112
623

 

THE MASS SPECTROMETRIC
ENVESTEGATION OF NEODYMIUM
MONO-THIO OXEDE

Thesis {or ”m Degree of M. 3.
MECE'EGAN STATE UNWEKSETY
Rebecca Moorhead Jacobs
1965

MICHIGAN STATE UNIVERSITY
DEQARTMESW .3 E ..

WHESQS

ABSTRACT

THE MASS SPECTROMETRIC INVESTIGATION
OF NEODYMIUM MONO-THIO OXIDE

by Rebecca Moorhead Jacobs

This investigation concerns the mass spectrometric vaporization
behavior of neodymium mono—thio oxide, NdZOZS. Over the temperature
range 20910 to 24650 K the vaporization behavior, the enthalpy and the
entropy are compared with those observed previously for neodymium
sesquioxide.

Neodymium mono-thio oxide vaporizes in this temperature region

according to the equation:
The vapor pressure of NdO can be expressed by the following linear

empirical equation fitted to the data by the method of least squares.

1n de0 .—. (-6. 598 .+_ 0.097)104/T + (17. 306 i 0.430)

From the second law of thermodynamics the following values were

obtained for the above reaction:
AH?98 = 209. 0 i 4. 0 kcal./mole
A5398 : 61.1 i 2.2 kcal. /mole

Values for the free energy functions of NdO (g) and NdZOZS (c) were

estimated and combined with published data for sulfur in order to

Rebecca Moorhead Jacobs

calculate the enthalpy for the above reaction according to the third law

of thermodynamics.
AH§98 == 210. 3 i 2.0 kcal./mole
For the reaction
;— Nd203 (c) 2: NdO (g) + %— o (g)
the literature value for the enthalpy is

AHzogs = 214.4 kcal./mole.

THE MASS SPECTROMETRIC INVESTIGATION
OF NEODYMIUM MONO-THIO OXIDE

By

Rebecca Moorhead Jacobs

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Chemistry

1965

AC KNOW LEDGMENTS

The author wishes to express her sincere thanks and appreciation
to Dr. Harry A. Eick for his guidance, help and encouragement through-
out the course of this research for without these this work would not
have been possible.

Gratitude and appreciation are also expressed to my colleagues,
Dr. Robert E. Gebelt, whose suggestions and help particularly during
the early part of this work were very much needed, and Kenneth Manske
whose assistance is appreciated.

A very deep sense of appreciation is extended to the author‘ 5
parents and to her husband, Lowell, without whose patience, encourage-
ment, enthusiasm and interest this work would not have resulted.

Financial assistance from the Atomic Energy Commission under

Contract AT-(ll-l)-716 and the National Science Foundation is grate—
fully acknowledged.

>:< 3:: >:< >:< >:< >:: >:< >:< >1: :1: >§c >1: ::< >:: :1: >§<

ii

TABLE OF CONTENTS

I.INTRODUCTION............ ......

A. Preface . . . . . . .
B. Incentive for This Work
C. Historical . .

II. THEORETICAL

A. Mass Spectrometer . .
1. Theory of the Time- of- Flight Mass
Spectrometer. . . ...... .
2.. Knudsen Source Region Theory ..... . . .
3. Calibration to Obtain Pressure Measurements
B. Thermodynamic Relationships Involved
1. Second Law Determination of the Heat of
Reaction . .
2.. Third Law Determination of the Heat of
Reaction .

III. EXPERIMENTAL

A. Preparation and Analysis of the Mono-thio Oxides .
B. Vaporization Experiments

1. Instrument . . . . . . . . . . . .....
Z. Crucible .
3. Temperature Measurement . ......

IV. RESULTS AND DISCUSSION .

A. Silver Sensitivity Factor .
B. COpper Vaporization Experiment .

I. Results . . .

2. Discussion. . . . . . . . .....
C. Appearance Potential Work . . .

iii

10
14

l4

17

19

19
21
21
Z3
Z4

2.7

Z7
Z8
Z8
Z9
31

TABLE OF CONTENTS - Continued

Page

D. Vaporization of SmZOZS . ........... . . . 32

1. Results ............... . . . . . . 32

2. Discussion . . . . . ..... . . . . . . . . . 35

E. Vaporization Behavior of NdZOZS ........... 35

F. Results of Other Work ............ . . . . 39

G. Analysis of Errors ..... . . . . . . . . . . . . . 40

H. Suggestions for Further Research . . . . . . . . . . 42
BIBLIOGRAPHY . ........... . . . . . . ..... . 43
APPENDICES .......... ‘ .............. . 46

iv

LIST OF TABLES

TABLE Page

I. Comparison of Literature and Experimental Values
for the Interplanar "d" Spacings for Hexagonal

NdzOzs and sszZS . o o o o o o o o o o o o . o o o o 22
II. Corrections for Pyrometer No. 1619073 . . . . . . 25
III. Third Law Treatment of Copper Vaporization Data . 30

IV. Comparison of Experimentally Determined Appear-
ance Potentials and Literature Values for Ionization
Potentials....................... 34

A. I. Calculated Cp Values Versus Temperature for LaO. 52

A. 11. Calculation of Afef for the reaction: g—Ndzozs (c) =
Nd0(g)+;—S(g)................... 57

A. III. Third Law Treatment of Vaporization Data for
NdZOZS........................ 58

FIGURE

LIST OF FIGURES

Diagram of Time—of-Flight Mass Spectrometer .

Simplified Electron Multiplier Schematic

Apparatus for the Preparation of Mono-thio Oxides .

Appearance Potential Curves .

Log INdO

vi

T versus 104/T for Vaporization of NdzOzS.

Page

20
33

37

APPENDIX

A.

l

. Z

. 3

. 4

LIST OF APPENDICES

Page
Calculation of the Silver Sensitivity Factor Using
the Integration Method . . . . . . . . . . . . . . . 47
Calculations Involved in the Treatment of the
Copper Va porization Data . . . . . . . . . . . . . 47
Correction of Experimental Enthalgy and Entropy
to Reference Temperature, 298.15 K . . . . . . 49
Sources of Free %nergy Functions for Third Law
Calculation of. AHzgg . . . . . . . . . . . . . . . . 54
The Samarium-Oxygen-Boron System . . . . . . . 66
Tabulation of Physical Constants . . . . . . . . . 69

vii

I. INTRODUCTION

A. Preface

This thesis reports the study of the vaporization of a compound
of the formula, NdzOzS. Values for the enthalpy and entropy changes
for the vaporization reaction of NdZOZS were determined, and a cursory
examination of the vaporization mode of SmZOZS was made. Similar
mono-thio oxides exist for most of the lanthanides and some of the
actinides. These compounds are formed by substituting a sulfur atom

for an oxygen atom in the appropriate sesquioxide, M303.

B. Incentive for This Work

Because the mono-thio oxides differ from the sesquioxides only
in that a sulfur atom has replaced an oxygen atom, it seemed that the
mono-thio oxides should vaporize like the sesquioxides, and that the
enthalpy and entropy changes for the vaporization would be comparable
to those of the sesquioxides. This research was undertaken to test

this hypothe sis .

C. Historical

Very little work other than their preparation has been reported
for the mono-thio oxides. The preparative work was carried out in
two separate laboratories. The preparation and crystal structure of
the mono-thio oxides of lanthanum, cerium and some of the actinides
have been reported by Zachariasen (l). The mono-thio oxides of all
of the lanthanides except cerium and promethium have been prepared

by Eick (2) who reports only their lattice parameters.

Two methods of preparation were used in this latter work. The
first utilized carbon disulfide and was used to prepare the lanthanum,
neodymium, samarium and europium mono-thio oxides, starting with
the respective sesquioxides. The second method which was used to
prepare the remaining species, involved pyrolyzing thioacetamide in
the presence of the hot sesquioxide.

Prior to my investigation, the vaporization behavior of only one
mono-thio oxide has been characterized. Wiedemeier (3) studied the
species CezOzS, using mass spectrometric and Knudsen effusion

techniques, and observed that it vaporizes according to equation (1).

CezOZS (c) = ZCeO (g) + S (g) (l).

Considerably more work has been done on the vaporization
behavior of the lanthanide sesquioxides. Since the behavior of the
mono-thio oxides is expected to parallel that of the sesquioxides, a
summary of the investigations on the latter seems in order.

Several workers have investigated the vaporization behavior of
the lanthanide sesquioxides. Chupka (4) investigated the La—LaZO3
system using mass spectrometric and effusion techniques and deter-
mined the dissociation energy of gaseous LaO. White e_:_t a_._l. (5)
studied the vaporization behavior of LaZO3 and NdzO3 in both tungsten
and tantalum effusion cells, and observed that the presence of tantalum
enhances the volatility of these oxides. This enhanced volatility has
been attributed to the formation of the gaseous species, TaO and TaOz.

Panish (6, 7) studied the vaporization of all lanthanide sesqui-
oxides, except those of lanthanum, cerium and promethium, and found
that with the exception of LuzO3, they fall into two groups. These
groups are identical with the cerium and yttrium groupings of the

lanthanides. Within each group the vaporization mode changes from

reaction (2) to reaction (3) with increasing atomic number of the

lanthanide metal.
M203 (c) = ZMO (g) + O (g) (2)

An abrupt change in the vaporization mode occurs after europium
and ytterbium, the lanthanide metals in which the 4f electron shell
is respectively half and completely filled. In the case of YbZO3 only
reaction (3) was observed, while for Lu203 only reaction (2) was
observed. These studies were made at temperatures ranging from
20000 to 25000 K by analyzing with a Bendix time-of-flight mass
spectrometer the species effusing from an iridium cell.

Goldstein and others (8) studied the vaporization of La203 and
Nd303 in order to determine the dissociation energies of LaO and NdO.
In so doing they also determined the heats of reaction for the vapori-
zation processes in the temperature range 20000 to 25000 K, using a

combination of Knudsen effusion and mass spectrometric techniques.

II. THEORETICAL

A. Mas s Spectrometer

1. Theory of the Time-of-flight Mass Spectrometer

The Bendix time-of-flight mass spectrometer used in this
research produces 10,000 spectra per second of any gaseous sample
which enters the ion source region. Unlike magnetic focusing mass
spectrometers in which no electric or magnetic fields are used to
enhance separation, mass separation in the time-of-flight instrument
results only from mass dependent velocities.

In the operation of this mass spectrometer a group of ions is
given an impulse of kinetic energy such that the ions are all directed
down a field-free drift tube toward a collector. Since all of the ions
receive an equal energy impulse, their respective velocities vary
according to their mass-to-charge ratios. The lighter ions have the
higher velocities. ’

Basically this time-of-flight mass spectrometer consists of
four regions (see Figure l): (l) the ionizing, (2) the accelerating,

(3) the field—free drift tube, and (4) the multiplier or collector regions.
In order to initiate each of the 10, 000 cycles which result in the 10, 000
spectra per second, an ionizing electron beam is created. The
electrons, which are emitted from a hot tungsten filament, are
normally prohibited from entering the ionizing region by a negative
voltage on the control grid. To produce the electron beam a 0. 25
microsecond positive pulse is applied to the control grid and the
electrons are drawn into the ionizing region through the narrow slits
in the control grid and electron collimator, and are directed to the

electron trap. .

"33882:... 8:. «.3213:an mo 53.3 J 8&2

 

 

 

 

 

 

 

 

 

    

a»...

L our 8838 a

{alga—30

.— u o . 3H» R 33 9.».qu Ewan 523.» lldlluwnnoo

« a. g 35/ v a .833». IE

IIIIIL .—r\ 30.." 60—385 J 57 0.3030
33. a . . s. f. .

25. Bad . w w .M . $23 a» I.
4—1 L. 4. _.U 6 +1313
.3353 9033 333333 93393
new 3a» 260.“ =3.U

 

woman—Mann! can #030

 

 

Samoa an;

This beam of electrons strikes the gaseous sample which enters
at a right angle to the electron beam. The kind and number of ions
produced from the sample molecules depends, at least in part, on the
energy of the electrons striking the molecules. This electron energy
is determined by the negative d. c. potential applied to the filament,
and can be varied between 0 and 100 volts.

From molecules struck by electrons of sufficient energy positive
and negative ions, and neutral radicals may be produced. In this
treatment only positive ions are considered. These ions collect in the
ionizing region which is bounded by the backing plate and the ion focus
grid. As soon as the electron beam is shut off, the ion focus grid,
normally at ground potential, is given a 2. 5 microsecond negative
pulse of about -270 volts. This pulse draws the ions through this grid
into the acceleration region. Here the ions are accelerated by a -2800
volt potential on the ion energy grid, and they drift into the field-free
drift region where they separate into packets since their velocities
vary according to their respective mass-to—charge ratios.

For singly charged ions:

K. E.=%-m v2 (4)

where K. E. is the kinetic energy, m is the mass of the ion and v is

its velocity. Thus,

p-

.1. ._
v = (2K. E./m)‘Z = constant/m2

and the velocity is inversely proportional to the square root of the
mass of the singly charged ion.

Since velocity is equal to the distance traveled divided by the
time necessary to travel this distance, the time-of-flight, t, is given

by equation (5)

1. 1
t = dmz /(2K.E.)T (5)

Where d is the distance and the time-of-flight is the time from ion-
ization to collection, and is thus a function of voltages and the
dimensions of the instrument.

Since the initial kinetic energies of the ions are neither equal,
nor zero, the ions in a packet do not have exactly equal energies.

This means that some ions of the same mass will arrive before, and
some after, the main group. This scatter reduces the resolution of
the instrument. A convenient measure of resolution is the largest
mass for which adjacent masses are separated essentially completely.
That is, that mass for which the time spread for ions of the same mass
just equals the time between adjacent masses. Wiley (9) discusses
how the ion gun (ion source) reduces this problem of a time spread for
ions of the same mass, and results in increased resolution.

After drifting down the drift tube in order to separate into the
various mass packets, the ions enter the multiplier by striking the
ion cathode. Electrons are ejected from the cathode surface by the
ion impact, and these electrons enter the space between the field and
dynode strips--glass electrodes with highly resistive semiconducting
coatings. A voltage gradient of approximately 300 volts exists between
these glass strips while a potential gradient of about 2800 volts exists
along the strips. They produce an electric field which is not perpen-
dicular to the strips (see Figure 2) . An external magnetic field is
applied perpendicular to the electric field.

Under these conditions of crossed electric and magnetic fields
electrons follow a cycloidal path along the dynode strip. Since the
potential gradient along the dynode strip varies from -2800 to —75
volts, the electrons gain energy after each cycloid. The impact of
these electrons striking the dynode strip knocks electrons from this
strip on a greater than one-to-one basis. These new electrons continue

to collide with the dynode strip, each time dislodging more electrons,

8

or; 83333. 8.383 3.395 .« «as:

>08 b

 

83.33.: J. 8n/
- ,/. i T /7 /nuts§5

 

 

 

0“" 1‘ 1 s I
. any, fl - a. . . -
agaflflgjki 1’ if] n /..,T.. il/L; .1
. 4‘ ‘4.“
nl 'II II. I!" s
— . . . a 4/ I . . 1 / I; ..A
i - .0”! .0 ti} III/II"

 

W a... 3:...

31m“:- a
«83 no 6310.3 nuance.“ Nun-Soon

until they reach a point known as the "jump-off line. " Here the
potential gradient is reversed by making the conductive jump—off line
10 volts negative with respect to the end of the dynode strip. Ahead
of the jump-off line the electrons strike the dynode strip before
reaching their zero energy or equipotential level. After the line they
reach the zero level before striking the dynode and continue to cycloid
in space until reaching the anode. By this time the multiplication may
be of the order of 105.

The electrons collecting on the anode set up a voltage pulse
which is amplified and fed to the oscilloscope input. Thus with the
proper timing set-up the different mass packets are displayed across
an oscilloscope screen.

The instrument instruction manual (10) gives a detailed description
of the timing circuits and the other electronics involved in the operation
of the mass spectrometer. There are several other sources of more
detailed and theoretical explanations (9, ll, 12). In addition to describ-
ing the ion source as mentioned previously, Wiley (9) gives the mathe-
matical equation for computing the ion flight time and discusses,

mathematically, the conditions for obtaining Optimum resolution.

2. Knudsen Source Region Theory

The sample to be analyzed can enter the source region in a
number of ways, depending in a large part upon the type of analysis
to be performed. For studying the vapor species above a solid sub-
stance at high temperatures the Knudsen cell inlet system is used.
When this inlet system is used, the vapors enter the source region at
right angles to the plane of the paper in Figure 1, instead of the path
shown in that figure.

The Knudsen cell is a crucible entirely closed except for a small

knife-edged hole in the top. If a solid contained in this crucible is in

10

equilibrium with its vapor species, the pressure of that species at
a given temperature will be the equilibrium vapor pressure at that
temperature. According to Knudsen theory (13), the escape of an
occasional molecule through a small thin-edged hole or orifice does
not affect the equilibrium inside the crucible. By determining the
number of molecules which escape in a given time, the pressure of
the vapor species can be determined using expression (6).

The partial pressure, P, of the vapor species of molecular
weight, M, is related to the number of moles, Z, of that species

effusing in time, t, through an orifice of area, SO, by the equation

L
P = (9.94 x 10-4) (Z/Sot)(21TMRT)2 (6)

where T is the absolute temperature inside the crucible and R is the
molar gas constant. If S0 is in square centimeters, t in seconds,
M in grams per mole, R in cubic centimeter-atmospheres per mole-
degree and T in absolute degrees, P will be in atmospheres.
However, when the Knudsen cell is used in conjunction with a
mass spectrometer, a measure of the number of molecules effusing
is obtained by recording the ion intensity. By calibrating the instrument
with a compound whose vapor pressure is known, the ion intensity can
be related to the partial pressure. A more detailed discussion of this
calibration procedure will be given in the next section. More detailed
treatments of Knudsen theory are given by Kent (14), Cater (l5) and
Carlson (l3). Cater (15) also discusses the use of a Knudsen cell in

a mass spectrometer.

3. Calibration to Obtain Pressure Measurements

In order to obtain any thermodynamic data beyond a second law
enthalpy change it is necessary to convert the observed intensities

into absolute vapor pressures.

11

There are two methods which have been used to obtain intensity-
pressure conversion factors. The first method which has been used
by Chupka (16, 17) is to volatilize at a known temperature, a substance,
usually silver, whose vapor pressure as a function of temperature is
well-known. The observed intensity is expressed in arbitrary intensity
units per atmosphere of the vapor species to obtain the sensitivity
factor, C. At any other temperature the pressure is then
equal to the observed intensity I divided by the sensitivity factor, C.

This relationship is shown in equation (7).

P = I/c (7)

If this sensitivity factor is to be used to obtain vapor pressures
of another substance certain correction factors must be applied.
These will be discussed later.

The second method of obtaining an intensity-pressure conversion
factor has been used also by Chupka (l7) and by Colin, Goldfinger and
Jeunehomme (18). This method involves the use of the Knudsen
equation to calculate the pressure of the substance being volatilized,

again usually silver. The sensitivity factor, 5- obtained from this

1’
method is given by equation (8).

.1— .1.
si = (1.006 x 103) (Mi/ZHR)‘ (s/ci)?1.ljrj4 A tj (8)

where Si is the sensitivity for species i in intensity unit-degrees/atm. ,
Mi is the molecular mass of species i, in grams per mole, R is the
universal gas constant in cubic centimeter—atmospheres per mole-
degree, 5 is the area of the effusion orifice in square centimeters,

t is the time necessary to evaporate the weight Si in grams of species
1, I1 is the measured intensity in arbitrary intensity units of species i,

T is the absolute temperature of the Knudsen cell, and j represents

the different temperatures which may be involved.

12

In order to make this calculation it is necessary to integrate
the intensity of the vapor species over the time necessary for the entire
sample to vaporize at the temperatures involved. The pressure Pi of

species i is then obtained from equation (9).

pi = IiT/Si (9)
and Si is related to C in equation (7) above by:

Si/T = c (10)

In discussing these two methods for determining the sensitivity of the
mass spectrometer, Chupka (17) says that they "have generally agreed
within 50%. "

These sensitivity factors must now be converted from the known
species to the species of interest.

Inghram, Chupka and Porter (16) have used the following equation:
PX = (1/rC)(0"Ag/o’x) (TX/TAg)(6Ag+/6X+) 11x1» (11)

where PX is the partial pressure of the gaseous species x, ix+ is an
ion of isotopic species of molecule x and mass i, r is the isotopic
fraction of molecules x having mass 1, TX and TAg are the absolute
temperatures of the cell during the observation of x+ and Ag+,
respectively. Silver was used for calibration. 6Ag+/6 x+ is a cor-
rection due to the different electron multiplier efficiencies of Ag+ and
x+, C is the sensitivity in arbitrary units of 1°7Ag+ ion intensity for
80 volt electrons per atmosphere of silver vapor. Iix+ is the observed
ion intensity of ix+ in the same units as were used for 1(”Ag +, and

0—A fl]; is the ratio of the ionization cross sections for silver and for
g

species x.

13

The factor used to correct for different electron multiplier
efficiencies for the different ion species is difficult to evaluate.
Inghram has seemingly done the only work in this field and his results
are elusive. In reference (16) he cites reference (19) which, however,
is not very enlightening and leads to the conclusion that a factor of
one is probably the best estimate until more definite data become
available. Since in our instrument only electrons are multiplied, a
value of unity was used.

The ionization cross sections for elements up to and including
barium were calculated by Otvos and Stevenson (20). Panish (6) ex-
tended their treatment to obtain a value of 73 for the ionization cross
sections of the lanthanide metals, and assumed that the ionization
cross sections of the lanthanide monoxides are equal to the sums of
the cross sections for the metals and oxygen. (Note: Otvos and
Stevenson show some evidence for the additivity of atomic cross
sections to obtain molecular cross sections for lower atomic weights.)

Chupka, Inghram and Porter (4) took the relative cross sections
of Ag+zLa+zLaO+ as 121:2. Based on the work of Otvos and Stevenson,
and of Panish, it seems that a better estimate would be 1:2:2, since
the cross section of silver is given as 34.8 and that of oxygen as 3. 29.
Thus a ratio of 34.8 : 73 : 76. 3 would be expected.

Several other factors which should be taken into consideration in
obtaining an intensity-pressure relationship are:

(l) The instrument should be focused to obtain the maximum
intensity for the ion species being observed.

(2) The appearance potential of the ion species should be con-
sidered, and if the electron energy at which the observation is being
made is not the energy which gives maximum intensity, a correction
should be applied. This refers to both the species used for calibration

and the species being observed.

14

B. Thermodynamic Relationships Involved

1. Second Law Determination of the Heat of Reaction
Consider the vaporization reaction at T0 K:
2‘— M2028 (c) = M0 (g) + 4 s (g) (12)

For this reaction

AFC - AHO TASO - RTI K (13)
T ' T T ‘ n

where K is the equilibrium constant and can be expressed in terms

of the partial pressures of the reaction products as such:

1
.. ‘2"
In this work it was not possible to measure PS due to the background
oxygen in the mass spectrometer. Thus if PS can be expressed in
terms of a constant, c2, times pMO’ then
3/2
= C
K (PMO) (15)
and
o o 0 3/2
AFT - AHT - TAST - -RT In C(PMO)
= —RT(3/Z In PMO+ln c) (16)
Solving this expression for In pMO gives:
0 o
In PMO .. - [AHT/( 3/2 R)]l/T + AST/( 3/2 R) - 2/3 In c (17)

. . o o
If 1t 15 assumed that AHT and AST are constant over the

temperature range in which the measurements are made, a graph of

In pMO versus l/T will describe a straight line from whose slope and
intercept AH; and A5; can be obtained for some average temperature

in the temperature range, if the constant c is evaluated.

15

o o . . .
T and AST are constant 1mp11es either

that the heat capacities of the gaseous products are equal to the heat

The assumption that AH

capacity of the condensed phase, or that the difference between the
heat capacities of the gaseous products and the condensed phase is a
constant over the temperature range. Since the former case cannot
be true, only the latter need be considered. The assumption of
constant heat capacity difference is often true within the limits of the

. 0 . .
experiment and hence a value for AH can be obtained from this

T
second law treatment.
The constant c in equation (17) can be evaluated by recalling the
Knudsen equation (6) which gives the partial pressure, P, of a species

effusing from an orifice of area, So’ in time t at a temperature of T

degrees Kelvin.

1_
P = (z /s t) (2TrM RT)2
S o

S S (18)
1
.. T
PMO (ZMO/sot) (2 n MMORT)
where ZS and ZMO represent the number of moles of S and of MO which

effuse through the orifice in time t, and are related to each other from

equation (12) by equation (19).

1
ZS 3 7 ZMO (19)

Since 50’ t and T are the same for both S and MO

 

 

l l
2' 2'
PS _ ZMo/Z Ms _ Ms 1 (20).
__ _ _ ,_ _
PMO ZMO MMOZ ZMMO
and
1
= 7’7 2
PS (PMO/21 (MS/MMO) ( 1)
thus
1
2_ 1 2'
c - ~2- (MS/MMO) (22)

16

For use with the mass spectrometer equation (17) must be
further modified. When positive ions are produced from neutral
species by electron impact as they are in the mass spectrometer, the

+
current collected, I , is given by equation (23) (21).

+
1 = n QIeLni (23)

where n is the efficiency of positive ion collection, 1. e., the ratio of
ions collected to those formed, Q is the ionization cross section in
cm}, Ie is the electron beam current, L is the active path length of
the electrons in cms, and ni is the concentration (in cm." 3) of neutral
species in the ionizing region.

From the ideal gas equation
r11 = PV/RT (24)

and so, assuming ideal behavior for the vapor species

+

I n QIeL(PV/RT) (25)

and

p (R/n QIeLV) I+T = k I+T (26)

Substituting this expression into (17) results in

O O
In LMOT .. ~[2AHT/3R]l/T + ZAST/3R - 2/3 In c - ln k (27)

The slope of an In IT versus l/T plot is then equal to -2AH;/3R and

AH; can be easily evaluated. The intercept of this plot is given by
2AS;/3R - 2/3 In c - ln k and in order to obtain ASE, c and k must
both be known.

0
T

will be accurate within the limits of intensity and temperature measure-

A moment‘s consideration of equation (27) will show that AH

ments, but that all of the assumptions and estimates are lumped into

the evaluation of AS; and hence the value of this quantity will be only as

good as these assumptions.

17

Comparing equations (9) and (26) shows that

k: 1/51- (28)
S'
ince O O T
AH = AH293 + f A C dT, (29)
T 298 P
heat capacity data must be available to obtain AH?” . In order to

calculate A8298 from the value of AS; obtained from the intercept of

the plot of In IT versus l/T equation (30) must be used.

0 O T
AS = AS -f A(C /T) dT. (30)
298 T 298 P

2. Third Law Determination of the Heat of Reaction

The slope obtained from a second law plot is very sensitive to
small errors in the measurement of the temperature. However, heats
of reaction obtained from a method based on the third law may be more
reliable since there is a decrease in the sensitivity to errors in
temperature measurements. It will be shown below that the third law
method involves the use of the free energy function. The decrease in
sensitivity to temperature errors is due to this use of the free energy
function which varies only slowly with temperature resulting in more
accurate interpolation between given values.

Starting with relationship (13) again,
AFOT = -RT In K (13)
and dividing both sides by T gives
0
AFT/T = -R In K

subtracting AHgge/T from both sides gives

0

T - AH§93)/T = -R In K - AHé’gg/T

(AF

but

sal o‘. ..Il»l>: 1....
a

 

 

18

O O
(AFT - AH298)/T -.- NF; - HEM/T

which is the change in the free energy function, Afef, for the reaction.
Thus

Afefz -R1nK— Angs/T (31)
solving equation (31) for AHga gives
A142,. = -TAfef - RT In K. (32)

Substituting expression (15) for K, equation (32) becomes

21113,, = -TAfef - RT(3/21n PMO + In C). (33)

III. EXPERIMENTAL

A. Preparation and Analysis of the Mono-thio Oxides

Neodymium and samarium sesquioxides of 99. 9% purity were
obtained from the Michigan Chemical Corporation, Saint Louis,
Michigan. The hydrogen sulfide, C. P. grade, came from The
Matheson Company, Inc. , Joliet, Illinois.

The apparatus which was used in preparing the mono-thio oxides
is shown in Figure 3. The inner and outer tubes were both Vycor and
Pyrex joined by a graded seal and a standard taper , respectively.
The entrance tube extended the length of the outer tube so that the gas
would be hot before it contacted the oxide.

The sesquioxide, contained in a quartz boat, was placed inside
the apparatus and the entire assembly was placed in a standard tube
furnace arranged so that the ground glass joint protruded from the
furnace. Hydrogen sulfide was swept across the oxide for three to four
hours at about 8000 as observed with a chromel-alumel thermocouple
and a potentiometer. After the sample had cooled to about 2000 oxygen
was swept across it for two to three hours as it was heated to about
5000. This was done in order to convert sulfur impurities to volatile
oxides, and to oxidize the sample to the sulfate. After cooling, once
more, hydrogen was swept through the apparatus for about one hour
as the sample was heated to 500-6000 to reduce the sulfate to the
mono-thio oxide.

The samarium mono-thio oxide was more difficult to prepare
than the neodymium mono-thio oxide and additional treatments with
the gases were required. An attempt to prepare holmium mono-thio

oxide from the sesquioxide by this method was unsuccessful and the

19

33nd 35:83. «a 83328.— 3» now an; .n. v.33.—

 

 

 

 

 

 

 

 

 

 

 

 

21

product gave an X-ray powder diffraction pattern identical with that
of the starting material.

The products were shown by X-ray powder diffraction to contain
only one phase. The interplanar "d” spacings compared favorably with
those reported by Eick (2). Table I shows a comparison of the observed
and the literature values for the interplanar "d" spacings.

A portion of neodymium mono-thio oxide was burned to the
sesquioxide. The weight loss corresponded to 9.03% sulfur whereas
9. 09% was expected. This was done only for neodymium mono-thio
oxide as samarium mono-thio oxide oxidized too slowly under the avail-

able conditions .

B. Vapo rization Expe riments

l. Instrument

The Bendix time-of-flight mass spectrometer equipped with a
high temperature Knudsen source region used for this investigation is
described in detail elsewhere (22) and so will be discussed only briefly.
The instrument is basically the Bendix Corporation Model 12-101 as
described in the instrument instruction manual (10). The output system
for the mass spectrometer consisted of two Bendix scanners model 321
which made it possible to monitor two peaks independently or to scan
two separate mass regions simultaneously. The output from these
scanners was displayed on both an ammeter and a Bausch and Lomb
Model V. O. M. 5 recorder. The entire spectrum was displayed on a
Tektronics type 545A oscilloscope equipped with a type CA plug-in unit.
The electron energy was monitored by a Digitec (United Systems
Corporation) digital d. c. voltmeter, model Z-ZOO B.

The source filament was maintained at 3. 0 amperes and the trap

current was operated in the regulate position at 0. 125 microamperes.

22

Table I. Comparison of Literature and Experimental Values for the
Interplanar "d" Spacings for Hexagonal NdzOzS and SmZOZS

 

 

 

 

 

 

ngOZS szOé

hkl literature observed hkl literature observed
002 3.395 X 3. 392 X 100 3. 372 X 3.389 X
101 3.053 3.048 101 3.012 3.032
102 2.408 2.404 102 2.378 2.390
003 2.263 2.256 003 2.238 2.249
110 1.973 1.968 110 1.947 1.960
111 1.895 1.938 111 1.870 1.875
004 1.698 1.698 112 1.688 1.694
201 1.657 1.645 201 1.635 1.647
104 1.520 1.513 202 1.506 1.518
113 1.487 1.478 113 1.467 1.480
005 1.358 1.357 203 1.346 1.357
114 1.287 1.289

121 1.208 1.209

300 1.139 1.141

123 1.122 1.122

 

Lattice parameters from reference (2):

O
3.946_+,0.001X c0 6.790i0.003A
3.8934: 0.0003 .8 c 6.68563: 0.0007 R

Nd: a
o

Sm. aO

The literature values for szOZS were calculated by Eick (2); those for
NdzOZS were calculated by this investigator using the lattice parameters
given above.

0

23

Most of the vaporization work was done at 14 electron volts since at
this setting the background in the region of mass 150 to mass 170

was negligible and the monoxide intensity was at a maximum for the
regulate position; at higher electron energies the monoxide begins to
fragment into metal and oxygen ions. The intensity at 14. 0 electron
volts was about 0. 4 that of the maximum obtained with the trap current
switch in the manual position. This maximum occurred at about 25
electron volts with the trap current about 1. 22 mic roamperes.

The silver sensitivity measurements were made at 30. 0 electron
volts with the trap current at 0.125 microamperes in the regulate
position. The intensity at this setting, which was a maximum for the
regulate position, was about 0. 25 that of the maximum obtained with
the trap current in the manual position. This maximum occurred at
about 30. 0 electron volts.

The Knudsen cell was heated by electron. bombardment and was
surrounded by a tungsten filament, which in turn was surrounded by
tantalum radiation shields. The voltage applied to the shielding was
slightly more negative than that applied to the filament. The negative
voltage applied to the filament and shields causes the electrons emitted
by the hot tungsten filament to go preferentially to the grounded crucible.
By varying the voltage on the filament, the energy of the bombarding
electrons could be either increased or decreased, thereby, either
increasing or decreasing the temperature of the crucible. The temper-
ature of the crucible was accurately controlled by the power supply and
was found to remain within the reading error of the pyrometer for at

least 45 minutes.

2 . Crucible

The Knudsen effusion cell was machined from molybdenum rod

and had an overall height of 0.75 inch (19. 0 mm) and a diameter of

24

0. 575 inch (14. 6 mm). The cavity was 0. 625 inch (15. 9 mm) in depth
and 0. 375 inch (9. 5 mm) in diameter. Holes were drilled in the bottom
of the crucible such that it could sit on the three tungsten legs of the
Knudsen source region.

A crucible lid, also molybdenum, drilled with a 0. 04 inch (1. 0 mm)
orifice, was machined precisely and sealed to the bottom portion of the
crucible by means of a shrink fit. By this technique the crucible became
essentially one piece and heated more uniformly in the mass spectrometer
Knudsen source region. This assembly was then outgassed by induction
heating in a vacuum line. The powdered sample was introduced through

the c rucible orifice .

3 . Tempe rature Measurement

Temperatures were measured with a Leeds and Northrup model
no. 8622—C disappearing filament optical pyrometer, serial number
1619073. This pyrometer had previously been calibrated by the National
Bureau of Standards and a table of scale readings versus the true
temperature (Table II) was obtained from them. From this table a
calibration curve was prepared by plotting the difference between the
true temperature and the scale reading, versus the scale reading.
Thus a correction could be applied to every observed temperature to
obtain the true temperature. This pyrometer was also checked against
a similar pyrometer with serial number 1640444 which is used only as
a standard. The agreement was within the reading error (i 10) when
sighting on the tungsten band lamp.

A 900 prism was mounted on the optical window of the mass
spectrometer so that the pyrometer could be mounted horizontally and
sighted on the crucible orifice. Prism and window corrections were
made by observing the apparent temperature, Twp, of the filament of

a tungsten band lamp near 15000 through the window and prism and

25

Table II. Corrections for Pyrometer No. 1619073

 

 

 

 

 

 

Low Range High Range X High Range
0 Scale TocNBS TOCScale TOCNBS TOCScale TOC:NBS
800 796 1100 1092 1500 1485
850 845 1200 1189 1600 1581
900 893 1300 1289 1700 1678
950 943 1400 1391 1800 1777
1000 994 1500 1494 1900 1876
1050 1045 1600 1598 2000 1977
1100 1097 1700 1702 2200 2183
1150 1149 1750 1753 2400 2390
1200 1201 2600 2597
1225 1226 2800 2804

 

26

comparing this to the temperature, T, observed for the lamp only.

The transmissivity correction Al/T was computed as follows:
Al/T = 1/TWp - l/T (34)

This transmissivity correction is approximately constant for all
temperatures and the true temperature, T, was obtained from the
experimentally observed temperature, To, by using the following

relationship:
1/T = 1/TO - Al/T (35)

The above-mentioned pyrometer correction had been applied previously
to the observed temperature to obtain To'

The observed temperatures were obtained by averaging several
readings after a constant temperature had been obtained.

Some difficulty, due apparently to reflected light within the mass
spectrometer itself, was experienced in observing the temperatures.
The use of a magnifying lens did not remove the problem as a match
between the pyrometer filament and the crucible orifice could not be
obtained. It was found that the problem of reflected light could be
minimized through careful adjustment of the pyrometer and prism.

Temperatures are believed to be good to i 50.

IV. RESULTS AND DISCUSSION

A. Silver Sensitivity Factor

It has been mentioned that there are two commonly used methods
for obtaining an intensity-pressure conversion factor, or a measure
of the sensitivity of the instrument. In this work silver was used to
obtain these factors. It was also mentioned earlier that Chupka (17)
found that these two methods agree within 50%, and that the integration
method usually is considered the more accurate.

In using the first method, the intensity of silver 107 at 1335.70 K
was observed and the average intensity was found to be 0. 0548 milli-
microamperes (mua). The pressure of silver at this temperature is
known (23) to be 2. 5 x 10‘5 atmospheres. The isotopic fraction of
silver 107 is 0. 5135 (24). The sensitivity for total silver in terms of

intensity units per atmosphere of vapor pressure is then:

0.0548/(0.5135 x 2. 5 x10-5):= 4.27 x103 mua./atmosphere

The second method is known as the integration method since the
total intensity must be observed over the time necessary to volatilize
a weighed quantity of silver. The sensitivity factor for total silver
was found by this method to be 3. 316 x 106 mua°deg. K/atm. or 2.48
x 103 mua. /atm. at 1335. 70 K. The calculation of this quantity is
shown in Appendix A. 1. Thus, as expected, the agreement between
these two methods of determining the sensitivity of the instrument is
within 50%. The value given by the integration method will be used in
converting the intensity data to pressures.

In determining this factor care had to be taken not to volatilize
excessive quantities of silver as it condensed on the boron nitride disc

27

28

Which served as a supporting structure’and insulated the shielding
supports from the electrically grounded tungsten legs which support
the crucible. This silver deposit caused the boron nitride disc to
conduct, and the potential difference between the shielding and the
crucible was destroyed. The bombarding electrons no longer went
preferentially to the crucible and the crucible temperatures attainable
were limited severely.

An attempt to determine the heat of vaporization of silver was
not successful since it vaporized at a temperature which was too low
to be determined accurately with a pyrometer. Increasing the tempera-
ture increased the vapor pressure to such an extent that the optical
window was coated rapidly making temperature measurements meaning-

less.

B. Copper Vaporization Experiment

1. Results

Due to the unsuccessful attempt to determine the heat of vapori-
zation of silver, intensity versus temperature measurements were
made using copper.

However, the conduction of the boron nitride disc caused by the
condensing of silver and copper disrupted this experiment and only
nine experimental points were obtained. Both of the copper isotopes,
masses 63 and 65, were observed, and assuming no mass discrimination
the sum of these two intensities represented the total COpper vapor'
pressure. The temperature range was 13970 K to 15710 K.

A second law treatment of the data gave

614?... = 66.8 :1: 2.8 kcal./g.f.w.

for the isotope of mass 63, and

A142)... = 64.8 :1: 1.2 kcal./g.f.w.

29

for the isotope of mass 65. None of the nine points was outside 2. 0
times the standard deviation. The literature (23) value of AH?500 is
75.8 kcal./g.f.w.

A third law treatment of the data is shown in Table III. Free
energy functions for copper were taken from Stull and Sinke (23).
The enthalpy of vaporization at 2980 K was found by this treatment
to be

ABS... = 81.24 i 0.43 kcal. /g.f.w.
whereas the literature (23) value is given as
o
AH298 = 81.10 kcal./g.f.w.

From the intercept of the second law treatment the entropy at
15000 K is
A823,... 2 18.61 :t 1.88 cal./deg./g.f.W.

The value from the literature (23) is
1351200 = 26. 96 cal. /deg. /g.f.w.

Appendix A. 2 gives the details of calculation for these quantities.

2. Discussion

Table III shows that the third law treatment gives good agreement
with the literature value. However, the second law values indicate
that there is an error of some kind. The agreement of the third law
value with the literature indicates an error in the temperature
dependence rather than in the pressure measurements.

The nature of this error is unknown. The following are a few
possibilities:

(I) A magnifying lens was used on the pyrometer and improper
focusing or some other problem involved with this lens may have

affected the temperature readings.

30

'I'II .*

 

 

 

3mw\H.mox
mvd «1*qu .o><
oméw quwv woomm ozhét eta: Kofidfi .INO. NUONmH
woéw NINHSV mvmwm wmmwé- etoH x Hw.mH omfio. mam“:
ovéw memov mmwvm omnoJ- eta: x meg: NNHO. >.~mm~
onéw owoo¢ mfiwmm namimt 6...: Xmmét vwco. m4)“:
ooéw Howmv ovnmm mmwfidn etofi ammo ohoo. Boom:
«44w 503$ “:qu >036- etoH Kowzv wmoc. $63;
«66w mp3; wofiom 0mm¢.m- eta: x 906 mvvoo. meme:
wodw msmw .3056. mmqodt .21on mo.m mNoo. CANE:
ow.ow «game. wwwhm mvmo .m- .m etofi x m; .A .998 mfioo . vfiomd
ABHM\~.mov$ A3mm\fi.mov A3ww\HmUv SO :Onm .ousmmoum ewfimcoucfi

memmd wow <81 a a: BM. nm on “63.35613 130a MOB

momQ coflmnfinomm> noQQoO mo Eoflbmouh 23A UHEH .HHH 3an

31

(2) Although it was not observed, the window may have become
sufficiently coated during the course of the experiment to affect the
temperature measurements (since the calculated entropy is low, this
hypothesis seems most likely).

(3) The temperature range may have been too narrow to give

representative results .

C. Appearance Potential Work

The appearance potential of a molecule is the minimum energy [

necessary to produce a given fragment from its parent neutral mole-

cule. For simple molecules, such as H20 and N3, where the fragment

 

 

i.‘ -2. k-

removed is an electron, the appearance potential is identical with the
ionization potential which is the energy necessary to remove an
electron from a given molecule.

In this work it was desirable to do some cursory appearance
potential work for two reasons:

(1) It was necessary to ascertain whether or not the Nd+ observable
in the spectrum under certain conditions was a product of the vaporization
reaction, or whether it was a secondary fragment produced by breaking
NdO+ into Nd+ and 0+.

(2) It was necessary to determine the maximum on the appearance
potential curve (intensity versus electron energy) in order to obtain a
factor to correct for the ion intensity measurements not being made
at the electron energy which gave the maximum intensity.

The instrument used in this work was not designed to make accurate
appearance potential measurements. The modifications which can be
made are discussed by White (2.5), Damoth (26), and Melton (27).
However, the measurements made on the instrument without modifications

were adequate for giving the desired information, i. e. , for differentiating

32

parent ions from secondary fragments, and for determining the maxi-
mum on the appearance potential curve.

Cursory observations of the intensity versus electron energy,
with the trap current switch in the manual position, that is, the trap
current was not maintained at a constant value, were made for H20,
N2, 1”Ag, 142Nd and 14‘7‘NdO. The Nd and NdO were obtained from the
vaporization of NdZOZS. The ion intensity was plotted versus the
electron energy on a linear scale, and the straight line portion of the
graph was extrapolated to zero intensity. This intercept was then
taken as the appearance potential. Figure 4 is an example of the type
of curves obtained.

The experimentally determined values of the appearance potentials
are shown in Table IV along with the literature (28) values for the ion-
ization potentials. The high value for 14"'Nd shows that it was not a
parent species in the vaporization of NdZOZS but a secondary fragment

+
from NdO .

D. Vaporization of szOzS

1. Results

A brief investigation was made into the vaporization mode of
szOZS. This compound was found to vaporize in the temperature

region 22400 K to 24000 K according to two reactions:

szozs (c) = 25m (g) + 2 o (g) + s (g) (36)
and

No evidence for SmS was found.
It was not possible to observe the oxygen in reaction (36) because

of the presence of masses 16 and 32 in the mass spectrometer background.

 

130 .
17o
' 160
150
140
130
120

110

 

amperes)

2

Ion Intensity (10"1

 

Figure 4.

142Ndo+

A
‘ F

d-

 

 

 

 

 

 

 

l
l
I
I
l
l - .
o a 10 12 14 16

'Bnergy (e. v.)

Appearance Potential Curves

34

Table IV. Comparison of Experimentally Determined Appearance
Potentials and Literature Values for Ionization Potentials

 

 

 

molecule experimental literature (28)
or atom value (ev.) value (ev.)
H20 13. 8 12. 6

N2 16. 9 15. 5

1°7Ag 10. 4 7. 5

Nd 21. 0 6. 3

NdO 5. 7 ---

 

 

 

35

Mass 16 results from the fragmentation of water and mass 32 is from
molecular oxygen.-

This cursory investigation indicates that the ion intensity ratio

+ + . . .
of Sm 'to SmO is between 1. 5 and 2. 0, posmbly, although not 11kely,
as high as 4. 0. However, there is no doubt that this ratio is consider-
ably greater than unity. Since the experiment was performed at a low

+ . .

electron energy (14 ev.) the amount of Sm arising from secondary

fragmentation is negligible .

2. Discussion

From the work of Panish (7) with samarium sesquioxide, I ex-

 

pected that SmZOZS would vaporize by these two modes. He found that
the ratio of the ion intensities of Sm+ and SmO+ for the vaporization of
the sesquioxide was between 0. 5 and l. O.

This, and the absence of any SmS in the vapor, indicate that the
Sm-S bond is significantly weakerthan the Sm-O bond in szOzS, and
that the presence of the sulfur atom in the SmZOZS lattice tends to

weaken the Sm-O bond relative to the bonds of Sm203.

E. Vaporization Behavior of NdzOZS

Neodymium mono-thio oxide, NdZOZS, was found to vaporize

solely according to the reaction:

in the temperature region 2.0910 K to 24650K.

Five separate experiments were performed on two different
preparations of NdZOzS. Intensity versus temperature measurements
were made for each of these experiments. Masses 158 and 160
(142Nd0 and 144NdO) were observed for experiments 1 and 5, masses

158 and 162 for experiment 2, mass 158 only for experiment 3 and

36

masses 160 and 162 for experiment 4.

The results of experiment 2 were discarded because of wide
variation in the individual data points. This experiment was only a
cursory examination and some SmZOZS remained in the crucible from
a previous experiment. Also it is believed that the instrument was
focused improperly for these masses during this observation.

The experimental intensities for the other experiments were
divided by the appropriate isotopic fraction in order to obtain a value
for the total intensity of all of the NdO isotopes. These intensity values L
were then multiplied by the absolute temperatures at which they were

observed, to obtain IT values. These IT values are recorded in

Table A. III.

 

The data were submitted to the CDC 3600 computer using a com-
puter program (22) which performed a least squares calculation and
determined the best straight line for a plot of ln IT versus l/T
(Figure 5). From the computer output the slope and intercept of this
plot and standard deviations of the slope and intercept were obtained.
The computer rejected all points outside two times the standard
deviation, then recalculated the slope, intercept and standard deviations.
It has been shown how the enthalpy and entropy changes can be calcu-
lated from the slope and intercept of an 1n IT versus l/T plot.

The values for the enthalpy and entropy changes for reaction (39)
%— NdzOZS (c) : NdO (g) +-§;-S (g) (39)

at an average temperature of 2273.4.O K obtained from this second law

treatment of the data are:
AHSZ73g4 : 196.66 i 2. 90 kcal. /mole of NdO formed

Asfi’m,4 : 50. 08 i 1.28 cal./deg./mole of NdO formed.

Of 204 data points submitted to determine these values, 14 points were

.10

 

37

O-RunIIm 158
g-RunInm 160
.- Run III mass 158
A-Runfllus 160

. A-Rnnflwnloz
D Q . o-mv-aulsa
Q-lmVnul 160

. _ - obtained from
Q '19 lent squares
" treatment of

the data

I»: rmr
as

104/ r c m. 0'1 Q Q

I L l I L I 1
4.10 4.20 4.30 4.40 4.50, 4.60 4.70

Figure 5. 1403 1’14 man 104/T for Vaporization of M2028

r11. E. O -. _”

 

 

38

found to deviate from the average by greater than two times the standard
deviation.

The changes in the enthalpy and entropy at 2273.40 K were cor-
rected to 298.150 K by the methods shown in Appendix A. 3. The

following values for these changes at 298.150 K resulted:
231-1293 ‘3 209. 0 j; 4. 0 kcal. /mole
A82” 2 61.1 j; 2.2 cal./deg./mole ’7

The vapor pressure of NdO at a mean temperature of 227 50 K is given 3

by equation (40).

.. _ J. 4 L
In PNdO .- (6.598 3. 0.097)10 /T + (17. 306 3. 0.430) (40)

 

.3
A third law value for [XI-1:93 was obtained from each of the 204 L_

1
data points according to equation (33), where c = (Ms/4M dc))-‘;-from

N
equation (22); M and M are the molecular weights of sulfur and

S NdO
neodymium monoxide, respectively.

IT values were converted to pressures of NdO by correcting the
silver sensitivity factor, 3. 016 x 10'7 atm. /mua. /deg., determined
previously, in the following manner:

kNdO L: (3. 016 x 10‘7)(appearance potential correction)
(ratio of cross sections)
The appearance potential correction corrects for not observing the
ions, Ag+ and NdO+, at the maxima on their respective appearance
potential curves, and is equal to 2. 5/4. 0. The cross sections of silver
and oxygen were obtained from reference (20) and that for Nd was
taken from reference (7). It was assumed that the cross section for
NdO was equal to the sum of the cross sections for Nd and 0. Apply—
ing these corrections the conversion factor for converting NdO, intensities

into vapor pressures becomes

kNdO : (3.016 x 10'7)(2.5/4.0)(34.8/76. 3) -.=

0. 85972 x10‘7 atm. /mua. /deg.

39

The sources of free energy functions are discussed in Appendix A.4.

The average value of AH?” obtained from the third law treatment is
o
AH298 = 210. 3 i 2.0 kcal./mole.

. . . . 0 .
Nine of the 1nd1v1dual values for AH298 dev1ated from the average value
by greater than two times the standard deviation. These points were
. O . .
omitted and the average AHzgg and standard dev1ation were recalculated

to obtain the value given here.

F. Results of Other Work

No values for the enthalpy and entropy changes for the vapori-
zation ofmono-thio oxides are reported in the literature. It has been
discussed that the behavior of the mono-thio oxides was expected to
be similar to that for the corresponding sesquioxides.

White e_3_t 211. (8, 29) have determined the heat of vaporization at

00 K for reaction (41).

g—Ndzo3 (c) = NdO (g) + 1—0 (g) (41)
He reports:

AH§307 :: 204. 5 _+_ 4.4 kcal. /mole.
Correcting to absolute zero, he obtains:

AH: 7: 214.0 kcal./mole

Correcting this value to 2980 K using (42)

O O O O O O
(2511,98 - AHO ) )— (Hzge - H0 )0 + (11298 - Ho )

H

0.42 kcal./mole (42)

gives

AH?” = 214.4 kcal./mole

40

for reaction (41). The heat contents for oxygen and neodymium

sesquioxide were obtained from references (2 3) and (29), respectively;

that for the monoxide was calculated as shown in Appendix A. 4.
Comparison of this AH?” for the vaporization of NdZO3 with the

values of AH?” obtained for the vaporization of NdzOzS in this work
AH?” (2nd law) 2 208. 95 kcal. ,/mole
AH?” (3rd law) 2 210. 3 kcal. /mole

suggests that similar behavior can indeed be expected for these two

compounds .

G. Analysis of Errors

It should be possible to determine from the expected errors in the
experimentally measured quantities whether or not the scatter in the
data plotted in Figure 5 is due to random errors in the measurements,
or if there are perhaps errors due to other factors.

The errors in the data of Figure 5 should be from two sources:

(1) errors in measured IT values

(2) errors in 104/T.

By analyzing the data for a single representative point an estimate of
the effect of these errors can be obtained. The point selected was
from experiment 5-mass 158 at 22760 K, 104/T 2‘ 4. 3942, IT =2 87.288.
The errors due to log LMOT from both effects will be analyzed separately
and the results combined into a single estimate of how much log IT can
be expected to deviate from the values calculated from equation (43).

In IT = (6. 598 i 0.097)104/T +(33.579:1-. 0.430)
01' (43)

log IT 2 (2.8653; 0.042)104/T + (14. 580 i 0.187)

41

The deviation in T is expected to be i 5° and that in 1 at 22760,
io.ooos mua. IT at 22760 K is then found to be 87. 288 i 3. 678.
The contribution of random errors in intensity and temperature

measurements is thus 4. 214%. The expected deviation in log IT is then
log(.04214IT + IT) - log (IT) = log (1.04214) 2 0.0179.

Thus the deviation in log IT due to errors from effect (1) is 0. 0179.

The deviation of log IT due to errors in the measurement of
104/T can be calculated by using equation (43). The estimated deviation
in temperature measurement is i 50. An error, AT, in the measured
temperature will correspond to a displacement along the log IT coordinate
of A104/T times the slope == AT/T‘Z x slope x 104, if AT is small relative
to T.

From equation (43) the slope is 2.8650 and at T = 22760 K, the
expected deviation in log IT is 0. 0277. The corresponding expected
deviation in IT due to errors in 104/T is thus 9.09%.

Combining these two deviations in log IT, gives for this point:

 

 

dev. in log IT =,f(0.0179)2 + (0.0277)? 2f? 2041 + 7.6729 x10.z
=f10.8770 x 10'7- 0.033.

 

The corresponding expected deviation in the value of IT is 8. 0%.

In the neighborhood of 104/T 4. 39, about two-thirds of the points
may be expected to lie within i 0. 033 of the value of log IT obtained
from equation (43). Examination of Figure 5 shows that while in the
immediate neighborhood of the chosen point this criterion is met, an
overall look at the graph seems to indicate that perhaps the data as a
whole do not meet this criterion. This means that there are other
sources of errors which have not been accounted for.

The following are possible additional sources of errors:

(1) The uncertainty in the intensity may be larger than supposed.

(2) The low temperature data must of necessity involve errors due

42

to the smaller number of ions, and hence lower intensities where
random fluctuations have a greater significance on the average intensity.

(3) On occasion deposits were seen to form around the orifice at
the higher temperatures. While no data were taken under these con-
ditions, it is possible that a deposit large enough to destroy Knudsen
conditions yet small enough to go undetected, formed at some time.

(4) Although it was felt that enough time had been allowed, it is
possible that some of the data were taken before thermal equilibrium
had been attained within the crucible.

(5) Due to the difficulty experienced in properly focusing the
pyrometer for temperature measurements, it is possible that the error

in the temperature is greater than supposed.

H. Suggestions for Further Research

It would be of interest to make a study of the vaporization modes
of all of the lanthanide mono-thio oxides and to compare their vapor-
ization behaviors with that found previously for the lanthanide
sesquioxides (7).

Thermodynamic functions for these reactions could also be
determined. However, in order to obtain reliable values for these
functions heat capacity data for the mono—thio oxides and for the
monoxides is needed.

If this work is to be done mass spectrometrically, the problem of
temperature measurement should be investigated and any reflected light
within the mass spectrometer Optical system should be eliminated.

A more accurate method of converting intensities to pressure is
also desirable. This would include an investigation of ionization cross
sections and of the efficiency of the multiplier for ions of different

masses.

10.

11.

12.

13.

14.

15.

BIBLIOGRAPHY

. W. H. Zachariasen, Acta. Cryst., 2, 60 (1949).

 

. H. A. Eick, J. Am. Chem. Soc., 80, 43 (1958).

 

. P. W. Gilles, Private Communication.

. W. A. Chupka, M. G. Inghram and R. F. Porter, J. Chem. Phys.,

 

24, 792 (1956).

. P. N. Walsh, H. W. Goldstein and D. White, J. Am. Ceram. Soc.,

 

43, 229 (1960).

. M. B. Panish, J. Chem. Phys., 34, 1079 (1961).

 

. M. B. Panish, J. Chem. Phys., 34, 2197 (1961).

 

. H. W. Goldstein, P. N. Walsh and D. White, J. Phys. Chem.,

 

65. 1400 (1961).

W. C. Wiley and I. H. McLaren, Rev. Sci. Inst., 26, 1150 (1955).

 

Instruction Manual for Model 12-100 or 12-101 Bendix Time-of—Flight

 

Mass Spectrometer, The Bendix Corporation, Cincinnati Division.

 

W. C. Wiley, Science, 124, 3226 (1956).

D. B. Harrington ”Time-of-Flight Mass Spectroscopy" in
Encyclopedia of Spectroscopy, C. F. Clark, Ed., Reinhold
Publishing Company, 1960.

 

K. D. Carlson, Argonne National Laboratory Report, ANL-6156
(1960).

 

R. A. Kent, Ph. D. Dissertation, Michigan State University, 1963.

E. D. Cater, P. W. Gilles and R. J. Thorn, Argonne National
Laboratory Report, ANL-6liO (1960), cf. J. Chem. Phys., 35’
608 (1961) and ibid., 619.

 

 

43

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

44

M. G. Inghram, W. A. Chupka and R. F. Porter, J. Chem. Pth”
23, 2159 (1955).

 

W. A. Chupka, M. G. Inghram, J. Phys. Chem., 59, 100 (1955).

 

R. Colin, P. Goldfinger and M. Jeunehomme, Nature, 187, 408
(1960).

U. S. Dept. of Commerce, NBS Circular 522, 257 (1953).

 

J. W. Otvos and D. P. Stevenson, J. Am. Chem. Soc., 78, 546
(1956).

 

R. Honig, J. Chem. Phys., 22, 126 (1954).

 

R. E. Gebelt, Ph. D. Dissertation, Michigan State University, 1965.

D. R. Stull and G. C. Sinke, Thermodynamic Properties of the
Elements, American Chemical Society, Washington, D. C., 1956.

 

Chart of the Nuclides, 2nd Edition, 1961.

D. White, A. Sommer, P. N. Walsh and H. W. Goldstein, "The
Application of The Time-of—Flight Mass Spectrometer to the Study
of Inorganic Materials at Elevated Temperatures" in Advances in
Mass Spectroscopy, vol. 2, Pergamon Press, New York, 1962.

 

 

Donald C. Damoth, Time-of—Flight Notes, Bendix Corporation
Cincinnati Division, Summer 1963.

 

C. E. Melton and W. H. Hamill, "Techniques for Studying Appear—
ance Potentials (RPD) and Ion-Molecule Reactions with the Bendix
TOF Mass Spectrometer, " preprint of talk given at Bendix TOF
MS Symposium, September, 1963.

Handbodk of Chemistry and Physics, 40th edition, Chemical Rubber

 

Publishing Company, Cleveland, Ohio, 1958-59.

D. White, P. N. Walsh, L. L. Ames and H. W. Goldstein, ”Thermo-
dynamics of Vaporization of the Rare-Earth Oxides at Elevated
Temperatures: Dissociation Energies of the Gaseous Monoxides"
from Thermocbrnamics of Nuclear Materials, International Atomic
Energy Agency, Vienna, Austria, 1962.

 

30.

31.

32.

33.

34.

35.

36.

37.

45

H. W. Goldstein, P. N. Walsh and D. White, J. Phys. Chem., 64,
1087 (1960).

 

G. M. Barrow, Physical Chemistrl, 109, McGraw-Hill, New York,
1961.

 

L. Akerlind, Arkiv. Fxsik, 19, 1 (1961); ibid., 22, 41 (1962).

 

P. N. Walsh, D. F. Dever and D. White, J. Phys. Chem., 65,
1410 (1961).

 

L. Brewer and M. S. Chandrasekharaiah, U. S. Atomic Energy
Commission Report UCRL-8713 (rev.) 1960.

 

K. S. Pitzer and L. Brewer, Revision of G. N. Lewis and M.
Randall, Thermodynamics, 2nd edition, McGraw-Hill, New York,
1950.

 

L. B. Pankratz, E. G. King and K. K. Kelley, U. S. Dept. of
Interior Bureau of Mines, Report of Investigations, RI-6033, 10
(1962).

B. H. Justice and E. F. Westrum, Jr., J. Phys. Chem., 61,
339 (1963).

 

APPENDICES

46

APPENDIX A

A. 1. Calculation of the Silver Sensitivity Factor Using the Integration
Method

From the strip chart recorder, the integrated intensity for 10-"Ag
was 1. 3213 mua.-min. at 1335.70 K plus 0.0664 mua. 'min. for the
warm-up period at an average temperature of 1280.1O K. The relative
abundance of the 107Ag isotope is 51. 35% (24). Thus the integrated
intensities for total silver would be 2. 5731 and 0.1293 mua. 'min.,
respectively.

From reference (18)

Si : (Mi/217R)%-(s/Gi) Z IijTJé—A tj (8)
where Mi, the gram-atomic weightJof silver, = 107.88 grams/g.atom,
Gi’ the sample weight, = 7.21x 10'3 grams, 8, the orifice area, 1:
8.758 x 10‘3 cm2, and (217R)f= 221.700 (cm3 atm./g. atom-deg.)T.

8.758 x10'3cm2

S _ (107.88 grams/g.atom)2—
‘“ 7.2 x10’3 grams

1 22. 700(cm3atm. /g. atom. deg)

 

 

'2' 60 seC./min

l .1.
times (2.5731 - 1335.7 %-+ 0.1293 ° 1280.1?) mua.min./deg.2

_ 1 1
S- 3.295 x103 cm2 deg.sec.mua./atm.2_gm.f

1

11

.1.
2

N'H

1.006 x103 gm.

/

Since 1 atm. 31.013 x106 gm./sec.zcm., latm
.1.
sec.cm.2.

Si 3 (3.295 x103) (1.006 x103)mua.deg.,/atm.
=; 3.316 x 106 mua.deg./atm. (A. l)

A. 2. Calculations Involved in the Treatment of the COpper Vaporization
Data

The factor, S to be used to convert the observed intensities

Cu'
of copper into pressures is given as follows, based upon equation (11)

which was taken from reference (16):

47

48

= S
sCu < Ag) ((Cu/ (Ag) (A. 2)
where SAg is the silver sensitivity factor, 3. 316 x 106 mua.deg. /atm. ,
(G‘Cu/O/Ag) = 18.4/34.8 = 0. 5287, is the ratio of the ionization cross

sections for copper and silver as found in reference (20), and the
ratio of the electron multiplier efficiencies for copper and silver was
taken as one.
Then,
SCu 2: (3. 316 x 106 mua.deg./atm.)(0. 5287) = 1.753 x106

mua. deg. /atm.

The pressure, P , is given by equation (A. 3).

Cu

=1 T = T 1. 10": T , o 0‘?
1:)Cu Cu Cu/SCu ICu Cu/ 753 x ICu Cu(5 7 XI )

atm. (A. 3)

For the reaction,
Cu (c) 3 Cu (g)

A143,. 2 TAfef - 2. 303 RT log Pcu.

Afef values were obtained by subtracting the free energy functions for
the ideal monatomic gas at TC from that for the reference state at
u

TCu’ as given by Stull and Sinke (23).
o

T
of the ln IT versus l/T plot as follows:

The entropy, AS , at 15000 K was determined from the intercept
o o
In p = -AHT/RT + AST/R, but In p = 1n ITk : 1n IT + 1n k
where k is the intensity-pressure conversion factor. From (A. 3),

k: 5.70 x 10-7 atm. and log k = -6.24416.

Then,
0 0
1n 1T = 'AHT/RT + AST/R - 2.303 (-6.24416) (A.4)
intercept = Asg/R + 14. 3802

experimental intercept = 23. 7474 for average temperature of 15000K

ASISOO = (23.7474 - 14.3802) R = 18.613 cal./g.f.w./deg. (A.5)

49

The literature value was obtained in the following way using values

from Stull and Sinke (23).

o o
ASISOO w (S 1500) ideal . - (81500) reference - 47.77 - 20.81
monatomic state
gas
2 26.96 cal./g.f.w./deg. (A.6)

A. 3 Correction of Experimental Enthalpy and Entropy to Reference

Temperature, 298.150 K

The experimental values for the heat and entropy of vaporization
are obtained for some average temperature in the second law treatment
of the data. In order to compare these with existing data for other
compounds this heat and entropy of vaporization must be corrected to
some reference temperature; in this work the temperature 298.150 K
will be used.

1. In order to perform this correction the following equation was

utilized for the enthalpy correction:

T
23143,. = AHO - f A c dT (29)
T 298 P
T
where f A deT is given by equation (A.7)
298
T T T
3 - C dT A.7
2158 A Cp dT Z£8 (Cp)products age ( p)reactants ( )
For the reaction
%—NdzOZS (c) = NdO (g) + %— S (g), (39)
T T T T
AC psz (C) dT+ (C) dT -1 (C) dT (A.8)
2J8 1}:st p5 29!; p NdO 22918 p Ndzozs
T T
0 o
a. C : dH 2 H -
218 ( p)S dT 29!, T HZ98
HOT - Hg” for the ideal monatomic gas can be found in Stull and Sinke (23)

and for T = 22750 K, H; - H398 = 10.23 kcal./g.atom.

50

T

b. (C (1T
2,(, P)NdO

The value of this integral was obtained by assuming that

(Cp)NdO : (Cp)LaO'

The heat capacity of LaO was calculated by assuming ideal behavior

and that
= + +

(Cp)total (Cp)trans (Cp)rot + (Cp)vib (Cp)e1ec (A° 9)
where

(C) = 2. 5 R

ptrans

(C p)rot : R

(C p)elec " O (S:e Nste)

(C) = Ruze /(e -1)-2 (A.10)

p vib

where u = E/kT and e is the vibrational energy level spacing (31).

Note: The assumption that (Cp )el ec z: 0 may not be valid in view of the
work of Akerlind (32) who has pfound that LaO has a 42 ground electronic
state, and of the work of White (33) who claims that a 7‘2 ground
electronic state for LaO has been reported. However, White's reference
is elusive, and until definite spectroscopic data on NdO is available, the
assumption that (Cp)elec 3 0 is probably as good as any.

Brewer and Chandrasekharaiah (34) report values for the vibrational

free energy function for LaO at several temperatures. Since (35)

F; - HO
0 .. _ _ -u
T vib .. Rln (1 e ), (A.11)

solving this equation for 11, using a value for -(fef) ib at some chosen
v

temperature, will lead to the calculation of e. (Cp)vib can then be

expressed as a function of T. This expression cannot be easily integrated

so a graphical integration must be performed.

Using Brewer's value for -(fef)vib at 2980 K of 0. 04, u is 3. 915

and 6 is 1. 610 x 10"13 ergs/molecule. This leads to the following

expression for u:

51

u .= 1166.7/T
-(£e£)vib at 2500° K is given as 1.96. Using this value, u is 0.467 and
e is 1. 620 x 10-13 ergs/molecule. The expression for u is then

u = 1167.5/T

An intermediate value of 1167 will be used for e/k. (Cp)total is then
given by equation (A. 12)
(c ) = 3.5 R + (1167/T)2 e l”’T/T R./(e”‘=”’/T .1)Z . (A. 12)
p total ‘ '

The values of this function at various temperatures are shown in
Table A. I.

A plot of Cp versus T followed by graphical integration between
the desired temperatures, resulted in the evaluation of the integral,

' (C )
248 p NdO
kcal. /mole.
T
. (C ) dT can be obtained if it is assumed that
29}; p NdzOzS

(Cp)NdzOZS = (Cp)ngO3°

dT. For T = 227 5° K the value of the integral is 17.14

C

Pankratz, King and Kelley (36) have investigated the high temperature
heat content of NdzO3 and found a minor thermal anomaly near 13950 K.

They give the following expressions for Cp:

Cp : 27.67 + 7.12 x10‘3 T - 2.84x105T‘2

(for 298° K to l395° K)
and (A. 13)
c x 37.20
p O O
(for 1395 K to 2000 K)

The heat absorption at the anomaly is 140 cal. /mole.

Thus,
T 1395 _
[(C) dT=f(27.67+7.12x10'3T—2.84x105T2)dT+
298 P Nd2025 298 T

140 cal./mole +1 37.20 dT.
1395

52

Table A. I. Calculated Cp Values Versus Temperature for LaO Where
cp = 3. 5R + (l127/T)2e“"7/TR/(e‘“WT—l)2

 

 

TO K Cp (cal. /mole)
300 7.59
350 7.80
400 7.97
450 8.12
500 8.24
600 8.42
700 8.54
900 8.68

1000 8.73

1200 8.79

1500 8.84

1700 8.87

2000 8.89

2200 8.90

2400 8.91

 

53

It was assumed that Cp above 20000 K is also 37. 20 cal. /mole/deg.

T T

(C ) dT=36.351 ‘1' 37.20 dT
2918 p Ndzozs ӣ5

For T = 2275° K, the value of this integral is 69.087 kcal./mole.
Then,

2215
f AdeT = 5410.23) + 17.14 - fi-(69.09) = -12.29 kcal./mole
298
and
AH?” '3 196. 66 - (—12.29)‘-= 208.95 kcal. /mole. (A. 14)

2. Equation (30) was used to correct the entropy to 298.150 K.
IT

A52,8 = As°- A(c T) dT (30)
T 298
where T T
A " - T

2.18 (C p/T) dT: 29],;(Cp/I)products T {£3 (Cp/ )reactants

For reaction (4)
'1' T T
A(C T)dT:=1 (c T) dT+ (c T) dT
23.3 P/ 2_?.9f8 P/ S 29fs P/ NdO
T
(C T) dT,
%-2£8P / Nd2028

T T o o

a. T)s d8 = S - S
291° p/ Tinf. T 298
s; and 82,, were taken from Stull and Sinke (23), and for T = 227 5° K,
sg-Szogg = 10.70 ca1./g.atom/deg. for the ideal monatomic gas.
T
T

b mfgc p/ )NdOdT

The value of this integral was obtained by assuming that

(Cp)NdO '5 (Cp)LaO

The heat capacity of LaO was calculated as shown previously, and a

plot of C p/T versus T was integrated graphically to obtain the value

of the integral. For T: 22750 K
T

C T) dT 1' 17.27 cal. molede .
2918<p/ r.o / / g

54

T
c. C T dT:
2J8 ( 15/ )NdzOzs

This integral was evaluated by assuming that

(CP)NdZOZS : (Cp)NdZO3

N

From the work of Pankratz, King and Kelley (36) which was discussed

previously
T 1395 _3
[(c /T) dT: f(27.67/T+7.12x10-3-2.84x105T )dT+
298 P Ndzo3 298
T F
140/1395 cal./mole/deg. +f 37.20/T dT (A.l5)
1395

Again it was assumed that Cp above 20000 K is 37. 20 cal./mole/deg.

This expression simplifies to
T

= . - .4
zgfsmp /T)Ndz3do dT 37 201nT 220 2

 

and for T = 2275° K, the value of this integral is 67.26 ca1./mole/deg.

Then,
2275
f AC p/T dT: 2—(10. 70) + 17. 27 - i—(67. 26): —11.01 cal./mo1e/deg.
298

and

A52,8 : 50.08 - (-11.01) = 61.09 cal /'rnole/deg.

A. 4 Souorces of Free Energy Functions for Third Law Calculation of

In order to treat the data by the third law method, free energy
functions must be obtained or estimated.
a. Free energy functions of sulfur

The free energy functions for sulfur were taken from Stull and
Sinke (23) for the ideal monatomic gas. A linear interpolation was made

where fef values were needed but not tabulated in the table.

b. Free energy functions for NdO
Free energy functions for NdO relative to 00 K are given by

White (29). The following relationship was used to relate these to 2980 K:

55

(F; - High/T = (F°T - HS’VT - (11:1,, - H31/T (A. 16)

o o . . . . .
(H298 - H0 ) is not available in the literature but was calculated in the
.'

following way.

0 o o o o o o 0
(H298 ‘ H0 )total = (H298 ' H0) + (H298 ' H0) t ‘1‘ (H293 " Ho) +

trans ro vib

O 0
(H298 "' I"Io )

elec (A. 17)
where (35) EA,
0 o '
H - :: .
( :98 (:3 trans 2 5 RT L
(H298 - Ho)rot 2 RT
0 o (A. 18)
(H298 - H0 )elec , = 0
o 0
(H298 ' H°)vib = RT u/(eu - 1) ._~

 

Brewer‘ s (34) value for (fef)v. for LaO was used to obtain a value for

1b
u as described in Appendix A. 3. u at 2980 K is 3. 915.
(H2398 - H51tota1= [3.5 + 3.915/(e3’91°'- 1)] (1.987)(298) = 2120 cal./mole.

c. Free energy functions for NdZOZS

The free energy functions for NdZOZS were obtained in the following
manner:
(fef)

= (fef) i—(fef) (reference -'r (fef)S (reference (A. 19)

NdZOZS ngO3 02 state) state)
The free energy functions for oxygen and sulfur were taken from Stull
and Sinke (23). The free energy functions for NdzO3 were calculated

using the following relationships:

0 O O 0
(F3. -H...)/T = (HT - Hzgsl/T - sT (£1.20)
T
sf} 2 52,. +ng Cp/T dT (A.21)
9

The function for Cp from reference (36) as given in equation (A. 13) was
used. Westrum (37) estimates $398 for NdzO3 as 37.87 cal.[ mole/deg.

Then, since all temperatures in this work were above 13950 K:

56

1395
s° : 37.87 +f (27.67 T"1 + 7.12 x10_3 - 2.84 x105 T'3) dT +
T 298 T
140/1395+ f 37.20/T dT
1395
s; = 86.95 + 37.201n T/1395 = 37.20 In T - 182.55

where 140/1395 gives the entropy of transition for the thermal anomaly
at l395° K.
Pankratz, King and Kelley (36) give the following equations to p711
represent the heat content of NdzO3. i
H; - Hi)... : 27.67 T + 3.56 x10“3 T2 + 2.84 x105 T"- 9, 519
(for the temperature range 2980 to 13950 K) (A. 22)

14;- 112,, : 37.20 T - 15,540

 

O - :
(for the temperature range 13950 to 2000 K) 5
Since all temperatures in this work were above 13950 K, only the

second expression was needed.
0
T

O

- H398 and sT

Substituting these expressions for H into equation
(A. 20) gives:

(Ff)r - Hip/T : 219.75 - 1.554 x104 T-l - 85.67 log T ca1./mole
for temperatures above 13950 K.
This expression was used to obtain free energy functions for ngO3.

The change in the free energy function for reaction (39)

i- NdZO.s (c) = Ndo (g) + %,—s (g) (39)

as a function of temperature is shown in Table A. II.

........fl.w Hmflflu. !

 

57

Ed ..

 

 

 

51%. $4- 3.3 8.3 39$. 83.
$43. $4- 3.? mos» mZ.m~. 85
35m- $4. 3.3 3.: £13 83
S .S. No.7 No.3 3.: 39$. 83.
$6...- mod- 3.3 3.3 38.2 8:
HH.oo- 8N.N- om.mm 6¢.o> mom.mm oooN

.36 No N 82 m
$38518 H6“ 4 H :31. coax- $65.“- cfll - :31- .M 68

33m 83m mam obsoumcoe

.moh .mou Homo“

N m N
a 0%: .m. - :6: + 082:6: a .M - :6: + :63“... 83

3 m .m + 3 062 n 3 m~o~621w E2863 6.: So “63 No 66383630 .2 .4 6369

58

 

 

 

 

 

 

podcflcoo

88¢.MHN Nos.ww «mm.m- 66m.wms wmw.mH 6.66HN Nos->H
Noo.oom o~6.¢w «mm.m- 66m.mms ws6.mm 6.66HN oos->H
wwv.oHN 666.mm oom.m- Hmo.wms mwo.ws H.6msm oosw>
686.oom wmw.¢w oo~.m- Hmo.mms A~¢.~N H.6msw wms->
owm.HHN oom.ow ¢o~.m- mom.>NH 68m.ms m.~msm oos->
mm>.osm omo.6m vom.m- moa.pms 86¢.NH o.NmH~ wmd->
Hmw.HHN «88.6w mom.m- mmo.wms vom.¢s o.HmHN N6H->H
moo.NHN mm~.>w No~.m- omo.wms 6m¢.ws o.HmHN oos->H
omm.mom ooo.¢w om~.m- HNH.>NH How.os N.>mHN N6H->H
mws.8om www.mw owH.m- HNH.>NH Nov.NN N.8mHN oos->b
NwN.HHN www.8m 68H.m- «no.6NH Nom.NH c.8mHN N6H->H
oH~.HHN Nsv.8w 68H.m- 486.6NH wmm.ms o.vmsm oo~->H
omm.wom NoN.¢w 68H.m- «no.6NH 46¢.om c.4mHN wms-az
888.80N 68H.6w 68H.m- 88¢.oms Hum.ms o.¢MsN o6H-H
som.mom wov.mw 68H.m- 488.6NH mum.hs o.vmsm me-H
¢o~.wo~ was.mw cps.m- 66~.6NH mwm.ps w.o~HN wm~-:u
wmo.vsm oo~.mw m6H.m- 8H6.6NH ws¢.w 6.6~H~ oosw>
888.80N mum.¢m mos.m- ps6.6Ns pmm.ws 6.6NHN wms->
mom.msm osw.oo mos.m- mmm.o- ¢o¢.~ 6.mNHN N6~->H
8N6.NHN Nm~.ow mos.m- mmm.6ms wsm.w 6.mNsN O6H->H
wsv.oHN 86¢.sw ovs.m- ooo.o- pmo.ss H.6HHN Nos->H
wm6.osm wo>.>m owd.m- moo.6~s 646.os H.6HHN o6H->H
wwN.HHN ov¢.mw 64s.m- voo.mws wmm.o c.8HHN wmsw>
psv.mom oom.mm 88H.m- H6O.6N~ osw.¢s o.mHH~ wms-:H
mom.mo~ mw6.mm mmH.m- www.mws Noo.¢~ w.mosm wms-:g
vom.po~ moo.mm 6~H.m- NHo.m~H oeu.mH o.¢oo~ oos-H
Nmo.oo~ 6w8.vw oss.m- Nuo.mms 886.4s o.¢oo~ me-H
wuv.¢o~ Nw6.~w NHH.M- www.¢- smo.om o.ooo~ mmfl-:s
A223 .Eoémwmd 02m 5 em SM. 6 5 em to as- .E M om. .Exm
mmOan no.“ mudQ CoSnuComw> wo “coebmogh. EMA ppflsrfi A: .< 2de

59

 

 

 

 

podcflcoo
mw¢.wom wvm.sw NwN.M- 88¢.omH wwa.o¢ m.mo- ~6H->H
Hwo.oom 6ww.sw NwN.m- 88¢.oms Num.mw m.mo~N 66H->H
mom.ssm mmv.ww Hmm.m- wsv.omH mom.om m.vo- o6H->.
pmw.oom oo».~w me.m- ws¢.oms wew.om m.¢o~N wms->.
smm.mHN om~.6w w8~.m- ¢o~.omH omo.m~ w.~o- Nos->H
w-.~HN woo.mw mpm.m- ¢o~.oms omo.¢~ ¢.~o- O6H->H
Hwo.oo~ N6N.mm w6N.m- www.8NH Hmw.¢m H.68HN mmH-H:
¢~H.osm mom.mw mow.m- www.o- oss.mm o.woum mms-:u
owo.oo~ 8mm.mw omN.m- Nos.ous wms.mm o.oo- o6H-H
omo.oom mum.mw omm.m- Noo.o~H Num.Nm o.oosm wm_-H
m6~.~HN www.cw mmm.m- 6mm.mNH N8H.om c.8wsm o6H-H
mH¢.NH~ «ms.6w mmm.m- 6mm.m~_ Hw~.s~ c.8wsm wms-H
66~.wom eve.mw mmm.m- ms¢.o- Hem.om 6.mmsm mmamzH
oms.osm mo».vw 64~.m- o>N.oNH oH6.m~ v.swsm oo~->
omo.oo~ 6oo.mw 6¢N.m- osm.¢~4 66¢.wm ¢.stm wmsw>
omm.oHN omm.¢w 6¢N.m- o»~.m- wmw.pm w.HwH~ oos->
N~6.wo~ oom.um 6¢N.m- onm.o- mo¢.mm ¢.Hw- wms->
som.ssm NH~.mw N¢N.m- ¢~H.oms wom.sm ~.m»H~ N6H->H
Nev.odm oom.8w N¢N.m- ¢NH.¢~H Nsm.mm N.wssm O6H->H
wm¢.osm wso.vw ovN.m- owo.o- www.mm H.88sm ~6H->H
wos.~ow 868.Hw ovm.m- owo.o- mm~.mm H.8>- o6H->H
mmo.HH~ oo~.mw wmm.m- Hmo.o~H ~48.NN H.68HN N6H->H
Nom.oHN osm.ww wmm.m- HNo.oNH 88m.mm H.68HN O6H->H
wow.oo~ va.vw NNN.M- ooo.w- www.8N 8.w6H~ wm~-:u
pwo.oo~ www.mw mN~.m- mom.wmu 8mm.pm c.86HN ooH-H
mso.oom osm.vm mNN.m- wom.m- ow8.mm c.86HN me-H
ESE .Hmoéammd 02m E .3 Q? 6 5 am to <9- E M om. .Exm

 

 

6623660 - E .< 633.

podcflao o

 

60

 

4M6.NM~ wwo.mw ovm.m- 66m.~mM «Mo.m¢ 6.«¢~N N6M->M
686.MMN omv.mw ovm.m- 66m.~mM c6».sm 6.66NN ooM->n
Moo.»o~ 6Mo.w> 6mm.m- MM¢.NMM Mmo.wo 6.M¢N~ NcM->M
smo.6om 666.8» 6mm.m- MMv.NmM mom.MoM 6.M¢N~ 66M->H
6m6.oM~ omo.fiw 6mm.m- 6mm.~mM Mm¢.6m m.o¢- 66M->
66M.oMN «wM.Mm 6mm.m- 6mm.~m~ 66M.M6 m.o¢~N me->
MMN.oM~ 66M.Mm omm.m- 68M.NmM M6¢.wm ~.8m- ~6M->M
moo.oo~ omm.ow omm.m- 68M.NmM NM6.66 ~.>m- 66M->M
www.0MN oM6.Mm omm.m- 68M.NMM mMM.6m ~.8m- me-:M
msw.MM~ omM.mw mNm.m- ooc.~mM MMM.¢¢ c.6mmm ~6M->M
owm.oMN wmw.Mm mmm.m- soo.~ms Mom.mm o.vm- 66M->H
mom.oM~ wmo.Mw m~m.m- 66¢.MmM va.mm o.mmmm o6M-H
va.oMN cow.sw mNm.m- wvs.MmM 6mm.mm o.mm- mmM-H
smo.oMN mom.mw mmm.m- N66.MMM 6M6.66 6.~m- 66MA>
mma.oom som.sw mmm.m- N68.MMM mm6.~m 6.Nm- mew>
Mm¢.~om 668.68 mMm.m- sw>.MmM moo.ow w.o-~ me-:M
66m.oM~ moo.~m 6Mm.m- >m6.MmM omv.om 6.>N- oosw>
”No.mom 66m.om osm.m- >m6.MmM www.mo 6.8NNN wms->.
6mm.wom vm6.ow oMm.m- 66¢.MMM 6M~.os m.¢-m me-:M
mmm.mMN N66.mw Mom.m- 66M.MmM 6mm.wm N.wMNN ~6M->M
666.MMN moo.vw Mom.m- N6M.MMM M6M.mm N.mMNN 66M->M
NMo.MMN www.mm mom.m- wmo.sms 6~o.6m H.6MNN 66M->
oom.oMN 666.Nm mom.m- mmo.dms MmM.6v H.6MNN wmu->
86M.wo~ NNm.ow 66N.m- mp6.0ms 666.6m o.mMNN mmM-:M
Mm¢.6MN vm~.6w so~.m- m~8.OmM mmw.- o.mMNN 66M-H
omo.mMN N6N.6w 68M.m- meo.omM N66.6N o.mMNN me-H
$363366: mommq 02m 5 em 83. 6 E HM M3 as- E M 6H. .axm

'i

 

 

6625660 .5 .< 6369

.. fishnfiuvfi. fl

61

Umficficoo

 

 

666.666 666.66 666.6- 666.666 666.66 6.6666 66Mw>
666.666 666.66 666.6- 666.666 666.666 6.6666 66M->.
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.66_ 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.66 6.6666 666-6
666.666 666.66 666.6- 666.666 666.66 6.6666 666-6
666.666 666 66 666.6- 666.666 666.666 6.6666 66M¢:H
666.666 666.66 666.6- 666.666 666 66 6.6666 66Mw>
666.666 666.66 666.6- 666.666 666.66 6.6666 666w>
666.666 666.66 666.6- 666.666 666.66 6.6666 66M->M
666.666 666.66 666.6- 666.666 666.66 6.6666 66M->6
666.666 666.66 666.6- 666.666 666.66 6.6666 66M->
666.666 666.66 666.6- 666.666 666.66 6.6666 666->.
666.666 666.66 666.6- 666.66M 666.66 6.6666 6664>
666.666 666.66 666.6- 666.666 666.66 6.6666 66Mw>
666.666 666.66 666.6- 666.666 666.66 6.6666 666-6
666.666 666.66 666.6- 666.666 666.66 6.6666 666-6
666.666 666.66 666.6- 666.666 666.66 6.6666 666-6
666.666 666.66 666.6- 666.666 666.66 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666636
666.666 666.66 666.6- 666.666 666 66 6.6666 66M->H
666.666 666.66 666.6- 666.666 666.66 6.6666 66M->H
666.666 666.66 666.6- 666.666 666.66 6.6666 666-26
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666 66 6.6666 66M->
666.666 666.66 666.6- 666.666 666.66 6.6666 66M->
666856666: momma 02m 5 Hm 6\6- 6 5 HM 666 46- .2 M 6.6 .66ng

 

 

 

6666:6666 - E .6. 636.6

lilfln.»fll.qfifl

 

 

62

 

 

O

poscficoo
666.666 666.66 666.6- 666.666 666.666 6.6666 66Mw>
666.666 666.66 666.6- 666.666 666.666 6.6666 66M->.
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666 66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 66M->H
666.666 666.66 666.6: 666.666 666.666 6.6666 66M->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-26
666.666 666.66 666.6- 666.666 666.666 6.6666 66M->
666.666 666.66 666.6- 666.666 666.666 6.6666 66Mw>
666.666 666.66 666.6- 666.666 666.66 6.6666 66M->H
666.666 666.66 666.6- 666.666 666.666 6.6666 66M->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-66
666.666 666.66 666.6- 666.666 666 66 6.6666 66M->6
666.666 666.66 666.6- 666.666 666.666 6.6666 66M->M
666.666 666 66 666.6- 666.666 666.66 6.6666 66M->
666.666 666.66 666.6- 666.666 666.666 6.6666 66Mw>
666.666 666.66 666.6- 666.666 666.66 6.6666 666-6
666.666 666.66 666 6- 666.666 666.66 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6:
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666 66 666.6- 666.666 666.666 6.6666 666-6
666.666 666 66 666.6- 666.666 666.66 6.6666 66M->M
666.666 666.66 666.6- 666.666 666.666 6.6666 66M->6
666.666 666.66 666.6- 666.666 666.66 6.6666 66Mw>
666.666 666 66 666.6- 666.666 666.666 6.6666 66M->
6668566666 66mm< 02m E em 66. 6 E .56 666 4.6- .: M .6 .axm

 

ill

6666:6666 - 5.66 636.6

63

P .v\I . _. E

 

 

 

 

O

 

66056636600
666.666 666.66 666.6- 666.666 666.666 6.6666 666-26
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6:
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666 666 6.6666 666.6;
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->
666.666 666.66 666.6- 666.666 666.666 6.6666 666->.
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666 66 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 6661>
666.666 666.66 666.6- 666.666 666.666 6.6666 666->
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.66 6.6666 666-6
666.666 666.66 666.6- 666.666 666.66 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666126
666.666 666.66 666.6- 666.666 666.666 6.6666 666136
666.666 666.66 666.6- 666.666 666 666 6.6666 666-26
6666856666: 66mm< 02m 5 am 66. 6 E .666 666 <6.- 66 M 6. .666m

 

 

662666660 - 6: .6. 66666.

64

“6662636600

 

 

 

0

666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->
666.666 666.66 666.6- 666.666 666.666 6.6666 666->
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->
666.666 666.66 666.6- 666.666 666 666 6.6666 666¢>
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666 666 6.6666 666-:6
666 666 666.66 666.6- 666.666 666.666 6.6666 666-26
666.666 666 66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-:6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6 6666 666->
666.666 666.66 666 6- 666.666 666 666 6.6666 666->
6666856666: 666M664 062m 5 6.66 6\6- 6 66 6.66 666 46.- 66 M .6. .3666

J

 

6626:6666 - :6 .6. 636.6.

65

 

 

666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666 66 666.6- 666.666 666.6666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666 66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666 666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-66
666.666 666.66 666.6- 666.666 666.666 6.6666 6664>
666.666 666.66 666.6- 666.666 666.666 6.6666 666A>
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666-6
666.666 666.66 666.6- 666.666 666.666 6.6666 666->6
666.666 666.66 666.6- 666.666 666 666 6.6666 666->6
666.666 666 66 666.6- 666.666 666.666 6.6666 666->
666.666 666.66 666.6- 666.666 666.666 6.6666 666->
6666.3 .6666: 66wm< 06,666 E 6.66 6\6- 6 66 .36 666 46.- .66 M 6. .3666

 

O

 

662666660 - 666 .6. 66666.

APPENDIX B

THE SAMARIUM-OXY GEN- BORON SYSTEM

B. 1 Introduction

 

A previous worker in this laboratory1 observed that the reaction
between mixtures of samarium sesquioxide and boron is a complicated
process dependent upon the composition of the starting materials, the
type of heating cell used, and several other factors. He performed
mass balance experiments on the reaction of samarium sesquioxide
with varying amounts of boron. One series of experiments was per-
formed by heating to 16500 for one hour pelletized mixtures contained
in a boron nitride lined molybdenum cell. The results of these experi-
ments indicated the loss of gaseous molecules whose boron to oxygen
ratio varied inversely with the composition of the reactants. These
data were interpreted to indicate that there is no one simple equation
for the reaction of samarium sesquioxide and boron under these con-
ditions.

Since Galloway did not have a mass spectrometer at his disposal,
he was unable to determine the molecular species which comprise this
boron oxide phase. This work was performed to observe, in a Bendix
Time-of—Flight mass spectrometer equipped with a high temperature
Knudsen source region, reaction products inferred from mass balance

data in the experiments described above.

 

1Gordon L. Galloway, Michigan State University, Ph. D.
dissertation, 1961.

66

67

B. 2 Experimental

 

The mass spectrometer available is basically a Bendix Time—of-
Flight instrument. The high temperature effusion cell and its power
supply are similar to those used by Everett G. Rauh at the Argonne
National Laboratory.

The molybdenum crucible used had an overall height of 19 mm.
and a diameter of 15. 8 mm. The cavity was 9. 5 mm. in depth and
9. 5 mm. in diameter. The lid had a 1. 0 mm. effusion hole. The
boron nitride liner was machined to fit snugly within the crucible.
Holes were drilled in the bottom of the crucible such that it could sit
on the tungsten legs of the Knudsen source region.

0. 250 gram samples of a mixture of the desired mole ratio;of
samarium sesquioxide and boron were pressed into pellets by applying
6000-8000 p. s. i. with a Carver Laboratory press. The samarium
sesquioxide (99. 9 % pure) was obtained from the Michigan Chemical
Corporation, St. Louis, Michigan, and was calcined to constant weight
in platinum crucibles. The boron (99. 5% pure) was obtained from
U. S. Borax and Chemical Corporation (100 mesh crystalline). The
residues from the mass spectrometer runs were analyzed by X—ray

powde r diffraction photography.

B. 3 Results

Pellets of mole ratios 1:5, 1:10, 1:15, and 1:20 samarium
sesquioxide to boron were analyzed. The pellets of 1:5 mole ratio
were found to lose samarium as Galloway had predicted. However,
no boron oxide species could be observed. It was discovered later
that the reaction proceeded to completion before a low enough pressure
could be obtained at 16500 such that the mass spectrometer could be

turned to a high sensitivity. Heating the sample region invariably

 

.lnlfld-d'1fll ~_

 

 

68

causes the pressure to rise, probably as a result of outgassing of the
cell and shielding. Since this entire region must be removed whenever
a new sample is inserted, outgas sing this region prior to sample
introduction is impossible.

The mass spectrometric observation of this reaction was abandoned.
If a more rapid means of pumping out the source region can be found,
further work on this project could be performed successfully. Closer
analysis of the data may tend to further support Galloway' 5 work but

will not result in the boron oxide phase compositions.

APPENDIX C

TABULATION OF PHYSICAL CONSTANTS

R = l. 987 cal/deg-mole

= 82.05 cc-atm/deg-mole
k = 1. 38031 x 10’16 ergs/deg-molecule
w = 3.1416

lnx : 2.303 logx

standard atmosphere = 1. 013 x 106 dynes/cmz

atomic weight of silver = 107.870
atomic weight of sulfur = 32. 064
atomic weight of neodymium = 144. 24

atomic weight of oxygen = 15. 999
isotopic abundance of lmAg = 51. 35%
isotopic abundance of 1“Nd = 27. 11%
isotopic abundance of 144Nd = 23. 85%
isotopic abundance of 146Nd = 17. 22%

69

CHEMISTFY HEM

 

 

NIVERSITY LIBRARIES

3083 2350

TAT

NS E

3 1293

MICHIGA

ii
0