Scholars' Mine
Doctoral Dissertations Student Theses and Dissertations
1971
Structure and properties of silver borate glasses
Edward Nashed Boulos
Follow this and additional works at: https://scholarsmine.mst.edu/doctoral_dissertations
Part of the Ceramic Materials Commons Department: Materials Science and Engineering
Recommended Citation Boulos, Edward Nashed, "Structure and properties of silver borate glasses" (1971). Doctoral Dissertations. 2280. https://scholarsmine.mst.edu/doctoral_dissertations/2280
This thesis is brought to you by Scholars' Mine, a service of the Missouri S&T Library and Learning Resources. This work is protected by U. S. Copyright Law. Unauthorized use including reproduction for redistribution requires the permission of the copyright holder. For more information, please contact [email protected]. STRUCTURE AND PROPERTIES OF SILVER BORATE GLASSES
by EDWARD NASHED BOULOS, 1941-
A DISSERTATION Presented to the Faculty of the Graduate School of the
UNIVERSITY OF MISSOURI - ROLLA
In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY in CERAMIC ENGINEERING 1971
Advisor ll
PUBLICATION THESIS OPTION
This thesis has been prepared in the style specified by the Journal of the American Ceramic Society. Pages 1-45 are submitted for publication in this journal. Appendix E, pages 78-85, was also submitted for publication in the same journal. Appendices A,B,C,D and F have been added for the purposes normal to thesis writing. iii
ABSTRACT
Clear glasses form in the system Ag o-B o up to about 2 2 3 35 mol.% (65 wt.%) Ag2o. Infrared absorption, thermal expansion and density data indicated an analogy to the Na o 2 B2o3 system. Pentaborate-triborate group pairs appear to be formed upon the addition of Ag2o to B2 o 3 up to 20 mol.% Ag2 o and diborate groups from 20 to 33 mol.% Ag2o. This inter pretation is supported by the comparison of the infrared absorption spectra of quenched and crystallized glasses.
One crystallization product ,Ag20.4B2o 3 , has been identified previously. A new compound starts to appear at 28 mol.% Ag2 o Silver is generally present as a network modifier like sodium. This was substantiated by the comparison of the molar volume of sodium and silver borate glasses. Above 27 mol.%
Ag o some atomic silver is assumed to be present. Below 2 15 mol.% Ag o exploratory studies indicate a two-phase struc 2 ture within an immiscibility gap. A low temperature internal friction peak in the glasses up to 28 mol.% Ag o corresponds with the alkali peak in other 2 glasses; a high temperature peak appearing in the 34 mol.%
Ag o glass is associated with the appearance of non-bridging 2 oxygen in the system. iv
ACKNOWLEDGEMENT
The author is deeply indebted to his advisor, Dr.
Norbert J. Kreidl whose capability and direction were major
factors in the accomplishment of this investigation.
Appreciation is also extended to Dr. Delbert E. Day for his help and valuable suggestions.
Thanks are also expressed to the Ceramic Engineering
Department, University of Missouri-Rolla for providing the equipments and facilities which made this research possible.
Gratitude is also extended to the Chairman of Chemistry
Department for his cooperation in allowing the use of the
Infrared Spectrophotometer.
The helpful discussion and assistance of Mr. M. Maklad
and Mr. J. Starling are also acknowledged. The author is indebted to the National Science Foundation
for financial support during part of his program.
Also, my deepest thanks go to my wife, Mervet, for her encouragement and patience during the undertaking of this
study. v
TABLE OF CONTENTS
Page
ABSTRACT...... iii
ACKNO~vLEDGEMEl'JTS. • • . . • . • ...... • ...... • ...... • . . . i v
LIST OF FIGURES...... vii
LIST OF TABLES. . . • ...... • • • . . . • . . . . • . . • . . . X
I. INTRODUCTION...... 1
II. EXPERIMENTAL...... 3
1. Sample Preparation ...... 3
2. X-Ray Diffraction Studies .....•..•....•.... 3
3. Thermal Expansion •.....••..••••••••.•..•..• 3
4 . Infrared Absorption .....•...•..•...... •... 4
5. Density and Holar Volume .•..•..•..•....•.•. 4
6 . Internal Friction ....•...... •...... 5
III. RESULTS AND DISCUSSION...... • . • . 6
1. Range of Glass Formation ...••.••.....•••••• 6
2. X-Ray Diffraction and DTA. • . . . . . • . • . • • • • . • • 8
3. Thermal Expansion. • . . . . • • • . • . . • • • ...... 11
4. Infrared Absorption ....•.•...••••..••.•••.. 18
5. Density, Molar Volume and Phase Separation ...... •...... •... 28
6. Internal Friction ..•••...... •..••...... 34
IV. CONCLUSIONS ••.•.•...... ·•····•·······•·•··•··· 39 41 V. REFERENCES ...... · . · · · · · · · · · · · · · · · · · · · · · 46 VI. APPENDICES ••••.... · •. · • · · • · · · · · · · · • · · · · • · · · · · · · A. X-Ray Diffraction ••...... •.....•.••.••••... 47
B. Thermal Expansion ••..••••.•.•..•..•..•...•. 53 vi
TABLE OF CONTENTS (continued)
Page C. Infrared Absorption...... 55
1. Theory ...... 55
2. Pellet Technique For Infrared Spectroscopy...... 57
3. Infrared Spectrophotometer ...... 58
4. Literature Review of Infrared Studies of Boric Oxide and Alkali Borate Glasses ...... 58
5 . Data...... 6 1
5.1. Semi-Quantitative Analysis ...... - 61
5.2. Infrared Absorption ...... 65
D. Density Measurements ...... 74
1. Theory...... 74
2. Procedure ...... 75
E. Mixed Cation Effects in Silver Borate Glasses...... 78
F. Internal Friction ...... 86
VII. VITA ...... · . · · .. · · · · · . · ..... · 88 vii
LIST OF FIGURES
Page
Figure No.
1 X-ray diffraction patterns of two devitri fied silver-borate glasses ...... •...... 9
2 DTA curves of two silver-borate glasses .... 10
3 Typical thermal expansion curves of silver- borate glasses ...... 13
4 Coefficients of thermal expansion for sil ver-borate glasses as a function of corn- position...... l4
5 Thermal expansion of lithium, silver, and sodium borate glasses in the 1:4 composi- tion range ...... 16
6 Transformation and softening temperatures, Tg and Ts, of silver-borate glasses as a function of composition ...... 17
7 Infrared spectra of sodium, lithium, and silver-borate glasses in the 1:4 composi- tion range...... • ...... 19
8 Infrared spectra of Ag 0.4B2o3 glass be- fore and after devitrificatlon ...... 20
9 Infrared spectra of Ag 0.2B 2o 3 glass be- fore and after devitrificatlon ...... 22
10 Infrared spectra of B203 glass compared with low silver content glass ...... 23
11 Infrared spectra of sodium and silver borate glasses before and after devitrifi- cation ...... 24
12 Infrared spectra of sodium and silver borate glasses in the 1:2 composition range ...... · · . · · · · · · · · · · 12
13 The borate groups ...... •...... 26
14 Variation of density and molar volume of silver and sodium borate glasses as a function of cornposi tion ...... •. 29 viii
LIST OF FIGURES (continued)
Figure No. Page
15 Variation of denstiy with composition in wt.% for silver-borate glasses...... 31
16 Transmission electron micrograph of glass containing 13 mol.% Ag mark indicates 2o, 5000 °A...... 33
17 Internal friction for XAg 0-(lOO-X) B 2 2 o glasses; frequency 0.5 Hz ...... 3 35
APPENDIX A
l Effect of increasing Ag20 on x-ray diffrac tion patterns of XAg 2o. (l00-X)B2o 3 glasses. 48 2 X-ray diffraction patterns of devitrified silver-borate glasses...... 49
3 x-ray diffraction patterns of devitrified silver-borate glasses ...... 50
4 Comparison of the 'd' values of Ag 20.4B2o 3 in single crystals and devitrified glass ... 51
APPENDIX B 1 Thermal expansion curves of silver borate glasses...... 54
APPENDIX C
l Effect of increasing Ag2o on the intensity of B04 and Ag 2o bands. Lx Ag 20· (100-x) .B o glasses] ...... ········ 62 2 3 2 Effect of increasing Ag 2o on the four co-ordination of boron. [xAg 2o· (100-x) n o glasses] ...... ··.········· 63 2 3 3 Effect of increasing Ag2 o on the intensity of the B0 bands. [xAg 0· (l00-x)B 4 2 2o 3 qlasses] ...... 64 ix
LIST OF FIGURES (continued)
Figure No. Page
4 Effect of increasing Ag o on the 2 infrared transmittance of XAg o. 2 (100-X) B o glasses ...... 2 3 66 5 Effect of increasing Ag o on the 2 infrared transmittance of XAg o. 2 ( 100-X) B o glasses •...... 67 2 3 6 Effect of increasing Ag o on the infrared 2 transmittance of XAg o. (l00-X)B o devit- 2 2 3 rified glasses ...... 68
7 Effect of increasing Ag 0 on the infrared transmittance of XAg o.tlOO-X)B o devit- 2 2 3 r i fie d g 1 as s e s ...... 69
Effect of increasing Ag on the infrared 8 2o transmittance of XAg (l00-X)B devit- 2o. 2 o 3 rified glasses ...... 70
9 Infrared transmittance curves of Ag2 o and H BO crystals ...... 71 3 3 APPENDIX E
l Internal friction curves for (l-x)Ag2 o. xNa o.4B o glasses; frequency 0.5 Hz ...... 82 2 2 3 2 Internal friction curves for (1-x) Na2o. xAg 0.4B o glasses; frequency 0.5 Hz ...... 83 2 2 3 3 Internal friction curves for (l-x)Ag2o. xcu 0.4B o glasses; frequency 0.5 Hz ...... 84 2 2 3 X
LIST OF TABLES Page Table I. Composition and color of silver-borate glasses...... 7 Table II. Coefficients of thermal expansion, transformation and softening temperatures, density and molar volume data...... 12
Appendix F Table I. Temperature and height above back-ground for low and high temperatures oeaks,in the xAg (100-x)B glasses ...... 87 2o. 2o 3 1
I. INTRODUCTION
Silver is a constituent of special glasses which have unusual properties and applications; knowledge of these glasses is limited to studies or researches motivated by special objectives. However, because of the documented or potential function of silver in photo-chromic, photo- sensitive, dosimeter and semiconductor glasses, extensive studies have been made on glasses in which traces of silver have been incorporated. The state of silver in these glasses has been studied by several researchers. Weyl(l) stated that if a small amount of silver oxide (0.1% Ag 2o) is introduced into silicate glass, Ag+ ions participate in the glass struc ture in a fashion similar to that of Na+ ions due to the similarity of their charge and size. Silver however, has a much lower ionization energy than sodium and, therefore, is more easily reduced. Electron spin resonance experi ments(2'3'4) have shown that the silver ion, Ag+, is capable of trapping both types of charge carriers, electrons as well . . h 1 . A+,- +,+ as holes, becom1ng e1ther t e neutra spec1es g or Ag respectively. (The notations +,- and +,+ symbolize the possi bility that the environment of Ag+ is maintained in the electronic process). In this respect, silver does not seem to behave very differently from many transition ions, for instance, iron or cerium. 2 5 6 Previous studies on the solubility of silver in glasses( ' ' ) showed dependence upon oxygen pressure, melting temperature 2 and composition of the base glass. It is difficult to melt silicate glasses containing appreciable amounts of silver due to their high melting temperature. On the other hand, substantial amounts of silver may be diffused into silicate glasses, in which case silver presumably replaces sodium. Westermann(?) obtained a silver aluminosilicate with 22%
Ag by melting kaolin with silver in an oxidizing atmos- 2o phere. Also, relatively large amounts of silver (Ag/P ratio of 1), can be introduced during the melting of phosphate glasses (S).
The purpose of this work was to study the structure and properties of silver borate glasses containing significant amounts of silver. 3
II. EXPERIMENTAL
(1) Sample Preparation:
The raw materials were certified reagent grade: silver nitrate, boric acid, sodium carbonate, and lithium carbonate.
The weight of each batch ranged from 40-70 grams. The batch materials were carefully ground, mixed and then melted in alumina crucibles for 60-80 minutes in an electric furnace
under a normal atmosphere. The alumina crucible was pre heated to temperatures ranged between 800 to 1000°C, small charges were made at a time and, to assure homogeneity,
the melt was stirred several times prior to each charge and after the completion of charging. Glasses were annealed at
temperatures ranging from 350-450°C and stored in black walled desiccators.
(2) X-Ray Diffraction Studies: A General Electric XRD-5 diffractometer with a copper target and nickel filter was used to obtain the diffraction patterns of powdered samples of both the glasses and the corresponding devitrified glasses. To induce crystallization the glass samples were heated at temperatures ranging from 500-580°C for a period of 100 hours. The temperature of crystallization was determined by
DTA technique.
(3) Thermal Expansion: The thermal expansion was determined from room temperature to the softening temperature of each glass.
An Orton Automatic Recording Dilatometer, with HR-100 x-y recorder, was used for this purpose. Rods 2(+0.002) inches 4 long with parallel ends were prepared, annealed, and then used for these measurements. The heating rate was l°C per minute. The method given by Green(g) for the evaluation of the coefficient of thermal expansion, transformation and softening temperatures was used. The parameters are evaluated from plots of the thermal expansion versus temperature as in fig. 3 where the slope of the line AB determines the coefficient of thermal expansion, the intersection AB-BC gives the glass transformation temperature at point B; and point c is the softening temperature, which is the temperature at which the rod shows first signs of sagging.
(4) Infrared Absorption: Infrared absorption of all glasses and corresponding devitrified samples was measured in the 2.5~ to 25~ range using a 337 Perkin-Elmer infrared spectrometer. Samples were prepared using both the film and KBr pellet techniques. Com- parison was made between the spectra of the silver borate glasses and lithium or sodium borate glasses of the same composition ratio, as well as between the glass and the corres- pending devitrified sample.
(5) Density and Molar Volume: Density was measured using the suspended-weight method based on the Archimedes principle. Xylene -- whose density had been carefully measured -- was used as the immersion liquid.
The samples were weighed on a Sartorius semi-micro balance enclosed in a constant temperature box, the tenperature being controlled at 25.6°C for all samples, with an accuracy of + 0.1°C. 5
The samples were suspended by a fine tungsten wire 0.08 mm. ln diameter. Densities were computed from equations described in detail elsewhere(lO). Corrections were made for the immersed and unirnrnersed parts of the suspended wire in xylene and air respectively as well as for the force due to surface tension. Three well annealed, bubble-free samples from different batches of the same composition were measured sep- arately and the average of their densities taken. The maximum error in each measurement was + 0.0005 g/cm 3 . [see Appendix D]
Density data was used to calculate the molar volume of all glass samples by using the molecular weight of the glasses in terms of x Ag (1-x) B . 2o. 2o 3 (6) Internal Friction: Internal friction was measured from -180°C to 250°C using an inverted torsion pendulum operating at about 0.5
Ilz. The apparatus and theory used in the internal friction 2 and activation energy measurements are explained elsewhere(l ).
Fibers approximately 0.5 rnrn. in diameter were drawn from the melt, annealed at the proper annealing temperature for 20 minutes, then cooled slowly to room temperature. 6
III. RESULTS AND DISCUSSION
(1) Range of Glass Formation:
Clear glasses were obtained in the system, xAg 2 o. (100-x)B , were x ranged from 0 to 35 mol.%. Metallic 2 o 3 silver was precipitated in the crucible for glasses containing more than 25 mol.%; 35 mol.% Ag was the maximum amount of 2 o silver incorporated into the glass; introducing higher per- centage into the batch always resulted in precipitation of metallic silver. This metallic silver accounts for the difference between batch and analytical compositions in the high silver containing glasses shown in Table I. However, the analysis does not account for the water content which, in
. 9- ( 11) b orate glasses, lS on the order of 0.1 to 0.5 wt.o. .
The glasses were hygroscopic, especially those rich in B2o3 , Those rich in Ag became becoming cloudy at the surface. 2o dark and developed a mirror-like layer on the surface after exposure to light for several hours. This mirror-like layer . . . t (2, 13) d 1 t . 1 h as been reported by var1ous 1nvest1ga ors an e ec r1ca conductivity measurements have shown that the films are 14 15 metallic in nature( ' ). However, glasses kept in black- walled desiccators showed no change in color for several months. All glasses so prepared contained no crystals detectable by x-ray diffraction. The diffuse x-ray diffraction bands for all glasses look alike. [see Appendix A] 7
TABLE I
Glass Ag?O Content Analysis Color No. Mol.% I'>Jt.% Mol.%
1 1 3.3 Colorless 2 2 6.4 " 3 5 14.9 4.88 II 4 7 20.0 " 5 10 27.0 "
6 13 33.2 II 7 15 37.0 Light Yellow
8 17 40.5 II 9 19 43.9 Golden Yellow 10 20 45.4 19.16 II
11 22 48.4 II
12 25 52.6 24.56 II 13 27 55.2 Brownish Red 14 30 58.8 28.61 "
15 33 62.1 30.21 II
16 36 65.2 34.07 Dark Brown
17 40 68.9 34.29 II
TABLE I. Composition and Color of Silver-Borate Glasses 8
(2) X-Ray Diffraction and DTA:
The x-ray diffraction patterns of devitrified samples in the composition range 5 to 27 mol.% Ag 2o showed crystals of Ag 0·4B as the predominant phase (fig. la). This is 2 2o 3 the only compound in the silver-borate system identified in the literature: it was first recognized by De-Carli(l6 ) and then by Willis and Hennessy(l 5 ) from studies on partial molar free energy and heat content curves; single crystals of this compound were prepared and identified by Krogh-Moe(l 7 ) using x-ray techniques.
Glasses containing 28 to 34 mol.% Ag2o showed a new type of diffraction pattern after heat treatment (fig. lb).
This pattern is characterized by peaks at the following 28 angles in the order of decreasing intensity: 17.9, 27.6, 31.2,
36.4, 35.6, 21.3, 29.8, 32.7, 38.7, 43.4. It was not possible to identify the corresponding compound by using the ASTH
cards. Similar diborate compounds have been found in both lithium and sodium borate
systems and infrared results (section III 4) indicate that
lithium and silver borate glasses in the 1:2 composition range,
contain the diborate network. The DTA curves for two glasses containing 20 and 28 mol.% Ag o are shown in fig. 2 as curves a and b respectively. 2 The exothermic peak beginning at 520°C with a maximum at
555°C (curve a) could correspond to the formation of the Ag 2o. 4B o compound. Curve b shows a formation temperature of 2 3 about 580°C, with an exothermic peak at 620°C, which could be
attributed to the compound Ag 20·2B2o 3 postulated by x-ray. > t- (/)- z w MOL. 0/o Ag20 t z- (b) 30 (GLASS NQ 15) 0 w 0 ~ 0 () w (a) 20 (GLASS NQ 10) ~
70 60 50 40 30 20 10 DEGREES, 2 9 Figure 1. X-ray diffraction patterns of two devitrified silver-borate glasses. ~ MOL. 0/o Ag20 u- ~ 0::: w :::c ~ 0 Xw l
100 200 300 400 500 600 700 800
TEMPERATURE, °C
Figure 2. DTA curves of two silver - borate glasses. f-J 0 11
The x-ray radial distribution for this system compared with the synthetic ideal peak is under investigation(lB).
This study on a glass containing 20 mol.% Ag showed that 2o
0 the oxygen-oxygen distances are 2.3 and 3.0 A , while the
0 cation-cation distances are 5.4 and 6.9 A •
(3) Thermal Expansion:
Figure 3 shows typical thermal expansion curves for four silver borate glasses. Similar curves were obtained for the other glasses in this series ranging from 0 to 35 rna 1.% Ag Table II contains the coefficients of expansion 2o. as well as the dilatornetric transformation and softening
(T , T ) • temperatures g s Like those of other alkali borate glasses, the coeffic- ients of thermal expansion of the silver borate glasses showed a distinct minimum at about 15 mol.% Ag 2o (fig. 4). Silver borate glasses, in general, show coefficients of expansion higher than those of the corresponding lithium borate glasses but lower than those of the corresponding sodium borate glasses. A plot of the thermal expansion versus temperature of the lithium, sodium, and silver borate glasses containing 20 mol.% is shown in fig. 5. The 6 6 thermal expansion coefficients are 7.5 x 10- , 9.0 x 10- and
8.2 x 10-6 crn/crn/°C for lithium, sodium, and silver borate
Data for the B o glass obtained here glasses, respectively. 2 3 using the rod-dilation method agree with the results of
Green( 9 ), Sarnsoen(l9 ), and Jankal( 2 0) who used other methods. 12
TABLE II
Glass Coeff. of Trans. Sof. Density Molar Vol. No. therm. Exp. temp. Temp. g/cm3 cm3 cm/cm/°C T °C Ts oc g -+ 0.0005 -6 B203 16.0xl0 210 260 1.8460 37.71 6 2 15.2xlo- 220 275 1.9094 37.31
3 2.1459 36.22 -6 4 10.7xl0 260 300 2.3216 34.87 -6 5 9.0xl0 295 330 2.5300 33.92
6 2.8245 32.11
8 3.1193 31.80 -6 10 8.2xl0 370 400 3.3112 30.41
11 3.4022 30.90 -6 12 9.5xl0 370 395 3.6516 29.96
13 3.6525 30.04
14 3.8142 30.40 -6 15 11.5xl0 360 380 4.0395 29.35 -6 16 12.3xl0 350 370 4.3082 29.71
TABLE II. Coefficients of Thermal Expansion, Transforma- tion and Softening Temperatures, Density and Molar Volume Data. l3
0.6 z 0 en 0.5 z a..
..._ z 18r------T------~------~------LLJ -(.) -LL. LL. LLJ 0 (.) z 0 en z ~ <.0 X 0 w >< _J (.) <( 0 ~ a:: ~' w (.) ::c..._ '~ (.) 0::: <( w z _J z <( LLJ ~ 10 20 30 40
Figure 4. Coefficients of thermal expansion for silver-borate glasses as a function of composition. 15
In agreement with Krogh-Moe( 2 l) and Uhlmann( 22), we find that the minimum in the expansion coefficient observed in this study, as well as in other alkali borate studies cannot b e exp 1 alne• d b y th e orlglna• • 1 Warren hypothesls• ( 2 3 I 2 4 ) which assumes saturation in the formation of tetrahedral
Bo units. The behavior of these thermal expansion curves, 4 which has always been associated with the term anomalous, does not give any clues on the association of Bo tetrahedra. 4 Krogh-Moe( 2 l) and Uhlmann( 22 ) gave an acceptable qualitative explanation for the broad minima in the thermal expansion curves, namely, that they are due to a competition between the formation of B0 tetrahedra, tending to decrease the 4 expansion coefficient, and the introduction of modifying cations, tending to increase it. More quantitative studies are needed to explain the thermal expansion behavior.
The transformation and softening temperatures as functions of the Ag o content are shown in fig. 6. There is a rapid 2 rise in both curves with Ag 2o addition up to 20 mol.% Ag 2o where they go through a maximum, to decrease with higher ( 20 ) and Green (g) found that ""Ja 0 B 0 Ag o content. Jankel l' I 2 2 - 2 3 Bao-B o and K o-B o glasses behaved similarly. 2 3 2 2 3 The variations in Tg and Ts as shown in fig. 6 indicate that the introduction of Ag 2o produces a more rigid structure The maximum with a corresponding increase in T g and T s . shown in the Tg curve at 20 mol.% Ag 2o may correspond to the congruent melting point of the Ag 2o-4n 2o 3 compound, the existence of which is confirmed by x-ray and DTA analyses. Such a 16
0.6
I- Li 0 · 4 8 0 GLASS z 2 2 3 0 2- Ag o · 4 8 0 GLASS (/) 2 2 3 z 0.5 3- Na 4 8 0 GLASS 2o · 2 3 ~ wX _J <( 0.4 :E a:: w I..._ 0.3 a:: <( w z -_J ..._ 0.2 z w u a: w 0.1 a.
100 200 300 400 500 TEMPERATURE, °C Figure 5. Thermal expansion of lithium,silver and sodium borate glasses in the 1:4 composition range. 17
500,------;.-----~------~------
400 u 0 w Tg 0::: :::::> ~ 0::: w a.. 200 :E w 1-
1000 10 20 30 40
0 MOL. /o Ag 20 Figure 6. Transformation and softening temperatures, T and T , of silver-borate glasses g s as a function of composition. 18
correlation between the variation of T and the shape of the g liquidus in the phase diagram has been noted before by Hyers and Felty( 25 ).
(4) Infrared Absorption:
Figure 7 shows the infrared spectra of sodium, lithium,
and silver tetraborate glasses from 4-25 ~· As is seen from the figure the absorption for silver borate glass is re-
markably similar to that for the other alkali borate glasses.
A complete assignment of infrared absorption bands to the
fundamental modes of vibration has been made for alkali borate glasses by different authors( 26 , 27 , 28 , 29,30).
Figures 8 and 9 show the considerable resemblance
between the infrared spectra of two silver borate glasses
and the corresponding devitrified glasses. Figure 10 corn-
pares the spectrum of B2o3 glass with that of a glass of low silver content. The three different types of spectra shown in figs. 8, 9, and 10 represent the spectra of all
Figure represents glasses containing 0 to 35 mol.% Ag 20. 10 the infrared spectra of the low silver content glasses up to Ag fig. 8 shovvs glasses between 10 to 25 mol.% 7 mol.% 2 o, shows the high silver content glasses. As Ag 2o and fig. 9 expected, a pronounced breading is exhibited in the spectrum of the glass compared with that of the corresponding devitri- fied sample, since in the glass the individual groups no longer have identical surrounding. Figure 11 compares the infrared spectra of devitrified sodium and silver tetraborate glasses. The same similarity has been demonstrated for all other glasses in this series. WAVE LENGTH, MICRONS
4.0 5.0 6.0 8.0 9.0 10.0 15.0 20.0
1- No 20 · 4820 3 GLASS ~ \\ \ 2- Ll20 · 48203 GLASS - \\1 3- Ag20 · 4 8203 GLASS w u z ._~ -::E en z 2500 2000 15001300 1200 1100 1000 900 800 700 600 500 400 FREQUENCY, cm- 1 Figure 7. Infrared spectra of sodium,lithium and silver-borate glasses in the 1:4 composition range. 1--' 1.0 WAVE LENGTH, MICRONS 4.0 5.0 6.0 8.0 ,, ...... , BEFORE HEAT TREATMENT ', GLASS NQ 10 ', \ -0~ Ag 0 · 48 0 AFTER HEAT TREATMENT \ 2 2 3 - ' \ w \ ~ () \ I \ \ \ II ' \ z \ I \ \ \ 2500 2000 1000 900 800 700 600 500 400 FREQUENCY cm- 1 Figure 8. Infrared spectra of Ag 20.4B2o3 glass before and after devitrification. tv 0 21 (30) Krogh-Moe has also found a strong resemblance in the infrared spectra of sodium borate and the corresponding lithium borate glasses and crystals. In 1938, Hibben( 3 l) observed that the Raman spectrum of borax is very similar to that of anhydrous sodium diborate glass; he concluded that the structures of these phases must be related. Since the structure of borax is known to consist of polyions where 50% of the boron atoms are fourfold coordinated, Hibben's results indicated that these polyions condense with- out losing their identity to three-dimensional networks when borax is dehydrated and fused to a glass. On such a basis, Krogh-Moe( 32 ) proposed his valuable group model for the structure of alkali borate glasses. [see Appendix C] The foregoing results and discussions enable us to suggest that the boron-oxygen network in silver borate glasses is similar to that of sodium and lithium borate glasses in the same molar composition range. -1 The bands appearing at 800, 1053, 944 ern shown as the dotted curve in fig. 8, are characteristic vibrations ( 2 9) . . of the so group ; these bands develop on the addltlon 4 32 of Ag o to n o . In agreement with Krogh-Moe's( ) 2 2 3 assumption that the structural model for boron oxide glass is a random three-dimensional network of B03 - triangles with a comparatively high fraction of six-membered boroxol rings( 33 ), (fig. 13 a), these bands can be explained as being due to the formation of triborate groups and penta- borate groups (fig. 13, band c). The absorption bands at 885, WAVE LENGTH, MICRONS 4.0 5.0 6.0 8.0 9.0 10.0 15.0 20.0 25.0 GLASS NQ. 15 ------BEFORE HEAT TREATMENT - Ag20 · 2820 3 --- AFTER HEAT TREATMENT -~ Ll.J , I u I I / z I I / I / ~ I I I I .... I I I - , I I ~ I I I I I ~~ II C/) I I ' I I / ' I z I \I I , ' ._/ 2500 2000 1500 1300 1200 1100 1000 900 800 700 600 500 400 FREQUENCY em -I Figure 9. Infrared spectra of Ag 20.2B2o3 glass before and after devitrification. (\,) (\,) WAVE LENGTH, MICRONS 4.0 5.0 6.0 8.0 9.0 10.0 15.0 20.0 25.0 .., ------GLASS NQ 3 I I ...... I l Ag20 ·198203 I I -...., 'I *- B2o3 GLASS 'I w I I I ,.- u I 1 I I z ', I \ , ~"-.. , ... I .... , .. , I ~' ' I' /' - - ~ -, / .__ \ / ..... , _, , 1// - ', / ~ ', / (/) z ', \ / 2500 2000 1500 1300 1200 1100 1000 900 800 700 600 500 400 FREQUENCY em -I Figure 10. Infrared spectra of B o glass compared with a low silver content glass. 2 3 !\.) w WAVE LENGTH, MICRONS 4.0 5.0 1- Na20 · 4 8203 AFTER HEAT TREATMENT 2-Na 0·4B 0 BEFORE HEAT TREATMENT '-... , 2 2 3 ...... ' ' - ' ' ' ' ...... ,~ ' \ ' ~ \ \ "' ' 2 \ ' w \ ', (.) \ \ ' \ f\ \ \ I \ z \ \ I ,,, I I \ \ I I \ ~/ I ' \ ' ' ,-~ I J II ' \ l ~ \ ,, -, I \ I \ I \ I I- \ I \ I - \ I \ I ~ ' I \ \ I ' I \ \ I en ' ~4 3 I f'l. ', \.,.., z \~ 1~1 I ...... '----... <( ', I I \ I a:: ', ,.--..{ I ..,_ ...... _,// ..._/ 3- Ag20 · 4 8203 AFTER HEAT TREATMENT (K Br PELLET) 4- Ag 20 · 48203 BEFORE HEAT TREATMENT 2500 2000 1500 1300 1200 1100 1000 900 800 700 600 500 400 1\) FREQUENCY, em -I *'" Figure 11. Infrared spectra of sodium and silver borate glasses before and after devitrification WAVE LENGTH, MICRONS 4.0 5.0 6.0 7.0 8.0 9.0 10.0 15.0 20.0 25.0 J' I I I I I I ,-..... , I ,,.... \ Na 0 · 2 8 0 GLASS 2 2 3 I ,, \ I ,"" \ ------Ag 0 · 2 8 0 GLASS r/ 2 2 3 I ~ ,"" ,,.. \ \ -0 ' \ I - \ \ \ w \ (.) \ \ \ z \ \ \ ~ \ I- \ - \ :;E \ (/) \ \ z \ <( \ 0::: ' I- (K Br PELLET) ~ 2500 2000 1500 1300 1200 1100 1000 900 800 700 600 500 400 FREQUENCY, cm- 1 Figure 12. Infrared spectra of sodium and silver borate glasses in the 1:2 composition range. I\) Ul (a) (b) (c) (d) BOROXOL PENTABORATE TRIBORATE DIBORATE • BORON ATOM 0 OXYGEN ATOM Figure 13. The borate groups. N "" 27 -1 719 and 1250 ern correspond to the v , v and the doubly- 1 2 degenerate v vibrations characteristic of the Bo units( 29 ). 3 3 The assignment of fundamental bands to the triborate and pentaborate groups is found to be a complex task, since the individual fundamentals appear to merge into broad bands mainly due to the interaction between these groups( 3 , 4 ). This increase in the B04 groups on the addition of alkali oxide 35 to about 33 mol.% was confirmed by NMR studies( ). NMR, also, showed that there are two types of B03 groups present from 10 to 30 mol.% alkali oxide; these could arise from forma- tion of the triborate and pentaborate groups suggested by Krogh-Moe( 34 ). X-ray study(l7 ) of the crystalline silver tetraborate, Ag o-4B o , suggested that it consists of two separate, 2 2 3 identical, three-dimensional, interlocking networks, each network being composed of triborate and pentaborate units which have been found previously in anhydrous cs2 0·3B 2o 3 and . (34) K 0·5B o respect1vely . The spectra shown in figs. 7, 2 2 3 9, and 11 thus prove that this borate network present in the crystalline silver tetraborate is the same for silver and sodium borate glasses, as well as their crystals in the 1:4 composition range. The infrared spectra of the silver borate glasses and corresponding devitrified samples in the composition range 27 to 35 mol.% Ag o, shown in fig. 9, were found to resemble 2 spectra of the anhydrous crystalline Li 20·2B2o 3 studied by Krogh-Moe< 34 ) who found them to consist of diborate units. 28 This suggests the presence of such diborate group in silver glasses (fig. 13d) within the given composition range. Figure 12 compares the spectra of Ag 0·2B o and Na 0·2B o 2 2 3 2 2 3 glasses, and suggests that the borate network of the two glasses in this composition range is the same; in other words, they both contain the diborate units. The infrared absorption spectra of the silver borate glasses show the 2.9~ and the 3.13~ bands due to structural and surface water respectively, the amount of which is expected to be approximately in the range given by Stevels et al. ( ll) . These ranges are from 0.40 to 0.25 wt.% H o 2 for the structural water and from 0.50 to 0.20 wt.% H o for 2 the surface water for glasses containing up to 20 mol.% Na o. 2 However, for higher alkali content, up to 28 mol.% Na o 2 there is no indication in the infrared spectra of the presence of surface water, while the structural water ranges from about 0.25 to 0.10 wt.% H 2o. As the Ag2o content is raised beyond 30 mol.% the water bands begin to increase again. It should be taken into consideration that hydrogen bonds play an important part in the atomic arrangement of the borate glasses, especially glasses of low alkali contents< 25 , 27 , 28 ). (5) Density, Molar Volume, and Phase Separation: Plots of density versus composition for silver borate glasses are shown in fig. 14, together with comparable sodium . (36) borate glasses measured by Sp1nner et al. • Both series of glasses were melted in air, their water content is reported in sec. III. 4. 29 4.5 Ag 20- 8 20 3 GLASS SYSTEM ---- Na2 0- 8 2 0 3 GLASS SYSTEM 39 (AFTER SPINNER et. al.) 4.0 3.5 r() 3.0 35 § rn -- E w 2.5 34 ~ () __, ::::> _J 'C) -<------33 0 ~ > J- ~ (f) 1.0 29 0.5 F~gure 14. Variation of density and molar volume of silver and sodium borate glasses as a function of composition. 30 Figure 15 shows the variation in density with composi- tion in wt.% for the silver borate glasses. Shaw and Uhlmann( 37 > have recently postulated that the compositional regions whose density-weight per cent composition plots show smooth, posi- tive curvature, with no extreme or inflections can be re- garded as those compositions most likely to phase-separate. Figure 15 shows that the presenceofanirnrniscibility region in the Ag system is likely. 2o-B 2o3 The change in the molar volume of glass as a function of composition was used by several authors as a tool(JB) for studying the distributions of different ions in the glass network. Figure 14 shows this relation for silver borate glasses compared with that for sodium borate glasses found in the literature( 36 ). The decrease in molar volume for both silver and sodium borate glasses shown in fig. 14 may be explained on the basis that the addition of Na2o or Ag 20 to B o increases the proportion of Bo units, which will 2 3 4 result in a more compact structure. In addition, these modifier ions (Na+ and Ag + ) , take interstitial positions ln the network resulting in a large increase in mass but only a slight increase in volume. These two effects can thus result in a sharp decrease in the molar volume. The lower rate of decrease in molar volume of the silver borate glasses which starts at about 27 mol.% Ag 2o may be attributed to the presence of higher amounts of the uncharged atomic silver. 31 rt) E ~ ~ ~ t- en z w Cl 2 'o 10 20 30 40 50 60 WT. 0/o Ag20 Figure 15. Variation of density with composition, in wt.%, for silver-borate glasses. 32 Electron microscope investigations of silver borate glasses indicated a two phase structure similar to that found in lithium and sodium borate glasses by Shaw and 39 Uhlmann( ). The extent of this immiscibility in different alkali borate glasses was reported as 2 to 18 mol.% for Li o-a o glass, 7 to 24 mol.% for Na o-B o glass, 2 to 2 2 3 2 2 3 22 mol.% for x o-B o glass, 2 to 16 mol.% for Rb 0-B o 2 2 3 2 2 3 glass and 2 to 20 mol.% for cs o-a o glass. To explore the 2 2 3 Ag 2o-B 2o 3 glass system three glasses containing 7, 13, and 20 mol.% Ag 2 o were examined by direct transmission electron microscopy. A square-edged diamond file was used to file flakes from a fresh surface prepared by breaking the sample just before viewing under the electron microscope. A fresh surface is important because of the possible reduction of surface silver due to light exposure. A detailed description of the technique and equipment used is discussed elsewhere( 40). All glasses before heat treatment and the heat-treated glass containing 20 mol.% Ag2o showed a single phase structure. The heat-treated glasses containing 7 and 13 mol.% Ag 2o showed a two-phase structure. Figure 16 shows the submicro- structure of the 13 mol.% Ag 2o glass which was held for two hours at 430°C and etched with dilute nitric acid for six seconds; the same type of microstructure was found for the No crystallization or change glass containing 7 mol.% Ag 2o. in color was observed after heat treatment for the two glasses that showed phase separation. The particle size of the Ag 2o a2o3 glasses is comparable with that of Li o-B o and Na o-B 2 2 3 2 2o 3 33 Pigure 16 . Transmission electron micrograph of glass containing 13 mol.% Ag2o, mark indicates 5000 °A. 34 glasses and much smaller than that of the Rb o-B o 2 2 3 and Cs glasses. 2o-B 2o3 Glasses containing up to 25 mol.% Ag o fluoresced 2 under uv radiation. Enough atomic silver is probably 2 present to account for this fluorescence( ). At compositions higher than 25 mol.% Ag2o fluorescence quenching occurs due to the aggregation of atomic silver. All the silver glasses on heat treatment or exposure to light for a long time, suffer reduction of Ag ions to atomic (metallic) silver which in turn tends to aggregate causing the disappearance (quenching) of fluorescence even in the low silver containing glasses. Similar effects have been observed in silicate glasses containing traces of silver( 2 ). (6) Internal Friction: Internal friction curves for binary silver borate glasses in the composition range 10 to 35 mol.% Ag 2o are ill- ustrated in fig. 17. Glasses containing 10 to 30 mol.% Ag o have one peak in the temperature range -180 to 250°C; 2 the peak height increases with increasing silver concentra- tion. Increasing the silver content decreases the peak 41 4 2 temperature, in analogy to the behav1or. o f a lk a l 1 . b orate ( ' ) and silicate glasses( 44 ). In general, however, the inter- nal friction peaks of silver borate glasses are at a much lower temperature than those of the corresponding alkali (41) . borate glasses . This behavior is due to the h1gh mobility of silver ions in glasses. Electrical conductivity studies( 45) have shown that silver ions have a higher mobility than the alkali ions. 9~~----~--~~----~----~------~ ------X = 10 8 rt) -----X = 15 0 ,... ---X= 20 )( '\ ------X = 28 I X = 34 I I 'o I 1I \ I"' I 1 ~ ' I \ z I I,,,I ' 0 I I - ' I \ I I' ..... \\ I \ I /"'-.. '\. At' u I ,,, - \ I I . ' h 1,' a:: l ' .. / /II 'I LL.. ~ , \ )< ..../ /: _J )';--/>· ...... __/ .. ;'j <[ \ /' ··"" / ' . _, / z I___ .A...... ___ _...... ,..,.,.,.. ~" 0:: _/ ,______...... / w / -- "' ..... z X Ag 20 · ( 100 -X) 820 3 I I I I I I I -.I I I I I .-.I I I I I I - I I I 1 1 I 0' I -100 TEMPERATURE, T oc Figure 17. Internal friction for xAg o. (100-x)B o glasses; 2 2 3 frequency 0.5 Hz. w lll 36 The activation energy calculated from internal friction measurements of the glass containing 15 mol.% Ag o was 2 found to equal 22.2 K cal./mol., as compared with the lit- erature value of 32.0 K cal./ mol. for the corresponding sodium borate glass( 46). A second peak at a higher tenperature lS observed in the glass containing 34 mol.% Ag 2o shown in fig. 17. A similar peak is found in the alkali silicate glasses and is usually attributed to the nonbridging oxygens present in the glass( 44 ) or to the interaction between alkali ions, bridging H+ ions and nonbridging oxygens( 42 , 43 ). The presence of nonbridging oxygens together with the OH group is offered as the explanation for the high temperature peak of glasses containing 34 mol.% Ag since N.M.R. ( 35) and 2o, other studies showed that nonbridging oxygens start to form around a composition of 33 mol.% alkali oxide and all these glasses contain appreciable amounts of water. The absence of the second peak in the silver or alkali borate glasses containing less than 33 mol.% reported in this study and ln. t h e 1'lterature (41,42,46) cou ld b e exp 1 alne . d on the assumption that the nonbridging oxygen is either not present or present in such low concentration as to be un- detectable by internal friction. The absence of the non- bridging oxygen peak was noticed in the aluminosilicate glasses with an aluminum to alkali ratio of one, in these glasses it is generally postulated that the addition of alumina to the alkali silicate converts the nonbridging oxygen 37 ions to bridging ones< 47 ). Also, fig. 17 shows no additional peaks as would be expected in the high silver content glasses; such peaks could occur due to relaxation process between atomic silver and the network. The absence of additional peaks in the internal friction curves could be explained as due to one of two facts: either atomic silver is present in very small concentration, or the internal friction peak due to these mobile silver atoms lies at a much lower temperature than the temperatures studied here. To check the effect of atomic silver on the internal friction curves, a glass containing 34 mol.% Ag 2o was exposed to uv and visible radiations for a long period of time until reducing more 0 Ag+ to atomic (metallic) Ag , thus changing its color from brown to black. An identical internal friction curve is reproduced with no additional internal friction peaks. Internal friction studies on mixed sodium silver borate glasses< 48 ) showed similarity to the mixed alkali silicate glasses< 12 ). This suggests that silver plays the role of a second alkali in these glasses. [see Appendix E] From the preceding discussions, it is seen that silver borate glasses behave in a fashion similar to that of the alkali borates. The appearance of the nonbridging oxygen peak in the internal friction curve for glasses containing more than 33 mol.% Ag 2o is an additional proof for the Krogh-Moe group model(34 ) which was also supported by NMR( 3 S) studies and which suggests the continuous formation of the B0 4 group 38 on the addition of alkali oxide to B until 33 mol.% 2o 3 alkali oxide is reached. 39 IV. CONCLUSIONS In the binary silver-borate system, glasses were formed within the composition range of 0 to 35 mol.% (65 wt.%) Infrared absorption, thermal expansion, density, molar volume, and internal friction studies indicated that in these glasses Ag plays a role similar to that of Na 2o 2 o and other alkali oxides. Thus, the addition of Ag to 2o B appears to result in the formation of pentaborate- 2o 3 20 triborate group pairs up to mol.% Ag2o. At 20 mol.% Ag the structure would consist mainly of two interpene 2o trating networks of alternating pentaborate and triborate 20 groups, and from to 33 mol.% Ag 2o of diborate groups. This interpretation is supported by the similarities between the x-ray spectra of the devitrified glass and those of the crystals in the 1:4 composition range, and between the infrared spectra of quenched and devitrified silver and sodium borate glasses. X-ray and DTA studies suggested the presence of a new compound, Ag 0-2B , which starts to appear at 28 mol.% Ag 2 2 o3 2o. A two-phase structure appeared in the silver borate glasses containing less than 15 mol.% Ag 2o as indicated by electron microscope and density studies. An internal friction peak, corresponding to the alkali peak was produced on the addition of Ag 2o to B2o 3 up to 28 Glasses containing 34 mol.% Ag 2o showed a second peak at a higher temperature, which is associated with the appearance of nonbridging oxygen in the system. 40 The change observed in the molar volume curve and the fluorescence quenching which occurred at compositions higher than 25 mol.% Ag 2o indicated the presence of some atomic silver in the high silver containing glasses. 41 V. REFERENCES (1) G.E. Rindone and TiJ.A. vJeyl, "Glasses as Electrolytes in Galvanic Cells," J. Amer. Ceramic Soc., 33 (3) 91-95 (1950). ( 2) W.A. Weyl, "Silver in Glasses: II," J. Soc. Glass Tech . , 2 9 ( 13 5 ) 2 91-3 8 9 T ( 19 4 5) . ( 3) ~>J. A. Weyl, "Metals in the 1\tomic State in Glasses," J. Phys. Chern., 57 pp. 753-756 (1953). (4) 0. Kubaschewski, "Diffusion of Silver in Glass," J.Z. Electrochem, 42 pp. 5-7 (1936). (Brit. Chern. Abs. A. 1936, 281). (5) G.M. Willis and F.L. Hennessy, "The System Ag o-B o ; its 2 2 3 Thermodynamic properties as A Slag Hodel," Trans. A.I.M.E., 19 7 pp. 13 6 7 ( 19 53) . (6) T. Maekawa, T. Yokokawa and K. Niwa, "Solubility of Ag o in Na o-B o Melts," Bul. Chern. Soc. Japan, 42 pp. 677- 2 2 2 3 681 (1969). (7) V.I. Westermann, "Uber die Aufnahme von Silberoxyd durch Oxyde und Oxydver bindungen bei hoheren Temperaturen," J.Z. Anorg. Allgem. Chern., 206 pp. 97-112, (1932). (8) E. Lell and N.J. Kreidl, "Radiation Effects on Complex Glasses," Proc. Cairo Solid State Conf., 1966, Ed. A. Bishay, Plenum Press, New York, (1966). (9) R.L. Green, "X-Ray Diffraction and Physical Properties of Potassium Borate Glasses," J. Amer. Ceram. Soc., 25 (2) 83-89 (1942). 42 (10) P.M. Vora, "Imperfections in the Ag-In System and Lattice Parameters of Cadmium Oxide," M.S. Thesis, U.M.R. (1970). (11) F.C. Everstein, J.M. Stevels, and H.I. Watermann, "The Density and Refractive Index of Vitreous Boron and Sodium Borate Glasses as Function of Compsoition," J. Phys. and Chern. Glasses, 1 (4) 123-133 (1960). (12) J.E. Shelby and D.E. Day, "Mechanical Relaxations in Mixed-Alkali Silicate Glasses: I, Results," J. Amer. Ceramic Soc., 52 (4) 169-174 (1969). (13) K.H. Sun and N.J. Kreidl, "Coloration of Glass by Radiation," Glass Ind., 33 pp. 590 (1950). (14) G.E. Rindone, "The Spontaneous Growth of Silver Films on Glasses of High Silver Content," J. Soc. Glass Tech., 37 pp. 124 T (1953). (15) B.I. Markin, "Electrical Conductivity of Argenta-Boric Glasses," J. Gen. Chern. (U.S.S.R.), 11 pp. 285-292 (1941). (16) F. DeCarli, "Anhydrous Berates of Silver, Barium, and Zinc," Atti. Accad. Lincei, 5 (6) 41-47 (1927); Cited from Chern. Abs., 21 pp. 1771 (1927). (17) J. Krogh-Moe, "The Crystal Structure of Silver Tetra- borate," Acta Cryst., 18 pp. 77 (1965). (18) E.N. Boulos, M. Hydlar, and N.J. Kreidl, To Be Published. (19) M.U. Samsoen, "Anomaly in the Expansion of B 2o3 Glass," Compt. Rend., 181 pp. 354 (1925). (20) E. Janckel, "Temperature of the Transformation Intervals of Glasses Formed from B o with Na o and BaO," Z. Elektrochem., 2 3 2 40 (76) 541 (1934); Ceramics Abs., 14 (6) 137 (1935). 43 ( 21) J. Krogh-Moe, "The Relation of Structure to Some Physical Properties of Vitreous and Molten Borates," Arkiv Kemi., 14 pp. 553 (1959). (22) D.R. Uhlmann and R.R. Shaw, "The Thermal Expansion of Alkali Borate Glasses and the Boric Oxide Anamoly," J. Non. Cry s t . So 1 . , 1 ( 5 ) 3 4 7 - 3 5 9 ( 19 6 9 ) . ( 2 3) B. E. Warren, "Summary of Work on ."1\tomic Arrangement in Glass," J. Arner. Ceram. Soc., 24 (8) 256-261 (1941). ( 2 4) J. Biscoe and B.E. Warren, "X-Ray Diffraction Study of Soda-Boric Oxide Glass," J. Arner. Ceramic Soc., 21 (8) pp. 287-293 (1938). (25) M.B. Hyers and E.J. Felty, "Structural Characteriza tions of Vitreous Inorganic Polymers by Thermal Studies," Mat. Res. Bull., 2 pp. 535-546 (1967). ( 2 6) S. Anderson, R.L. Bohon, and D.D. Kimpton, "Infrared Spectra and Atomic Arrangement in Fused Boron Oxide and Soda Borate Glasses," J. Amer. Ceramic Soc., 38 (10) 370-377 (1955). (27) P.E. Jellyman and J.P. Procter, "Infrared Reflection Spectra of Glasses," J. Soc. Glass Tech., 39 pp. 173-192 T (1955). (28) H. Moore and P.~-J. McMillan, "Study of Glasses Consisting of the Oxides of Elements of Low Atomic Weight: II," J. Soc. Glass Tech., 40 pp. 97-138 T (1956). (29) R.V. Adams and R.W. Douglas, "The Absorption of Infrared Radiation and The Structure of Glasses," Glastech. Ber. V, International Glass Congress Heft VII pp. 12-24 (1959). 44 (30) J. Krogh-Moe, "The Infrared Spectra of Some Vitreous and Crystalline Borates," Ark. Kemi, 12 pp. 475 (1958). (31) J.H. Hibben, "The Constitution of Some Boric Oxide Compounds," Arner. J. Sci., 35 (5) 113-135 (1938). (32) J. Krogh-Moe, "New Evidence on the Boron Co-ordination in Alkali Borate Glasses," J. Phys. Chern. Glasses, 3 (1) 1-6 (1962). (33) J. Krogh-Moe, "The Structure of Vitreous and Liquid Boron Oxide," J. Non. Cryst. Sol., 1 (4) 269-284 (1969). (34) J. Krogh-Moe, "Interpretation of the Infrared Spectra of Boron and Alkali Borate Glasses," J. Phys. Chern. Glasses, 6 (2) 46-54 (1965). (35) P.J. Bray and J.G. O'Keefe, "NMR Investigations of .Alkali Borate Glasses," J. Phys. Chern. Glasses, 4 (2) 37 (1963). (36) L. Shartsis, W. Capps, and S. Spinner, "Density and Expansivity of Alkali Borate Glasses," J. Arner. Ceramic Soc., 36 (2) 35-43 (1953). (37) R.R. Shaw and D.R. Uhlmann, "Effect of Phase Separation on the Properties of Simple Glasses: I," J. Non. Cryst. Sol., 1 (6) 474-:-498 (1969). (38) A.M. Bishay and M. Maklad, "Radiation Induced Optical Absorption in Lead Borate Glasses in Relation to Structure Changes," J. Phys. Chern. Glasses, 7 (5) 149-156 (1966). (39) R.R. Shaw and D.R. Uhlmann, "Subliquidus Inuniscibility in Binary Alkali Borates," J. Amer. Ceramic Soc., 51 (7) 377-382 (1968). 45 ( 4 0) M.S. Maklad and N.J. Kreidl, "Some Effects of OH groups on Sodium Silicate Glasses," Proc. IX International Congress on Glass, To Be Published in Sept. 1971. (41) V.H. Karsch and E. Janckek, "Theoretrische and Jl1echanische Untersuchungen an Alkaliboratesglas," Glastechn. Ber., 34 Jahrg., Heft 8 pp. 597-903. (1961). (42) M. Coenen, "Mechanical Relaxation of Silicate Glasses with Eutectic Composition," Z. Electrochem, 65 (10) 903-908 (1961). (43) H. Coenen, "Influence of Anisotropy on the Relaxation of Silicate Glasses and General Systematic of Damping Maxima in Glasses," Phy. of Non-Cryst. Sol., Proc. of International Conf., Delft, July (1964), pp. 445-60. ( 4 4) R.J. Ryder and G.E. Rindone, "Internal Friction of Single Alkali Silicate Glasses Containing Alkaline Earth Oxide: II," J. Amer. Ceramic Soc., 44 (ll) pp. 532-537 (1961). ( 4 5) B.I. Markin, "Electrical Conductivity of Argenta-Boric Glasses," J. Gen. Chern. (USSR) ll, pp. 285-292, (1941). ( 4 6) H. DeWaal, "On the Boric Oxide Anaornaly in Nabal Glasses," J. Phys. Chern. Glasses, 10 (3) 101-107 (1969). ( 4 7) D.E. Day and W.E. Steinkamp, "Mechanical Damping Spec- trum for Mixed Alkali R 0.Al o .6sio Glasses," J. Amer. 2 2 3 2 Ceramic Soc., 52 (ll) 571-574 (1969). ( 4 8) E.N. Boulos and N.J. Kreidl, "Hixed Cation Effects 1n Silver Borate Glasses," Submitted for Publication in the ,J. Amer. Ceramic Soc. 46 VI. APPENDICES 47 APPENDIX A X-Ray Diffraction Figure (1) shows the x-ray diffraction patterns for the Ag glasses in the composition range 0 to 35 mol.% Ag 2o-B2 o3 2o. As is seen in this figure, the diffuse x-ray patterns for all glasses in this series are alike. Figures (2 and 3) show the x-ray patterns for the d ev1"t r1"f" 1e d g 1 asses, heated 100 hours at 550 °~.c Discussion of the results shown in these figures are found in section III.2. Figure (4) gives a comparison of the 'd' values of devitrified Ag 0.4B glass and those of the compound Ag 0.4B 2 2o 3 2 2o 3 ( 1 ) computed from Krogh-Moe's paper . 48 Mol.% Ag20 35 :>-! E-1 H 30 U)z li:l zE-1 H Cl J:il Cl 0:: 20 0 u ~ 70 60 50 40 30 20 10 DEGREES, 2 e Figure 1. Effect of increasing Ag2o on the x-ray diffraction patterns of xAg (100-x)B glasses. 2o. 2o 3 Mol.% Ag 0 2 23 :>t 8 H U}z rLI 8z H 0 rLI 50 ~ 30 20 10 DEGREES,26 Figure 2. X-ray diffraction p0tterns of devitrified silver-borate glasses. ol:::. \.0 Mol.% Ag 0 2 34.5 ~ e-c H Ul 34 r:qz ze-c H 0 r:q ~ 0 0 ~ 50 40 30 20 10 Ul DEGREES, 2e 0 Figure 3. X-ray diffraction patterns of devitrified silver-borate glasses. 51 - - - U) (Y") (Y") U) 0 ..-... 0 .:X: Q) N....:l 1- ~N ~ 0 ~ t...9 ~ E-1 ~ ~ . U) I . . 0 >; ..c: 0 8 - NP:: tJI N H tJI u 0 tJI > .:X: H .:X: J:il -~ Q ·- - - Ul 0 .:X:... - - - ,_ - - (Y") - --:...... ::: - - - - 1- ·- N Figure 4. Comparison of the 'd' values of Ag o.4B o single crystals and 2 2 3 devitrified glass. 52 REFERENCES FOR APPENDIX A (1) J. Krogh-Moe,"The Crystal Structure of Silver Tetra Borate," Acta Cryst., 18 pp. 77 (1965). 53 APPENDIX B Thermal Expansion Figure (1) shows a typical thermal expansion curve for some silver-borate glasses. The compositions of the corresponding glass numbers shown in this figure are listed on page ( 7 ) . 54 0.7r-----~------~------~------~------~ 0.6 X=O.O X=7 § 0.5 H (f) ~ P-1 :X: ~ H 0.4 ~ ::r:~ E-1 X=28 p:: ~ z~ 0.3 H H E-1z ~u p:: X=lO ~ 0.2 ~ 0.1 0 100 500 0 TEMPERATURE, c Figure 1. Thermal expansion curves of silver borate glasses. 55 APPENDIX C Infrared Absorption 1. Theory The study of infrared absorption spectra in glasses can give much information about their structural building units. These structural units have natural vibrational frequencies which correspond to wavelengths in the 5-100 ~ region of the spectra. Spectra in the infrared region arise from the absorption of definite quanta of radiation as a result of transitions between certain vibrational or rotational energy levels. Transitions between vibrational energy levels within the same electronic level give rise to spectra in the near infrared region, while spectra in the far infrared region arise from transitions between rotations within the same vibrational level. An isolated assembly of N atoms can have as many as 3N-6 different fundamental modes of vibration. These normal modes are due to either a change in bond lengths (stretching) or angle (bending) of the mole cule; they appear as fundamental bands and are characterized by high intensities and sharpness. All general vibrational motions which a molecule may undergo can be resolved into either one or a combination of these normal modes. However, not all transitions between different energy levels are infrared-active. There are two selection rules restricting the activity of such transitions: a) In order for molecules to absorb infrared radiation as vibrational excitation energy, there must be a change in 56 the dipole moment of the molecule as it vibrates. Therefore, any change in direction or magnitude of the dipole during a vibration gives rise to an oscillating dipole which can interact with the oscillating electric field component of the infrared radiation giving rise to the absorption of this radiation. b) For a harmonic oscillator, only transitions for which 6n = +1 can occur, where n is the vibrational or rotational quantum number. The frequency corresponding to the transition from state n to n is called the "fundamental 0 1 frequency". However, since most molecules are not perfect harmonic oscillators, this selection rule breaks down and transitions corresponding to 6n = +2 and +3 do occur. Such transitions are referred to as the "first" and "second" overtones. In crystalline compounds atoms have both short range and long range order. The lattice vibrations of the surround- ing atom will affect the normal frequencies characteristic of a certain point group.* Sharp, well resolved bands are characteristic for crystalline lattice vibrations. In the vitreous state, where long range order is absent, the asymmetry caused by the random arrangement of the vitreous * All molecules can be classified into one or other of the point groups. Each point group is a collection of all the symmetry operations that can be carried out on a molecule belonging to this group. 57 network will change the dipole moment. Accordingly, vibrations that are normally infrared-inactive in the case of the isolated structural groups, become active in the vitreous state. Furthermore, since any small variation in the surroundings of a group affects a shift in the fundamental frequencies, the randomness of the network, where structural units do not have entirely identical surroundings as they do in the crystalline lattice, leads to broadening and overlapping of the absorption bands. 2. Pellet Technique For Infrared Spectroscopy: The KBr pellet technique(l, 2 )was used to prepare the samples for infrared study.The glass samples were first ground to a fine powder in an agate mortar. The powder was then sieved (200-360 mesh) to insure uniformity of particle size. Each pellet consisted of 1% glass powder, the rest being spectroscopic KBr powder; a total weight of approximately 0.2 grams gave a suitable pellet thickness. The weighed powders were thoroughly mixed, then placed in a die and kept under vacuum for 15 minutes, and finally a pressure of 20,000 lbs was applied. The pellets thus prepared were transparent. Other techniques for sample preparation are: a) Very thin films ( 3 ) (blown as bubbles from molten glass). 4 b) Diluted fine powder in oil (Nujol) suspension( ). 58 3. Infrared Spectrophotometer: The infrared absorption curve was obtained by placing the transparent pellet in a Perkin Elmer 337, infrared double beam grating spectrophotometer. (Ranges from 2.5~ -1 -1 to 25~ or 4000cm to 400crn ) . One grating operates in the range from 2.5~ to 8.33~ (4000cm-l to 1200cm-1 ) and the -1 -1 other from 7.5~ to a 5~ (1333cm to 400cm ) . 4. Literature Review of Infrared Studies of Boric Oxide and Alkali Borate Glasses. Several attempts have been made to interpret the infrared spectra of boron oxide and binary alkali borate glasses. Krogh-Moe( 5 ) recently reviewed infrared studies He proposed that the best structural model for boron oxide glass is a random three-dimensional network of Bo -triangles with a comparatively high fraction of six 3 membered rings (boroxol rings) ; this model has been supported {6) recently by the x-ray study of Warren and coworkers . Anderson, Bohon and Kimpton(?) measured the infrared absorption spectra of a series of Na2o-B2o 3 glasses. They stated that as Na o was added to the B glass, the number 2 2o 3 of B0 tetrahedra and B-0-B linkages increased. After a corn- 4 position of 15% Na o was reached, the atomic arrangement 2 changed decidedly. They concluded that the hydrogen bonds play an important part in the atomic arrangement of the glass of low sodium oxide content. 59 8 Jellyman and Proctor( ) measured the infrared re- flection spectra of a series of binary borate glasses in the wavelength region 1-15~. Their analysis was made by observing the change and appearance of new bands in the spectra in the composition range from pure B o to 2 3 They confirmed the hypothesis that all the additional oxygen atoms, from the addition of Na o, are 2 accomodated by the formation of boron tetrahedra until one- fifth of the boron atoms are tetrahedrally co-ordinated. Additional oxygen atoms are thereafter accomodated as non- bridging oxygen ions. 9 Moore and McMillan( ) studied the infrared spectra of different alkali borate glasses, and assigned certain fre- quencies to modes of B03 and Bo4 groups. As in the study by Anderson et al, (?) the spectra obtained by Moore and McMillan showed serious water contamination evident from the appearance of a strong OH stretching frequency in their spectra at about 3~. Adams and Douglas(lO)assigned the B0 and B0 units 4 3 The frequency to the Td and n 3h point groups respectively. of the fundamental bands for B0 and B0 groups are given in 3 4 the following table: B0 B0 3 4 -1 -1 \)1 11.3~ (885 ern ) \)1 12.5~ (800cm ) -1 -1 \)2 13.9~ ( 719cm ) 9.5~ (1053cm ) -1 -1 6.9~ (1449cm ) \)3 10.6~ (944cm ) -1 \)3 8.0~ (1250cm ) 60 Other bands in the infrared spectra of borate glasses were attributed to overtones or combinations, or to the presence of water in the glass. Moore and McMillan( 9 ) observed a displacement of the modes of vibrations of Bo groups to 3 lower frequency when an alkali oxide was added. This was attributed to the development of a more open type structure due to the replacement of the original B0 groups by B0 3 4 groups. Krogh-Moe( 4 ,ll) presented a group model for the structure of alkali borate glasses based mainly on the in- frared study of the glasses and a series of anhydrous Ll. 0.2B (13) , 2 2o3 These compounds are characterized by the presence of pentaborate, diborate, triborate and pentaborate-triborate groups respectively. These borate groups are shown in fig. 13 page (26). From the analogy of the infrared spectra of alkali borate glasses and those of the corresponding crystalline (15) compounds, Krogh-Moe deducted that the structure of these glasses consists of two interpenetrating networks. In pure boron oxide glass these networks are built up of boroxol groups. The addition of one molecule of alkali oxide to boron oxide results in the formation of one tri- borate group and one pentaborate group, which tend to occupy adjacent positions in the network. The formation of tri- borate-pentaborate pairs continues upon further alkali add- ition until a concentration of 20 mol.% alkali oxide is 61 approached. At 20 mol.% the structure consists mainly of two interpenetrating networks of alternating pentaborate and triborate groups. The diborate groups presumably start to form even before the 20 mol.% composition is reached. On further alkali additions, the number of diborate groups will increase at the expense of the pentaborate pairs. At 33 mol.% alkali oxide, the glass will mainly consist of an interpenetrating network of diborate groups. The number of boron atoms in fourfold co-ordination (NB ) is given by 04 the equation: X NB04 = 100-X where X is the molar percentage of alkali oxide; this equa- tion is obeyed up to 33 mol.% alkali oxide. 6 Krogh-Moe model is supported by NMR(l ) and other studies. 5. Data 5. 1. Semi-Quantitative Analysis: Many of the bands characteristic of Ag 2o (700,880, -1 1050, 1450 ern ) overlap with most of those of boron (B04 and B0 ) ; it was thus necessary to restrict the semi 3 quantitative studies to the singlet frequency of B04 (v at 800 crn-l), to the triply degenerate frequencies 1 -1 -1 of Bo cv at 944 ern ) , and to the absorption at 535 em 4 3 characteristic of Ag This ratio between the optical 2o. densities at 944 crn-l and 535 crn-l versus mol.% Ag 2o is shown in fig. (1). Two maxima are observed in the curve at about 20 and 33 mol.% Ag there is a continuous increase 2o; 't'J 't:l 1:: 1:: co co ..0 ..0 H ...... 4 .. 0 '<3' N 0 tJi l1l ...... ~ ...... -1 ...-1 I I 3 8 8 u u '<:!' 1.[) '<:!' ("'") 0'\ 1.[) E-i 8 ~ ~ 2 ~ ~ E-i E-i H H (/)z (/)z rr.:J rr.:J Q Q 1 H ~ ~ u u H H 8 8 p. p. 0 0 0 5 10 15 20 25 30 35 MOL.% Ag 2o Figure 1. Effect of increasing Ag o on the intensity of B0 group 2 4 and Ag o bands. [ xAg o. (100-x)B o glasses] 2 2 2 3 0'\ 1\) '0 1.0 !:! cu 0. 8 ~I= ...... r-1 I E; 0 oq< lu J oq< "'' 0.6 0\ 8 8 F:!! F:!! ~ ~ 8 8 0. 4 H H z(J) z(J) j:ij j:ij Q Q 0.2 ~u ~u H H 8 8 ~ ~ 0 0 0 5 10 15 20 25 30 35 MOL.% Ag 0 2 Figure 2. Effect of increasing Ag 2o on the four-co-ordination of boron. ( xAg o. (100-x)B o glasses] 2 2 3 0'1 w 64 lJ") Q) (Y') .s:: ..j...l lH 0 ?1 ..j...l 0 •.-I (Y') U) ~ Q) U) ..j...l Q) ~ U) ·.-I U) lJ") Ctj N Q) ....-! .s:: tJ> ..j...l 0 (Y') N ~ 0 0 N ~ m 0 0 N dP N X . tJ> I H ~ 0 0 0 ::E: tJ> ....-! ~ ·.-I . lJ") U) 0 ....-! Ctj N ,._,Q) tJ> 0 ~ ~ ·.-I lH U) 0 'd s:: ..j...l Ctj 0 .0 Q) lH -.::t' lH 0 ~ m . (Y') ,._,Q) 0 ::l tJ> N ·.-I ~ 65 in Bo 4 concentration until the composition of 33 mol.% Ag 2o is reached. Also the ratio between the optical density -1 -1 of Bo at 944 ern to Bo at 1250 ern fig. (2), and 4 3 1 -1 -1 between Bo I at 944 ern to Bo II at 800 ern fig. (3), 4 4 1 showed maxima around 20 mol.% Ag o. 2 The maxima observed in figs. (11 2 and 3) around the 20 mol.% Ag o may indicate 2 the formation of a compound at this composition. This has been confirmed by x-ray and D.T.A. study as shown in section III 2. The first part of fig. (3) shows a decrease in the This may be explained as being due to an increase in the number of non-bridging oxygen in the B0 4 group region. Also, the increase in this ratio at corn- positions higher than 25 mol.% Ag may be due to an in 2o crease of the diborate groups at the expense of a decrease in the pentaborate-triborate groups. This hypothesis has been tested by other authors for different glasses(l?). 5. 2. Infrared Absorption: Figures ( 4 and 5) sho\<7 the infrared absorption of some silver borate glasses which were investigated in the composition range 0 to 35 mol.% Ag 0 1 and which were not 2 shown in section III 4. Figures (6, 7 and 8) are the infrared spectra of the corresponding devitrified glasses. Figure (9) show the infrared absorption spectra for WAVE LENGTH, MICRONS 2.5 3.0 5.0 6.0 8.0 9.0 10.0 15.0 20.0 25.0 ...... dP rilu ~ 8 8 H ~ {f) ~ 8 000 3500 3000 1500 1200 1100 1000 700 600 500 400 FREQUENCY, CM-l Figure 4. Effect of increasing Ag 2o on the infrared-transmittance of xAg (100-x)B glasses. 2o. 2o3 0'\ 0'\ WAVE LENGTH, MICRONS 2.5 3.0 5.0 6.0 8.0 9.0 10.0 15.0 20.0 25.0 GLASSES -o\O ...... X rr:lu ~ E-t E-t H ~ (/) ~E-t 4000 3500 3000 1500 1200 1100 1000 900 800 700 600 500 400 FREQUENCY, CM-l Figure 5. Effect of increasing Ag 2o on the infrared-transmittance of xAg (l00-x)B glasses. 2o. 2o3 0'1 "-l WAVE LENGTH, MICRONS 2.5 DEVITRIFIED GLASSES dP rilu ~ 8 8 H ::8 U) ~ 8 4000 3500 3000 1500 1200 1100 1000 900 800 700 600 500 400 FREQUENCY,CM-l Figure 6. Effect of increasing Ag 2o on the infrared-transmittance of xAg 2o. (100-x)B2o3 devitrified glasses. co"' WAVE LENGTH, MICRONS 2.5 3.0 5.0 6.0 8.0 9.0 10.0 15.0 20.0 25.0 DEVITRIFIED GLASSES -dP u~ ~ 8 8 H :E: U) ~ 8 400 Figure 7. Effect of increasing Ag 2o on the infrared-transmittance of xAg 2o. (100-x)B2o3 devitrified glasses. m \.0 WAVE LENGTH, MICRONS 2.5 3.0 5.0 6.0 8.0 9.0 10.0 15.0 20.0 25.0 DEVITRIFIED GLASSES dP ji:l u X ~ 8 8 H ~ (f) ~ 8 4000 3500 3000 1500 400 FREQUENCY, CM-l Figure 8. Effect of increasing Ag 2o on the infrared-transmittance of xAg 2o. (100-x)B2o3 devitrified glasses. -..J 0 WAVE LENGTH 1 MICRONS 2.5 3.0 5.0 6.0 8.0 9.0 10.0 15.0 20.0 25.0 H Bo 3 3 dP (CRYSTAL) rLlu ~ E-4 E-4 H ::8 U) ~ 8 000 3500 3000 500 400 FREQUENCY I CM-l Figure 9. Infrared-transmittance curves of Ag o and H Bo crystals. 2 3 3 -...] I-' 72 REFERENCES FOR APPENDIX C (1) V.U. Schiedt and H. Reinwein, "Zur Infrarot-Spektros kopie von Aminosauren," Z. Naturf., 7 [B] 270-277 (1952). (2) H.M. Stimson and M.J. O'Donnell, "The Infrared and Ultraviolet Absorption Spectra of Cytosine and Isocytosine in the Solid State," J. Amer. Chern. Soc., 74 [7] 1805-1808 (1952). (3) N.F. Borrelli, B.D. McSwain and G. Jen Su, "The Infrared Spectra of Vitreous Boron Oxide and Sodium Borate Glasses," J. Phys. Chern. Glasses, 4 [1] 11-21 (1963). (4) J. Krogh-Moe, "Interpretation of the Infrared Spectra of Boron and Alkali Borate Glasses," J. Phys. Chern. Glasses, 6 [2] 46-54 (1965). ( 5) J. Krogh·-Moe, "The Structure of Vitreous and Liquid Boron Oxide," J. Non. Cryst. Sol., 1 [4] 269-284 (1969). (6) R.L. Mozzi and B.E. Warren, "The Structure of Vitreous Boron Oxide," J. Appl. Cryst., 3 [4] 251 (1970). (7) s. Anderson, R.L. Bohon and D.D. Kimpton, "Infrared Spectra and Atomic Arrangement in Fused Boron Oxide and Soda Borate Glasses, "J. Amer. Ceram. Soc., 38 [10] 370-377 (1955). (8) P.E. Jellyman and J.P. Proctor, "Infrared Reflection Spectra of Glasses," J. Soc. Glass Tech., 39 pp. 173-192T (1955). (9) H. Moore and w. McMillan, "Study of Glasses Consisting of the Oxides of Elements of Low Atomic Weight: II, "J. Soc. Glass Tech., 40 pp. 97-138T (1956). 73 ( 10) R.V. Adams and R.W. Douglas, "The Absorption of Infrared Radiation and the Structure of Glasses," Glastech. Ber. V, International Glass Congress, Heft VII pp. 12-24 (1959). (11) J. Krogh-Moe, "Structural Interpretation of Melting Point Depression in the Sodium Borate System, "J. Phys. Chern. Glasses, 3 [1] 1-6 (1962). ( 12) \'J. H. Zachariasen and H. A. P let tinger, "Refinement of the Structure of Potassium Pentaborate Tetrahydrate," Acta Cryst., 16 pp. 376-379 (1963). (13) J. Krogh-Moe, "The Crystal Structure of Lithium Di- Acta Cryst., 15 pp. 190-193 (1962). (14) J. Krogh-Moe, "The Crystal Structure of Cesium Triborate, cs20. 3B203, II Acta Cryst., 13 pp. 889-892 (1960). (15) J. Krogh-Moe, "The Crystal Structure of Silver Tetra- borate," Acta Cryst., 18 pp. 77 (1965). (16) P.J. Bray and J.G. O'Keefe, "NMR Investigations of Alkali Borate Glasses," J. Phys. Chern. Glasses, 4[2] 37 (1963). (17) A.M. Bishay and s. Arafa, "Gamma-Induced Absorption and Structural Studies of Arsenic Borate Glasses," J. Amer. Ceram. Soc., 49 [8] 423-430 (1960). 74 APPENDIX D Density Measurements 1. Theory The density of the sample at t°C is given by: w 1 (dt - d ) + d {1) = 1 g g where wl g is the weight of the sample in air, w2 g is the weight of sample + suspension in air, w3 g is the weight of the sample + suspension in xylene, dt g/cc is the density of xylene at t°C, 1 d g/cc is the density of air at room g temperature. w3 is actually made up of: 1. weight of sample alone in xylene, 2. weight of unimmersed suspension in air, 3. weight of immersed suspension in xylene; this is equal to the weight of the immersed suspension in air minus the weight loss of the immersed suspen- sian in xylene, 4. force due to surface tension. The weight loss due to the immersed suspension in xylene given in item 3 is equal to: wt. of suspension in air X density of xylene density of suspension material w (wt. of sample in xylene) 3 = + (wt. of unimmersed suspension in air) + (wt. of immersed suspension in air) 75 (wt. of immersed suspension in air density of suspension material X density of xylene) + force due to surface tension i.e w3 = (wt. of sample in xylene) + (wt. of whole suspension wire in air) - ~ 1 - ~2 = wt. of immersed suspension in air where ~l density of suspension material ~ 2 = force due to surface tension ·~ 1 ' can be determined as follows ( 2 ) where w is the weight of the sample + immersed part of 4 suspension,dT is the density of the suspension material (tangsten) = 19.3 g/cc Using the result given in reference (1), the density of xylene at 25.6°C was found to be equal to 0.860 g/cc. Also ~ 2 was calculated in the same manner as given in this reference and was found to equal 0.00063 g . . equation (1) can be written as: (dt - d ) + d 1 g g Procedure: The weight w of each glass sample in air was first 1 determined, followed by the weight w2 of the samples when suspended by a tungsten wire. The glasses were free of bubbles 76 as seen by the naked eye and an optical microscope. Their surface was clean and they were kept in xylene for some time before measurements because of their hygroscopic nature. Next a beaker with the xylene, containing the sample, was put into a desiccator and de-aerated for a few minutes. The beaker was then transferred to the bridge of the balance and allowed to attain the balance temperature. The sample was next suspended from the hook of the balance pan and the weight of the sample in xylene w noted. 3 This procedure was repeated several times and the average weight noted for the same temperature. Between each weighing operation, enough time was allowed for the xylene vapor to attain equilibrium with the enclosed air. To heat the xylene to the wanted temperature, 25.6°C, a stirring rod was warmed on a hot plate and the xylene stirred with it. Precautions were taken to allow the suspended sample to reach temperature equilibrium with the xylene. After the weighings, the suspension wire was cut just above the liquid line, then the sample and attached piece of wire were dried and weighed in air w4 . 77 REFERENCES FOR APPENDIX D (1) P.M. Vera, "Imperfections in the Ag-In System and Lattice Parameters of Cadmium Oxide," M.S. Thesis, U.M.R. (l970). 78 APPENDIX E This appendix was submitted for publication in the Journal of the American Ceramic Society. Mixed Cation Effect in Silver Borate Glasses 79 The internal friction peaks in simple alkali and mixed alkali silicate glasses have been studied intensively in ( 1 , 2 , 6 , 7 , 8) . th e pas t f ew years . Llterature reports about the internal friction of mixed alkali borate glasses are scarce. However, electrical conductivity measurements show that mixed alkali borate glasses behave in the same fashion as mixed alkali silicate glasses: there is a sharp drop in the conductivity with a minimum in the conductivity isotherms when one alkali is replaced by another( 3 ). In the course of our study of the structure and properties of binary silver borate glasses( 4 ) internal friction studies were conducted to determine the effect of small substitu- tions of sodium in silver, and silver in sodium glasses. Also a glass 0.95 Ag o·0.05 cu 0·4 B o was examined. 2 2 2 3 Silver is generally considered to behave as a modifier similar to alkali in the glass structure. The apparatus and the ( 1) method used were the same as those used by Day et. al . Internal friction curves for the binary silver borate glass Ag 0·4 B o and the corresponding glass with a small 2 2 3 sodium substitution for silver 0.95 Ag o·0.05 Na o·4 B o 2 2 2 3 are shown as solid and dashed dotted lines respectively in fig. 1. Internal friction curves for the binary sodium borate glass Na 0·4 B o and the corresponding glass with 2 2 3 a small silver substitution for sodium are similarly illustrated in fig. 2. In both cases the low temperature "alkali peak" has decreased and shifted to a higher temperature in the mixed glasses. And in both cases a second peak at a 80 higher temperature is observed in the mixed sodium-silver glasses. The dashed curves in figures 1 and 2 show the development of this high temperature peak for the glass 0.5 Ag 0·0.5 Na 0·4 B o with its equal shares in Na and 2 2 2 3 Ag. Similar large high temperature peaks ("mixed alkali peaks") have been observed in mixed alkali silicate glasses and attributed by Day and coworkers(l) to a new relaxation involving both alkalis. The same explanation is offered for the mixed alkali silver borate glasses. The high tempera- ture peak observed in one-alkali silicate glasses, frequently called the "non-bridging oxygen" peak, and somewhat over- lapping with the mixed alkali peak in some silicate composi- tions, is not present in the borate compositions reported here. A second high temperature peak in binary silver borate 4 glasses is only observed when more than 33% Ag o are present( ) 2 Fig. 3 shows the internal friction curve for a glass 0.95 Ag o-0.05 cu 0·4 B o in comparison with that of the 2 2 2 3 binary silver borate glass. Again, the low temperature "alkali peak" has decreased and shifted toward higher temp- erature and a pronounced high temperature "mixed alkali peak" has appeared in the mixed copper silver borate glass. Sim- ilar observations have been made in silicate glasses contain- ing copper and alkal1. by Wey 1 et. al. (S) and Wh1"te(lO). In conclusion, mixed silver sodium (and copper) borate glasses exhibit internal friction like mixed alkali silicate glasses. Silver and copper participate in relaxation processes 81 just like second alkalis. 82 ...... 0 0 .... ,, J"() ' ' \ ---- -...... - ' ...... _ ,,,' - ...... -.... , ...... '' ' ' " ' \ ' ' ...... , I .... , .... 1 ...... _ ...... J"() -\ ...... 0 ...... C\J J"() ' ...... m 0 ' t- C\J ) ',, .. ~. m ' , w ' ' 0::: 0 ~ \ C\J ' \ o:::> 0 w~--: 0 \ <( z C\J \ 0 ' \ 0::: l() w z I a... 0 . 'I J"() l() I ~ 0 I 0 d I w C\J 0 I t- CD C\J 0 ~ C) C\J <{ C) <{ 0 l() C\J L() C) ~ . <{ 0 0 I • I • I 1 3 INTERNAL FRICTION, Q- x 10 Figure 1. Internal friction curves for (l-x)Ag2o.xNa20.4B2o 3 glasses; frequency 0. 5 Hz. 83 0 0 1"0 I I 1"0 0 C\J 1"0 m 0 ...... v C\J .. .. CD ...... w ¢ 0 \ a:: ' \ C\J \ ::::> Ol \ 0 \ <{ C\J ~ 0' 0:: lO <{ '\ .. '\ w 0 . \ a_ 1"0 I{) \ 0 . 1 ::?i 0 0 I w C\J ' I co 0 I t-- C\J 0 I ¢ 0 C\J • I z 0 \ : ' I 0 lO z C\l 0 m. I{). z 0 0 I I I• -I INTERNAL FRICTION, Figure 2. Internal friction curves for (l-x)Na 0.xAg 0.4B o glasses; frequency 0.5 Hz. 2 2 2 3 84 ' ...... ' ' \ ' \ I I / / _,.; ./ , "" J'() 0 (\J CD ~. 0 (\J ::J .. u a::lJ.J tO ::> 0 ' 1- rt') . ' ...... <( 0 0. ' ' a:: C\J ', CD 0 ' lJ.J q- (\J ' ' Cl.. 0' ' ' ~ <2: ' \ lJ.J 0 (\J LO \ 1-- (]') ' \ 0' . \ <( 0 \ \ ' '' '' I 'I I I 0 ------~------._------~------~----~------~----__.00 co I() -I 1 3 INTERNAL FRICTION, Q- x 10 Figure 3. Internal friction curves for (l-x)Ag o.xcu 0.4B o 2 2 glasses; frequency 0.5 Hz. 2 3 85 REFERENCES FOR APPENDIX E (1) J.E. Shelby, Jr., and D.E. Day, "Mechanical Relaxa- tions in Mixed-Alkali Silicate Glasses: I." J. Arner. Ceram. soc • ' 52 [ 4 J 16 9 -7 4 ( 19 6 9) . (2) Emil Deeg, "Relationship Between Structure and Meehan- ical-Acoustical Properties of Single Glasses: N." Glastech. Ber. 31 [6] 229-40 (1958). (3) K.A. Kostayan, "Investigation of the Conductivity Neutralization Effect in Fused Borate Glasses." Structure of Glass, Vol. 2, pp. 234-236, Consultants Bureau, New York (1960). (4) E.N. Boulos and N.J. Kreidl, "Structure and Properties of Silver Borate Glasses," to be published. (5) L. c. Hoffman and W.A. Weyl, "A Survey of the Effect of Composition on the Internal Friction of Glass," Glass Indus try 3 8 [ 2 ] 81 - 8 5 , 1 0 4 ( 19 5 7 ) . (6) R.J. Ryder and G.E. Rindone, "Internal Friction of Single Alkali Silicate Glasses Containing Alkaline-Earth Oxides: II." (7) H.K. Sheybany, '"l'he Structure of Mixed-Alkali Silicate Glasses." Verres et Refract. 2 pp. 127-299 (1948). (8) W.A. Weyl, and E.c. Marboe, "The Constitution of Glasses." Vol. II, pp. 470-507, Interscience (1964). (9) G.L. McVay and D.E. Day, "Diffusion and Internal Fric- tion in Na-Rb Silicate Glasses," J. Am. Cer. Soc. 53 [9] 508-513 (1970). . ( 10) P.L. White, "Mechanical Relaxations in Copper Aluminosilicate Glasses." To be published in J. Amer. Cer. Soc. 86 APPENDIX F Internal Friction The internal friction peak heights and temperatures for the silver borate glasses studied and shown in fig. (17) section III 6 are given in table I. 87 LOW TEMP. PEAK HIGH TEMP. PEAK Glass 'Alkali Peak' 'Nonbridg. Oxygen Peak' Temp. Height Temp. Height oc o-1 xlo3 oc o-lxlo3 x=lO 199 0.30 -- -- x=l5 105 2.25 -- -- x=20 16 3.70 -- -- x=28 - 78 5.45 80 0.20 x=34 -100 4.90 85 1.00 TABLE I. Temperature and height above back-ground for the low and high temperatures peaks, in the 88 VII. VITA Edward Nashed Boulos was born on May 19, 1941, in Cairo, Egypt. He received his primary and secondary education in Damanhour, Egypt, and a Bachelor of Science degree in Chemistry and Physics from Cairo University in May 1963. Following graduation from University, he worked as advisor of glass technology, Industrial Control Department, Ministry of Industry, Cairo, Egypt; at the same time, he was enrolled as part-time graduate student in the American University, Cairo, from which he received his M.S. degree in Solid State Science on June 1966. From November 1967 until the present, he attended graduate school at the University of Missouri-Rolla studying for a Ph.D. degree in Ceramic Engineering. August 1967, he was married to the former Mervet Saleh, Egypt. They now have one daughter Nermine.