
—

L-2
—

IQ-8 + LiF tr a c e
M-3b
VL-k
N-Uf
—

3Q-5 + LiF tra ce
3Q-6 -t- LiF tra ce
liQ-5
W -5

Combination (3)
Li^eF.^ + CaBeF4 + BeF2
IL -liia
LL-lUe
MM-12a
MM-12c
MM-12d
MM-12f
M-l
M-2b
H-3
0 - 2b
—
—
I4Q-2
5Q-3

The r e s u lt s are shown g r a p h ic a lly in Figure 2 7 , a p o r tio n o f the
ternary diagram near the L i2BeF4 p o in t.

The d if f e r e n t phase com binations

are in d ic a te d by the d if f e r e n t symbols p lo t te d a t the com position p o in ts .
Since the com p osition s are s u b je c t to some u n c e r ta in tie s the boundaries
o f th e s e r eg io n s are not sharp when p lo t te d on a la r g e s c a le .

It is

apparent, however, th a t the phase r e a c tio n s d iv id e the system in to sev e r a l
r e g io n s .

The p r o je c tio n s o f l i n e s o f tw o -fo ld sa tu r a tio n have been in d i ­

cated to perm it comparison o f predicted, and a c tu a l s o l i d i f i c a t i o n b e­
haviou r.

I t w i l l be noted th a t m elts showing a forb id d en phase ( i n

a d d itio n to BeF2) are th ose ly in g q u ite c lo s e to the in v a r ia n t p o in t s .
In gen era l the agreement i s q u ite s a t is f a c t o r y co n sid e rin g the un­
c e r t a in t ie s and the apparent te n d e n c ie s fo r form ation o f ''forbidden'*
p h a se s .

T h is a g r e e m e n t fu r n ish e s a d d itio n a l con firm ation fo r the

254

GO

flO
co

co

OJ

<fl> •

m

flu

CM

ar

OJ
U J *0

oc >.
o a:
x
OJ

co

255

in te r p r e ta tio n o f the therm al d ata.

A d e ta ile d c o n sid e r a tio n o f the

data shows th a t the u ltim a te components o f th e s o l i d i f i e d m e lts are de­
termined by th e lo c a t io n o f the o r ig in a l m elt in a p a r tic u la r com position
t r ia n g le , corresponding to one o f the th ree com binations noted above.
The anomalous behaviour o f se v e r a l m e lts ly in g alm ost on a boundary lin e
between two such r e g io n s i s u n d ersta n d a b le.

The thermal data i s a lso

in general agreement w ith the r e s u l t s o f the X-ray exam ination.

Thus

s o l i d i f i c a t i o n by the tim e a p a r tic u la r in v a r ia n t p o in t i s reached
precludes th e form ation o f another s o lid component, which appears o n ly i f
the m elt i s capable o f reach in g another nearby in v a r ie n t p o in t.

A number

o f r e s u lt s o f t h is typ e are ap p aren t, the I4Q s e r ie s bein g an example ,
where the m e lts showed, a t r a n s it io n through th ree in v a r ia n t p o in ts as the
com position was s h i f t e d .

The therm al and s o lid phase s tu d ie s showed

corresponding r e s u lt s in agreement w ith the p r e d ic te d behaviour.
One a d d itio n a l X-ray a n a ly s is o f p a r tic u la r in t e r e s t was th a t made
on a sample o f m elt S-h.

The com position value o f t h i s m elt was intended

to be th a t s u ita b le fo r a t e s t o f the "Klarkreuz" method d isc u sse d in
Appendix 5 .

The m elt was prepared c a r e f u lly ( p .273) from L i2BeF4 and CaF2

o f high p u r it y .

The c a lc u la te d com p osition was 25 mole percen t CaF2 ,

25 mole p ercen t BeF2 and

mole p ercen t L iF , corresponding to th e p o in t

o f in t e r s e c t io n o f a li n e drawn between L i2BeF4 and. CaF2 , and o f a lin e
drawn between LiF and CaBeF4 .

A fte r use the a n a ly t ic a l com position was

found to be 2 5 .1 mole p e r ce n t BeF2 , 25. 7 mole p ercen t CaF2 and k 9 .2 mole
p ercent L iF.

I t was th e r e fo r e p o s s ib le th a t some departure had occurred

from the d e sir e d valu e during th ree p reced in g therm al a n a ly s e s .

256

X-ray a n a ly s is o f m elt S-l; gave th e fo llo w in g inform ation on the
c o n s titu e n t p h a se s.
Strong CaF2 c o n te n t.
Moderate L i2BeF4 c o n te n t.
No LiF c o n te n t.
No CaBeF4 c o n te n t.
Trace BeF2
This evidence th a t L i2BeF4 and CaF2 may c o e x is t , w h ile LiF and
CaBeF4 do not show any such ten d en cy, shows c o n c lu s iv e ly th a t the L iFCaBeF4 l i n e does not rep resen t a "q u a si-b in a ry s e c tio n " .

Such a r e s u lt

would be exp ected from the thermal r e s u lt s fo r the 'ternary system , which
show two l i n e s o f tw o -fo ld s a tu r a tio n c u ttin g a cro ss the l in e jo in in g
LiF and CaBeF4 .

Such l i n e s o f tw o -fo ld s a tu r a tio n cannot pass in a con­

tin uous manner a cro ss a "quasi-binary" s e c t io n .
The f a c t th a t CaF2 and L i2BeF4 are the c o e x is tin g s o lid phases does
not n e c e s s a r ily mean th a t the s e c tio n jo in in g th ese p o in ts i s " q u asibinary" .

The r e s u lt m erely in d ic a te s th a t i t might b e , sin c e the

"Klarkreuz" method ( i n the presence o f p e r it e c t ic r e a c tio n s) can only
in d ic a te th e absence o f p o s s i b i l i t y , and not th e d e f in it e presence o f a
"quasi-binary" s e c t io n .

T his in form ation i s th e r e fo r e not in con tra­

d ic t io n w ith th e thermal r e s u lt s which show a lin e o f tw o -fo ld sa tu r a tio n
c u ttin g a c ro ss the s e c tio n L i2BeF4-CaF3(,

The s li g h t tra ce o f 3eF 2 noted

i s b e lie v e d to have been due o n ly to s l i g h t l y Im perfect phase r e a c tio n s
at th e tem perature o f s o l i d i f i c a t i o n .

APPENDIX Ik

257

APPENDIX lb
ANALYTICAL METHODS
The major p o r tio n o f th e a n a ly t ic a l work req u ired the d eterm ination
o f calciu m , b ery lliu m and lith iu m , s in g ly and in com bin ations.

Most o f

the a n a ly ses were concerned w ith m ixtu res o f the flu o r id e s o f th e se
m e ta ls .

As a n a ly t ic a l methods fo r flu o r in e are n e ith e r con ven ien t nor

a c cu ra te, the d eterm in ation o f t h i s component was always in d ir e c t by
c a lc u la tio n from the w eigh t o f m eta ls found.
flu o r id e s were p resen t

T h is assumed th a t on ly

and in e a se s where t h i s was not e n t ir e ly tru e,

because o f d egradation o f the m e lts , th e error became apparent when the
c a lc u la te d w eight p ercen ta g es f a i l e d to t o t a l to 100%.
Due to in te r fe r e n c e s o f the c a t io n s , the procedure n e c e s s a r ily was
one in which B e, Ca and L i were sep arated c o n s e c u tiv e ly from the same
sam ple.

The flu o r in e must be co m p letely removed by v o l a t i l i z a t i o n u sin g

p e r c h lo r ic a c id fo r decom position o f the f lu o r id e s .

O therw ise, b erylliu m

w i l l be in co m p letely p r e c ip ita te d due to form ation o f a so lu b le flu o r id e
com plex, and calcium w i l l be p a r t ia l ly l e f t in in s o lu b le form during the
b erylliu m se p a r a tio n .

The sep a ra tio n o f b erylliu m as the hydrated o x id e ,

perm its a subsequent se p a r a tio n o f calcium as the o x a la t e .

The f i l t r a t e

at t h i s p o in t must be tr e a te d to remove a co n sid era b le accum ulation o f
ammonium s a l t s p r io r to the lith iu m se p a r a tio n .

For th e l a t t e r procedure,

a sep a ra tio n in th e form, o f a complex m ixture o f lith iu m p e r io d a te s was
made, f in is h in g w ith an e m p irica l v olu m etric d eterm in ation o f the lith iu m
based on an io d im e tr ic method.

258

The d if f e r e n t procedures are d isc u sse d b r ie f ly below and v a r ia tio n s
are m entioned in co n n ectio n w ith th e a n a ly t ic a l r e s u lt s in Appendix 15.

l ) Decom position o f F lu o rid es
I t was n e c essa ry to make a t l e a s t two v o l a t i l i z a t i o n s u sin g 80$
p e r c h lo r ic a c id (about 10-15 ml. fo r each ste p fo r 1 g , powdered sam ples)
to be rea so n a b ly c e r ta in o f com plete decom position .
found to be more s a t is f a c t o r y than c r u c ib le s .

Platinum d ish e s were

As the decom position must

be c a r r ie d out near maximum tem perature to be e f f e c t i v e , and as the e v o lu ­
tio n o f hydrogen flu o r id e i s v ig o ro u s y e t e r r a t ic , some s p a tte r in g lo s s e s
undoubtedly occu r.

Two evap oration s r ep resen t the p r a c t ic a l lim it as

the p r o c e ss has u s u a lly caused some creepage to the rim a t t h is s ta g e .
The evap oration s were continued to the sa tu r a tio n p o in t , the d ish co n ten ts
were p a r t i a l l y c o o le d , and then were r in se d in to beakers fo r the b ery lliu m
se p a ra tio n .

2) B erylliu m S ep aration and D eterm ination
The above s o lu t io n , c o n ta in in g a moderate amount o f f r e e p e r c h lo r ic
a c id , was d ilu te d to approxim ately 175 m l. and th e b erylliu m p r e c ip ita te d
as the very voluminous hydrated o x id e , u sin g on ly a s u f f i c ie n t e x ce ss o f
ammonium hydroxide to perm it a d e te c ta b le odor.

W hile i t would be de­

s ir a b le to u se a much la r g e r volume to reduce o c c lu sio n e r r o r s , the f i l ­
t r a t io n , reco v ery and washing o p era tio n s req uired approxim ately two h o u rs,
even w ith t h i s sm all volume .

Recovery o f m a ter ia l adhering te n a c io u s ly

259

to the beaker w a lls was ach ieved by th ree su c c e s siv e treatm ents w ith
d ilu te h yd roch loric a c id and r e p r e c ip ita t io n s w ith d ilu t e ammonium
h yd roxid e.

These clean -u p liq u id s and a t o t a l o f about 150 ml. o f s l i g h t l y

ammoniacal wash w ater were used in washing the p r e c ip it a t e which was f i l ­
te r e d u sin g Whatman No . IfL p a p er .

The p r e c ip it a t e e q u iv a len t to about

0 .1 g.BeO co m p letely f i l l e d t h i s s iz e o f paper.
A fte r d ryin g a t 1 1 0 °, the paper and co n ten ts were tr a n sfe r r e d to
preweighed platinum c r u c ib le s , u sin g a c id if ie d f i l t e r paper policem en
to remove any m a ter ia l adhering to the fu n n e l.

The c r u c ib le s were ig n ite d

a t the maximum heat o f F ish e r burners to convert to BeO.
The w eigh in g o p era tio n s fo r t h i s m a ter ia l must take in to account
i t s extrem ely hygroscopic c h a r a c te r .

The fo llo w in g system was used to

overcome the d i f f i c u l t y , p a r tic u la r care b ein g n e c essa ry to keep the tim ­
in g r e p ro d u c ib le .

The hot c r u c ib le was removed from, the flame and im­

m ed ia tely c o o led on an aluminum, p la te under an in v e r te d fu n n el su p p lied
w ith a flow o f d r ie d n itr o g e n .

During the c o o lin g p e r io d , about two

m in u tes, the p reced in g c r u c ib le was weighed and the succeeding c r u c ib le
was s ta r te d on i t s c o o lin g c y c le .

W eighing was c a rr ie d out a t maximum

p o s s ib le speed (two to th ree swings o f the b a la n c e ), u sin g p r e s e t w e ig h ts.
E ith er one or two p relim in a ry w eigh in gs o f each c r u c ib le were n ecessary
to o b ta in th e approximate w eight w ith s u f f i c ie n t accuracy so th a t the
f i n a l w eighing might be p o s s ib le w ith no more than a very minor r id er
admustment.

The ta r e w eigh t o f the c r u c ib le s was obtained in an e x a c tly

sim ila r manner.

260

By t h is method a w eight o f the BeO resid u e was ob tain ed which was
b e lie v e d to be in error by no more than 0 .0 0 0 5 g .

T his was n o t, however,

e n t ir e ly s a t is f a c t o r y s in c e the maximum p r a c t ic a l r e sid u e to handle ( in
p r e c ip it a t e form) amounted to o n ly about 0 .1 g.

Hence the f i n a l p er­

cen tage error m ight r e a d ily be 0 .2 to 0 .3 mole p e r c e n t, w ith p o s s ib le
r e la t iv e err o r s o f 0 .5 mole p e r c e n t.

The p o s s ib le e rro rs due to o c c lu ­

sio n o f oth er io n s i s not known, se v e r a l attem pts u sin g c o n tr o ls r e s u l t ­
in g in both p o s it iv e and n e g a tiv e r e s u l t s , w ith about th e same d e v ia tio n s
shown by a n a ly se s made in the absence o f oth er m etal io n s .

3) Calcium S eparation and D eterm ination
A standard o x a la te sep a ra tio n was c a rr ie d out on the f i l t r a t e o f the
beryllium se p a r a tio n , u sin g a volume o f approxim ately I4.OO m l.

The s o lu ­

t i o n , alread y c o n ta in in g co n sid era b le ammonium c h lo r id e , was s u f f i c i e n t l y
a c id if ie d to perm it h old in g a l l o x a la te in s o lu tio n when a la rg e e x ce ss
o f ammonium o x a la te was added.

D ilu te ammonium hydroxide was then added

u n t il th e f i r s t permanent p r e c ip it a t e o f calcium o x a la te appeared, and
a fte r about 15 m inutes a d d itio n a l d ilu t e ammonium hydroxide was added
w ith s t ir r in g u n t il the m ethyl red endpoint was reached.
The o x a la te p r e c ip it a t e was d ig e s t e d , c o o le d , f i l t e r e d , and washed
in the u su a l manner w ith JiO m l. o f c o ld water u sin g G ooch-asbestos f i l t e r ­
in g c r u c ib le s .

The d eterm in ation was com pleted v o lu m e tr ic a lly by d is s o lv ­

in g the p r e c ip it a t e in d ilu t e s u lfu r ic a cid and t i t r a t i n g in a volume o f
about 200 m l.a t about 60° w ith standard potassium permanganate s o lu tio n .
C on cen tration s o f the l a t t e r ranged from 0 .0 5 to 0 .2 N and e ith e r 10 m l.

261

or $0 m l. b u r e tte s were u se d , depending upon th e requirem ents o f the
p a r tic u la r samples b ein g an alyzed .
J . T. Baker " S p e c ia l1’ grade o f calcium carbonate was used as the
primary stand ard .

C ontrol samples were c a r r ie d through the e n tir e p ro cess

and o th e r s were run in which a l l s te p s p r io r to o x a la te p r e c ip it a t io n
were o m itte d .

Although good checks were o b ta in ed in g e n e r a l, in d ic a tin g

good calcium r e c o v e r y , an o c c a sio n a l "wild" r e s u l t in d ic a te d the in a d v isa ­
b i l i t y o f b a sin g n o r m a litie s on the c o n tr o l samples .

A ccordin gly the

sta n d a r d iz a tio n s were based on sim ple con version o f the carbonate to
o x a la te under th e same c o n d itio n s used in the a n a ly t ic a l calcium separa­
tio n s .

Such sta n d a r d iz a tio n s u s u a lly checked to w ith in 1-2 p a r ts per

th ousand.
The accuracy o f the a c tu a l calcium d eterm in ation s was in f e r io r to
t h i s , as judged by the r e p r o d u c ib ility o f r e s u lt s on d u p lic a te samples
c a r r ie d through th e e n tir e p r o c e s s .
thousand seems r e a so n a b le .

An error o f about 2 to 3 p a r ts per

W hile t h i s does not c re a te any very s i g n i f i ­

cant error in c a lc u la tin g com position s near the L i2BeF4 p o in t , i t exceeds
the d e s ir a b le maximum error near the CaBeF4 p o in t where the r e s u lt in g
c a lc u la te d com p osition may be in error by 0 .3 to O.Lj. mole p e r c e n t.

I4.)

Lithium. Sep aration and D eterm ination

The f i l t r a t e o f the calcium determ ination was con cen trated and
tr e a te d to remove th e la r g e co n cen tra tio n o f ammonium and o x a la te s a l t s
which o th erw ise might in t e r fe r e in the lith iu m d eterm in a tio n .

To o b ta in

262

com plete decom position o f th ese s a l t s , in c lu d in g o x a la t e s , a t steam bath
tem perature i t was found n ecessa ry to make rep ea ted treatm ents w ith
mixed n i t r i c and h y d roch loric a c id s .
mended fo r t h i s s t e p .)

(Free bromine has a lso been recom­

The o x a la te s do not decompose a t a l l r e a d ily under

simple n i t r i c a c id tr e a tm e n t, which does however e f f e c t i v e l y a tta c k the
ammonium io n .
The r e s id u a l m a ter ia l was converted to the p er ch lo ra te form and
tr a n sfe r r e d to a 100 m l. volu m etric f la s k from which a liq u o ts were taken
fo r th e lith iu m p r e c ip it a t io n .

These a liq u o ts were evaporated to dryness

on a sand bath to remove e x c e s s p e r c h lo r ic a c id .
The lith iu m p r e c ip it a t io n fo llo w e d the method o f Rogers and C aley
( 132) w hich r e q u ir e s th e absence o f a l l m etal io n s oth er than the a l k a l i
m etals and low ammonium io n c o n c e n tr a tio n s.

Lithium was p r e c ip ita te d by

dropwise a d d itio n o f a s o lu tio n o f potassium p erio d a te (KI04) d is s o lv e d
in 5 molar potassium h y d ro x id e.

The p r e c ip ita te has no d e f in it e com posi­

tio n and c o n s is t s o f a m ixture o f lith iu m p e r io d a te s ( ? ) , w ith the app roxi­
mate com position corresponding to L is I 0 6 .

The c o n d itio n s o f p r e c ip it a t io n ,

d ig e s tio n , f i l t e r i n g and washing must a cco rd in g ly be r i g i d l y m aintained
to o b ta in c o n s is t e n t r e s u l t s .
The p r e c ip it a t e was th en d is s o lv e d in d ilu te s u lfu r ic a c id , io d in e
was lib e r a te d by the a d d itio n o f e x c e ss potassium io d id e , and th e io d in e
was determ ined v o lu m e tr ic a lly in th e u su a l manner, g e n e ra lly w ith standard
sodium, t h io s u lf a t e s o lu t io n .

The t i t e r o f the l a t t e r was determ ined by

carryin g out s im ila r p r e c ip it a t io n s and t i t r a t i o n s o f a liq u o t samples o f
known lith iu m con ten t which were prepared u sin g "J . T. Baker Analyzed®
lith iu m carbonate as the primary stan d ard .

263

The o r ig in a l method c a l l s fo r d is s o lv in g the p r e c ip it a t e in "5 m l.
o f N H3S04".

In t h i s work, even w ith more than 10 m l. o f 2N H3S04 , the

p r e c ip it a t e would not com p letely d is s o lv e in any reasonab le le n g th o f
t im e .

I t was found to be much more e f f e c t iv e to add a con cen trated

potassium io d id e s o lu tio n to the m ixture o f a c id , p r e c ip ita te and
a sb e sto s s lu r r y in the Gooch f i l t e r i n g c r u c ib le somewhat in advance o f
titr a tio n .

T h is procedure i s open to c r it ic is m because o f p o s s ib le io d in e

v o la tiliz a tio n .

I t was fou nd, however, th a t th e a c tu a l recovery was

always much b e tt e r than on p r e c ip it a t e s handled in the u sual manner.

The

l a t t e r la c k s th e e f f e c t i v e d is s o lv in g power o f th e com bination o f acid
and co n cen trated potassium io d id e s o lu t io n .
The method was m od ified by u sin g a p o ly e th y le n e b o t t le - p ip e t t e com­
b in a tio n fo r sto ra g e and d isp e n sin g o f th e p r e c ip it a t in g r e a g e n t.

I t was

thought th a t some o f the in h eren t error in th e method m ight be due to
the form ation o f r e l a t i v e l y in s o lu b le lith iu m carb on ate, which might then
become coated w ith the p e r io d a te complex during the p r e c ip it a t io n .

The

reagen t was th e r e fo r e p r o te c ted a g a in st carbon d io x id e from the atm osphere.
An attem pt was made to prepare ca rb o n a te -fr e e reagent s o lu tio n by e le c t r o ­
l y t i c p rep a ra tio n o f potassium -m ercury amalgam, but the p rocess was not
s a t is f a c t o r y fo r the produ ction o f s o lu tio n s w ith co n cen tra tio n s g r ea ter
than 1 to 2 m olar.
O ptional io d in e e v o lu tio n and t i t r a t i o n methods are g iv en in the
o r ig in a l r e p o r t, s u ite d to la r g e r or sm aller amounts o f p r e c ip it a t e ,
which in v o lv e a r se n ite t i t r a t i o n w ith b u ffe r s o lu tio n added b efo re or
a ft e r a d d itio n o f potassium io d id e .

The t h io s u lf a t e method was found

261;

to be much more s a t is f a c t o r y , sin c e th e p r e c ip ita te amount cou ld u s u a lly
be p r e d ic te d and the s iz e o f the a liq u o t governed a c c o r d in g ly .

Then

th e potassium io d id e m ight be used w ith the a cid to o b ta in the superior
d is s o lv in g a c tio n noted above.
The method i s lim it e d to approxim ately $0 mg. o f L i by the voluminous
nature o f the p r e c ip it a t e and th e washing req u irem en ts, the tr a n s fe r and
washing o p era tio n s during f i l t r a t i o n b ein g c a rr ie d out w ith o n ly fou r 2 m l.
p o r tio n s o f k N KOH .

O bviously th e washing o p era tio n must be e f f e c t iv e

to p reven t in te r fe r e n c e by the e x c e ss p r e c ip ita tin g agent w ith i t s h igh
p erio d a te c o n te n t.

A c tu a lly the washing a c tio n i s much more e f f e c t iv e

than would be e x p e c te d , provided the wash s o lu tio n i s a p p lie d w ith a
p ip e tte in a c a r e fu l and sy stem a tic manner.
The e r r o r s in h eren t in such an em p irica l method are apparent.

Rogers

and C aley in d ic a te d t h a t , i n g e n e r a l, 1 to 2 p ercent error might be ex­
p e c te d .

T his was confirm ed in t h i s work.

I t was found th a t a given s e t

o f r e p lic a t e sam ples might g iv e r e s u lt s d if f e r in g by no more than 1% .
Another s e t o f the same samples run on a subsequent o c ca sio n might d i f f e r
by 2% from th e f i r s t r e s u l t s , although again d if f e r in g by no more than 1%
among th e s e t .

C onseq uently, fo r th ese a n a ly s e s , numerous a liq u o ts were

taken on a t l e a s t two d if f e r e n t o c ca sio n s fo r each sample in order to
p a r t ia l ly overcome t h i s d i f f i c u l t y .

For the sta n d a rd iza tio n s a la r g e

number o f checks were made whenever th ese a n a ly ses were perform ed.

The

t i t e r was u s u a lly determ ined to an average value su b jec t to probably l e s s
than 1% e r r o r .

The a c tu a l lith iu m a n a ly ses were su b je c t to p o s s ib le

errors o f at l e a s t 1%.

T his rep resen ted a se r io u s d i f f i c u l t y in the

265

reg io n o f the L i^eF,^ p o in t where the corresponding error in th e c a lc u ­
la te d com position amounted to about 0 .3 mole p e r c e n t.

APPENDIX 15

266

APPENDIX 15
ANALYTICAL RESULTS
PART A
ANALYSIS OF CALCIUM FLUORIDE
Methods g e n e r a lly were based on the N a tio n a l Bureau o f Standards
method (103) fo r lower grade ty p es o f flu o r s p a r .

T ypical r e s u lt s fo r

the m a te r ia l used in t h i s work were as fo llo w s :
1) Carbonate by E v o lu tio n and A b so rp tio n . (71)
Sample 5.000 gr.,

Gain in w t . due to CO2 = 0 .0005 gr.

% CaC03 = 0 . 0 2 .
2) A cid so lu b le 11carbonate" from a c e t ic a c id e x tr a c tio n .
Based on calcium c o n te n t o f e x t r a c t , % e x tr a c ta b le "CaCOg1'
= ca. 1. 0.
When the method was m o d ified by adding some LiF to the
a c e t ic a c id p r io r to e x tr a c tio n to reduce s o l u b i l i t y o f
CaF2 , th e fo llo w in g r e s u lt was o b ta in ed .

Based on calcium

con ten t o f e x tr a c t, % e x tr a c ta b le nCaC03* = 0 .1
3) S i l i c a by HF E v o lu tio n M ethod.
R e su lts — Unmeasurable, or s l i g h t l y n e g a t iv e ,^ s i lic a
li.) Heavy M etals by H2S P r e c ip ita tio n in A cid and A lk a lin e S o lu tio n .
Unweighable t r a c e s , estim a ted 0 .003$.
5) Calcium by p r e c ip it a t io n as o x a la te and t i t r a t i o n w ith standard
potassium permanganate.

267

Date o f
A n a ly sis

Sample
w t. ,
g.

V ol.
KMri)4 ,
ml.

1 2 /1 8 /k9
12/20/1+9
12/22/1+9
12/22/1+9
3 /1 8 /5 0
5 /2 9 /5 0

0 .U900
0.5000
0.5000
0.5000
0.1000
0 .0950

16.38
17.52
1+7.51
k l .U7
U9.55
1*7.23

KMnP4
S o ln .
No .
2
2
2
2
k
k

Wt. Percent
CaF 2

See Note
No.

97. 8
1 0 0 . 1+
1 0 0 .1+
100 .2
99. 8
100.5

1
2
3
3
3 , 1*
3j k

N otes ( l ) Sample e x tr a c te d w ith a c e t ic a c id . Volume shown c o n s iste d
b5 .9k m l. in regu lar t i t r a t i o n and 0 .kb m l. in t i t r a t i o n o f
CaC20l+ p r e c ip ita te d from the e x t r a c t . T otal = 1+6.38 ml.
(2 ) Same comment, excep t k l .01 + 0 . 5 l = k 7 .52 m l.
( 3) These sam ples were n o t ex tr a cte d w ith a c e tic a c id .
(U) The sample used was ,,CaF2j Batch 3/1 0 /5 0 ® , which was used
fo r a l l subsequent requirem ents fo r calcium f l u o r i d e .
(5) S ta n d a rd iza tio n o f KMn04 T it e r s .
S o ln . #2 — 0 .0 1 0 5 5 g . CaF2/m l. — See p a rt G fo r d e t a i l s .
S o ln . #k — On 3 /1 8 /5 0 , 0.002016 g . CaF3/m l.
See Part H
— On 5 /2 9 /5 0 , 0.002020 g. CaP2/m l.
fo r d e t a i l s .
I t i s b e lie v e d in view o f the "Carbonate by E volution" r e s u l t s , th a t
most o f th e calcium in the a c e t ic a cid e x tr a c t was d erived from CaF2
ra th er than CaC03 im p u rity .

See L undell (103) fo r data in t h i s r e s p e c t.

The gen era l accuracy o f the method fo r calcium i s no b e tte r than - 0 .2%
a t b e s t when the sample must be tr e a te d by p e r c h lo r ic a cid evaporation
w ith p o s s ib le a tten d a n t l o s s e s .

268

PART B
ANALYSIS OF BERYLLIUM FLUORIDE AND “BERYLLIUM CARBONATE"
A n a ly sis o f B erylliu m F lu o r id e .

(Brush B eryllium Corp. p ro d u ct.)

Method - P e r c h lo r ic a c id e v o lu tio n o f flu o r in e u sin g platinum
d ish and p r e c ip it a t io n o f Be(OH)2 by ammonium hydroxide.
Hydroxide f i l t e r e d , washed and ig n ite d and weighed as BeO
w ith p r e ca u tio n s n o ted under s e c tio n on gen eral a n a ly t ic a l
m ethods.

P r e c ip it a t io n o f ir o n and aluminum by 8-hydroxy-

q u in o lin e p r io r to p r e c ip it a t io n o f the hydroxide gave o n ly
a n e g lig ib le tr a c e o f co lo red p r e c ip ita te when f i l t e r e d .
Sample W t. ,
0 .2 9 0 3
0.2077

Wt. BeO,

% P u r ity o f
BeF 2

0.1530
0 .1 1 1 2

99.0
100.5

A n a ly sis o f "B eryllium Carbonate" (Kahlbaum)
Method - 8-h ydroxyquin oline p r e c ip it a t io n to p rim a rily separate ir o n
and aluminumj th en b ery lliu m p r e c ip ita te d as Be(0H)3 and ig n ite d
and weighed as BeO.
Sample
W t. , g .

Wt. o f
Wt. BeO,
Oxine p p t . ,
g.
g.

Wt. % Fe

Wt. % BeC03

1 .0 0 0

0 .0 1 5 2

0.3582

0 .1 9

99.0

1.000

0.0173

OJ551

0.21

98.0

N otes - Sample oven d r ie d a t 120° . Oxine p r e c ip it a t e la r g e ly due to
F e . R e su lts c a lc u la te d as percen t Fe and percen t BeC03 ,
alth ough l a t t e r compound probably does not e x i s t , b u t i s
r e a l l y a b a sic s a l t .

269

The r e s u lt s o f a n a ly se s o f th re e samples o f BeF2 prepared by e v o lu ­
t io n o f NH4F from (NH4) 2BeF4 (w ith o u t adding another flu o r id e ) are given
in P art H o f t h i s Appendix.

The v a lu e s obtained ( 9 9 3% , 1 0 1 .3 $ , 99.2$

and 98.5$) probably were c o n tr o lle d somewhat by the presence o f carbon as
an im p urity from c o n ta c t w ith th e carbon c r u c ib le s used fo r the e v o lu tio n
p rocess.

T h is carbon was e ith e r d estroyed by u sin g added n i t r i c a cid

(during th e p e r c h lo r ic a c id tr e a tm e n t), or was removed by f i l t r a t i o n , so
th a t s l i g h t l y low r e s u l t s would be expected u n le s s i t s ash co n ten t was
u n u su ally h ig h .

270

FART C
ANALYSIS OF AMMONIUM FLUOBERYLLATE PREPARATIONS
Method — The p roced u re, as given in Appendix i l l , c o n s iste d o f v o l a t i l i z ­
in g flu o r in e by p e r c h lo r ic a c id ev a p o ra tio n , p r e c ip it a t io n as the
hydrated b erylliu m o x id e , and ig n i t io n to BeO . W eighings were made
w ith s u ita b le p r e c a u tio n s .
These samples were analyzed in con ju n ction w ith the m elt samples and
the d e t a i l s are rep orted in the p a rt d e a lin g w ith th ose sam ples.

The data

ta b u la ted below summarize the r e s u l t s o b ta in ed .
Sample
D e sig n a tio n

Average
P ercen t P u rity

For D e t a ils See-

(NH4) 2BeF4
Lot L -l

75.

Table C, Part I,

(NH4)sBeF4
Lot L -2 ,3

65.

(NH4) 2BeF4
1 /1 9 /5 1
F i r s t fr a c tio n

98.

(NH4) 2BeF4
1 /1 9 /5 1
L a st fr a c t io n

76.

(NH4) 2BeF4
Lot M

9 9 .5

(NH4) 3BeF4
Lot J-2

9 8 .8

271

PART D
ANALYSIS OF BERYLLIUM BASIC ACETATE PREPARATIONS
Method — S in ce th e se sam ples con tain ed no f lu o r in e , the flu o r in e v o l a t i ­
l i z a t i o n step was o m itte d .

The m a ter ia l was converted to the

n it r a t e by b o ilin g w ith n it r i c a c id .

Then, to make the s it u a t io n

comparable w ith the o th er b ery lliu m a n a ly s e s , the n it r a t e was con­
verted to p e r ch lo ra te by b o ilin g w ith 80$ p e r c h lo r ic a c id .

The

u su a l method o f p r e c ip it a t io n o f b ery lliu m hydroxide and ig n i t io n to
BeO fo llo w e d .
Comments — On many o f th e e a r ly a n a ly se s th e p u r ity was low app arently
because o f entrapped a c e t ic a c id .

U nless d r ie d a t higher tempera­

tu r e s than 110° th ere seems to be incom plete removal o f the a c id as
may be see n from th e example shown below .
That th ere may have been in te r fe r e n c e due to oth er cau ses would
appear from a comparison o f the l a s t two a n a ly ses fo r which the
samples d if fe r e d o n ly by two a d d itio n a l r e c r y s t a l li z a t io n s .

Date o f
A n a ly sis

Sample
D esig n a tio n

2/2 7 /5 1

M ixture
Prod. 1 /2 /5 1
P rod. 3 /2 1 /5 1
R edried
R e c r y s t.

3 /2 1 /5 1
k /2 k /$ l

N otes —

CD

Sample
Drying
Temp. , C.

Sample
w■
t .,
g.

Wt. BeO,
g.

110
110
110
180
180

C . 24OOQ
0 .I4OOO
1.000
0 .I4OOO
0 .11)00

0 .0858
0 .08]|1
0 .2 2 2 5
0.0 9 2 1
0.0990

% P u r ity

87.3
85.5
90 .3
9 3 .5
100.5

See Note
1
2
3
h

The s ta r tin g m a ter ia l was a m ixture o f the fo llo w in g
co m p o sitio n , which was tr e a te d to convert to p erch lo ra te
form and su b jected to the r eg u la r a n a ly t ic a l scheme.

272

M ixture

0 .1000 g . CaCO3
1.000 g. L i2C03
0 .I4OOO g .B ery lliu m b a sic a c e t a t e , P r o d .1 /2 /5 1 .

(2 ) The treatm ent was th e same as ( 1) except no other
m etal io n s were p r e se n t.
(3 ) The same m a ter ia l as Prod. 3 /2 1 /5 1 * excep t dried a t a
high er tem p eratu re.
(Ij) Same as ( 3) excep t w ith two a d d itio n a l r e c r y s t a l l i z a ­
tio n s o f the Prod. 3 /2 1 /5 1 end d ried a t high er
tem p eratu re.
(5) The c a lc u la te d p u r ity i s based on the formula Be0(CH3G00) 6
fo r b ery lliu m b a sic a c e t a t e .

273

PART E
ANALYSIS OF L i3BeF4 :lH 20 PREPARATION AND DETERMINATION
OF SOLUBILITY OF L i2BeF4
The s t a r t in g m a te r ia l was prepared from a r e l a t i v e l y pure f i r s t
crop o f (NH4) 2BeF4 w ith an a n a ly s is o f 9 8 . b% p u r it y .

A known w eight o f

LiF was added to a weighed q u a n tity o f (NH4) 2BeF4 , and the NH4F was
v o l a t i l i z e d by h e a tin g .

During p rep a ra tio n o f the L i2BeF4 i t was found

n ecessa ry to use 1 .8 # e x c e s s o f the (NH4) 2BeF4 (assum ing 100# p u rity )
to o b ta in th e t h e o r e t ic a l w eight o f p rod u ct, which checks w e l l , w ith in
th e accuracy o f the a n a ly s is .

T his p rod u ct, c a lle d L i2BeF4 , 1 /1 9 /5 1 ,

was analyzed to co n ta in 3 3 .h mole p ercen t BeF2 .
Appendix, P art j ) .

(See Sample S in t h is

T h is m a te r ia l, th e r e fo r e , was checked in th ree ways

as bein g very c lo s e to L i2BeF4 i n com p osition .
A sample o f 3 9 .0 g. o f the above m a ter ia l was d is s o lv e d in approxi­
m ately 3 l i t e r s o f w ater a t room tem perature w ith n e g lig ib le r esid u e on
filtr a tio n .

On evap oration a t reduced p ressu re no d e p o s itio n o f m a ter ia l

was observed u n t i l the volume reached about 1500 m l.

The approximate

s o l u b i l i t y o f L i2BeF4 a t room tem perature was th e r e fo r e about 26 g / l i t e r .
A fte r about 3 weeks a d d itio n a l ev a p o ra tio n , the c r y s t a ls and mother
liq u o r were allow ed to stand a t room temperature fo r about 5 weeks b efo re
f i l t r a t i o n , washing and a ir d r y in g .
L i2BeF4 sIH20 , 3 /2 0 /5 1 .

T his hydrated product was marked

A 20 m l. sample o f the u n d ilu ted mother liq u o r

was a lso taken w ith a p ip e t t e a f t e r f i l t r a t i o n .

The a n a ly se s o f th ese

samples are in d ic a te d b elow , the procedure used b ein g th a t d escrib ed in
Appendix l.h.

27k

LiaBeF4:lH 20
Wt. sample ( a i r d ried ) 1.1000

g .

Wt. a f t e r drying (180°) 0 .9h70

g .

(N ote-ab out 2 % o f w ater was ev o lv ed between 110 and 180° .)
Wt. w ater l o s t = 0 .1 5 3 0 g .
W t. BeO r e sid u e = 0 .2323 g .
V o l. Na2S20 3 "B" = i f f . 81 m l. (f o r 10 m l. a liq u o t o f 100 m l.)
T ite r o f Na2S^D3 "B« = 0 .0 0 1 0 6 g . L iF /m l. on 3 /2 1 /5 1
(S ee P a rt J o f t h is app en dix.)
C a lc u la te d on anhydrous and hydrated b a ses:
Wt. P ercen t
H20
LiF
BeF2
Hydrated
Anhydrous

1 3 .9
—

1+6.1
53.5

Sum o f
Wt. P ercent

3 9 .7
1+6.1

9 9 .7
9 9 .6

Mole P ercen t
E3°
LiF
BeF2
23.0
—

52.2 21+.8
6?. 8 3 2 .2

D e sp ite th e good reco v ery v a lu e s , the r e s u lt s d if f e r from theory by
about 1% w ith a probable error o f about 0 .5 $ .

The d iffe r e n c e in the ca se

o f the hydrated m a te r ia l can o n ly be p a r t i a l l y ex p la in ed by some dehydra­
t io n during the se p a ra tio n and a ir drying o p e r a tio n s .
Su bsequently the b alan ce o f th e hydrated product was used to prepare
M elt d and a f t e r therm al a n a ly s is , t h is m a ter ia l was sampled.

The r e s u lt s

(See P art I o f t h i s appendix) were as fo llo w s :
Wt. P ercen t
LiF
BeF2
5 3 .7

1+7.1

Sum o f
Wt. P ercen t
100 . 8

Mole P ercen t
LiF
BeF2
67.5

3 2 .5

The agreement was reasonab ly good w ith th e precedin g a n a ly s is , but
s t i l l showed disagreem ent w ith th eo ry beyond the probable erro r o f about
+ 0 .5 mole p e r c e n t.

275

The r e s u lt s o b ta in e d on th e s o l u b i l i t y determ in ation by a n a ly s is o f
the mother liq u o r sample are shown below .
Mother Liquor Sample ( 20 m l.) .
On drying t h i s m a ter ia l p r io r to p e r c h lo r ic a c id trea tm en t, oxyflu o r id e form ation occurred and hence the w eigh t ob tain ed was m eaningless
as an in d ic a tio n o f s o l u b i l i t y .

The u su a l Be and L i a n a ly ses were c a rried

out to o b ta in th e fo llo w in g r e s u l t s .
Wt. BeO resid u e = 0 .1 7 5 2 g.
V o l. Na2S^Q3 "B” used — 22 .lj.2 m l. fo r 10 m l. a liq u o t o f 100 m l. volume.
1±I|..90 m l. fo r 20 m l. a liq u o t o f 100 m l. volume .
T ite r o f Na2S20 3 = 0 .0 0 1 0 6 g . L iF /m l. on I4./ 2I4./5 1 .
(S ee P a rt J o f t h i s appendix.)
C a lc u la tin g the m oles o f LiF and BeF2 p r esen t in the sample volume
and c o n v e rtin g to the b a s is o f a l i t e r o f sam ple, the fo llo w in g r e s u lt s
were o b ta in e d .
In 20 m l. Sample
g.
m oles
LiF

0 .2 3 8

Mole R atio LiF/BeF2

0 .0 0 9 1 8

Per l i t e r ,
g.
1 1 .9

1 .3 1 /1
BeF2 0.3292

0.0070C
T o ta l

1 6 .5
2 8 .1;

Assuming a 2 to 1 m olal r a tio and a s o l u b i l i t y o f 0.00918 moles o f
LiF per 20 m l ., th e t o t a l w eight o f LiF + BeF2 should be 2 2 .7 g . / l i t e r .
I f th e s t a r t in g m a ter ia l (LiaBeF4 , 1 /1 9 /5 1 ) co n ta in ed a s l i g h t ex cess
o f BeF2 , t h i s would e x p la in the f a ilu r e o f th e mother liq u o r to have the
proper 2 to 1 m olal r a tio o f LiF to BeF2 .

On t h i s b a s is a s o l u b i l i t y o f

276

L i3BeF4 in w ater o f approxim ately 23 g . / l i t e r a t room temperature i s
ob tain ed in f a i r l y good agreement w ith the roughly estim a ted valu e o f
about 26 g . / l i t e r ob tain ed during ev a p o ra tio n .
T his v a lu e i s about tw ice th e value g iv en by Ray (128) in the o r i g i ­
n a l r ep o rt on th e p rep a ra tio n o f th e compound, namel5^ 1 3 .2 g . / l i t e r .

277

FART F
DISCUSSION OF ANALYTICAL ACCURACY AND CORRECTION OF
CALCULATED COMPOSITIONS
The in a c c u r a c ie s o f th e a n a ly ses fo r the determ ination o f B e, Ca and
L i have been e stim a ted in Appendix llj. in connection w ith the d is c u s sio n
o f th e m ethods.

The e rr o r s are dependent to some e x te n t on the amount o f

the c o n s titu e n t p r e s e n t, but as a g en era l r u le th e fo llo w in g p o s s ib le
e rro rs may be g iv e n .
CaF2 ,

*

0 .2 to 0 .3 w t. p e r c e n t.

BeF2 ,

*

0 .3 to 0 .5 w t. p e r c e n t.

L iF ,

*

0 . 5 to 1 .0 w t. p e r c e n t.

The e f f e c t o f th e s e

erro rs in w eigh t p ercen t i s d if f e r e n t when the

o v e r - a ll com p osition s are expressed in mole p e r c e n ta g e s, and i s dependent
a lso upon th e r eg io n o f -the tern a ry system under c o n s id e r a tio n .

In the

r eg io n s o f most c r i t i c a l in t e r e s t , near the com position p o in ts o f CaBeF4
and L i2BeF4 , the r e s u lt a n t p o s s ib le erro rs are estim ated as b ein g a t l e a s t
* 0 .3 to 0.1; mole p ercen t w ith t 0 .5 mole p ercen t as a c o n ser v a tiv e e s t i ­
mate .
For a few c a s e s o f very c r i t i c a l n a tu re, in which undegraded samples
were a v a ila b le , and on which a number o f check d eterm in ation s were made,
i t i s b e lie v e d th e probable erro r was no g rea ter than ± 0 .2 5 mole p e r ce n t.
In view o f th e se a n a ly t ic a l u n c e r t a in t ie s , the c o r r e c tio n o f m elt
com position s was made o n ly when the c a lc u la te d com positions d if fe r e d from
the a n a l y t i c a l com position s by more than the estim a ted a n a ly t ic a l e r r o r .

278

In g e n e ra l th e ca lcu la ted , com p osition s were considered more r e l i a b l e ,
u n le s s a confirm atory s e r ie s o f a n a ly se s in d ic a te d a uniform d e v ia tio n
o f the co m p o sitio n s.

In a few such c a s e s , the com positions were then

r e c a lc u la te d to o b ta in agreement w ith th e a n a ly s e s .

U su a lly an average

value was used in th e few c a se s o f d e v ia tio n s o f 1 to 1 .5 mole p e r c e n t,
rath er than a c ce p tin g th e a n a ly t ic a l v a lu e .

Most a n a ly ses agreed

w ith in 1 mole p ercen t w ith the c a lc u la te d co m p o sitio n s, so th a t the
a n a ly ses p r im a r ily served to check the co m p o sitio n , rather than to
determine i t .
I n the ta b u la tio n s o f the a n a ly ses o f m elt samples the c a lc u la te d
com position i s shown fo r comparison w ith the a n a ly t ic a l r e s u lt s .

Many

o f th e r e s u lt s are given to the n ea rest 0 .1 mole p e r c e n t, but as noted
above, t h is im p lie d gen eral accuracy was not o b ta in ed .

In some c a s e s ,

however, where o n ly a sm all amount o f a component had been added, the
com position w ith r e s p e c t to t h i s component was a c tu a lly known to - 0 .0 1
mole p e r c e n t.

I t h a s, th e r e fo r e , been g e n e r a lly n ecessa ry to r e ta in

at l e a s t one u n ce r ta in fig u r e in p r e se n tin g the r e s u l t s .

The u n ce r ta in ty

in r e la t iv e com position s i s q u ite o fte n sm aller than 0 .1 mole p e r c e n t,
so th a t from t h is sta n d p o in t, th e number o f Ms ig n if i c a n t ” fig u r e s i s
ju s tifie d .

279

PART G
SYSTEM CaF2-LiF ANALYSES

The a n a ly t ic a l r e s u lt s shown in the fo llo w in g ta b u la tio n req u ire
the fo llo w in g exp lan atory n o t e s .
Samples Ek , EB and EG were ob tain ed by dip pin g samples o f the
e u te c tic liq u id from the s o li d if y in g m elt during the in v a r ia n t r e a tio n
a t the e u t e c t ic p o in t .

N a tu r a lly such samples might have contained

some p r e v io u s ly d e p o site d s o l i d .

The m e lts in q u estio n were A } B and G.

Sample PE c o n s is te d o f CaF2 c r y s t a ls c o lle c t e d from the s t ir r in g
rod p r io r to the e u t e c t ic r e a c tio n on m elt E.
The o th er samples were, r e p r e s e n ta tiv e o f the s o l i d i f i e d m e lts as a
w h o le .
The “t h e o r e t ic a l com position" in d ic a te d i s the c a lc u la te d com posi­
t io n based on th e w eigh ts o f CaF2 and LiF used in preparing the m e lts .
Inform ation on th e sta n d a rd iza tio n o f the potassium permanganate
fo llo w s th e ta b u la tio n .
The a n a ly t ic a l methods were as d escrib ed in Appendix l h , except
th a t lith iu m d eterm in ation s were made o n ly on the samples marked CL.

________

On
rd
oCM S
CM

On O n

-d-d

On O n

ON

-d-d

O CM
CM CM

CM :

CM CM

CM CM

CM

CO f O

On o n
i—1 i 1

—

o o
CM CM

KMn04 V ol.,
ml.

O UN
nO CO -d-d
O -d O oo
_d -d -d _d

CM 03
iH CM
• •
d On
-d -d

OO
8 8oCM

o
8 • CM•

d iH

•

»

H CO

8 8

O CM

d i—I

*

CM

•

1— 1

.

CO CO

'ON
o• o*

CM

CM
CO

CM CM
CM CM

-= t -3 *

1

o

CO
•
to

•

rH
UN
#
CO

Sample,
No.
Date

r—
i—{

-d
jd^
i—l
i—1
PQ

•

-d

o
o
o
UN

rH

orH

•

r

o•

O*

-d
CM

On
CM

UN

CO OJ

NO IN-

no

o o
UN UN

or - oin-

C\J
OJ

O c-

ON CM
IN- IN-

nO

-d
o

NO
ON

MO MO

UN UN
- d -d

UN
-d

i—i
UN

On

°o go
T
-TNUN

OO

O J OJ

CM t o
CM H

•
•
O n On
CM CM

O•
O

*

*

CD

o

O
r—

CM
CM

*

_=r
•

ON

-d
<o
I—I
I—I

ON

-d
UN

j—
I
i—I
(—I
PC)
Pm

MO CM

CO CM

u n u n

C

-d _d
_d _d

"LA
CO

o o
o o
o o

o o
o o
o o
O n On

MO MO

UN UN

o

i—I rH

o o

OO

OO

On

-d

On

On

-d

—
d

ON

-_ d

On

o•

to

ON
*
UN
CM

•
CO

CO

-d

•
NO
CM

H

pd
to
I—I
I—I
o

MO MO

-d-d

On

-d

-d

i—
J

iH
<o
<o
CM

-d-

-d

CM

CM

*

co
i—

l

1— 1

pH

On

S

ON

fO•
CO
<o

-P

On O n

€

H

d
d
C
O
<D
Ccj
u
o
o
CM

UN f —
•
•
ON <N-

(—i
un

l>-

O CM
CM CM

Sample Wt.,
g.

A nalytical
C om position,
Mole % CaFo

T heoretical
Composition 3
Mole % CaFa

Date o f
A n a ly sis

280

fO

o
o
i—
i

O'

o
o

o
o
o

1—1

rH

i—i
UN

i—1
UN

i—1
UN

n O*

NO

NO

vcf

MO

NO

d
CD
d

id
CD
-d

d
CD
O
rH

O

On

UN

i—I

i—I

i—I
i—I

On
CM

UN

O

O
r—

281

KMn04 S ta n d a rd iza tio n s
Standard KMr04 S o lu tio n §2 used on a l l sam ples.

Standardized a g a in st

CaC20 4 ob tain ed from primary standard CaC03 , 9 9 .91% p u r it y , Baker S p ec ia l
Grade .
S o lu tio n
No.

W t. CaC03;
g-

V o l. KMrD4?
m l.

2
2
2
2

0 .6 2 0 0
0.6200
0.6 2 0 0
0.6200

5 5 .7 6
55.77
55.71
55.73

2
2

0.6000
0.6000

56.22
56.21

avg.
avg.
avg.
avg.

CaF2 T ite r t
g ./m l.

Ave . CaFo T ite r , Date
/
g ./m l.

0 .01055
0 .01055
o .01056
o .C1055

o .01095

1 2 /1 1 /5 9
1 2 /1 8 /5 9
1 2 /2 2 /5 9
1 2 /2 7 /5 9

0 .01012
0 .01012

0 .01012

9 /1 9 /5 1

282

PART H
SYSTEM CaF2-BeF2 ANALYSES

The fo llo w in g comments app ly to the ta b u la ted a n a ly t ic a l r e s u lt s fo r
th is system .

The methods used were th o se described in Appendix 15.

In th e c a lc u la t io n o f the com position s from the a n a ly t ic a l r e s u lt s
i t was fr e q u e n tly found th a t th e sum o f the w eight p ercen tages did not
c lo s e ly equal 100$.

In some c a se s t h is was app arently due to im p u r itie s

o r ig in a tin g in the u se o f carbon c r u c ib le s and in other c a se s h y d r o ly sis
product contam ination seemed to be the ca u se.

C a lc u la tio n s o f mole

p ercen tages were th e r e fo r e made in th e fo llo w in g ways.
For the a n a ly ses dated 2 /2 3 /5 0 and 2 /2 7 /5 1 the r e s u lt s were c a lc u la te d
as i f the sums o f th e w eigh t p ercen ts were equal to 100 .
For the a n a ly se s o f 3 /1 8 /5 0 and 5 /2 9 /5 0 the d e v ia tio n from 100$ was
too g r e a t, so th e com p osition s were c a lc u la te d assuming the w eight percent
o f CaF2 to be c o r r e c t and o b ta in in g the w eight p ercen t o f BeF2 by d if fe r e n c e .
This i s j u s t i f i e d by the g e n e r a lly b e tt e r accuracy o f the Ca d eterm in a tio n ,
and the fa c t th a t CaF2 seems l e s s su b jec t to h y d r o ly sis e f f e c t s .
The accuracy o f the r e s u lt s i s d e f i n i t e l y l e s s than i s im p lied by
the number o f s ig n if i c a n t fig u r e s r eta in e d in the c a lc u la t io n s .

The CaF2

an alyses are probably p r e c is e to no b e tte r than 1 or 2 p a r ts per thousand.
The probable error on the BeF2 a n a ly se s i s dependent to a la rg e e x te n t on
the s i z e o f the r e sid u e o f BeO w eighed.

For p r a c tic a l reasons connected

w ith th e f i l t r a t i o n and washing o f the Be(0H )2 p r e c ip ita te t h i s resid u e
has to be kept s m a ll.

Probable error should not be worse than 5 p a r ts per

283

1000 on the w eigh in g o p e r a tio n s , but the presence o f BeO in the o r ig in a l
sample to g e th er w ith oth er im p u r itie s ap p aren tly g r e a t ly in crea sed the
t o t a l error in many c a s e s .
Some o f the sam ples in the ta b u la tio n were s p e c ia l sam ples, not
r e p r e s e n ta tiv e o f th e whole m e lt, as in d ic a te d in the fo o t n o te s .
r e p r e se n ta tiv e a n a ly se s o f BeF2 products are in clu d ed .

Some

These m a ter ia ls

were produced in th e e a r ly s ta g e s o f the work when the NH4F e v o lu tio n
from th e (NH4) 2BeF4 was c a r r ie d out w ithout a d d itio n o f the other flu o r id e
components.
The column headed “T h e o r e tic a l Comp.11 l i s t s the c a lc u la te d composi­
tio n s based on in p u ts to th e m e lt s .

P a r tic u la r ly in the e a r lie r work

t h is value was u n ce r ta in because o f i n a b i l i t y to a c c u r a te ly weigh the
BeF2 products which were u se d .
The column headed "Sum o f Wt. Percents" g iv e s a f a i r l y good in d ic a ­
tio n o f th e r e l i a b i l i t y o f the a n a ly ses and the ex ten t o f degradation
o f th e sam ples.

Some o f the high r e s u lt s on the samples marked w ith an

a s t e r is k are p u re ly due to the assum ption th a t flu o r id e s were p r e s e n t,
whereas such was not the c a s e ,

BeO b ein g o b v io u sly p r e s e n t.

28b

‘a

.o

CM CO

0

HA

0
C
"—H HA Op
i—
I CM

O H

r0 O O
Eh S

CP

ha

o

cp -H

h

HA

0O

-Hf

CM

o
-= t

ha
ha

HA

O

O

ha

Ha

-H
nO

rt
o

f esS. 01
rH
0 1*4
cx3 rH cu
0 OO
rH

-■

o
nO

S

•

CN CM CO rH O n CO

CO -c ? rH MO H

H

C M O N C O N O t P H C— M3 rH HA C— O CO O ON K— NQ —H
C P - H n t C— 0— C— HA C— 0— C— —H
HA—
nO nO

«

•

•

Ha tp

—I CM

I

•

O

•

•

HO o

i—I i—I CM

C p n o O - r p _ j - HA r p CM H \ K - nO H a nQ

w

-p
«H
o
0
O
-P o
5 u
0
C
O

H

• CK* H • nO• nO
• J•- n •O
rH

PH

OJ

+>’ ‘a

|

^ o

I

• M
*
I

I II

0

|3

m

C M rH C M H A rp O -C K M A

CM O n H \C M H A H A H 3

O n CO HA HA H a HA O

Q nO HA
. p. HA CM
. Q
C
O CM CM o
HA CO CD OO nO CO GO CO l A NO NO NO m o in- k -

q

' A\ H' J
H

rH CM Cp

rH O-NO HA HA o h a
CMHA H a H A O

HA O

O

o
CKn H
i—I oq

h aCO
c oQq
HA

HA CO O

O

K- CP CM nO o o O n

. . • •

I

I

CK CO NO HA CKOO NO
i—I CM —Ct
rH c p

I HA K— O
0 - ON

c p CM H A i—I O

O

■ s

O HA
O

CD

no

On O r i O N C p
HA nO HA C— O

CO CM O n O - - d O M3

H O O P — O N tH rH C O c p

c p k -H H A O n c o
•

•

•

•

•

•

On r p
• •

- 4 - H O H C m o d On nO
CO O
CMCM CM i—I CO CO
i—I
i—I i—I rH i—I i—I i—I
iH

O cm nO

cphan

cm c p

CM i—I -Cj" K— C p rH O n - H O N -H NO C K _ j- CK q
C P CM CM rH c p CM rH H H f U A J r p _ H CM C p

I—I

•O
0

o

o

H M3
c p CO CO CK O n n o C— c p i—|
CP —cd" n o o n r p n o cm n o
o
CO MD O rH H H CM CM CM
O O H O O O O O OOOO

O O CM
HAHANO
i—I O rH
O r iH

O

O

O

cp
H A O n n O c O H A O IA -H A C O
H A -C f o
On o H A n O CO O'- O ON
IK C O O N X IN O O n H C A H O n O n
O O O O O O r H c P r H

CM HA O
O N CpH A
O n CO O n
O O O

CO C—
C -O
0— CO
O O

•

|3 pq bo
0

ONHACM

C O O nnQ

O N r l ON rH - H (H C p U A ON C K C O - J - J o
On O CD CD NO ON CO 0 ^ NO O n O n O n H \ * J

Hv

- P

W

ha h

HA HA
c p c p CO H a C p i—| CM O

* 1*4

-P

H J

HA c p O ~ H O O
C P CM CO H A -= f _Z* C p

i—I

OJ

O

O n HA O CM c p HA CM H r H C p O N C U H A O N r H - H H A H A
ON O O O n O O
O O O O
C K O r H O O O
O Q
i—I i—I
i—I i—I i—I i—I r H i —I
r l i—I i—I i—I i—I i—I i—I

rH

rH r p

^ 9 h ?

On On n O

ON On o no CO O CK
On On O On On O On

O

Q

O

O

O

O

O

O

O

O

O

O

O

O

O

O O O O

-V

I—I

•

Pr-P

‘

1 is M
CO

ha

_

$i—1 i—I

CM CM CM CM CM

O

O

O

O

O

O

q

ha

C P r H t H C M C p r P O r H H

H

o

O

O

O

O

O

H

H

O

O

O

O

O

O O O O O O
O O O O O O
O O O O O O ,

- CM_ci_ct_n_n

CM CM

O O O

O O O O

-l!i« AC
H W H

A*
S ^ O P h

O

O

O

^

W

O

O

O

O O

O O
O O
o p

~'~

haha

O O

CK O
0

H

CM CM
* *'<ri p q o
OJ CM

i1

[*04 &
H
0

CQ PP

Q

w

*'t
Hr o

CM CM
&
* 'p | t d
0} OJ

O

CO

O

$

ha

rH

0

rH

+3 cO
ctf 0

P «si

rp
cm

=

s

s

«» A<
Eh fx! >H INI - P
■<

10
*4 f*04
cq cq

CO
<H -H

p>

O

HA

CD

On
CMS

cp

HA

s s n “ i—|

S

next page

o £

Continued

fH 04

285

cm

‘S
o
p

0
rt ft
O
- e

cd

0

to

co

co
U \

On o
O -= t
l O - S " l-O l- O T - O

CO

o

O H
r£ O O
EH

H

to
H

H

''A

o

ft

ft
O

£

•

S

P

CD
O

3

Si

co On On 1_o q co
i—
1 i—1 COI CM(—Ii I

CD
Pu,

co

O

CM
P

•

• M
<
Q

*

CO U \
t- o n o

’3 O

O

O

CO 0 -

^

H

O

f i r l

*

» S B

x

!?

»

Q

•

O

o

rl H rl

C— 1 - 0 r H

♦

o

«

C O -= f

•

•

CM \ A j c o
m nq no no

H• I P • \l Ao • c Oo • 0NO0 *
•
•
-rt -rt to On H i—I o
t—On rtl" i-H1 - 0 co-rt
(—Ii—I CMi—I
- r t ON

CM

' ftCD

NO CO

.

* I

CM

-rt O- On

P

3

PQ

CM NO
- r t CO

CO I
CO

C~— n O CM
C O C O CO

p
3

• oCD
CQ

.

CM Q
NO M J
CO rH
•
W)

1A
CO
CM

.

.

CMNO CM
On c—to
I

>

0

CD

CQ

P

CD

rt
0ft
rt
I
£

p

8

r—I

& +>

O IP O O O
O

bD
*

CO

♦

CO O
*

O

•

•

O

rt
O

•

OO OOOOo

*

O
cm
N O I— I I—1

CD
H

'I5
pq

I1

I I I

o

Q

W OO

s

£

CO

S

S

05

C
O

o

CO

CD H
p
rt
rt £

« -=tj

On
CM

1 -0

rH

to
s

S

1 -0
CM

CM

P

rt
•H

rrt

0

•

£
•rl
to P

O P

•N

*•»

•-

ft rt

E i—1
TO O
0
O

o

P
rH
0

0 6
0 0
I—I rt
f tP

I ft
CO O

0
O
s
P

o
u \

ft

0
rt

3

S

CQ

* rt

rrt

rt

•rt

p

I
I
f t

•rt

^£
Irt
£ s
E g o
o = s = == = e o
O r t rt •
ft
rt P ft
■ra
0
ft o
o
rrt
p
rt rt
H = c = = = e s i—I rrt 0 rt
0 0 £
rt
rt 0 i—I rt £ &
!*
rt
S ft rt -rt
cq ^
e
E §rt P •
o
r t O to - P
rt
O rt
o o rt
ft
ft rH CD o
cxj - p
0 rt
0
i—
I £ rw
t
£ i—
bO 0
0 I

I—I
I—I
rt
£

S|<

CQ

ft
ft
rt

0
S
£
O
rt
P
CO

o

P

•H

P

rrt

■s
rt
°
C O rtD S ftftX ^ O ft
<rt O'—'
P
H
0 =
= = E = = srft

ft
•rt
O
£
rt
o
o
rt
p
ft

rt
ft
£
0
rt
rt

bO

£

8

o

C
D *
rH *

•H
£

ft

1 -0 On O

P i—I i—I I r-H O rH
OOO
o o o

standard

rt

co

C
D

•rt
P
rt
P
£

the d ifferen t

Os O

_£- lA 1-0-S VO1-01-0

of

co

CM -r tf O
•
♦
•

tite r s

nO

•

for

OJ "LA CO
•
•
•

rt
rt
ft

See following tabulation
of KMr04 u sed .

H <D C®
rt i—
I ft
Sort
3 O

s o lu tio n s

P

I

286

S ta n d a rd iza tio n o f D iff e r e n t KMi£)4 S o lu tio n s
Date o f Use
and Standard­
iz a t io n

W t. CaC03
g.

KMr£)4 V o l.,
m l.

S o lu tio n
N o.

2/2 3 /5 0

0 .2 5 0 0

98.70

2 /2 3 /5 0

0.2500

98.60

#3

3 /1 8 /5 0

0.1 2 0 0

66.37

#6

Average T ite r ,
g . CaF2/m l.

0 .001976

0.002016
3 /1 8 /5 0

0 .1 2 0 0

66.30

,#6

5 /2 9 /5 0

0.1200

U6 .J4O

#6

5 /2 9 /5 0

0 .1 2 0 0

6 6 .2 0

#6

2/2 2 /5 1

0.6000;

2 /2 7 /5 1

0.6000

2 /2 7 /5 1

0 .002020

ca.

c a . 235.0
66.19

#2

66.33

#2

0.00200

0 .01011
0 .6 0 0 0

287

PART I
SYSTEM LiF-BeF2 ANALYSES
The com p osition s o f some o f the samples o f the above bin ary system
were checked by th e a n a ly s is o f d eriv ed tern a ry sam ples, u sin g g ra p h ica l
e x tr a p o la tio n to determ ine the o r ig in a l com position o f th e m elt in the
binary system .

The v a lu e s o b ta in ed in t h i s manner are g iv en in Table A.

In a d d itio n c e r ta in b in ary samples were an a ly zed , p a r tic u la r ly th ose
r e la te d to s e r ie s o f m e lts in th e v i c i n i t y o f 3 3 .3 mole p ercen t BeF2 .
These sam ples were analyzed

fo r b e r y lliu m and lith iu m co n te n ts and the

a n a ly tic a l d ata and r e s u l t s

are ta b u la te d in Table B .

In Table C r e s u lt s are ta b u la te d fo r m a te r ia ls such as BeF2 and
(NH4) 2BeF4 used in preparing the m elts o f

the LiF-BeF2 system .

The a n a ly se s fo llo w e d the procedure d escrib ed in Appendix lit .

As noted

th e r e , a l l lith iu m sample s o lu tio n s were d ilu te d to 100 m l. volume b efore
tak in g a liq u o t s .
The accuracy o f th e a n a ly se s i s l e s s than I s im p lied by th e number o f
s ig n if ic a n t f ig u r e s r e ta in e d in the c a lc u la t io n s .

The p o s s ib le a n a ly t ic a l

error i s as g rea t as 0 .5 mole p ercen t fo r samples o f m elts which were used
fo r a s e r ie s o f thermal a n a ly s e s .

For a few c a s e s , where the m e lts were

sampled b efo re u s e , or where the c o n d itio n s o f use were l e s s s e v e r e , the
accuracy i s b e t t e r , but in no case i s the p o s s ib le error l e s s than 0 .2 5
i
mole p e r c e n t.
The column headed "Bum o f W t. Percents" g iv es an in d ic a tio n o f the
r e la t iv e r e l i a b i l i t y o f the r e s u l t s , or the e x te n t o f m elt d egrad ation .

288

In T ab les D and E the r e s u lt s and data have been ta b u la ted w ith
r eferen ce to th e d eterm in ation o f th e t i t e r s o f th e d if fe r e n t standard
s o lu tio n s u se d .

TABLE A
COMPOSITIONS OBTAINED BY GRAPHICAL EXTRAPOLATION
FROM ANALYZED TERNARY SAMPLES

B inary
M elt

C a lc 'd Comp.
Based on ¥ t .
o f In p u t,
Mole % BeF2

C alc *d Comp.
Based on
A n a ly sis o f
Inpu t,
Mole % BeF

E xtrapolated
A n a ly tic a l
C om position,
Mole % BeF2

Based on A n a ly sis
o f Ternary
Sample Number
(S ee P art J)

2

A
B
C
D
E
F
FF
G
H
I
J

3 .3
7 .9
1 1 .8
1 3 .8
1 8 .8
22.0
23.0
2 1 .3
2 7 .9
2 9 .3
3 2 .3

k.2
8 .1
12 .0
IS.h
19.0
22.3
23.0
23.8
2 6 .9
28.7
3 1 .7

U.o
8 .1
12.0
13.0
19.0
22.0
23.0
23.0
28.0
3 0 .3
33.2

A-8
B-7
C-7
D-7
2 -5
F-h
FF-5
G-6

H-5
1-5

J-5

289
n

Ph
ppi
o

o
mi
o

Pq
H' 0
-p m
d

d
o

NO U N CO
•
•
ON CM rH
o n o n ON

•

rXj 8 * ^
0 M 0

•

i—1w H
rt rt
O
o m
S

-c t
CM
ON

s

•

rt rt
o

o

0
HO
rt
P<

-P

£(1)

0'&5.

ip ,

•

«

®

E H

H«

r—
j (O, 0 m

doom

®

on• on•

- p f CM CM

■OJ O s

on o n on

O*

U\♦

CM
ON

on

c—•

O

ON

vO
ON

CM

£

Cs-

XA
•
i— 1
fA

ON

•

•

<oD

co
CM

Tj
(D

g

•rH
-P

m

d

-p

rt

OJ

o

o

S

rt

CD
P-.

C
O

o

1
CM I P *
•
•
OHO
0 ,0 0
HH H

CM
•
ON
On

rt etJ rt

x>

oj o
o
- r t •rH
CO
jd

•rH

I—

GO
•

OO
•
ON
On

o

O
<—1

CO
•

co

O
1

O
i—

«

O

O

i— i

ON

ON
•

O
O
I—1

O
o
rH

£
CM
CD

p?
&
PJ

PQ
W

Hi

Eh

cn

bO

d
•H
O -P
d
d
m
CD
OJ M

TO CD
CD - P

CO O

tz>

d

m <3 w l /

hO
>

.
hO

>

uo

£q
m

m
m *h
H

d

o -p

>

^

ffl

m

hO
>

<

O »M)IA

O H

CO pcj =3j

hO
>

■<1

c o o c~- U b o N - U UN CM O
b - C J cvj OO
n
CM |H O n CO CO CO U N CO ON CM oj ip

O -P

J —K

e

o

r£
H

Xj

o

•

.

Eh

C B EH

o

o

o
•H

o
*rl

o

/-s

X—
N

o

.
O

C
O

0

o

I

- CD
d
cd
H CO
O »;<

pq
§

I

/r-N

•

•

o
*H

*H

r-*—
C
Eh

Eh

b.0
!>
<

>
<

O

r»-H

hO
i>
<3

hO

UN U N O O U N - d
ON
U N C O p - CO

ON CP-

CN NO CO

■ CM CM CM
J A H O 0 \ 0 H O —U U N U N UN ON ON r O P C - IP - U N U N L A P P
O-I U \ UN _ d - UN UN UN CM CM CM CM CM CM CM i—I H i—I C\J CM CM CM CM CM i—I i—I

CO

$

kH

1

-P
O
• d
H cr*
O •H •H
> rH t-P
<$

•

o

P 0
13 m

*
bO

•N
#
£
-P

<D

Q

O
I—I O
I—Io
I—I Io
—I
PON CM CO
p - O n c o cm

UN-Cj U ju d CM CM CM CM

UN

UN

ON

UN

8
CM

CO
CM

UN

UN

o

CM

OOOO

o

•N
0
i— 1 •
•
CH -P
£ ■ bO
rt *

OOOO
OOOO
oooq

O

to

i—I i—I i—I rH

•
-p
rH o
0
IS

HKPiCtJ

H W ON CO
C UN t—1

o
o

T3

UN

UN

UN

rH

On
UN
CM
CM
•

i—1
On

o

o

o
o
o
1
—1

O
O

o

o
CM
•
o

-U
CM
•
o

o
o
o

UN
P-

o
o

o
o
o

i—i

o

rH

CM

TJ
u t

r
tJ
P-

•

"d
o

1—1

p-

on

•

Q
rH

d)

o

ft

CM

ft

-

-d

P

O CD H

ft
o

O

S

£.

CM

ft

f t CD f t
Cd g H O
P O O CQ

p
Up

•St

cp

Os
•

os
*

ft

rp

cp

ft

U p

•
1
—1
up

o
u p

o

-ft

•
rp
rp

-ft

•
rp
rp

rp

sO

up

rp

up

O
O

Os
OS

CO
Os

0s
Os

Os
Os

Os
Os

*3"
v-'
o
•H
ft
EH

3*

3

r —.
CD

^—,
ft

--—s
CD

O
•rl
ft
H
.
bO
>
ft

O
•H
ft
H
.
bO
>
ft

1

1

1

»

CD

value

=3 O g

rp
rp

rp

<D

H W
nj cti
PQ

Os

co
-p
ft

as o f

*'<D
o
P +5 m
o 2
T5
Oh'SS.

p

bQ
•>

bO
>
ft

«v

o ^

o o

sO U P

ft p

CD H

o

>

00 g

CO f t -ad

ft
o
P
p
O'
•H *H CD
o
> H ft ft
ft

o
•
CD bO
m

<u
i—i •
&t ft *

•
UP

CO
f t
rp
•
O

co
-p

p

f t

•
o

CD S

g

u p

ft ft ft ft ft

o
8
ft

£ JS: bO

p u p

s O sO sQ sO s o

o

•V

u p rp sO

• •U

rp

UP

up o rp
up f t Os

ft
ft O
f t CM CPJ

u p

f t f t CM
CM CM -ft

o o
f t cm

sO
ft
Os
CM
•

SO
ft
up
ft
•

O
O
o

o
8
I—I

O
o
8
o

ft

ft

CO

rp
f t
rp
•
O

o

o

CM
O
CQ
ft

CM

CM

o
CQ

O
CQ
ft

ft

bO
i—1
>
•
•
ft
i—1 i—1l—1
(
—1r—
CM O S CO C P O - P CM

ft

II ft f t f t f t CM II
ft ft f t CM f t
O
i—1

q
ft

in
O
ft

ft

p
o

bO

bO

> OO

f t I—1 1—1

IP­
Os
Os
i—1
•

o o >
f t CM f t

ft
i—I
UP

o

CD u p

f t up

I OO

PQ

rp

f t CM

following the reagent solution
date of u se. See Table D.

p ft P
•rl
O -P • CD
P P P CD
q CD H CO
CD b O o S
ft f t CO K

The le tte r
particular

"Cl CD
CD ft
CQ O

bO

id e n tifie s

its

O • 0)
-P o
E
!s ^
P
CD
CO
ft

titer

S

O

291

TABLE C
MATERIALS USED IN MELT PREPARATION, LiF-BeF2 SYSTEM

Sample
D esign ation

Sample W t. ,
(«See n ote)
. g.

w t . BeO,
g.

(NH4) 2BeF4 ,
L -l

M ixt. A
0.7500
M ix t. B

(NH4) 2BeF4 ,
L -2 ,3
(NH4)
4j
1 /1 9 /5 1
F ir s t F r a c tio n

% P u r ity
BeF2 or(EH4) 2BeF4

Avg. %
P u rity

0 .1165
0 .1 1 3 8
0.1160

75.2
73.lt
7U.9

75.

0.7500
M ixt. C

0 .0980
0 .1 0 3 0

63.3
6 6 .5

65.

M ix t. D

0 .2 0 3 1

98. h

M ix t. E

0 .2010

97.U

Same, except
L ast F ra ctio n

1.000

0 .1 5 7 6

76.2

76.

(NH4) 2BeF4
J-2

0.5000

0 .1021

9 8 .8

99.

0.7500
0.7500

0 .1108
0 .1 0 9 7

7 1 .5
70.7

71.

0.2903
0.2077

0.1530
0 .1112

99.0
100 .5

9 9 .5

(NH4)2BeF4 ,
M
BeF2 , Brush
B eryllium
Corp. P rod u ct.

98.

M ixtures r e fe r r e d to above were c o n tr o l samples (fo r stand ard .
z a tio n s ) in to which (NH4) 2BeF4 was introduced to check on in t e r ­
fe r e n c e s . See Table E fo r d e t a il s o f m ix tu res.

292

TABLE D
SUMMARY OF TITER VALUES OF STANDARD SOLUTIONS

Date
Used

S t d . KMn04 S o ln . # 2 ,
g . CaFa/m l.

S td . A r se n ite S o ln .,
g . LIF /m l.

S td . Na2S203 S o ln .,
g . LiF /m l.

7 /1 2 /5 0

0.01056

0.001503

-----

7 /2 7 /5 1

0.01011

0.0011*77 ( f )

-----

3 /1 0 /5 1

0 .01012

0 .0011*8

-----

3 /2 1 /5 1

0 .01012

0.0011*65 (a)

k/2h/$±

0 .01012

5 /1 5 /5 1
9 /1 2 /5 1

(e )

0 .0 0 1 0 6

(a)

—

0 .0 0 1 0 6

(b)

0 .01010

-----

0 .001065 (c )

0 .01012

—

0.00101*

Note — The l e t t e r s ( a ) , ( b ) , e tc id e n t if y the t i t e r v a lu es
to be used w ith r e s u lt s g iv en in Table B.

(d)

293
CO

o

o

cr

•P •HCM
1—1 p R
o P
R 0 R

o

i—1 O
• cd
*
bD >

bO

t>
<4

P*H
a

R

R

b-P

CA

NO

NO

PI­

R

0

bO

cti

R

O

£

0

P
O'
©

SAMPLES

FOR

L iF -B e F P

SYSTEM

A NALYSES

H
•H
P

P

O

o
-d

bQ CD

a

w

HR

OD

rH o

P -p
> H p
CD CD
r bo

o

O

CM

CM
-R

R

R

R

CA
co co

CO 1 A

co
CA

CO

CM nO

CA

o\ p j

On R

CM NO

-

>

M

P

0

•H - P
O p

0?
O

ra

%

R 0 <£(

I
0?
O
w

I

CM

P
i P CD
PsR bO

I
CM

bO

>

8

R
CM

O
H
O

P
P pCD *
oRH

1A

Ti—
-AI

p

O
O
CM H

o
CM

O

CM

°CM H9

c r -H

R-<
P
•rl
ts

ra

•H R

0

C
D

Xi X}

i—i

CD CD ■s

O

H

e

■H
o

O
R

i—l
0
•H
P
P

o

O

AND

CONTROL

0

p

I—I
0

>

S T A N D A R D IZ A T IO N

NO
R

H

JD
0
H

CM
CM

0

CD
CD

0

CO

CO co

P
0
0
P
R
•H
R

H
0

i—I

i—
i

R

R
R

•H
O

CO

C-CA
R

R

CMfc.
R H
CD R
O
*
• bO
bO
CA
nO O
i a t -a
O H
H O
O O
*

*
o o
II 1!
M*1
O
cl
P O
ra

o

C
D ^
i— I

*

RP

bD

f ^

8 8O 8dr\ o
o
1A
H Pf IP CA
ooo

CO

O

o
8

o

o

X-A

o
o
CM
o

S
S <3

o
o
s
o

8o

O
O

88 8

O
O

o

O H

O
O

o
H

P P
0 ©
p p
•H H
Eh Eh

O

CD
CD

i—I
R
g

CO

P
O

R
©
m
CM
x—N

S

(0 o
°
Q
0

£
X

•R

9 ,
«
-H

S

.

C
O

/
o
O

C
O
o
8
a

CO

(0

o
o

cl

H
H

o
9 ,
*ri

R

R
M CM
O P H

rH

•p
> ®P
bO ra

H

R

O

29b
<0

oo
s>

o

T -r\

0 «H
•
i—I O rH
• 05
g

fcuO |>
>

-rl

\D
P t

•

CA

5-T\

P t

vO
P t

C
—
P t

r—
pt

bO

next

>

page

p

CP rH

W •

Continued

o

WOO

oQ

P t CM CA Q

vO O lA M J

T3

hO 0

IP— IP— C—- P—
P tP tP f P t

fl ra

•rH &

o

vo

ca

c—

0

vO

tUO

<E|

p j

•

VO*

vO*

P t

P t

Q vO O

v O CA CM r l
P t 0 - 0 0 CM

CA

OJ

•

•

P flA lW

•

[>- o> o- tr^

^

fcuO

GP
Oc

cm

01

I01

<8
*=!

«aj

•=*!

•H

ft

£

0

bo

0

td
o

>
p

o

s

i—I

£

0 •
QH

• M
<
HQ
2

pH P

g

*

g ^ bO

i—1

o

0
CO

o
o

s

o
o
o

o

o
o
o

o
o
o

T3

K
Ora
<1
p

p
0
p
•H

0

p

H

•H

E0jt

pc-,
0
0

0

H

Oh

C
O

g
£•H
a

0)
0) o
o
°n
♦rH
o pi
o

CM

/—*

<0 oCO
o Si
•rH
o i-p
w

mCM

II«

o
o
o
o

ICM)
Ora
<
p

P

0
p

0

p
•H
Eh

•H
Eh

rH
u\

vf

+3
G
O
o

o
CM
vO
Pt

•hH
E

&h

0

Eh

vO
P t

CO

r~I

w

I—I

UK
CM

co

r—I P t
o
I—I
o
o O

P t

o
o

ft

>P
hO hO

o

CM

H
rH
s
CO
o

0
0

vO

P t

Q

ft

'S3 i-H
o
bO

■h8
E

0

VO

CM

CM

tso

o
0
rH

■
A
&8
H C
<A

P t

o
i—i

O

O

s

CK

O fi i—I

i—I

S

p

2P S
•CHP
H
<3

>

ICM
oW

ch

CO

o
Si
•H
i-i

o

ra ipP

W) ra

>

0

C S

co

CM

i
—i
• p CM
a

o

to

o
<8
o

<0 o
o Si
•H
s
o i-q

WO
to ra
< ffi

295
rv
pi

O
P
*
1—1 P
0 G
t> ©
1
—1
• a!
bO >
> 'H
pi
<
O'
ft

O
0

CM
•H
PI
ft
•
O H
g
bO

Ox

CXI

c-—
-p f

VO

Ox

UX
VO

vQ

r—1

©

O

bO
cd

ft

■£

©

VO

SP

co

XJ
©
Pi
G
•H
P
G
o
o

OJ
ox

"0
«
•H P>
O
bfl ©

g

I—I

>o_ T2J -cP r
CD ©
Pd W)
<3

bo
c
© "H -p
C
©
Em P3 bo
© ■=*!

Pi

D
O P to q
X> O -Pi- vO M3

O O

IX - CO OO IX - O -

ox OX

H

rH

OX OX

'- p f - P f - P t p f

t
CM
o
w
<=d

8 S S

o
O

IX-VO VO
vO vO vO

o x cxj cm
OX o x OX

O
OO
0\H1A

OOO
o o o_

U X V O VOVO v O VO

OX OX OX

•

•

«

0 UX
O xco

ox ox ox

o
•H
P
Em

o
•H
PP

JP

EM

EM

ux

O

ux

H
o

> -PG
P
o
G

H

©

cr -H,

•H PI

1—I

cn

fCtU

cd PI PI

o

O

bO

• bO bO
bO
UX vO

r—I

1—I

■UX

CXI

(XI

VO
-P f

VO

CJ
OHO
H O O OX
O O * VO
• ’O
° ° II _pf
II M II
tdB
O,
o tMCO O
«l
o
2IP Oto ©

•

bO
CO

O

o

g

T3

©
p p p
•H
tH
•H
Em EH EM

©

B

•H
-P

a

VO

o

00
•

-pf

©

i—I
|

co

©

G
pi
■£
a

o

P)

•H
PI

o

p)

S
3
o

PI

o
S3
o

P)

8M
•H
PI

<+M to I—I
(XI

OP
H

• Pi

bo ra
> ©
ft

PI
50

UX
vO
o
I—I
o
o,
o
II

P)

O
O

$

s

o

O

°©
CO

0

CM

ci

fp

G
©
p

•H
Eh

1—!

O
o

Em

OJ

s
s
o.

O
JvO O
r - l - P f 1— I

ft

©
+3

H

Em

1—1

ux
<+m © u \

p)
O
o

o

p)

S
3
o

P)

O
o
PI
•H

PI

O P 1—I
• pi ux
bO ©

> ©c
caj o

296

#\
(0

o

po o*HOJ
•p p
C
p 0
o
ft *
> Pcd O p
e

* >

bO •rH bO

ft

cA
(A

nO
nO

0

O ’p
*
W

CA
CA

o

O fA

nO nO

nO nO nO

O J CM

P

R SS

O

lA H A

o

O

•rH

£

*>

CA
CA

cA
CA

O

o
•rH

O
O
o
CA
i—1 P
IA T A

ft

EH

Eh

O
•H
-v_t
Eh

■LA

1A

O
i—I

ft

EH

ft

nO

P

o
1—I

■Hft

• -tf
H Q
*
O H
> S £

"LA "CA
CM CM

O
■H
Eh

TA

MD

O P
bO bO
CM
i
—1
o
1
—I
o.
o

"
LA o
OJ
CM
ft

%

NO

■LA

ft

O
i
—1
o OJ
o OJ
o fNOt

OJ
NO

II
0
rH
•
ft P bO

s

o
o
q

CO

o

H

o
o

oj]

o
o

CO

9

cd

§

o
i—i

ft

-p

B

EH

•rH
-P

o

0
0

U

i— 1

P

£
°

<0

ft
.3
S

Cl

o
9

o

<0

o
o

O

nj

•H

Eh

p

PA
ft Wf t
O A

CM

■
bO
cof t

>

ft

o
o

o
o

o

NO

o

A
•H

co

10

C

O

0

0

CD

m
£

c*
O ,

nP.

u

ft

ft

"LA
CM O

= 0.01012

Q R
CT -H

8

O

1
—1 p.
01 f £t

O
> a
4-^
p 0

o

A

0
if t

3
O

o

o
9

°0J
•H

P

o

cO
o
°«

co

O
9
o

p

X

CO

o

ft

•rH

NaaSa0 a = 0 .OOlOi; g. Liff/m l.

NO

CA CA fA
CA fA CA

nO
nO

A-

O T-A

p h o

On
ft

Titer

<ACA

P

O-

g. CaF2/m l.

r~ -

co f t

KMrf)4

1A N Q
nO nO

bO
£
0 •H P
O C
0
ft
Eh 0 bO
ft

iH

On On

Titer

P

"LAI—I

•rH

°0J
•H
p

Avg. of
R e su lts
on 9/12/51

ft
bO 0
£ ra
•H ft *
O
o £ p e
> ft £
0 0
ft bO
*=3

Om
Q5 O
1A
O

297

FART J
SISTM CaF2-LiF-BeF3 ANALYSES
The a n a ly t ic a l methods used were th o se d escrib ed in Appendix l l i .
In ta b u la tin g the d a ta , con d en sation was n e c e ssa r y , so th a t the f o llo w ­
in g n o tes e x p la in c e r t a in fe a tu r e s o f th e ta b u la tio n s .
Table A i s s e c t io n a liz e d accordin g to the date o f a n a ly s is , so th a t
th e t i t e r s o f the standard s o lu tio n s (a s o f th a t d ate) may be obtained
from the summary ta b le (B) a t the end o f t h is p art o f the appendix.
In th e column headed "Vol. Reducing Agent" , the symbol "T" or "R"
s i g n i f i e s u se o f sodium t h io s u lf a t e or a r s e n ite s o lu tio n .
In the columns under "C om position, ¥ t . P ercen t" , the value fo r
lithium - flu o r id e has been o m itte d , i t b ein g understood th a t i t i s o b ta in ­
able b y d ed u ction o f the w eigh t p e r ce n ts o f calcium and b ery lliu m flu o r id e s
from th e "SumWt. Percent" v a lu e .
S im ila r ly in th e columns o f "Com position, Mole P ercen t" , the value fo r
lith iu m flu o r id e i s o b ta in a b le by d iffe r e n c e from 1 0 0 .
The a liq u o t v a lu es in the lithium , d eterm in ation are not shown, u s u a lly
a volume between 1 and 20 m l. b ein g taken to a d ju st the amount o f p r e c ip i­
t a t e to a con ven ien t t i t r a t i o n v a lu e .

The value shown in the column

"V ol. Reducing Agent" i s given on the b a s is o f u sin g 10 m l. a liq u o ts in a l l
c a s e s , based on se v e r a l t i t r a t i o n s .
For the r e s u lt s on 7 /1 2 /5 0 o n ly , th e c a lc u la tio n o f the mole p ercen ta g es
was made a ft e r f i r s t a d ju stin g the w eight p ercen ta g es to a t o t a l o f 100.
On th e oth er d a t e s , the mole p ercen ta g es were c a lc u la te d as i f th e sums o f
th e w eigh t p e r ce n ta g es were equal to 1 0 0 , w ith ou t adjustm ent.

298

- P j IN- P

O • P * P • P • P• o*- o• P» O
• "UN
• •CA
C\J
CM ON PR P MD P H CM

i—i
<3
o

o
OJ

C\1 OJ

•H

UN

o-PU N O

C\J i—| i—j r —| |—| )—| i—I i—f

O O P P

I rHO -oNO UNPO NO
I rH
U N P P P P

H H 0— ON
CM rH

■P

^ e r g 's

SAMPLES

On

o* P

cm

on

o - O' o*

CM P

H

P

O'

CO CH CM nO NO nO C O P O

o o

O

•

CM
•

•

o UNP P
•

•

•

On

•

•

I O - P P O P

P P P P

I P P P U N O N P O N P

co o n

r—I O

nQUNCM

O n O n U N O - O n c ^ - O n ON

«

o 'o

CM CM ON i—I ON UN O

ON UN CO (H U N UN H CM p P
ot—I i—I CM CM OJ CM OJ OJ

NO o - P O n
p p p p

R

IN- U N P

NO O n O n P

o

On O iH ononcoonhOOnOn

P p
»

•

•

•

•

•

ON ON ON C
— rl P

•

•

•

p

•

•

•

o o- o n p

I o - r— P

s|« I p

•

- P On U N O ON C 1
P P U N P P

;J( I rH

R E O ©
■a! o g m

UN ON ON P

•

«, I CM P

rH O CD O
J

03 PL.H fa

P O

CM

O- C --P P

CM ON CO P "
CM rH

CM M3 O CM PR CM ONNO pj" X CO
CM CM CM CM i—I i—I i—| i—! i—It—I t —I

O rl p
•H - P CD
*P -H CR
!>s CO

ANALYTICAL RESULTS FOR TERNARY SISTM

on

next

-p 0]
c3 o

P

* * « • • • • • • • •
CM U N On O NO O n O n O O N P co
H H rlH C M O J C M C M

P

UN on P

•

cm

ONP

CO
-p

R -p
CO H=»

(D

u
CR

o #

NO CM c o tH o

O H

i—I H

R -P
•O
rH R
© a
-P o '
•C
rO
HRo
©
o P-t 0}
a •
E -P ©
O
CO is pq

Q

P O JC M O N O H O N C M

r l i—I r l r l H i—I

VfN UN
P t
UN
tH c O n r l O n H - P on O cm iH

C— CM rH ON

UNO H U O O O

l> - p O- P
O'* CM H

P U N P 'P

n

O O ^ 'O

ON CM CM ON CM CM CM

pONOOpvOO

O nO nO U N O N C M

CM CM CM O

rl

ON UN CM C— CM U N nQ ON
I P N

•

I

o

*

^ C

J

I CM t H

ON UN CM ON

NO NO NO NO nQ

UN
O CM t H P C M p O J

UnUnP
On o n
no

O-COOO-P

CM iH U N U N ON

CM CM CM CM

j
ooqqqoqq
I H i—It—I H i—It—It—It—I

o
o© o
rH rH H H

O nO O O O O O O O O n O n

CD

bR ■r\
•XR -P
R
1
—1O ©
o R
> "©0 XbO
PR
•
1
—1
o £ rH
>
E

p

•

i

•

•

•

•

»

•

ONUN

»

«

O - ON NO Q
P U N UN NO

NO O N N O Q P H P O N p
P U n UN nD ONONONCM ON

tH O NP" nO
( * » •
* • • • • • • • * «
0)
P P CN X P P O N U N N O N Oi—
1 R j CM CM ON H
U N NO U N U N P P
ON ONCOJ CM
rH i—I
OJ

-P H P O
» 0 ' 0 © D
co • • • R • • » •
fR O O—C CM O UN rH §R= ON
f-R H
CM &

ON UN i—| O O P ^ C - 0 \ 1 T \ 0
X

X

cm o n cm cm cm o n o n

o

<D

ON CM O - P o P C O O n O C O

p
S

*
X
o

CH

CQ

■s

i—I

u

p U N o f f l c o m

o -c o c o

■p

o— ©

On P ■P
<

on

H

O C O o UN U \ UN Q
ON ON CO p ' c— ON CO
nO nO H P O
OHC
O O t—i 1—I i—I OJ CM

CMP H CM
UN CM On CM -P
UN H P OJ c3
CM CM H ON 0)

C
PO
P P
CM

O O O O

O O O O O O O *O
O

P

■P

CD

P

-**
X© • •

o

O O Q O O O O UN UN O O
O O O O U N U N U N OJ CM O O

co

O N C M C M C M rH rH H H H H H

rH O O rH co

o

q

q

o

q

o

o

o

o

•

•

.................................................

•

CM
P

c3
R

• * • * Or
©
R
-p
^ °o o
S o RCD
o
O R O O
R
o o-UN O O'

o

W

O - P C—CZVCM C M P H
X CM
P p P P

r—I

©

O UN O UN
i—| nO ON O O CNQ3 O i—| rH rH t—I

O O O O O O O O O Q O

CR-P
E is to

U N O N O O ONUNpON
■ P O O rlO U M N 'O

on

P

o

•N
•O
-p © •
[3 PQ hO

W
CD

O N P C— CN ON O n nO p
• « • • • • « • • * •
X
CO nO O n U N O N O n i H On P
X P ' CO
cm O n o —oo _ P o n - P on o n cm cm r
{ 1
o
rH O

•

♦

•

P O 1—* C3N CM O
UN
J r H r H C M O - CM P
ON
C M
C M O N O PU N P
CM t—I ON CM t—I rH UN t—I

P
O Q ' U N O O O Q O O
Q O O O O O O O O

& q p

CD

q o

o

o o

i—I H O i—I tH i—It—It—It—I

_a

R

co
I

CD

1-1
&

CO

fa

o

UN

)JN
CM
H 0 3 0 0 1 > U

\4

page

P

UN
I P

UN UN UN

I I I I I I fa 1 I I I
O-pCQOQWftri^COcCMH)

t—I P"
CMnO ON i—I i—1
. I I I I
CM W) M bfl M

X
UN

#

O

„
H P
t—| nO O n r—I i—I

P

I

I

I

I

P O N P
CQ

ON ilO t j ) fcO bO o

I

X

I

I

1SI P>

Continued

<D m pu
R o
O & v OJ
CD E H fa
O O CD
o S m

299

O
£
0

P t

P t

p f

LA LA LA O

r H CM C O n O

i—1

*' * * *

O
•H

w ft

u

o

ft

O

O

0

H
0

£
*0

CANO

or, O

o CO

O

•

•

•

lr \( M f t f t

O

i—I

«

i—I

f t

IP - C
•

•

CO

•

0—

CM n O

•

•

*

l a

•

•

C A P t r H CM C A P t P t

CM

C O NO A - O
L A NO f t L A
CM

LA

P f M3 O

f t P t

CN

L A C A L A O

ao
CA

N O CO H r H C O CM C A
P t P t c a c a CM c a c a

CN

P t L A NO L A
c a CA CA CM

0
p

•

• • * »C A«iC At CM* C A•

ao

CO 0 O

o x a o o c o

ca

CM

cn l a

l a

p t no

f t

f t

L A NO L A

fH

0

^C K

OJ
0

P

ft

-i.

CM CM p f

E

• * * * £t>8
f v 0 0 3 \0

£ H ft
o
o

•

O n CM P t O

•

'L A

PL,

ft 0

•

ao o o a l a O
CM c a C A c a P t

p fc \J n c n

0G ^CD
O
£

•

i—I i—I

CVJ o A p f

OO nQ L A f t

EH

C
O
0 •H
o -P
• r t •H
-P 0
r> o

»

0" ft

f t CD 03
CD £ rH f t
o

•

P T CM CM C A

CD

OO P t

•

•

IN - CM r r + O

•

•

•

*

•

P tC M

•

•

lA C M O O s H P tO O O \N ftC M

*

»

«

•

•

»

•

•

•

«

f t

•

•

•

rH

P - O N N O L A O O OO c o
f t O N CM CM C N P t P t O
O L A nO
CM CM c a c A P t P t C A c a P t P t C A c a CM C A c A O A c A c A A

C— L A
A O O

NO CM rH NO C A L A f t

N' °

£
£

0

w

P

"S
ft

i—
I rH C i—
IP

0
o
Pi

-p

•

*

p f M3 H
O, Q
Q

*

•

£

(0

-P
£
0

o
o

01
f t
cD

-p

C

P f O

L A N O CM P t P— i—I f t

O

LA

2 3i—I
rH

3 2i—I2rr+tH
2 3i—I 2rH
rH

O r H O Q O O O O r H r H H O Q

q
qooqoQoq o
ooq
i—I i—I i— I i—I i—I i—I i—I i—I H
i—I t —I i—I t —I

L A NO L A
CO
P f CO C O C N C N n O M3 C A ^ M O
r H P t M3 C A P t P t MO MO r — C N r H c O N D p t N a a o a o o o o C N r H

P rH L A IO itJ

• • • •0
'•Q
H Q 0A 00
M3 L A m 3 n o

0

O ft
Q.
W '

rH i—ICVlCVJ CVJ C\l

o

p
o
•H
-P
•rl

i—I

O

i—Ii—I i—I i—I 0

0

ft

ON

f t

c a -H
Q
f t

i—I i—I i— I cm o n Q n P— P—
C A C A i—1 CM
CM CM

P
r—I
0
* • « « £ ?

01

"LA C O I A

Is ft

C N L A C O O nO M O

* * * * *

O

C A CM CM C A

O N P t nO C O C N O L A L A
rH

CA P t l a L A
rH rH rH P t

LA

M 3 CM "LA C A

f t
0

f t

ft‘ ft- A 0 N *H - ^- 0« C
K A A O» A• O
'0
* • #
• •
MO3 NO f t n
nO
L A L A CA CA n
P f P -t-t p f P - t P -+t P - +t P pt CM

CM P t P— CACMCM P t P t O N o
P t LA O
C A C A P t P t L A L A P T P t L A NO P t —- l "

j >j
0

fcUD
ft
• H -P
o

£

£

o
>

0 H

£
f t S
=■ E H f t f t f t
L A CM O
P -O A -G O
O
CM A - CM C M LA O

•H

^
J - H ^ O C O

•

CQ

* • • * sU

b.O ^ f t i— I C O O "LA $
0 *0
rH r—
I

TJ

ft

H Q
0 -v O

i—I
o

LA O
C N CM
MO L A L A M3

-P 0
PQ

o

o

**
P-

CM C O A - C A O O O O n L A CO CO P - I A C N
CM r H CM C A P t C A CM C A P t CM CM CM r H

AC—

ft]

H CA O CA
i—I CMi—I P t

rH
CA

O

H

p CM

CO
f t
CA
CM

CN
„ .

ff tt

f t L A CO O
N O CO f t L O

f t CO C A N O CO O n O
rH

LA

CA C A L A P t
1— I 1—I r H — t

CM
LA
f t
CA

CN
CO
CO
CM

CM
QO,
ao
1—I

rH
O
P t
CM

f t
P t
H
CM

CO
CA
P t
CM

CA
O

CM

CO
P t
P t
CM

CM CM
f t LA
P f P t
CM CM

ooo o O o

0

0

0

0

0

0

0

o

0

0

0

0

0

0

8

0

0

0

0

0

o
o
o

0

0

0

0

O

o

•

•

•

•

O n cm a CM

ca

ca

rH
LA
f tC
rH

C A P t NO
CA C A C O
O o
cm
H CM CM

•

•

•

•

CO
CA
CM
CM

•

•

f t

O

CO O

co co

f t

• O N» CM• CO• L•A c• a •L A

CN
CA
P t
CM

CA

•

P f
rH
PT
rH

P
£

0

rH
*
f t p

£CD

o

0
0
0
£

M
o

0

L A P f CM P j ' C M O r H CO
CM CM CM C A C M C A C M CM

O

rH CM P f rH P
ON O L A P f
c o L A c a nO
H
H
rH rH

• O

|S

f t =
=
=
f t f t f t f t f t f t £
=
»
C M N O n O L A A -3
o L A O
O Q L A O
P — O N C O r H l a NO C N r H ON A N D CD f t

•

H
£

>

A A
CO CO

f t
o
ON

ExO

ft

O
o
o

O
o
o

o

o

rH

rH

rH

£
O’
0
0

rH

0

0

0

8

o

8 8888

ft

rH

rH O

rH

O

8 8 0 LN o
. ^ ^ 0

O

. 0

O
. 0

o

o
.

2

rH

1—I o

p f

o
f t NO C N H
CN
P f LA
I
I
I
I
I
1
1 cy a? op cy 3?
o
P -I c m C A N O f t ( X >

o o o q
o

o

1—11—11—1 1—1

1—I rH rH 1—I

£
0
0

0

0

ft

rH

ft

ft

LA

*

.rH
O H
rH

I

.SSI

C A ESI

M3

0Q

1

I

O

•

ft ft |S

|L A
-ft
P t
rH
rH
[CM L A L A
I P t
I

CA. ftI ftI ft
ft gI gS

is;I

ft

I
cy
ON

1—I1—InQ
1—I1—I 1—I
I
a*

I
I _
cy c y q f

O

CM c A

1—I r H

1—I

I

next page

LA

0 \ nO l a cm

Continued

f t f t P t GO p f

CD

£

cD

-P
CD

ft ft
CVJ LA D— LA

N

-p

i—i

300

•v-p
cD O
O *H
H -P

N

I—I CO c

© P
o o
Jh

H ©
r i
C\J CM

r l

C\l A

A
LA A

LA A- H

CM O
CO

A

rH A H

vO H

A

A

00

m •

-P -H ©
0
k

CO PL,
O

S Oe 't!
O ^
fl)

^
H

o

0« ^
C O 'A
CUft

A
CM A

O p , ffl «

A

P

A

CM OO O -

A

r - 'O
P 'H ©P H
OO
C\l A P
P r o r o

•

H
A

P f

s m

A

On4

•

*

no

•

cm

»

A

O

*

•

A
A

S 8 8
II •I II O
OQ
O
I II
* O* O*

cD H

53 H

I—I C A - A - C— nO
A CM CM CM CM CM

A

O

O

O

fc©

' p
«
fi f t
o © CD
cD •rH O O
O -p u
• H H. . ©
P
W CL,
f>s O

p c q r o i f t l A O ®
On
« « • * * • • •
C M C O H H C M A C M C M
CM CM
no

i—i

£

rl
^

f t 5) N
S H ft,

A

A

On

AH AH A A A
£

•

•

•

•

NO C CO i—I A
p
cm A
p
a

o ©
s
m

A

P P

•

A

<

O
•

CM CM O
A A A

»

H
A

On A

©
«

•

I—I
o
co

A no

»

!

«

CM A A A A A
A CM CM CM CM CM

© H

•p ^
’d ft

ca

© -rl

-p
•

o
ift
©

•

•

cm

• H'
P>
•H
co

©

pi

H

'a

©

•

4 H

«

0

i—i

•

•

4

•

O
o

O
q

•

#

•

ON o
on o

i—I i— I

*

o
o

•

O
q

i— I H

O
o

H

H

CM A
O H

CN
CM

A q A O
A A p p

A

OO p

A

A

•

•

A

•

P

•

•

no

LA

CM

A

w H
!h

A

rH
rH
A C O CO n P

rH O

P

ONCACO

s©
C
•H p
•
LA LA O P
O
O C rH s
COCOC— O O P
^ {
• *
• • •
P
£ MO p
L A C— CO
" d t©
H CM ( A C M ©

§?« 5
© <t|

• ^
«

A
H

C— CM

A
P

A c
CO p

o
C O A A
• • «
A n O CO
4 4 H

A

SS

CM CM

O

£

On A A c m - P A P
CM CM CM p
P
_P P

•

C— O n c A CM - p
fA P f

On A P
O -P O
A
CM i—I CM

• o
©

l a la

& ffl M

nO nO

© *

O

O

A

80

H
•
f t p
•
£ 12 t©
CO

O

Q
O

H

H

O

O

O

•

•

»

CM

A
H

o

A H CM CM CM O
LA i I H H

•

A
CM
•

CM A
H H

*

•

o

o

o

o
o

o
o

o

•

•

o

o

o
A

a

c o cm

A

P A C O

O o
o o
o

o
o
o

O

O

O H

H i—I

O

O

o

O

o

O

H

O
O
H

O

©
H

CO

j-

r—no oo

CM NO NO

A
p

I

I

I

I

I q 2 C?

hD© (J| S

a-

m
co

I 3 P ©

H

H 4 H

ft -=4

LA

I

I

W

A

no\o „

i—I i—I CM LO ,—1 |—| CM n O

'a
A

I
©

I
©

I
®

©

O

o

O

O H
A , A

H
A

H
A

H
A

H
A

CO

p
Ps
© r-j
- P CD
CD p

H

O

o
o
O

O

>-0

O

ttQ

ft -H
ft

O

o

P
H A A
O
CO A - CO ON O

o

O

•

0 00 0 00

o

•P
CO

O
O

oH A
H A H A
A A O A
i—Ii—I CM i—I

P
H O
i—I i—I I—I

O

CM

co

A A cm
A P co

O cm P
a
o
H On n O A - L A
r H O CM A L A
CM CM CM CM^ i—|
O

•

H

*

O

O N-P H
CO P
A

•

A

rH
LA

i—I A

S S 8

A
On A

o

CM A

w
H

vO A A
A A A

~P A

p A

,

o

o

§

On A

a

0 \ O
A p

l a

p

Cm

o

pq

CM A

A 4
P - P

m c m c o c m p a

A

On O

o
-P
CO

i— I i— I i— I

Z— O n H L A C
——
4 -P" O n
- p CM A A P P P CM

©

m

ft

p

•

CO A

E of

A P

0}

ft

•

O -P
o
^

o

•

P

©

i*

>

•

noLA

o o
Sft

o ft

o

•

a
a
o
o
™ o
o
h
p q q q o q o o
i—I H H i —I i—I H H H

ft
o PC

•

P

A
A
p
H
o
o

I of

P -P
CO ^

•©

CM A

© O

A
A CO
P P
H H
O O
o
q

Part

£

r l ©

A
o
A
H
O
o

S ta n d a rd iz a tio n

C

a p p e n d ix .

On g o A

th is

«

O

I

I

I

©

©

- P

I

— Table

«

H

values

O

of

C o
<4 o

A
A - CM A

A

CM A O H P A
H CM H CM CM H
Z—C
M
-A A P A

©
O
SH
4

£

301
PART K
ANALYSIS OF ELECTROLYTIC SAMPLES
As a means o f determ ining the com position o f th e s o l i d i f i e d le a d
cathodes from th e e l e c t r o l y s i s samples the fo llo w in g a n a ly t ic a l scheme was
u se d .

The o b j e c tiv e was to com p letely remove a l l le a d by appropriate

sep a ra tio n s in order to p reven t in te r fe r e n c e w ith subsequent b e r y lliu m ,
calcium and lith iu m a n a ly s e s .
A fte r removing a l l tr a c e s o f s o l i d i f i e d m elt by scrap in g the sam ples,
sh avin gs o f the cathode sam ples were prepared.

One gram samples were d i s ­

s o lv e d i n mixed n i t r i c and h yd ro ch lo ric a c id s a t steam bath tem perature and
were evaporated ju s t to dryness th ree tim es w ith subsequent a d d itio n o f
h yd roch loric a c id .

The r e sid u e was taken in to s o lu tio n in d ilu t e hyd roch loric

a c id and allow ed to slo w ly evaporate (under a watch g la s s ) on th e steam bath
u n t i l le a d c h lo r id e c r y s t a ls commenced to form „ The s o lu tio n was allow ed
to c o o l and th e major p o r tio n o f th e le a d was removed by f i l t r a t i o n , as le a d
c h lo r id e , u sin g c o ld d ilu t e h yd ro ch lo ric a c id a s a wash liq u id .

The le a d

c h lo r id e was r e ta in e d fo r subsequent com bination w ith the sm all amount o f
le a d rem aining in th e f i l t r a t e .
The f i l t r a t e was converted to th e n it r a t e form , was tak en to d ry n ess,
and s u f f i c i e n t n i t r i c a c id was added to b rin g the a c id co n cen tra tio n to
about 10 m l. o f d il u t e n i t r i c a c id per 100 m l. o f s o lu t io n .

T his s o lu tio n

was e le c t r o ly s e d a t low current d e n s ity w ith s t i r r i n g , u sin g platinum
e le c t r o d e s , to a n o d ic a lly d e p o s it Pb02 .
The s o lu t io n th a t remained was converted to the p e r ch lo ra te
form and (o m ittin g the flu o r id e decom position ste p ) b e r y lliu m , calcium
and lith iu m a n a ly se s were

302

perform ed.

Much sm a ller volumes were used in l i n e w ith the low v a lu es

a n tic ip a te d fo r the '’im p u r itie s 11 in th e lea d ca th o d e.

A sample o f the

o r ig in a l c h e m ic a lly pure le a d was used as a p a r a lle l c o n tr o l sam ple.
The d e p o s it o f Pb0 2 was converted to c h lo r id e form and was combined
w ith th e p r e v io u s ly separated PbCl2 to form a s o lu tio n which was taken
to d ryness to remove e x c e ss HC1.

The r esid u e was r e d is s o lv e d and the

le a d was p r e c ip ita te d and reco v ered as PbS) 4 u sin g Gooch f i l t e r i n g cru­
c ib le s .

The c r u c ib le and p r e c ip it a t e were ig n ite d a t 550° and weighed

as PbS04 .
The r e s u l t s are ta b u la te d below:

Samples

O r ig .
Lead

"A fter
1EL-1"

Sample w e ig h t, g .
1.0062
1.0012
V o l. KMrD4 , m l.
< 0 .0 5
< 0 ,0 5
< 0.0003
Wt. BeO, g .
< 0 .0 0 0 3
0 ,1 0 avg . 0 .1 5 avg.
V ol. Na2S203 , ml."
<0 .0 2
< 0 .0 2
W t. % Ca
W t. % Be
< 0 .0 3
< 0 .0 3
¥ t . % Li
o. ok
0 .0 3
Wt Xbl. t PbS04
16.6198
19.290k
Wt. MT Xbl.
15.1668
11.199k
Wt. PbS04
1.6550
1 .6510
9 8 .8
Wt. % Pb
99.3
N otes:
^

"A fter
2EL-1”

"After
6EL-1"

1 .OO36
< 0 .0 $
< 0 .0 0 0 3
0.1k avg.
< 0 .0 2
< 0 .0 3
O.Oli
18 .8366
17.3833
1.1533
99.0

"After
$EL-1"

1 .0021+
1.0033
< 0 .0 $
< 0 .0 $
< 0 .0 0 0 3
< 0.0003
0 . 1$ avg . 0 .10 a v g .
< 0 .0 2
< 0 .0 2
< 0 .0 3
< 0 .0 3
0 .0 3
0 .0 6
19.21+20
15.9969
1 7.7916
16 .$603
1 .6 5 6 6
1 . 1501;
9 8 .8
9 9.3

S o lu tio n s — KMn04 = 0.01012 g . CaFo/ m l .
Na2S20 3 = 0 .0 2 0 8 3 g . LiF/m l •
10 m l. a liq u o ts o f 100 m l. volume.

In view o f the f a c t th a t th e a n a ly t ic a l samples d id n o t d if f e r s i g ­
n i f i c a n t l y from th e c o n tr o l sam ple, i t was concluded th a t no ap p reciab le
Ca, Be or L i con ten t was p r e se n t in the sam ples.

The o v e r - a ll reco v ery

303

was rea so n a b ly h igh c o n sid e r in g a l l th e m a n ip u la tio n s.

Other sam ples ,

c o n ta in in g th e th ree elem ent ^3 which were run a t the same tim e gave s a t i s ­
fa c to r y r e s u l t s .

APPENDIX 16

30k

APPENDIX 16
CONDUCTANCE MEASUREMENTS

Rough c o n d u c tiv ity d eterm in a tio n s were made on s e v e r a l tern a ry m elt
m ixtu res p r io r to e l e c t r o l y t i c work.

P rim a r ily t h is in fo rm a tio n was de­

s ir e d o n ly in order th a t th e cause o f any high c e l l r e s is ta n c e during
e l e c t r o l y s i s cou ld be p ro p erly a ttr ib u te d to e ith e r a conductance e f f e c t
or a r e s u lt o f e le c tr o d e p o la r iz a t io n .

The equipment and gen eral accu r­

acy have been d is c u s se d in the s e c tio n d e a lin g w ith th e e l e c t r o l y t i c
study o f th e te r n a r y sy stem .
In th e r e s u lt s th a t fo llo w showing r e s is ta n c e o f the c e l l as a fu n c­
t io n o f tem perature and co m p o sitio n , th ere i s some obvious discordance
in th a t rep ro d u cib le r e s u l t s were not obtained as a m elt was heated and
c o o le d .

T h is may have been the r e s u lt o f unknown film r e s is ta n c e on the

e le c t r o d e s , or c o n c e iv a b ly could be due to a change in the m olecular com­
p l e x i t y o f the m elt in a tim e dependent manner.
The r e s u lt s ta b u la te d show th e r e s is ta n c e o f p a ir s o f m e lts d if f e r in g
in com p osition by o n ly a s l i g h t amount o f added CaF2 .

Some o f th e v a lu es

r e f e r to th e s o l i d form , as the m e lts were heated and co o led ( in some
c a se s ) through th e m e ltin g p o in t .

305

M elt 1 EL
R
e
s
i
s
t ,,
Temp.,
Temp., R e s is t
ohms
°c.
ohms
°c.
1+30
hb3
. 1+1+8
1+51
1+53
1+51+
U55
1+55
1+56
1+56
1+57
1+63

3 6.0
23 .0
1 6 .0
1 3 .5
9 .0
7 .0
5 .8
5 .1
1+.2
1+.0
3 .8
3 .0

1+69
1+76
1+86
505
509
1+82
1+65
1+55
1+1+1+
1+33
1+28

M elt 2EL
(H eated to 600° and
a llow ed to c o o l)
R e s is t
Temp%y
ohms
°c.
600
510
1+70
1+1+8
1+1+0
1+35

0.3
0.5
0 .5
0 .6
0 .7
0 .8

M elt 1EL-1
Temp. ,
R e s is t.
°c.
obros

2 .0
1 .0
0 .1+
0 . 1+
0 .1+
0 . 1+
0 .5
o .5
0 .6
1 .5
5 .0

1+50
1+56
1+63
1+72
1+81
I+90
502
521
531+

3 .6
3 .0
2 . 1+
1 .7
1 .1
0 .5
0 .1+
0 .3
0 .3

M elt 2EL-1
Temp. } R e s i s t . ,
ohms
°c.
1+52
1+73
1+90
506
52o
560
580
580
560
550
51+0

9 .0
6 .5
5.1+
3 .8
2 .8
2 .2
1 .8
1 .6
1 .7
1 .6
1 .8

Temp. , R e s is t
°C .
ohms
520
1+89
1+66
1+1+6
1+35
1+30
1+20
1+10
1+00
390
385
375

Continued n ex t page

2 .1
2 .2
2 .5
2 .8
3 .2
3 .9
5 .5
7 .8
1 2 .0
2 7 .5
5 7 .0
1 1 2 .0

3 0 6

M elt 3 EL
Temp. ,
R e s is t. ,
°C .
ohms

M elt 3EL-1
Temp.,
R e s is t.,
°C .
ohms

565
590
521
510
595
552
503
3 20

335
350
375
520
575
5io
580
530
365

3.5
2 .8
2 .3
2.2
2.5
2 .5
2 .6
2 2 .0

5 .0
3.7
5.5
1 .5
0 .9
0 .7
0..8
1 .5
1 .6

337

LITERATURE CITED
R e f e r e n c e h a s b e e n in c lu d e d i n c e r t a i n c a s e s t o m ore
a c c e s s i b l e l i t e r a t u r e , w here t h e o r i g i n a l r e f e r e n c e i s
u n a v a i l a b l e l o c a l l y , o r to C h em ical A b s t r a c t s w here
la n g u a g e a l s o p r e s e n t s a d i f f i c u l t y .

1 . A cker, German P a t . 1 1 0 , 558 (1 8 9 8 ).
2 . A ndersen,
3 . A rn d t,

(S ee E n gelhardt, p. 1 5 8 , r e f . 50) .

0 . and Bowen, N. L . , Z . anorg. Chem., 8 7 , 283 (1 9 1 5 ).

K. a n d W illn e r , K ., Z . E lek tro ch em ., 1 5 , 21 6 (1908) .

5 . B adger, A. E. and P ittm an , W. C ., Ceram. I n d ., 28, 218 (1937) .
5 . B a lc e , S . , U n iv . P h ilip p in e s N atural and A pplied S c i . B u l l . , 5, 119-21
(193k) . (S ee C. A ., 28, 7135® [1 9 3 5 ]) .
6 . B a n c r o ft, W. D ., '’A p p lied C o llo id C hem istry,” 1 s t . E d i t . , McGraw-Hill
Book C o .,I n c ., New York, 1921, p . 275.
7 . Bergman, A . G. and Bukhalova, G. A ., Zhur. Obshei Khim ., 1 9 , 603-11
(1 9 5 9 ); J . Gen. Chem. USSR, 1 £ , 553-62 (1 9 5 9 ). (See C. A ., 5 5 ,
9293a [ 1 9 5 0 ] ) .
8 . B ooth ,

H. S . and T orrey, G. G ., J . P hys. Chem., 3 5 , 2565-77 (1 9 3 1 ).

9 . B race,
P . H ., Chem. M et. E n g ., 2 5 , 105 (1 9 2 1 ); Trans
chem. S o c ., 3 7 , 565 (1 9 2 0 ).
1 0. B re to n , E. J . andWaggaman, W. H ., U. S . P at

Amer.E le c tr o -

2 ,5 1 0 ,0 5 3 (1 9 5 6 ).

1 1. Brew er, L ., "Chemistry and M etallu rgy o f M iscella n eo u s M a te r ia ls ,
Thermo dynamics," Paper No. 7 , McGraw-Hill Book Co. ,In c , New York,
1950.
1 2 . Brockman, C. J . , " E le ctro c h e m istr y , P r in c ip le s and P r a c tic e ,"
D. Van N ostrand C o ., I n c . , New York, 1931, p . 265.
13

Bukhalova, G. A . and Bergman, A. G ., D okladyA kad. Nauk SSSR, 6 6 ,
67-70 (1 9 5 9 ). (S ee C. A ., 5 5 , 9235a [1 9 5 0 ] ).

1 5 . Bunn, C . ¥ . , "Chemical C ry sta llo g ra p h y ," Oxford P r e s s , 1 9 5 5 , pp. 379-382.
15

B usk, R. S . , Leman, D. L . and C asey, J . U ., J . M e ta ls, 1 8 8 , $ 7 , T rans.
955-51 (1 9 5 0 ).

308

1 6 . Chipman, J . , D isc u ssio n s Faraday S o c.,1 9 H 8 , # ii, 2 3 -2 9 .
1 3 , to O h [1 9 lt9 ]).

(S ee C. A .,

1 7 . Compagnie de p r o d u its chim iques e t e le c tr o m e ta llu r g iq u e s A la is ,
Froges e t Camargue, Germ. P a t. 7 la ,l;7 3 (19H-3); B r i t . P a t.
582,0110 (19U 6). (S ee C . A ., HO, 31H8 [ 19I46] and lj l , 1199f [19li7]) .
18 . C u b ic io t t i, D. D ., U. S . Atomic Energy Commission, MDDC 1058 (19H 7).
1 9 . C u b ic io t t i, D. D. and Thurmond, C. D ., J . Amer, Chem. S o c . , 7 1 , 21H9
(1 9 U 9 ).
~
2 0 . Damiens, A ., '’T r a ite de Chemie M in era le, 11 V o l. I , Masson e t C ie ,
P a r is , 1931, p p . laii-U 2it.
2 1 . D a n n e ll, H ., and Stockem , L ., Z . E lektrochem . , 1 1 , 209 (1 9 0 5 )j 3 £ ,
lk $ (1 9 2 U ).
22 . DeBoer, J . H. and F a s t , J . D ., Z . anorg
(1930) .

a llg em . Chem., 1 8 7 , 177-89

23. D e lim a r s k ii, I . E . and K o l o t t i, A. A ., Zhur. F i z . Khim., 2 3 , H37-HO
(19U9) . (See C. A ., U3, 6519* [19U9]) .
2 l|. D e lim a r s k ii, Y . K. and K o l o t t i, A. A ., I b i d . , 2 3 , 339—U1 (19H 9).
(S ee C. A ., 1 3 , 6Q89f [1 9 U 9 ]).
2 5 . D e lim a r s k ii, I . K. and S k o b e ts, E. M ., J . P h y s. Chem. USSR, 2 0 ,
1005-10 (1924-6) . (S ee C. A ., U l, 23Hla [1 9 H 7 ]).
2 6 . Domange, L ., Ann,chim. J , 225-97 (1 9 3 7 ).
2 7 . D rossbach, P . ,
"Elektrochem ie geschm olzener S a lz e ," J u liu s S p rin g er,
B e r lin , 1938. (R eprinted by EdwardsB r o s ,, Ann A rbor, 19H 3).
2 8 . D rossbach, P . , I b id . , pp. 1 - 2 , 2 7 -2 8 .
2 9 . D rossbach, P . , Z . E lek troch em ., 3 6 ,

179 (1 9 3 0 ).

3 0 . D rossbach, P . ,

I b i d . , H3, 895 (1937) .

3 1 . D rossbach, P . ,

p . 27 o f t e x t , r e f . 2 7 .

3 2 . D rossbach, P . , Z . E lek tro ch em ., 3 3 ,

111* (1 9 2 7 ).

3 3 . D rossbach, P . , p . 119 o f t e x t , r e f .
r e f . HO) .

27.

(See a lso E n gelh ard t, p . 1 3 6 ,

3U. D rossbach, P . , Z . E lek troch em ., 55, 3 8 -h l (1951) .

309

3 5. D rossbach, P . , p p . 1 1 8 , 1 1 9 , 139 o f t e x t , r e f . 2 7 .
3 6 . Durganov, E . P . and Bergman, A. G ., J . P h y s. Chem. USSR, 22 , 625-32

3 7 . Durganov, E . P . and Bergman, A. G ., D okladyA kad. Nauk SSSR, 7 5 ,
815-18 (1 9 5 0 ).
~
3 8 . D eutsche Gold- und S ilb e r -S c h e id e a n s ta lt vormals R o e s s le r , French
P a t . 822,225 (1 9 3 7 ) . (See C. A ., 3 2 , 52903 [1 9 3 8 ] ).
3 9 . Emley, E . F . , U. S . P a t s . 2 ,5 5 2 ,9 1 5 (1958) and 2 ,5 5 2 ,8 9 5 (1 9 5 8 ).
(S ee C. A ., 5 3 , 1308a,b [1 9 5 9 ] ).
5 0 . E n g elh a rd t, V ., "Handbuch der Technischen E l e k t r o c h e m i e V o l . I l l ,
Akadem. V e r la g ., L e ip z ig , 1935. (R eprinted by Edwards B r o s .,
Ann A rbor, 1 9 5 5 .) .
5 1 . E n gelh ard t, V . , I b i d . , p . 187.
5 2 . E n gelh ard t, V ., I b i d . , pp . 221, 225.
5 3 . E n gelh ard t, V ., I b i d . , p . 178.
5 5 . E n gelh ard t, V . , I b i d . , p . 550.
5 5 . E n g elh a rd t, V ., I b i d . , p . 156 f f .
56. E n gelh ard t, V ., I b i d . , pp . 158, 1 5 3 -1 5 5 .
5 7 . Feldmann, Germ. P a t s . 5 9 ,9 1 5 ( 1887) and 50,370 (1 8 8 8 ).
(S ee E n g elh a rd t, p . 2 2 1 , r e f . 5 0 ).
5 8 . F ra ry , F . C . and B adger, ¥ . L . , T ra n s. Amer. E lectrochem . Soc . ,
1 6 , 185 (1 9 0 9 ) .
5 9 . Fremy, E ., A nn.chim . p h y s ., 5 7 , [ 3 ] , 5 (1 8 5 6 ).
r e f . 20) .

(See Damiens, p . 5 1 6 ,

5 0 . F u sey a , G ., M ori, M. and Imamura, H ., J . S oc. Chem.
In d . Japan, 3 6 ,
S u ppl. b in d in g 17 5 -6 (1 9 3 3 ). (S ee C. A ., 2 7 , 366O6 [1 9 3 3 ]) .
5 1 . G ibbs, J . ¥ . , T rans. Conn. A ca d ., 3 , 108-258 ( 1876)$ 353-525 ( 1878) .
See “S c i e n t i f i c Papers o f J . W illa rd Gibbs," V o l. I , LongmansGreen and C o ., New York, 1906.
5 2 . G oldschm idt, V. M ., Geochemische V e r te ilu n g sg e se tz e der E lem ente,
7 and 8 , O s lo , 1926 and 1927.

310

5 3 . Gore , G ., P h i l . T r a n s ., 1 6 0 , 227 (1870 ); 1 6 1 , 321 (1871)
(See Dam iens, p . &L6 , r e f . 20) .
51.

Gromakov,S , D. , J . P h ys. Chem. USSR, 2 k, 61|1 -9 (1950) .

55.

Grube, G. and Henne, H .aZ . E lek tro ch em ., 3 6 , 129-138 (1 9 3 0 ).

5 6 . Grube, G. and J a i s l e , J . , Z . E lek troch em ., 3 3 , U8l (1 9 2 8 ).
5 7 . Grube, G ., Z. E lek tro ch em ., 3 3 , I+83 ( 1927) .
58.

Grube, G. and Rftdel, W ., %. anorg. a llg em . Chem., 233, 375-88 (1921;).

59.

G u e r tle r , W ., M e ta ll und E r z , _8 , 192-5 (1 9 2 0 ).

60. Guntz, Comptes R en d ., 1 1 7 , 732 (1893) .
r e f . I4O) .

(S ee E ngelhardt, p . 122.

61. H a ll, C. M ., U. S . P a t. 1;00,766 (1 8 8 6 ).
6 2 . Hampel, C. A ., Chem. Eng. News, 2 7 , 2l;20 (19U9) .
6 3 . H anaw alt, J . D ., R inn , H. ¥ . and F r e v e l, L . K ., In d . and Eng. Chem.,
A n a ly t ic a l E d itio n , 1 0 , I48O (1938) .
61;. H anawalt, J . D ., R inn, H . ¥ . and F r e v e l, I . K ., I b i d . , p . 1;90 .
65. H a n ffs te n g e l, K. and Hanemann, H ., Z . M etallkunde, 3 0 , 50-1 (1938) .
(S ee C . A ., 3 2 , 70019 [1 9 3 8 ]) .
6 6 . Harvey, W. G ., Chem. and M e ta ll. E n g ., 3 2 , 5 7 3 -6 (1925) .
6 7 . Haven, Y ., R e c . t r a v . c h im ., 69, 1505-18 (1950) .
A b s tr a c ts , $ 3 #1 $, 3883 X T 951]).

(See N uclear S cien ce

6 8 . Heyne, G ., Angew. Chem., b6 , l;73-77 (1 9 3 3 ).
69. H ig n e t t, T . P . and S i e g e l , M. R ., In d . Eng. Chem., U l, 2U93-8 (19U9) .
7 0. H ille b r a n d , ¥ . F . and L u n d e ll, G. E . F . , "Applied In organ ic A n a ly sis,"
John W iley and S on s, New York, 1929, p . 116.
71. H ille b r a n d , W. F . and L u n d e ll, G. E . F . , I b i d . , p . 623.
72. Hodge

73

W ., J o f f e e , R. I . and Gonser, B . W.j Rand Corp. Report R-123
( B a t t e lle Mem. I n s t . ) . (S ee N uclear S cien ce A b str a cts, 3, 80-1
[19U9] and C. A ., UL;, l8 7 5 d [1 9 5 0 ] ).

H onigsehm id, 0 . , and Johannsen, T ., Z. N a tu r fo r sc h g ., 1 , 650 (19U6) .
(See C. A ., h1 , h3hS f [1 9 U 7 ]).

311

7^. I I * in s k i (H ym sk i) , V. P . and A n tip in e , P . F . , T rav. i n s t . ru sse
chim . a p p l., 1 9 2 5 , #81+, 3 -8 ; Chemie e t I n d u s t r ie , 1 7 , 601
(1 9 2 7 )j Ann. i n s t . a n a l. p h y s. c h im ., 3 , 1+61+. (19267; (See C. A .,
2 1 , 21+369, 3801s [1 9 2 7 ]) .
7 5 . I l l i g , K ., H o s e n fie ld , M. and F is c h e r , H ., ¥ i s s . V e r o f f e n t lic h .
Siem ens-Konzern, 8 [ 1 ] , 30-kl (1 9 2 9 ). (See C . A
21+, 23799
[1 9 3 0 ]).
—
7 6 . J a e g e r , G ., U. S . P a t. 2 ,1 2 5 ,0 2 6 (1 9 3 8 ).
77. J e llin e k , K. and C zerw in sk i, J . , Z . p h y sik . Chem'., 1 1 0 , 192 (1921+).
7 8 . Kausch, 0 . , " F lu ssa u re, k ie s e lf lu s s a u r e und deren M e ta lls a ltz e ,"
F . Enke, S t u t t g a r t , 1 9 3 6 , p . 95.
79. K e lle y , K. K ., "C ontrib ution s to th e Data on T h e o re tic a l M eta llu rg y ,
X ,” B u ll, 1+76, Bur. o f M in es, U . S . D ep t, o f I n t e r io r , 191+9.
8 0 . K je llg r e n , B ., U. S . P a t. 2 ,3 8 1 ,2 9 1 (191+5).
[191+5]).

(See C. A ., 3 9 , h & l f

81. Klemm, ¥ . , "FIAT Review o f German S c ie n c e , Inorganic C hem istry,"
P art V, 191+8, p . 196.
82 . Klemm, ¥ . , B e y e r sd o r fe r , K. and O ryschkew itsch, J . , Z . anorg. Chem.,
2 5 6 , 25-36 (191+8) .
8 3 . K o n ish i, Y ., J . S o c . Chem. In d . J a p a n ., 3 9 , Su ppl. b in d in g 209-11
(1936) . (S ee C . A ., 3 0 , 6275 [193^ !) 81;. Kreemann, R. and P r e ssfr eu n d , E . , Z . M etallkunde, 1 3 , 19-29 (1 9 2 1 ).
8 5 . Kremann, R ., F r it s c h , J . and R Ie b le , R ., Z . M etallkunde, 1 3 , 66-73
(1 9 2 1 ).
86.

Kremann, R. , and

Mehr, A . , Z . M etallku nde, 1 2 , 1+1+1+-55

87.

Kryagova, A . I . ,

J . Gen. Chem. USSR, 1 7 , 23-6 (191+7) .

88.

Kubaschewski, 0 .

and W a lter, A ., Z . E lek tro ch em ., 1+5,

89.

Kubaschewski, 0 .

and V i l l a , H„, Z . E lek troch em ., 5 3 , 32-1+0 (191+9).

90

(1920) .

732-1+0 (1 9 3 9 ).

K u rzn iec, E ., B u l l , in t e r n , acad. p o lo n a is e , 1931&, 3 1 -5 8 .
(S ee C. A ., 2 5 , 1+7719 [1931]) .

9 1 . Lange, N. A ., "Handbook o f C h em istry,” 7th E d itio n , Handbook P u b .,
I n c . , Sandusky, O hio, 191+9, p . 853.

312

9 2 . L atim er, ¥ . M. and H ildebrand, J . H ., "Reference Book o f Inorganic
C hem istry," R evised E d i t . , M acmillan C o ., New York, 19k0 ,
PP. 3 7 -3 9 .
9 3 . L atim er, W. M. and H ild eb ran d, J . H ., I b id . , p . 9 8 .

9h. Lebeau, P . and Damiens, A . , Comptes R end., 1 8 1 , 917 (1 9 2 5 ).
9 5 . Lebteau, P . and Damiens, A . ,
B u ll. s o c . encour. in d . n a t . , 1 3 9 , 15
(1 9 2 7 ) . (See Dam iens, p . !;22, r e f . 20) .
9 6 . Lebeau, P . , T h e sis (1 8 9 8 ). See M e llo r , "Comprehensive T r e a tis e on
In organ ic and T h e o r e tic a l C hem istry," V o l. IV , Longmans-Green
and C o ., London, 1923, p . 229.
9 7 . LeBoucher, L , , A nales s o c . esp an . f i s . qu im ., 3 2 , 11U2-56 (193U)
(See C . A ., 2 £ , 32086 [1935]) .
98. L in n e ll, R. H. and H aen d ler, H. M ., J . P hys. and C o llo id . Chem., 52
819-27 (1 9 U 8 ).
9 9 . L orenz, R ., "Die E le c tr o ly s e geschm olzener S a lz e ," (1905-1906) and
w ith F . K a u fle r , "E lectrochem ie geschm olzener Salze" (1 9 0 9 ).
(S ee E n g elh a rd t, r e f . UO).
100. L orenz, R ., and E i t e l , ¥ . , Z . anorg. Chem., 91,
101. L oren z, R ., Z . E lek tro ch em .,

(1 9 1 5 ).

277 (1901) .

102. L oren z, R .,
von H evesy, G ., and W o lff, E . , Z . p h y sik . Chem.,7 6 ,
732 (1 9 1 1 ) .
103. L u n d e ll, G. E . F . , Hoffman, J . I . and B r ig h t, H. A ., "Chemical
A n a ly sis o f Iro n and S t e e l," John W iley and Sons, New Y ork,
1 9 3 1 , PP. 5 6 2 -7 2 .
10U. Marden, J . W. and R ic h , M. N ., In d . Eng. Chem., 1 2 , 651-6 (1920) .
1 05. Marsh, J . S . , " P r in c ip le s o f Phase Diagrams," McGraw-Hill Book C o ., Inc
New York, 1 9 3 5 .
106. M arsh, J . S . , I b i d . , p . 1 5 7 .
107. M a r tin e tte ,

S is t e r Mary, J . Chem. E d ., 2 6 , 101-103 (1 9 ^ 9 ).

1 08. M ashovetz, V. P . and Lundina, Z . F . , Proc . F i r s t A ll-U n io n C onf. on
Nonaqueous S o lu tio n s , 1 9 3 5 , pp . 1 9 1 -2 1 2 . (See C. A ., 3 0 , U0711
[1 9 3 6 ]) .

313

1 0 9 . M asing, G ., "Ternare System e," Akadem. V e r la g ., L e ip z ig , 1933.
1 1 1 . M e llo r , J . ¥ . , '‘Comprehensive T r e a tis e on Inorganic and T h e o r e tic a l
C hem istry," Longmans-Green and C o ., London, 1923.
1 1 2 . M e t a llg e s . A . G B r i t . P a t. 3 1 8 ,5 9 9 (1928) .
20639 [ 1 9 3 0 ] ) .

(See C. A ., 2h,
—

1 1 3 . M oissan , H.^ Comptes R en d ., 1 0 2 , l%b3 (1886) .(See Damiens, pp.
t o f f , r e f . 20) .
“
ll U . M oldenhauer, W. and A ndersen, J . , Z . E lek troch em ., 1 9 , W ; (1913) .
115. Muthmann, ¥ . , H o fe r , H. and W e iss, L . , A n n ., 320 , 237 ( 1901) .
116. N a y lo r , B . F . , J . Amer. Chem. S o c ., 67, 150-2 (1 9 U 5 ).
1 17. Neumann, B ., and B ergve,

E . , Z . E lek tro ch em ., 2 0 , 187 (191U) .

118. Neumann, B. and R ic h te r , H ., Z. E lek troch em ., 3 1 , M31 (1 9 2 5 ).
119. Newman, A . B . and Brown, G. G ., In d . Eng. Chem., 2 2 , 995 (1 9 3 0 ).
120. N ov o selo v a , A . V ., L ev in a , M. E . , Simanov, Y. P . and Zhasmin, A . G .,
J . Gen. Chem. USSR, li± , 3 8 5 4 0 2 (19-UU) .
1 21. O 'D a n iel, H. and T s c h e is c h w ili, N a tu r w isse n -sc h a ften , 3 1 , 209-10
( 19)+3) . (A lso 0 * D a n ie l, Zement, 19^41, H eft 1^.0, p . 1 .)
(S ee T h ilo and Lehmann, r e f . 166) .
1 2 2 . Oshorg, H ., "Monograph-Lithi'um, T h e o r e tic a l S tu d ie s and P r a c t ic a l
A p p lic a tio n s ," The E lectro ch em ica l S o c ., New York, 1 9 3 5 , p . lU .
1 2 3 . Osborg, H ., I b i d . , p . 65.

12k. Osborn, E . F . , A b s tr a c t s , 1 1 8 th . Amer. Chem. S o c. M eetin g, p . 27L,
S e p t ., 1950.
1 2 5 . Osborn, E . F . e t a l . , J . Amer. Ceramic S o c ., 3 3 , 1 5 2 -9 , 160-7
(1950) .
126. P a r so n s, C . L . , Z . anorg. Chem., l £ , UOO-22 (19014.
J . Amer. Chem. S o c ., 2 8 , 555 [ 1 9 0 6 ] ) .

(See a ls o

1 27. R athenau, Z . E le ctr o c h e m ., 1 0 , 508 (190U) .
1 2 8 . R ay, N . N ., Z . anorg. a llg e m . Chemie, 2 0 1 , 289-300 (1931).
1 2 9 . R ay, N. N ., I b i d . , 205, 257-67 (19321.

314

1 3 0 . R ogers, B . A . , T ra n sla tio n o f ”Ternary System s” by G. M asing,
R einhold Pub. C o ., New York (1 9 4 4 ).
1 3 1 . R ogers, B . A . , I b i d . , p . 57.
1 3 2 . R ogers, I . B . and Ca3.ey, E . R
1 5 , 209-11 ( 1943) .

In d . Eng. Chem.. A n a ly tic a l E d itio n ,

133. Roozeboom, Schreinem akers, A ten and Buchner, D ie Heterogenen
G leich gew ich te vom Standpunkte der P h a sen leh re, 3 V o ls .,
F r ie d r ic h V iew eg. und Sohne, Braunschweig (1 9 0 1 -1 9 1 8 ).
(D ie Ternaren G le ic h g e w ich te , V o l. I l l , 1911 or 1913.)
1 3 4 . Roy, D. M ., Roy, R. and Osborn, E. F . , J . Amer. Ceramic S o c .,
3 3 , 85-90 ( 1950) .
1 3 5 . R u ff, 0 . and Hartmann, H ., Z.

anorg. a llg em . Chem., 1 3 3 , 29 (1 9 2 4 ).

1 3 6 . R u ff, 0 . and B r in tz in g e r , H .,

I b i d . , 1 2 9 , 267-75 (1 9 2 3 ).

1 3 7 . R u ff, 0 . and LeBoucher, I . , I b i d . , 2 1 9 , 376-81 (1 9 3 4 ).
1 3 8 . Rumpf, E . , Z . p h y sik . Chem., A b t. B , r{_s 11^8-54 (1930) .
1 3 9 . S a k a i, K ., J . E lectrochem . Soc. Japan, 1 8 , 1 2 -1 4 (1 9 5 0 ).
1 4 0 . S ark er, P . and Ray, N. N ., J . In d ian Chem. S o c ., 6 , 987-90 (1 9 2 9 ).
1 4 1 . S a r k iso v , E . S . , D okladyA kad. Nauk SSSR, 6 0 , 37 1 -4 (1 9 4 8 ).
(S ee C. A ., 4 2 , 71l4d [ 1 9 4 8 ] ) .
1 4 2 . Schm idt, J . M ., B u l l . S o c . Chim. F rance, 3 9 , 1686-1703 (1 9 2 6 ).
1 4 3 . Schmitz-Dumont, 0 . and E ., Z . anorg. Chem., 2 5 2 , 329 (1 9 4 4 ).
144. Schumb, W. C ., e t a l . , In d . Eng. Chem., 3 9 , 244-288 (1 9 4 7 ).
1 45. S e tte r b e r g , C ., A n n ., 2 1 1 , 100-116 (1 8 8 2 ).
146

Z ech er, G ., C la u s in g , P . and deBoer, J . H ., Z. a n o rg . a llg em . Chem.,
1 6 0 , 128 (1927) .

1 4 8 . S c h p ic h in e t s k ii, E . S . and P o g o d in , S . A ., I s v e s t . S ektora F i z . Khim A n a l., I n s t . O bshchei i N eorg. Khim., Akad. Nauk SSSR,
1 5 , 8 8 -9 5 , 96-111 (1 9 4 7 ) . (See C. A ., 4 4 , 5 7 8 3 f, 57831 [ 1 9 5 0 ] ) .
1 49. S i l b e r , P . and Ish aq u e, M ., Comptes r e n d ., 2 3 2 , 1485-7 (1 9 5 1 ).

315

15*3* Sim ons, J . H. e t a l . , "F luorine C hem istry," V o l. I , Academic P r e s s ,
New York, 1950.
1 5 1 . Sim ons, J . H. e t a l . , I b i d . , p . 3 0 2 .
1 5 2 . Sim ons, J . H ., J . E lectrochem . S o c ., £ 5 , # 2 , U7-6? ( 19)49) .
1 5 3 . Sm ith, D. P . , Z. anorg. Chem., 5 6 , 109-142 (1908) .
15U. Ssyrom jatnikow , R. R ., Chem. Z e n t r a lb la t t , 1 9 3 2 , I I , 3 6 1 5 .
1 5 5 . S tock b arger, D. C. and B lanchard, A. A ., U. S .- P a t . 2 , 1498,186
(1 9 5 0 )j (S ee C. A .,
[£ 1 0 i [1 9 5 0 ] ).
1 5 6 . S tock b arger, D. C ., D is c u ss io n s Faraday S o c ie t y , 19h9 , #5* 29U—9 .
1 57. S tock b arger, D. C ., D is c u ss io n s Faraday S o c ie ty , 19149, # 5 , 299-306.
1 5 8 . S tr u k tu r b e r ic h t, Z . K r i s t . , V o l. I I , 1 9 3 7 , p . 2I48 and V o l. I l l ,
19313 p . 27I4; Akad. V e r la g ., L e ip z ig .
1 5 9 . S tr u k tu r b e r ic h t, I b i d . , V o l. I , 1 9 3 1 , PP. 7 3 , 9 7 -1 0 0 , I 8I45 V o l. IV,
1 9 3 8 , p . 91; V o l. V , 19I4O, p . 37; V o l. V I I , 19U3, P . 6 7 .
1 60. T ab les A n n u elles de C on stan tes e t
Donnees Numeriques, H , P a rt 2 ,
# 2 0 , G a u th ie r -V illa r s , P a r i s , 1937, p . U 8-19, p . I48- 2 5 ,r e f . 1+9.
161 • I b i d -» # 2 8 * P- 7 1 - 9 , p . 7 1 -3 3 , r e f . [488.
1 6 2 . I b i d . , # 2 8 , p . 7 1 - 9 , p . 7 1 -3 1 , r e f . 532 to E . R inck, Comptes. R en d .,
1 9 2 , 1378 (1 9 3 1 ).
1 6 3 . I b i d . , # 2 8 , p . 7 1 -9 , p . 7 1 -3 1 , r e f . 1+88.
I6I4. T a c c h in i, G ., G azz. chim. i t a l . , 5)4, 777-80 ( 192I4) .

1 6 5 . T a y lo r , C. S . , T rans. Amer. E lectrochem . S o c ., U7, P r e p r in t II4 (1 9 2 5 ).
1 6 6 . T h ilo , E . and Lehmann, H. A ., Z . anorg. Chem.., 2 5 8 , 332-55 (1 9 ^ 9 ).
1 67. T h ilo , E . and Schroder, H ., Z . p h y sik . Chem., 1 9 7 , 39-62 (1951) .
1 6 8 . Thomson, J . J . — See E n gelh ard t, p . 298, r e f . I4O.
1 6 9 . Todd, S . S . , J . Amer. Chem. S o c ., 7 1 , h l l ^ - 1 6 (19)49).
170

Trautz . M ., Chem. Z eitu n g , 3 5 , 1096 (1911) and Z. E lek tro ch em ., 2 1 ,
130 (1 9 1 5 )• ( See E n g elh a rd t, p . 1 3 1 , r e f . U o ).

316

1 7 1 . Val'dm an, G ., M e t a llu r g y 1 2 , 85-92 (1 9 3 7 ).
[1 9 3 8 ]).
“

(See C. A . 3 2 , 69522

1 ? 2 . V e n tu r e llo , G .^ A tti. acad. s c i . T o rin o , C la sse s c i . f i s . m at. n a t . ,
7 6 , I , 556-63 (191*1). (See C. A . 3 7 , 29827 [191*3]).
1 7 3 . W arren, B . E . and H i l l , C . F . , Z . K r i s t . , 8 9 , 1*81-6 (1931*) .
17l*. W ohler, L . and Rodewald, G ., Z. anorg. Chem., 6 1 , 51* (1909) .
175. Z e d le r , R. E . and Luman, R. I . , U . S . Atom ic.Energy Commission,
MDDC-1155 (191*7) .
1 7 6 . Z i n t l , E . and Haucke, ¥ . , Z . E lek tro ch em ., k, 10 lj.-ll (1 9 3 8 ).
1 7 7 . Z i n t l , E . and Udgard, A ., Z . anorg. a llg em . Chem., 2l*0, 150-6 (1 9 3 9 ).

CaBeF,

7 oq

QO

•zz
MOLE % BeF2

>-

FIGURE 12
BeR,- RICH REGION

BeR

CaR

FIGURE

15

TERNARY SYSTEM
CaF.
B eE

CaBeF,

(Oo

600

Oo
Oq a

\<?
375'

_i F

MOLE PERCENT BeF.

BeR

600

<Oo

80 o

LiF

MOLE % BeF,

B eF
FIGURE

L i F —RICH REGION

FIGURE 13
- R I C H REGION