THE CRYSTALLIIATiON OF AMORPHOUS
SIUCA USING SALT CATALYSTS

Thesis for the Degree of M. S.
MICHIGAN STATE UNIVERSITY
Paui D. Baranyai
1954

filESI‘

LIBRARY

Michigan State
University

 

ABSTRACT
THE CRYSTALLIZATION OF

F AmHmWSEEA
USING SALT CATALYSTS

by Paul D. Baranyai

It is known that the three common polymorphs of silica; quartz,
tridymite, and cristobalite, crystallize within a large range of
temperatures at atmospheric pressure. Each polymorph is stable within
a certain range of temperatures. Tridymite and cristobalite are
characteristic of high temperature environments.

An investigation of the various polymorphs was conducted using
amorphous silica gel and various salts as catalysts to determine whether
the two high temperature forms could be made to crystallize at tempera-
tures below their respective thermal stability range. All work was
carried out under one atmosphere of pressure at temperatures ranging -
from 600°C. to 900°C. Various concentrations of each salt were used.
Each sample was heated for two hours.

‘There has been much work conducted at elevated temperatures and
pressures for long durations of time. However, there has been little
research conducted at the temperatures used in this study. Nor has there
been much work dealing with the action of various salts toward the cry—
stallization of amorphous silica gel.

The amorphous silica gel was prepared fresh from sodium silicate
and hydrochloric acid. The gel was washed thoroughly with de-ionized

water and was dehydrated at a low temperature. The powdered gel and

Paul D. Baranyai

salts were heated in covered platinum crucibles, and then x-ray
powder photographs were taken of each sample. Each polymorph pre-
sent was identified by the x-ray photographs. A spectrographic
analysis was also made to determine whether any of the salts had
entered into the atomic structure of the silica.

The results showed that at atmospheric pressure and at tempera-
tures ranging from 600°C. to 900°C. the various polymorphs of silica
can be crystallized by using various salts as catalysts. The poly-
morph which was formed predominately in all the samplesets was cris-
tobalite — the highest-temperature polymorph. It was also found that
sodium had entered into the structure of cristobalite. The open struc-

ture of cristobalite had been presumably stabilized by the sodium atoms.

THE CRYSTALLIZATION 0F
ADDRPHOUS S IL ICA

USING SALT CATALYSTS

BY

Paul D. Baranyai

A THESIS

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

MASTER OF SCIENCE

Department of Geology

1964

ACKNOWLEDGEMENTS

I would like to express my grateful appreciation to Dr. Harold
Stonehouse for his suggestion of the problem and for his constructive
suggestions during the course of the research and for his critical
review of the manuscript.

I would also like to thank.Dr. Hinze and Dr. Zinn for their
constructive criticism of the manuscript.

A special note of appreciation is for my wife, Jan, who provided

a constant source of encouragement throughout the entire research.

ii

TABLE OF CONTENTS
Purpose of Study
Introduction
Phase Relationship
Review of Early Literature
Experimental Procedure
X-ray Diffraction Theory
X-ray Data
Interpretation of X-ray Powder Photographs
Spectrographic Analysis
Procedure
Conclusion
Suggestions for Future Study

Bibliography

iii

Page
Page
Page
Page
Page
Page
Page
Page
Page
Page
Page
Page

Page

ll

18

23

24

44

45

48

50

51

LIST OF TABLES

Table 1 'Stability Range for the Polymorphs of SilicaPage

Table 11 .Structure of the Polymorphs of Silica
Table III High Pressure Crystallization

Table IV yCommercial Brands of Silica Gel

Table V IPercentage Loss of Weight

Table VI T-Ray Data

Table VII.Frystallization Products

Table VIIIPolymorphic Inversions

Table Ix,'Density Increase with Temperature

Table x pectrographic Wave Lengths

Table XI Spectrographic Analysis

iv

"T‘F—_—_-

Page
Page
Page
Page
Page
Page
Page
Page
Page

Page

ll

14, 15

23

36, 37, 38
39, 40

43

45

46

Figure
Figure
Figure
Figure
Figure
Figure

Figure

0

LIST OF FIGURES
Phase Relationships
Isothermal Rate Curves
X-Ray Radiation Wave Lengths
,Generation of a Lattice Net
Conditions for Scattering-In-Phase
iProduction of a Powder Diagram

I

'Cylindrical Film Arrangement

Page
Page
Page
Page
Page
Page

Page

5

9

17

18

19

20

21

PURPOSE OF STUDY

This study was undertaken to investigate the crystallization
of the polymorphs of silica from the amorphous state by the use of
various salts. Varying concentrations of the following salts were
mixed with the amorphous silica gel and heated at temperatures rang-
ing from 600°C. to 900°C.: sodium chloride, calcium chloride, lithium
chloride and potassium bromide. All of the heating was performed
under one atmosphere of pressure. These four salts were chosen be-
cause they are common impurities which are found in the polymorphs
of silica.

Although this study may not have direct geologiq implications,
it is important to note that while the temperatures used in this study
were below the stability range for tridymite and cristobalite, these
were the resultant crystallization products.

Studies of the crystallization of silica gel using various

catalysts are few at the temperatures used in this research.

INTRODUCTION

Silica, SiOz, exists in a number of polymorphs which are stable
at different temperatures and pressures. Among these polymorphs are
quartz, which is stable at ordinary conditions, tridymite and the high
temperature polymorph, cristobalite. Along with the above forms,
three less common forms exist, which are metastable at ordinary condi-
tions. These are keatite, coesite and stishovite.

The following table lists each polymorph and the thermal stability
range at one atmosphere of pressure.

Table I

(Frondel, 1962, Page 1)

 

 

Name Thermal Stability Range-
one atmosphere in 0C

a Quartz Stable below 573

/ Quartz Stable 573-870

Tridymite Stable 870-1470

If tridymite Below 117

A tridymite 117-163

a( tridymite 163-1570

Cristobalite Stable 1470-1720

04 cristobalite Below 200

/ cristobalite 200-1720

Keatite Metastable at ordinary conditions
Coesite Metastable at ordinary conditions
Stishovite Metastable at ordinary conditions

Historically, quartz as a name was used first in the Middle Ages
(Frondel, 1962) in Saxony for massive vein quartz. However, it was
not until the 1850's that an adequate method of study was available.

Tridymite, the first polymorph of silica to be recognized in
addition to quartz was described by von Roth in 1868, and he described
cristobalite in 1884. Since the first acknowledgment of these poly-
morphs of silica, other individual polymorphs closely related in
crystal structure have been identified. Merion in 1884 was the first
to observe the high and low forms of tridymite. In 1890, Mallard
described the high and low forms of cristobalite. Fenner, who in the
early 1900's investigated the stability relations of these polymorphs,
was the first to define the existence of middle tridymite. It was not
until much later that additional polymorphs of silica were synthesized
in the laboratory. The first of these was Coesite, synthesized in 1953
by L. Coes. Coesite was obtained originally in the temperature range
between SOD-800°C and at 35,000 atmospheres for periods of about fif-
teen hours. P. P. Keat synthesized Keatite in 1954. Keatite was
obtained over a range of temperatures from 380-585°C and 5,000:18,000

p.s.i. Then in 1961 Stishov and Popova synthesized Stishovite.

 

Stishovite was first synthesized at temperatures between 1200-1400°C
and a pressure reported to be about 160 kilobars. Since the initial
syntheses of these polymorphs, two have been identified in nature.
Coesite was found in nature at Meteor Crater, Arizona. It occurs
abundantly in sheared and compressed areas of sandstone as a fine
grained nearly istropic matrix in which the fractured quartz are em-

bedded. Stishovite has also been found in association with coesite

and silica glass at Meteor Crater, Arizona, and in other areas.

PHASE RELATIONSHIP
The phase relationship between the three common polymorphs is

given in Figure 1.

 

   

 

 

 

Figure l
I ,l’ l I | I
Liquid ///
1800 ._ x’ _
/
”T
*+———-Cristobalite
1500 _._._.,' ..
I, flquartz
I
44___——Tridymite
1200 ' .—
c, F'T
o
a /
m /
*5 /
g 900 / __
H
8. J Coesite
E
E!
600 I-— .1
I
O<quartz /
I
...300 _ /l _+
I
/
I
o I ’I I J I I
0 10 20 30 40 50 60

Pressure, kilobars

(Frondel, 1962, page 5)
CK quartz is stable under atmospheric pressure at temperatures
up to about 573°C. At this temperature it undergoes a reversible rapid

inversion (displacive) to ,3 quartz.

Closely related in structure to ax quartz, /3 quartz is stable
at atmospheric pressure at temperatures from 573°C up to 870°C.
Over 870°C and up to 1470°C the stable polymorph at atmospheric
pressure is tridymite. The type of inversion between/9 quartz and
tridymite at 870°C is of the sluggish (reconstructive)type. ‘/9
cristobalite is the stable polymorph of silica at atmospheric pres-
sure at temperatures from 1470°C to its melting point 1723°C.

The differences in free energy between the three polymorphs is
small and the stability relations apparently can be altered by the
entrance of H20 or of cations into interstital positions. The metast-
able formation of the relatively open, highly symmetric structure of
cristobalite apparently is favored in this way over tridymite and
quartz.

STRUCTURE OF THE POLYMDRPHS
(Frondel, 1962)
Table II
Quartz-Hexagonal .(lattice dimension)
Ao=4.90290 i 0.0003kx
Co=5.39365 i_0.0003kx
Temperature 18°C
A°=kx (1.00202)
c1,tridymite-Hexagonal
Ao=5.04A°
Co=8.24A°
/€ cristobalite-Isometric

_ O
AO-7.1285A

Temperature 250°C

REVIEW OF EARLY LITERATURE

Much work has been conducted on the presence of silica in fresh
water, sea water and also silica in hot-spring waters. K. B.
Krauskopf (1956) reviews the early literature on low temperature dis-
solution and precipitation of silica, and comes to some conclusions
on the solubility and precipitation of silica in different geologic
environments.

1) The solubility of amorphous silica is little
affected by changes of pH in the range of 0-9,
but increases rapidly as the pH rises above 9.
In other words, silica is no more soluble in
very dilute alkali than in acid, contrary to a
common geologic assumption.

2) Silica is added to natural waters by,
a. volcanoes and springs associated with vol-
canic activity and by b. dissolution of silica
and silicates during weathering. Probably most
of the silica from weathering is in true solu-
tion. Volcanic and hot springs silica may be
initially in true solution, but if the concen-
tration is high, it will partly change to the
collodial form as the solutions cool.

However, his work makes no mention of the polymorphs of silica.
Whether or not these were formed during dissolution of the silica is
not known. White, Brannock and Murata (1956) noted the presence of
cristobalite (opal) precipitated in hot-spring waters at Steamboat
Spring, Nevada. They make this statement.

The relations of amorphous opal, of opal with
an x-ray pattern of cristobalite and of true cris-
tobalite are not known. (White, 1956)

Tridymite was found by Anderson forming in lavas that were being

altered by sulphuric acid at temperatures close to 90°C.

White made this closing observation:

The origin of cristobalite and tridymite in

hot-springs is not clear.
formed metastably at temperatures far below

The minerals have

equilibrium temperatures and in contact with

waters that were strongly supersaturated in
silica with respect to the stable mineral

quartz. (White, 1956)

Carr and Fyfe attempted crystallization of amorphous silica at

pressures ranging from 15,000 p.s.i. to 59,000 p.s.i. The tempera-

tures varied from 325°C to 448°C.

at 59,000 p.s.i. to 840 hours at 15,000 p.s.i.

partial list of their results.

The time varied from six hours

Table III gives a

 

Table III
Pressure (p.s.i.) T°C t (hours) Products
59,000 440 9 Quartz
59,000 330 9 Poor Cristobalite
45,000 440 9 Poor Cristobalite
45,000 440 12 Keatite+Quartz
45,000 330 15 Keatite+Quartz
30,000 440 96 Quartz
30,000 330 120 Quartz
15,000 438 576 Cristobalite

It should be noted that tridymite was not formed in any of their

experiments in sufficient quantity to appear on x-ray photographs.

J. F. Corwin (1953) and others conducted a study of the effects of
flouride and alkali on the transformation of amorphous silica to cris-

tobalite or quartz.

When silica glass was heated with water or aqueous

solution at 400°C and 340 atmospheres, quartz was formed when flouride

or strong alkali was present.

even when sodium or potassium chlorides were present, only cristobalite

was formed.

However, in acidic or neutral solutions,

Apparent 2 content of
cristobalite

It was the authors belief that quartz was formed only from the ions of
orthosilicic acid. Such ions are present in solution above pH 10-12.
The less ionic forms, or molecular orthosilicic acid, lead to the for-
mation of alpha cristobalite.

A. G. Verduch carried out experiments at temperatures ranging from
945-10850C. Figure 2 shows isothermal rate curves of the transforma-

tion of amorphous silica to cristobalite.

 

 

 

 

 

 

 

 

Figure‘g
40 1085°c.
103500. 0
1 10 .
30 I O C 985°C 0
945°C.

20
10

0 1 I ‘ 1 _1 J

10 20 30 40 50
Time (hours) (Verduch, A. 0., p. 428, Fig. 2)

W. G. Hill and Rustum Roy performed experiments in an attempt to
determine theqp/Qinversion in cristobalite. It was found that the
inversion was variable and depended on the structure of the starting
material and on the temperature and length of heat-treatment. The
temperatures involved were in the 1500°C to 1600°C range. The dura-
tion of these experiments ranged from % hour to 76 hours. Only the

silica gel was heated.

10

There were no fluxing agents used. The resulting x-ray patterns
showed, in the case of the heated silica gel, only cristobalite.

Although there has been much work carried on with silica gel, to
the author's knowledge little work has been done at the temperatures
used in this work. Nor, has there been much work dealing with the

action of various salts toward the crystallization of silica gel.

EXPERIMENTAL PROCEDURE

Nhny commercial brands of silica gel are available; for example,
"Celite"-John-Mansville Company, "Hi-Sil"-Pittsburg Plate Glass,
Estersil-Product Information Bulletin, E. I. duPont de Nemours and
Company, Fine Silica Gels-Davison Chemical Corporation Product Bulle-
tin (Iler, R. K.). However, the silica gel used in this work was
prepared by the author. By doing this, a truly amorphous gel was ob-
tained, as was shown by the absence of lines in an x-ray powder pho-
tograph taken of the gel. The commercial brands may or may not be
truly amorphous as shown by Table IV.

Table IV

"Celite" - 89.7% Si0 3.7% A1 0 Ignition loss 3.7%

2 2
3.0% Ca0 Ignition loss 14.0%

2,

"Hi-s 11" - 860570 $1.02,
Estersil - 85-90% Si02
Fine silica gels - 99% minimum.SiO2

The gel was prepared by mixing a solution of NaZSiO and hydro-

3
chloric acid. The sodium silicate solution was prepared from the cry-
stal form obtained from the General Chemical Division of the Allied
Chemical and Dye Corporation. The maximum limits of contained impuri-
ties were as follows: Chloride (Cl)-0.005%, sulfate (804)-0.01%, heavy
metals (as Pb)-0.001% and iron (Fe)-0.005%. The solution was prepared
to give a density of 1.15 grams/cc. The concentrated hydrochloric

acid was prepared to obtain a 6N solution. According to Babor and
Lehrman, (1957), the acid should be added to the sodium silicate, solu-

tion. However, when 100ml of acid was poured into 100ml of sodium sili-

cate, accompanied by continuous stirring, a firm gel did not develop.

ll

12

After 24 hours, the gel was still not firm, so 10ml of sodium sili-
cate was poured into 10ml of 6N HCL. As the Na28i03 was poured into
the acid, the mixture was continuously stirred. In about 5 minutes
the solution had set to a firm gel, according to this idealized reac-
tion, NaZSiO3 + 2HC1 --- H28i03 + 2NaCL. It was decided to make small
batches of gel (50 m1 - total) rather than one large batch. The reason
for this was ease of preparation and more thorough washing. One liter
of NaZSiO3 was prepared and filtered free of any precipitated silica
before each batch was prepared.

When the gel set, they were cut in half and washed with de-ionized
water. Distilled water was used by Zielke (1950), but as a spectogra-
phic analysis was to be made, the de-ionized water would afford the
greater purity. The gels were washed until they were free of the Cl-
as was shown by the absence of a precipitate with silver nitrate. They
were then placed in changes of de-ionized water for three days. The
water was frequently changed during this time. It was shown in Smits
thesis (1926) that sodium chloride in solution acts to increase the
solubility of gelatinous silica. Periodic checks were made with the
silver nitrate and were, in all cases, negative, which was shown by the
lack of a percipitate. At the end of this period, the gels were allowed
to dry at room temperature for 24 hours. Finally, the gels were dehy-
drated in the electric furnace for four hours at 360°C. A sample of.
this powder was prepared, in a manner to be discussed later, for x-ray

diffraction analysis. This sample was shown to be amorphous by an

absence of lines and the presence of a diffuse halo.

13

Silica gel was then prepared for heating with the four salts,
sodium chloride, calcium chloride, lithium chloride and potassium
bromide.

Extreme care was taken in assuring as complete a mixture as
possible between the silica and the salts. Each sample of gel and
salt were ground together for one half hour. In all, eight sets
were made - four sets at a 2% concentration of each salt, and four
sets at a 5% concentration of each salE.

Because the furnace used could not be controlled more accurately
than i_10°C., the temperature interval used throughout the heating
was 60°C. The initial temperature was 600°C and rose in five incre-
ments of 60°C each to 900°C., the maximum temperature reached in all
cases. The mixtures of silica and salt were heated in covered plati-
num crucibles. Platinum was used because of its inactivity toward
the salts and its high melting point, 1769.3°C.

Zielke (1950) stated that upon washing his heated samples, very
little, if any, salt remained. He concluded that much of the salt
must have escaped during the heating period. Because of this, he
heated his sample of CaCl2 + Si02 in a closed copper tube. It should
be pointed out that the author found that it was necessary to wash
his heated samples three and four times before the salt was complete-
1y washed away. Because of this, it was decided that no special
heating receptacle was necessary and that the covered platinum cruci-
bles would be sufficient for this work.

A ninth set was made containing a 2% concentration of each of

the four salts and these were heated at 100°C intervals.

14

Each sample within the nine sets were heated for two hour durations.
At the end of heating each set, the samples were washed with de-
ionized water until free from the salt.

Each sample was weighed before heating and after heating. The
loss of weight, as shown in Table V can be attributed to loss of

H20 and a small loss of the salt.

 

 

 

 

 

Table V
Sample Temperature Weight before Weight after %_Loss
SQ heating;(gms) heatiggg(gms)

l) 2%KBr a. 600 0.3437 0.3408 8.5
b. 660 0.3383 0.3346 11.0

c. 770 0.3384 0.3347 11.0

d. 780 0.3377 0.3334 12.6

e. 840 0.3385 0.3336 13.1

f. 900 0.4650 0.3601 13.3

2) 2%LiCl a. 600 0.3392 0.3110 8.4
b. 660 0.3412 0.3040 10.9

c. 720 0.3396 0.3026 10.9

d. 780 0.4462 0.2934 12.8

e. 840 0.3317 0.2851 14.1

f. 900 0.3899 0.3370 13.6

3) 2%NaCl a. 600 0.3387 0.3109 8.3
b. 660 0.3414 0.3185 6.7

c. 720 0.3386 0.3116 8.0

d. 780 0.3402 0.3023 11.1

e. 840 0.3353 0.2944 12.2

f. 900 0.3850 0.3352 12.4

4) 2%CaCl a. 600 0.3395 0.3143 7.5
b. 660 0.3374 0.3087 9.5

c. 720 0.3414 0.3061 10.3

d. 780 0.3437 0.3086 10.3

e. 840 0.3383 0.3007 11.1

f. 900 0.3786 0.3341 11.8

15

Table V (con't.)

 

 

 

 

 

Sample Temperature Weight before Weight after %_Loss
39, heating (gms), heating (gms)

5) 5%KBr a. 600 0.3600 0.3476 3.5
b. 660 0.3548 0.3310 6.8

c. 720 0.3504 0.3056 12.8

d. 780 0.3546 0.3030 14.6

e. 840 0.3580 0.3080 14.0

f. 900 0.3610 0.3100 14.2

6) 5%LiC1 a. 600 0.3496 0.3115 10.1
b. 660 0.3506 0.3114 11.2

c. 720 0.3492 0.3030 13.3

d. 780 0.3514 0.2966 15.6

e. 840 0.3498 0.2954 15.4

f. 900 0.3887 0.3275 15.7

7) 5%NaC1 a. 600 0.3647 0.3401 6.8
b. 660 0.3627 0.3357 8.5

c. 720 0.4678 0.3320 9.8

d. 780 0.3650 0.3114 14.7

e. 840 0.3670 0.2976 19.1

f. 900 0.4432 0.3762 14.9

8) 5%CaC12 a. 600 0.3656 0.3524 8.7
b. 660 0.3577 0.3544 9.1

c. 720 0.3606 0.3574 8.9

d. 780 0.3628 0.3583 12.3

e. 840 0.3633 0.3585 13.1

f. 900 0.5244 0.5182 11.9

As each set was completed, samples were prepared for analysis
using x-ray powder methods. The samples were ground to a uniform
size--approximately 200 to 300 mesh.

There are essentially three methods of preparing a powder mount.
(Azaroff, 1958)

1. By fashioning the sample itself into a
suitable shape,

2. By coating the outside of a thin fiber
with the specimen powder,

3. By placing the powder inside a capillary
holder.

16

Of the three methods, the first one was used in preparing the
specimens for this work. It was the most suitable, because it in-
troduced no material other than the sample in the path of the x-ray
beam. The use of a clear nail polish was necessary to act as a
suitable binder for the powder. A few milligrams of the powder were
mixed with the polish until a homogenous paste was obtained. A uni-
formly shaped rod was formed by rolling the sample between two pieces
of ground glass. The thickness of the rod was approximately 0.5 mm.
and 10mm. in length.

A Debye-Scherrer type camera was used to make the x-ray photo-
graphs. The film was mounted according to the Straumanis-Ievins
method. The diameter of the camera used was 114.6 mm. For this dia-
meter, the conversion factor from millimeters distance on the film
to angle of arc is 10 per mm. The film used was the NO-SCREEN MEDICAL
x-ray SAFETY film. The width of the film is 1 3/8 inches. The
development of film followed a set procedure: a period of 4 minutes
in the developer, 30 seconds in the stop and 10 minutes in the fixer
at 70°C. An exposure time of 5.2 hours was found to give the best
picture.

The target element used was copper. Copper produces K<x_ radia-
tion of 1.5418 (Ko( = 1/3 (Ktx2 + 2K 0(1) ). The K0( 2 has a wave
length of 1.54050. The K; radiation is quite short, being only

1.39217. The filter element used was nickel.

17

As can be seen from Figure 3, RM 1 and K (x 2 have wave

lengths which are quite close.

Figure 3_

 

 

 

 

(Azaroff, 1958, p. 33)
Because of this it is customary to group these together and to
assign to the combination a wave length which is the intensity--
weighted mean of the two wave lengths. Namely K(X_= 1/3
(Ko( 2 + 2K «1). The nickel filter was used to absorb the K/

radiation.

X-RAY DIFFRACTION THEORY
In the crystalline state, the atoms are arranged in patterns
which are characterized by periodic repetition in three directions.
A consideration of repetition in two directions will enhance the
understanding of 3 dimensional repetition. First, consider a single
geometric point (A). This point can be translated in two directions,

say t and t2 (Figure 4a). Continued translation of this point

1

(Figure 4b) generates an infinite collection of points.

Figure 4_

a b
t2 ' -——————9/f
AIZZ:_____, A ZZ:::::::_+.________,
t1

This collection of geometrical points produce what is called a lattice
net. If the pattern is repeated in three directions, a space lattice
is generated. In a crystal, where the pattern is a group of atoms not
points, the repetition defined by the three conjugate translations is
called a crystal structure. The periodic repetition of this pattern in
three directions produces an identical atom at each point of the lattice.
This simple pattern can be termed a lattice array of atoms. A single
crystal can be regarded as several lattice arrays of atoms.

The lattice array of atoms can be regarded as an infinite stack of
parallel, equally spaced planes. Figure 5 shows that the path length

from the incoming wave front, to the plane, and to the scattered wave

front is longer in the case of the lower plane.

18

19

Figure 5

 

 

 

(Azaroff, 1958, p. 8)
The greater path difference,2§., is equal to ABC.

AS = ABC

2A8 (AB = BC. ABC = ZAB)

AB = dhkl Sin 9

The total path difference is

AS = 2AB

= Zdhkl Sin 0
If both of these planes are to scatter in phase, the path difference,
.43, must be an integral number of wave lengths. That is, n)\ , where
n is an integer. Therefore, the conditions for scattering-in-phase is
nx= A)
This condition for scattering-in-phase is known as Bragg's

law. The positions of x-ray reflections on the film depend on
the shape and type of the crystal unit cell and the relative in-

tensities of these reflections depend on the arrangement of the

20

atoms with the cell.
The principles involved in the production of a powder diagram

can be shown in the simplified arrangement shown in Figure 6.

Figure 6_

photographic film

    
 

 

x-ray beam

¥

.T

sample

 

(Azaroff, 1958, p.13)

21

By using a cylindrical type of film arrangement, a much greater
range of 20 can be recorded. As shown in Figure 7, the distance

S between similar arcs corresponds to 46.

Figurell

 

 

 

 

 

 

 

 

(Azaroff, 1958, page 25)

S = R - 49
or 9 = S/4R
49 = S/R
0 = (1/4R)S radians
0 = (180ffl' ' iR)S ; (R = 180/1T mm. in radians)
9 = (180/TT ' k ' 771180)S = s/4 degrees

22

Once the arc length S is determined, Sin 6 can immediately
be found, and from this, d, can be found by a simple calculation.
When the d spacings have been calculated from several lines on
the powder film, the lines are arranged according to their rela-
tive intensities. Once this is done, the INDEX TO THE POWDER DATA
FILE (ASTM) can be used to identify the mineral. In the case of
this paper, it was used as a means of identifying the various

polymorphs of silica.

Table VI

X-RAY DATA

most intense lines

4.26

1.85

2.49

2.53

1.82

1.57

2.85

1.64

TRIDYMITE 4 (EXPERIMENTAL D --SPAC INGS )

QUARTZ
o( 3.34
f 3.42
CRISTOBALITE
OK 4.04
,5 4.15
4.30
4.30
4.39
KEATITE
3.42
COESITE
3.099
srrsuovrrs
2.959
SILICON
3.14

4.08

3.81

4.12

3.33

2.722

1.530

1.92

3.81

4.08

3.73

3.11

2.705

1.981

1.64

23

INTERPRETATION OF X-RAY POWDER PHOTOGRAPHS
Set I_ SiO2 + 2%KBr
a) ‘QQQPC
The most intense line on the photograph is that of alpha
tridymite. In all, two lines of alpha tridymite are present,
4.29 and 3.82.. Along with this, alpha cristobalite is also pre-
sent. The lines measured show the most intense line of cristoba-
lite, 4.04, and the second most intense line of cristobalite, 2.45.
The amorphous silica is present also as is indicated by a
wide diffuse halo.
b) gggfc
Two lines are evident at this temperature. These lines
are found to be that of alpha cristobalite. Amorphous silica is
also present.
c) 1299c
The three most intense lines of alpha cristobalite are
present at this temperature. Along with this polymorph, there
appears another line which coincides with a line for keatite (?).

A wide diffuse halo is again present showing the presence of

amorphous silica.

d) 780°C
Only one line can be measured on this photograph. This

line is the most intense line of alpha cristobalite. Again

amorphous silica was detected. No other polymorph is present.

24

mi

25

e) 8499C
All three lines of alpha cristobalite are present along
with the halo for amorphous silica.
f) _9_99_°c
Alpha tridymite is present with alpha cristobalite.

Amorphous silica is also present.

26

Set 2 SiO

—— 2

+ 2%LiC1
a) 999°C
One line is present and is the strongest line of alpha
cristobalite. A strong halo of amorphous silica is also present.
b) 999°C
Alpha quartz is present in this sample as all three lines
are present. Amorphous silica is again present. No other poly-
morph is indicated.
c) 299°C
Alpha quartz is again present. There is no evidence of
either cristobalite or tridymite. Amorphous silica is present.
d) 7_8_0_°c
In this film, the two strongest lines of alpha cristoba-
lite are present. With the exception of amorphous silica, there
is no other polymorph or silica present.
e) 999°C
For the first time, Beta cristobalite is present. TWO of
the three most intense lines of beta cristobalite are present.
Amorphous silica is also present.
f) _9_9_g°c
The last sample of 900°C shows all three lines for alpha

tridymite to be present. Amorphous silica is also evident.

27
933.9_ Si02 + 2%NaC1
a) 999°C
This sample has no sign of any of the three polymorphs
of silica. Only a strong diffuse halo is present, indicating
an amorphous substance.
b) 969°C
Two lines are present. These are the two most intense
lines of alpha cristobalite. Amorphous silica is also present.
c) 199°C
Four line determinations are made of this sample. It
was found that the three most intense lines for alpha cristoba-
lite are present. Also, the line for keatite is evident d=
3.13. (7) The diffuse halo for amorphous silica is present.
d) 121390
Three lines were measured. Two lines belong to the
alpha cristobalite polymorph of silica. The third line appears
again to belong to the polymorph keatite d= 3.33 (?). Amor-
phous silica is again present. The 3.33 may also be that of
alpha quartz.
e) 999°C
Two lines of alpha cristobalite are present being the
two most intense. Also, the most intense line of alpha tridy-
mite is present. Amorphous silica is also present.
f) 9_og°c
A total of four lines were measured. The d spacings

for alpha tridymite are present in three lines. The d spacing

28

of alpha cristobalite is present in the fourth line. Amorphous

silica is once again present.

29

Set 9_ SiO2 + 2%Ca012
a) _6_09°C
There is no crystallinity present in this sample. The
amorphous silica halo is again present.
b) 999°C
One line is present and it is determined that it is that
of alpha cristobalite. Also, the amorphous silica is present.
c) 199°C
Three lines of alpha cristobalite are present. The line
for keatite is again present along with the halo of amorphous
silica.
d) 299°C
Alpha cristobalite is present in the sample. The three
most intense lines are present. Another line measured appears to
belong to the most intense line of coesite (777). Here again,
amorphous silica is present as the diffuse halo.
e) 999°C
After measuring the three most intense lines on the pow-
der photograph, it is determined that alpha cristobalite is
present. Two other lines are measured and one appears to be the
polymorph keatite (7). The halo is present indicating the pre-
sence of amorphous silica.
f) MOC
Three strong lines of alpha cristobalite were measured.

Again keatite appears (7) to be present. Amorphous silica is

still present in the sample.

30

§9£_9_ 8102 + 5%KBr
a) 991°C
Two lines were measured and the calculation of the d-
.spacings show the alpha cristobalite polymorph to be present.
Amorphous silica is also present.
b) 999°C
Alpha cristobalite is also present in this sample. No
other polymorphs are present. Amorphous silica is present.
c) 199°C
The three most intense lines of alpha cristobalite are
present. Also, there appears again the line for the polymorph
keatite (7). The presence of amorphous silica is again noted.
d) 299°C
The only polymorph present in this sample is alpha
cristobalite. Amorphous silica is again present.
e) 999°C
Although the most intense line of alpha cristobalite is
present, alpha tridymite seems to be more evident in this sample.
The halo of amorphous silica is also present.
f) 39°C
Again the most intense line of alpha cristobalite is pre-
sent. Alpha tridymite is again more evident. Amorphous silica

is again present.

31

§9£_9_ 8102 + 5%LiC1
a) 999°C
The three most intense lines of alpha quartz are deter-
mined from the d-spacings. No other polymorph of silica is
present. The amorphous silica is again present.
b) 96_0°C
Alpha quartz is again the only polymorph of silica pre-
_sent. The diffuse halo for silica is present.
c) 299°C
Amorphous silica is present. The only polymorph of
silica is again alpha quartz.
d) 199°C
Two polymorphs exist in this sample along with the amor-
phous silica. One is alpha quartz, while the other is alpha
tridymite. No cristobalite is present.
e) 999°C
The most intense lines of alpha cristobalite are present.
One line of quartz is also present. Amorphous silica is again
present.
f) 9_09°C
Alpha cristobalite is determined by calculating three
lines. These are the three most intense lines on the powder
photograph. Again amorphous silica is present. Along with the
amorphous silica and alpha cristobalite, a line of keatite (7)

is again found.

32

SiO2 + 5%NaC1

a) 999°C

m
(D
n
[\I

Only alpha cristobalite is found in this sample. The
two most intense lines are present. Again amorphous silica is
found to be present.

b) 999°C

One line of keatite (7) is again found. Amorphous si-
lica is present. The three most intense lines of alpha cristo-
balite are also present in this sample.

c) 999°C

The diffuse halo of amorphous silica is again noted.
Calculations of the three most intense lines on the film show
alpha cristobalite to be present again. Also, the one line for
keatite (7) is present.

d) 7_80_°C

Alpha cristobalite is present in this sample. Amorphous
silica is shown again by the diffuse halo" The most intense line
of alpha tridymite is also present. Again one line of the poly-
morph keatite is present.

e) 999°C

Two polymorphs of silica are present in this sample

along with amorphous silica. Both alpha cristobalite and alpha

tridymite are present.

33

f) 900°C
Again the amorphous silica is present. Alpha tridymite
is represented by its three most intense lines. Alpha cristoba-

lite is present also in this sample.

34

§99_9_ SiO2 + 5%Ca012
a) 999°C
Only amorphous silica is present in this sample.
b) 96_0°C
No polymorphs of silica are found in this sample. Only
the amorphous silica exists.
c) 999°C
Amorphous silica is in this sample also. The three
most intense lines of alpha cristobalite are found to be present.
There also is a slight line which is interpreted to be that of
alpha quartz.
d) 199°C
Alpha cristobalite is again present. It is determined
from its three most intense lines which are present. Again,
there is the presence of amorphous silica. One line of keatite
(7) is also present.
e) 999°C
The polymorph keatite (7) again is represented by one
line along with the diffuse halo for amorphous silica. Three
intense lines for alpha cristobalite are found to be present.
f) 999°C
This sample also contains alpha cristobalite represented
by its three most intense lines. Keatite (7) and amorphous sili-

ca are also present.

35

U)

('D

n-
lxo

3102 + 2% (Ca012 + NaCl + 1101 + KBr)

a) 600°C

Alpha cristobalite is present. Amorphous silica is
also present. One list of quartz is observed. Also, there
appears to be a line for keatite (7).

b) 199°C

Amorphous silica is again present. One line of quartz
is present. There are three distinct lines of alpha cristoba-
lite present.

c) 999°C

Two intense lines of alpha cristobalite are present
along with the amorphous silica. Also, three lines of alpha
tridymite are present.

d) 999°C

Alpha tridymite is indicated by three strong lines.

Alpha cristobalite is also represented. The diffuse halo of

the amorphous silica is also present.

Set

600

660

720

780

840

900

0(ITIDYMITE
and
CKCRISTOBALITE

CKCRISTOBALITE

CKCRISTOBALITE
and
7KEATITE

0(CRISTOBALITE

dCRISTOBALITE

0(TRIDYMITE
and

o(CRISTOBALITE

36

Table VII

II

OKCRISTOBALITE

«QUARTZ.

CKQUARTZ

CKCRISTOBALITE

[CRISTOBALITE

TRIDYMITE

III IV

AMORPHOUS AMORPHOUS

“CRISTOBALITE 0(CRISTOBALITE

CXCRISTOBALITE 0(CRISTOBALITE
and and
7KEATITE 7KEATITE

CKCRISTOBALITE C(CRISTOBALITE

and and
KEATITE 7COESITE
or
QUARTZ

cx CRISTOBALITE 0(CRISTOBALITE
and and
CL TRIDYMITE 7KEATITE

o( TRIDYMITE OLCRISTOBALITE
and and
0(CRISTOBALITE 7KEATITE

Set

600

660

720

780

840

900

37_

Table VII (con't.)

0(CRISTOBALITE

0(CRIS TOBALITE

“CRIS TOBALITE
and

7 KEATITE

KCRISTOBALITE

(X TRIDYMITE
and
0(CRIS TOBALITE

0LTRIDYMITE
and
(XCRIS TOBALITE

VI VI I VI I I
(X QUARTZ (XCRIS TOBAL ITE AWRPHOUS
(XCRIS TOBAL ITE
0(QUARTZ and AMORPHOUS
7 KEATITE

0(CRIS TOBALITE OACRISTOBALITE
0(QUARTZ and and
7KEATITE OLQUARTZ

0( QUARTZ 0(CRIS TOBAL ITE OQCRIS TOBALITE

and and and
(X TRIDYMITE (XTRIDYMITE 7KEATITE
7KEATITE

(X CRIS TOBALITE OiCRIS TOBALITE 0(CRIS TOBALITE
and and and
ofi QUARTZ N TRIDYMITE 7KEATITE

NRISNBALIE M TRIDYMITE 0(CRIS TOBALITE
and and and
7KEATITE OLCRISTOBALITE 7KEATITE

38

Table VII (con't.)

Set
IX
°c

D(CRIS TOBALITE
and
600 o( QUARTZ
and
7KEATITE

660

7 00
7 20 (XCRIS TOBALITE
and
(X QUARTZ

800
780 OLTRIDYMITE
and
(XCRISTOBALITE

840

(X TRIDYMITE
900 and
oLCRISTOBALITE

39

The temperatures involved in this research were within the
stability range for D( quartz 573-87000 with the exception of
the 900°C temperature. This temperature was within the stabi-
lity range for tridymite (870-14700C).

The significance of the temperature range used becomes evi-
dent upon examination of Table VII. For the intervals from
600-84000, cristobalite is the prominent polymorph of silica
present. Again, quartz is stable in this range of temperatures.
Therefore, this suggests that cristobalite formed directly upon
heating with the catalysts. This is true, because quartz, being
the stable polymorph, could not invert to cristobalite at these
temperatures.

According to Robert B. Sosman (1927), there are six possible
permutations: l) quartz to tridymite, 2) quartz to cristobalite,
3) cristobalite to tridymite, 4) cristobalite to quartz, 5) tri-
dymite to cristobalite, 6) tridymite to quartz.

Refering to Table VIII, we can see that the three polymorphs
of silica can each undergo three transformations at varying tem-
peratures. These transformations (inversions) are of the recon-

structive type.

 

 

 

Table VIII
Temperature Rangg Quartz Tridygite Cristobalite
Below 870°C No change To cristo- To tridymite

balite
To quartz To quartz

40

Table VIII (con't.)

 

 

Temperature Range Quartz Tridygite Cristobalite
870-1470°C To cristo- No change To tridymite
balite

To tridymite

1470°C to melting To cristo- To cristo- No change
balite balite

Three principles which should be kept in mind when study-
ing the transformations of silica are:

1) Under conditions of constant and uniform
temperatures and pressure, a stable form
will show no tendency to change into an
unstable form.

2) Under conditions of constant temperatures
and pressure, an unstable form may change
either to the form stable or into some
other unstable form.

3) under conditions where the temperatures
are not constant or uniform, or are not
uniform over all parts of the material
under observation, almost any form or
combinations of forms, stable or unstable,
may be found.

Considering the sets with a 2% concentration (I-IV), cry-
stallization occurred in only set I and II at 600°C. The melt-
ing points of the salts used were low enough to fall in this
temperature. While those salts used in sets III and IV had
higher melting points than 600°C. However, this trend does not
follow through. At 660°C crystallization was evident. Although
not strongly evident, as only one line was present in III, IV

at the 660°C temperature. By 720°C, the crystallization was

strongly present in each set. Cristobalite is the dominant

41

polymorph of silica up to and including 840°C. At 900°C the domi-
nant polymorph is tridymite. However, some cristobalite is again
present. Tridymite could not invert to cristobalite for it is
stable at this temperature. Therefore, either tridymite formed
directly with cristobalite or most of the cristobalite inverted to
tridymite. The latter does not seem probable, in as much as the
change from one polymorphic form to another is extremely slow and
sluggish. This is shown by the existence of tridymite and cristo-
balite at ordinary temperature. (Mason, 1952)

In sets V-VIII, cristobalite is the dominant polymorph up to
and including the 720°C temperature. The amount of crystallization,
however, has increased at the lower temperature, with the exception
of set VIII, as is indicated by an increase in the number of lines
present. At the temperature of 780°C, both cristobalite and tridy-
mite are present. Neither of these are the stable polymorph at this
temperature. In set VI quartz is present with tridymite. Here the
tridymite may have formed first and inverted to quartz, or both quartz
and the unstable polymorph tridymite formed together. In set VII,
both unstable polymorphs of silica are present, cristobalite being
the more abundant. Again, either these polymorphs formed separately,
or cristobalite formed and some inverted to tridymite (the temperature
involved being more conducive to an inversion from cristobalite to
tridymite.) However, as stated previously, this does not seem to be
the case. Set VIII appears to consist of the polymorph cristobalite.

Only the polymorph keatite (7) appears to be present with cristobalite.

42

The presence of keatite was determined from one line measurement
(d=3.13) and should not be considered an accurate identification.
(Although the d- spacing 3.13 is evident in many samples.)

Set IX is found to contain cristobalite and some quartz up to
and including 700°C. In the temperature from BOO-900°C, tridymite
is the abundant polymorph with some cristobalite. At 800°C both
are unstable and they may form separately or cristobalite may in-
vert to tridymite. At 900°C tridymite is the stable polymorph and
would not invert to cristobalite. Therefore, cristobalite must
have formed independently and may have inverted in part to tridy-
mite. Again, the assumption that both polymorphs formed separately
may be more valid.

In all sets, it was noted, that amorphous silica exists.

While weighing the samples after they had been heated, it was
noticed that the higher temperature samples appeared to be more
heavy than the samples at lower temperatures. With this in mind, it
was decided to run a volume to weight analysis for two samples in
each set. These samples were 600°C and 900°C. Each sample was
ground in an agate mortar to a uniform size of approximately 200 -
300 mesh. An equal volume of each sample was taken. The volume was
calculated to be 0.2000 cc. Table IX gives the difference in weight

of each set.

43

Table IX

Sample 600°C weight 900°C weight weight increase

(gas) (gas) (ggs)
1 0.0152 0.0627 0.0475
2 0.0131 0.0720 0.0589
3 0.0142 0.0648 0.0506
4 0.0160 0.0478 0.0318
5 0.0146 0.0703 0.0557
6 0.0150 0.0597 0.0447
7 0.0288 0.0740 0.0452
8 0.0124 0.0347 0.0223

In each case, the 900°C sample was more dense than the 600°C
sample. This would be consistent with the increase in cry-
stallinity from 600-900°C, that was found in each set. The
atoms became more closely packed as the temperature increased

giving rise to observed increase in density.

SPECTROGRAPHIC ANALYSIS

Prior to the preparation of the samples for x-ray analyses,
it was noted that the samples were washed free from the salts. It
was decided to do a qualitative analyses of the samples to deter-
mine whether the salts had entered into the structure of the poly-
morphs. The samples were again washed with de-ionized water to
assure complete removal of the salts. A test for the salts after
the first washing showed the absence of any salts in the samples.
The samples were washed twice more and then allowed to dry at room
temperature for a 24 hour period.

Each sample was weighed and ground in an agate mortar to a
size of approximately 150-200 mesh. To each sample an equal amount
of spectrographic graphite was added. This mixture was then stirred
together in a glass vile and shaken together for five minutes, to
assure a uniform mixture.

Spectrographic carbon electrodes were used. A % inch outside
diameter rod was used and a shallow cavity approximately 0.3mm. deep
was drilled into the rod. The samples were placed in the holder and
packed tightly by pressing the sample into the holder.

The author desired to identify only one element in each sample.
It was therefore decided to record the Spectra of each of the salts
being sought. This would provide a rapid means of identifying the
salts if they were present in the sample. Also, mixtures of the
salts and silica gel were prepared as above and the spectra of these

were also recorded.

44

PROCEDURE
Before each salt was run, iron electrodes were used in the
desired wave length of each salt to produce an iron spectra for
purpose of orientation on the plate. Each salt was then run three
times for varying exposure times and per cent transmission. Table

X shows the wave lengths used along with times and per cent trans-

 

 

 

mission.
Table X

Sample Wave lgpgth (A0) 9_Transmission Time in seconds
1. Fe 4200-5200 80 5
2. CaClz 4200-5200 80 20
3. CaClz 4200-5200 90 20
4. CaClz 4200-5200 100 20
5. Fe 5700-6600 80 5
6. NaCl 5700-6600 80 20
7. NaCl 5700-6600 90 20
8. NaCl 5700-6600 100 15
9. Fe 6700-7600 80 5
10. LiCl 6700-7600 80 20
11. LiCl 6700-7600 90 20
12. LiCl 6700-7600 100 15
13. Fe 7600-8400 80 5
14. KBr 7600-8400 80 20
15. KBr 7600-8400 90 20
16. KBr 7600-8400 100 15

Eastman Kodak Spectrum Analysis Film #1 was used for recording
the spectrum. It was developed in D-19 developer, at 70°F. for 2%
minutes, stopped for 30 seconds and fixed for 8 minutes, washed and

dried.

45

46

The samples were arced under the following conditions.

Excitation: Low voltage continuous
Voltage: 250 volts

Current: 10 amps

Resistance: 0

Inductance: 50

Capacitance: 20 mfds.

Time: 20 seconds

Transmission: 80%

Slit width: 20

Each sample was run at the wave length of the salt sought after.

Each sample showed the distinct lines for Silicon and Carbon. Table

XI gives the results for each set.

11

III

IV

VI

VII

VIII

Table XI

W

No potassium could be found in any of the six
samples.

There appears to be a very slight trace of
lithium present. (7)

A trace of sodium appears in all samples.

No trace of calcium present.

No trace of potassium is present.

There might be a slight trace of lithium here. (7)
A trace of sodium is present.

No calcium is present.

47

The result of the spectrochemical analysis appears to be
consistent with a remark made by Brian Mason, (1952).

Cristobalite has been found with a consider-
able amount of sodium in interstitial solid
solution, and the sodium atoms presumably
stabilize the open structure of this mineral.

CONCLUSION

The immediate conclusion which was reached from this study was
that the introduction of various salts promoted the crystallization
of the amorphorous silica gel. This was demonstrated by the fact that
the silica gel when heated alone did not crystallize at a temperature
of 900°C. It was noted that crystallization of the silica gel took
place at or above the melting points of the salts used.

The high-temperature polymorph of silica, cristobalite, appears
to be the dominant crystallization product at nearly all temperatures
below 900°C. At 900°C. some of the cristobalite, which crystallized
first, has inverted to tridymite.

It appears that crystallization will occur at temperatures slight-
1y above the melting point of the salts involved. It is conceivable,
therefore, that either of the high-temperature polymorphs of silica
could be formed in nature at temperatures far lower than here-to-fore
expected. Perhaps the use of tridymite or cristobalite as geologic
thermometers is not wise.

The crystallization obtained in this research was conducted in a
dry state. Very little water was available to enter into the crystalli-
zation. In nature, however, there is an abundance of water which would
tend to lower the temperature at which these polymorphs might crystal-
lize. Concentrations of the salts may also be greater in nature than
those used in this research.

The samples were heated for a relatively short duration of time
(two hours). It is logical therefore to assume that at longer periods
of exposure to temperatures below 600°C., crystallization of either

tridymite of cristobalite would occur.

48

49

It was seen in the spectrographic analysis of the samples that
sodium was found to be present at all six temperature ranges in the
polymorph cristobalite. This is consistent with what is found in
nature. It appears that the sodium atoms stabilize the open struc-
ture of cristobalite and considerable amounts of sodium have been

found in interstitial solid solution with cristobalite.

SUGGESTIONS FOR FUTURE STUDY

In the situation where both tridymite and cristobalite exist
together, either tridymite formed directly with cristobalite or some
of the cristobalite inverted to tridymite. The change from one poly-
morphic form to another is rather slow and sluggish. In any future
research of this nature, the samples should be heated for longer
durations.

It is also suggested that greater concentrations of the various
salts be used. Concentrations of at least 10% should be tried. The
various salts should be mixed with each other.

In as much as this research was conducted in a dry state, perhaps
by heating the silica and salts in an atmosphere containing water vapor
the temperatures used in this research could be lowered.

Also, an attempt should be made to heat the samples under two or
three atmospheres of pressure.

By the use of a differential thermal analysis (D.T.A.) apparatus,
the various inversion points could be recorded. Also, by using this
apparatus, various pressures, temperatures and atmospheres could be
obtained.

If the temperature increment could be controlled more precisely,
then smaller increments should be used in the 600°C to 900°C range.
This would be the advantage of the D.T.A. apparatus. A constant record
of temperatures would be available and more discrete sampling could be

had.

50

LE

‘Il'p-V ‘i a; m .niw

 

#1.... gummulmflu -mA 'hh’m-l _h
ml". —-

. . ‘.‘h .
y._’,"’-, r. ‘

n-An p In;

Tl'd

.T. :m. oi Inuvuxx- -.
#4104; g:-

 

.-Ti ‘.’ ..

    
 

 

"ram. ’

V;;.—h§‘$.‘._*

- ~Lw.a—.—.-*_. ””043j'i"
’ -3 . _ I
H . . . . . _. q. - — .

BIBLIOGRAPHY

Ahrens, L. H., 1950, Spectrochemical Analysis, Addison-Wesley Press,
Inc.

Azaroff, Leonid V., and Buerger, Martin J., 1958, The Powder Method
in X-Ray Crystallogrgphy, McGraw-Hill Book Company, Inc.

 

Babor and Lehrman, 1957, Textbook of Chemistry

Berey, L. G., and Thompson, R. M., 1962, "X-Ray Powder Data for Ore
Minerals," Geologic Society of America Memoir 85, The Peacock
Atlas, New York

Brindley, George W. (Editor), 1957, Index to the X-RayfiPowder Data
File, American Society for Testing Materials

Carr, R. M. and Fyfe, W. S., 1958, Vol. 43, "Some Observations on
the Crystallization of Amorphous Silica," The American
Mineralogist

 

 

Corwin, J. F., Herzog, H. H., Own, G. E., Yalman, R. G. and Swinnerton,
A. C., 1953, 75:3933, Journal of the American Chemical Society

Frondel, Clifford, 1962, The System of Mineralogy of James Dwight Dana
and Edward Salisbury Dana, Seventh Edition, John Wiley and Sons

Henry, N. F. M., Lipson, H., and Wooster, W. A., 1960, The Interpreta-
tion of X-Ray Diffraction Photographs, MacMillan and Company,
Ltd., New York

 

Hill, V. G. and Roy, Rustum, "Silica Structure Studies, Pt. V., The
Variable Inversion in Cristobalite," Journal of the American
Ceramic Society, Volume 41, Part 12, pp. 532-537

Iler, R. K., 1955, The Colloid Chemistry of Silica and the Silicates,
Cornell University Press, New York

Krauskopk, Konrad, 1956, Geochimica et Cosmochimic Acta, 10:1-25,
Pergamon Press, New York

Lange, Norbert Adolph, 1956, Handbook of Chemistry, Handbook Publishers,
Inc.

Mason, Brian, 1952, Principles of Geochemistry, John Wiley and Sons,
New York, pp. 86-89

Smits, Benjamin, 1926, "Silica Gel," Ph.D Thesis, Michigan State
College

51

52

Sosman, Robert B., 1926, The Propgrties of Silica, American Chemical
Society Menograph Series, The Chemical Catalog Company, Inc.,
New York

Verduch, A. G., "Kinetics of Cristobalite Formation from Silica Acid,"
Journal of the American Ceramic Society, 41:427-432, Part 11

White, Donald E., Brannock, W. W., and Murata, K. J., "Silica in Hot-
Spring Waters," Geochimica et Cosmochimica Acta, 10:27-59

Zielke, Clyde W., 1950, "The Salt Catalyzed Degelation of Silica Gel,"
B.S. Thesis, University of Wisconsin

 

 

 

 

 

 

 

 

 

 

 

.1: i
I
Y t l.
. '2
. .
j.
, .
. . .
,. .» L .
L" . -
' \ \
.

 

min-.4...-

AM damn-g .31....4' .

3
n:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

—

 

 

Inst ‘33 1E

 

 

 

 

 

» . full . .1. 9 _.. I wiwr 1.1. I Mm 1.1 . I I i . Il.0.c.1.M.5 {III
\ x . . x mm. .wr .
.. a? a
\ . a...
. c ._

a o r. F

8 e 8

A
x. \ \\ I / t
,, f r f
/ \, / x (\ I \\ 4 \

 

 

 

 

        

Amoéfhéils

 

Amorphous

 

 

 

 

 

ISE 0m.

 

   

"1111111111111!fiiflflifllfiiflifliflfflifilES