A RADIOCHEMICAL STUDY OF THE MECHANISM OF POLISHING GLASS by JOHN GORDON SMITH £ 0 p. I A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n the Department of Chemis t ry We accept t h i s t h e s i s as conforming t o the s tandard r e q u i r e d f r^m - candida tes f o r the degree of MASTER OF ARTS/ . Members of the Department of Chemist ry THE UNIVERSITY OF BRITISH COLUMBIA September, 1951 I l l ABSTRACT I n an e f f o r t t o determine whether g la s s flows when p o l i s h e d uranium g lasses were fused t o non uranium g lasses and p o l i s h i n g s c a r r i e d out i n a d i r e c t i o n from the a c t i v e t o the i n a c t i v e s i d e . S o d a - l e a d - s i l i c a t e and phosphate g lasses were p o l i s h e d w i t h rouge and c e r i c o x i d e . To l o c a t e any of the uranium g lass wh ich may have been t r a n s f e r r e d dur ing the p o l i s h i n g , nuc lea r t r a c k p l a t e s were used . Thus by autoradiographs taken before and a f t e r p o l i s h i n g , any a lpha p a r t i c l e s from the uranium g lass t r a n s f e r r e d t o the s i d e of the non uranium g lass would have r e g i s t e r e d on the developed p l a t e s . This method was oapable of d e t e c t i n g a 4 .5 A° flowed l a y e r or ch ips of uranium g lass 0.019 mms i n d iameter . No evidence of any f l o w , g rea te r than over a d i s t a n c e of 0.2 mms, was found by the a u t h o r s . Experiments were a l s o c a r r i e d out u s i n g r a d i o a c t i v e c e r i c o x i d e . Thus i f the contac t temperatures reached dur ing p o l i s h i n g approached the so f t en ing po in t of the g l a s s i t was reasoned t ha t the agent might, have become fused w i t h the p o l i s h e d su r f ace . By -7 u s i n g t h i s method i t would have been p o s s i b l e t o de tec t 2.3 X 10 grams of Ce02 spread over 6 crn^ on the sur face of the g l a s s . By assuming t ha t the Ce0 2 would fuse w i t h and become par t of the g l a s s i t was p o s s i b l e t o de tec t a 15 A° l a y e r of t h i s changed g l a s s , i f 10% of t h i s changed g lass was CeOg. The authors found no evidence of any f u s i o n of the C e 0 2 w i t h the g las s under the c o n d i t i o n s of p o l i s h i n g used i n t h i s p r o j e c t . TABLE OF CONTENTS Page INTRODUCTION 1 GENERAL METHOD OF ATTACK 5 EXPERIMENTAL 7 (a) PREPARATION OF GLASSES (l) Silicate Glasses Fig. I Sohematic Diagram of Electric Resistance Furnace (b) GRINDING AND POLISHING EQUIPMENT Photograph I Polishing Apparatus Photograph II Apparatus for Polishing with Radioactive Cerio Oxide (o) AUTORADIOGRAPHIC MEASUREMENTS (1) Emulsions (2) Development of Plates (3) Sensitivity of the Photographic Method (4) Determination of Polishing Compound Transferred to Glass Autoradiographs of Glasses Polished with Radioactive Cerio Oxide (2) Phosphate Glasses RESULTS 21 DISCUSSION FUTURE WORK BIBLIOGRAPHY 35 39 40 II ACKNOWLEDGMENT Grateful thanks are extended to Dr. J.G. Hooley, under whose direction this investigation was carried out. Through his knowledge and experience i n the fields of radioactivity and glass, Dr. Hooley imparted invaluable advice in the development of the techniques used i n this project. INTRODUCTION Before polishing glass, the surface is ground to the desired contour with an abrasive such as emery, alumina, or carborundum slurried with water on a revolving iron form. This process is recognized to be.one of chipping particles of glass away from the surface and leaves the latter rough and translucent. In the ensuing polishing, a finer powder is used, slurried with water on moving f e l t or wax. The powder particles must be hard and insoluble and have a melting point higher than the softening point of the glass being polished. They must also not be too fine as shown by Sir H. Jackson's failure to polish 1 glass with the very fine powder obtained from an iron arc. Rouge and ceric oxide are both in common use today and were used in our work. There are three schools of thought about the mechanism of the polishing operation. The f i r s t claims that i t i s simply a continuation of the grinding operation, a chipping off process, continued to such a degree of fineness that the reflected light waves cannot show any measurable amount of straying. The second believes that the friotional heat of polishing raises the surface temperature to the softening point of the glass and thereby smooths the surface by an actual flow of glass. The third claims a combination of these two mechanisms. - 2 -The evidence for these alternatives will now be considered. POLISHING IS A CONTINUATION OF THE GRINDING PROCESS 2 Sir George Beilby reported that the flawed layer on orystals, such as caloite, was vastly tougher than the crystalline material below. He interpreted this toughening as being due to a non-crystalline flowed layer. This layer was considered to be tougher for i t was able to resist the cutting action of the agent by which 3 the original surface had been satisfactorily ground. Preston, however, found nothing of this sort with glass for he was unable to detect any progressive toughening of the glass with prolonged polishing. He found that material came off steadily at the same rate whether the glass was quite "grey" or had been polished for many hours. From this he concluded that the flowed film, i f i t existed at a l l , had no direct influence on the process and that mechanical abrasion, i.e., the complete removal of glass was the conspicuous feature. Further evidence supporting this mechanism may be obtained 4 from the method Strong recommends for the polishing of precision lenses and prisms. He uses wax instead of felt as the base, and recommends that the contour of the wax base be altered in order to remove any high spots which may have been left on the surface after the i n i t i a l polishing. This prooedure of altering the wax base is continued until the surfaoe becomes planar. This method of e l i m i n a t i n g h i g h spots suggests tha t the important f e a t u r e of the p o l i s h i n g ope ra t i on i s t ha t of a removal of g l a s s . THE SURFACE IS SMOOTHED BY AM ACTUAL FLOW OF GLASS 5 Dreyer and E r t e l used a technique which af forded a means of d i f f e r e n t i a t i n g between a su r face whioh had been p o l i s h e d and a su r face wh ich had been ground. They a p p l i e d a l i g h t p o l a r i z i n g s o l u t i o n c a l l e d "po lacoa t " t o the sur face be ing s t u d i e d . This s o l u t i o n on d r y i n g takes on the o r i e n t a t i o n of the sur face w i t h wh ich i t i s i n c o n t a c t . When t h i s s o l u t i o n was a p p l i e d onto a p o l i s h e d sur face a un i fo rm p o l a r i z e d f i l m r e s u l t e d and l e f t a v i s i b l e r ecord of the su r f ace , w h i l e w i t h a rough sur face the p o l a r i z a t i o n was non-uniform. They removed a po l i shed l a y e r on p l a t e g lass by rubbing w i t h w h i t i n g , an a b r a s i v e m a t e r i a l . Then, when "po lacoa t " was a p p l i e d t o t h i s s u r f a c e , the marks made i n the o r i g i n a l g r i n d i n g of the g lass reappeared. They exp la ined t h i s by a f l o w , i . e . , the o r i g i n a l g r i n d i n g marks were f lowed over du r ing p o l i s h i n g . Another argument i n favour of the flowed l a y e r i s the s o -c a l l e d burn wh ich i s u s u a l l y made when the t a b l e i s a l lowed t o 1 run too d r y . Burns may be i n c r u s t a t i o n s of rouge on the g lass s u r f a c e . Chemical t e s t s show these t o ' b e i n s o l u b l e i n a l l organi s o l v e n t s , and i n a great many cases may be removed by h y d r o f l u o r i c - 4 -acid only. However, burns are not necessarily incrustations of rouge and in some cases rouge may be absent. A burn is readily recognizable for i t is always raised above the surface of the glass. In view of this i t would appear that they may be particles of glass which have been removed during the polishing and re-attached later due to high contact temperatures. Hence i t may be reasoned that i f such high contact temperatures are reached during polishing i t may be possible that there is a smoothing action at work. POLISHING IS A COMBINATION OF FLOW AMD REMOVAL 6 Bayer, in 1938, pointed out that when an attempt was made to improve a poor polish by further polishing, the effort proved fruitless. He interpreted this as a different process at work during polishing than during grinding. He also stated, " With proper fine grinding, after twenty minutes of polishing there will be no unevenness, but only scattered depressions of grinding holes. The surface has already assumed an incipient shine and appears to the layman as 'finished'. At closer observation, however, there is s t i l l a general blueishness on the glass, which disappears with further working. This is occasioned by the flowing process, which takes place with dry polishing material, when the temperature of the glass surface is so high that shell edges of extraordinary small height extenuate into the shell oups ". Those who claim the operation is a combination of the two processes maintain that glass is removed during the i n i t i a l stages of polishing. They consider the wet polishing to be a continua-tion of the grinding, the difference being one of rate. From such considerations i t is evident that further polishing w i l l not improve the resulting surface. For i f the areas which require polishing l i e in surface depressions they would remain untouched during subsequent polishing. They believe that these areas are, however, flowed over during the dry polishing. GENERAL METHOD OF ATTACK To determine whether glass flows during polishing a piece of radioactive glass was fused to a non radioactive sample and polishing carried out in a direction from the radioaotive side to the non radioactive side. Autoradiographs taken before and after polishing were used to show whether active glass flowed over to the inactive side. Uranium was used as the source of radioactivity because i t is an alpha emittor. One reason for - 6 -using an alpha emittor was because of the very short range of these particles. Hence alphas originating below the surface would not activate the emulsion for they would be effectively absorbed in the glass. This would result in a sharp boundary between the two glasses on the ensuing autoradiograph. If a beta emittor had been used a distinct boundary would not have resulted because beta particles originating below the surface would have caused a darkening on the side of the inactive glass in the resulting autoradiograph. The other reason that an alpha emittor was chosen is due to the fact that alpha particles register as distinct individual tracks in nuclear traok plates. This affords a very sensitive method to detect minute amounts of such radioactive materials. Soda-lead-silicate and phosphate glasses were the types polished and were chosen because of their relatively low softening points. Another phase of this project was that of using a radioactive polishing oompound. Thus i f sufficiently high contact tempera-tures were reached during the polishing i t was reasoned that the agent might become fused with the glass, indicating that a flow would be possible. Polishings were carried out using radio-141 active ceric oxide. After neutron irradiation Ce emits beta particles and gamma rays. A Geiger counter was used for detect-ing beta particles emitted from the radioactive agent incorporated - 7 -with the polished glass. In cases where there was a sufficient amount of activity transferred to the surface autoradiographs of the sample were taken to determine the distribution of this activity. EXPERIMENTAL PREPARATION OF GLASSES (a) Silicate Glasses Fig. 1 illustrates the electric resistance furnace used to prepare the soda-iead-silicate glasses. The resistance wire was 50 mil Platinum 10% Rhodium imbedded in high grade alumdum cement. Some of the advantages this alloy possesses over pure platinum are: the tensile strength (at the operating temperature of the furnace) of the alloy is considerably higher than that of 100$ platinums the rate of vaporization is less than platinum; and the melting point of this alloy is I860 °C whereas that of 100$ platinum is 1773 °C. The temperature of the furnace was controlled by a Wheeloo Potentiotrol to ± 20 °C. Glasses of the following compositions were prepared: TYPE "A" Non Uranium Glass SiOg 56.5 per cent SCHEMATIC DIAGRAM OF ELECTRIC RESISTANCE FURNACE (To follow page 7) KEY TO DIAGRAM Pot eati omet er. Chromel - Alumel Thermocouple leading to inside of oven. Oven - clay core consisting of sixteen turns of the alloy, diameter - 9 cms height - 17 cms Insulation - Finely divided clay. Size of circular furnace diameter - 60 cms height - 43 oms Chromel - Alumel Thermocouple leading to outside of oven. Ammeter. Fuse. Cut-off Control. Variac. Source - 110 volts. To Potentiotrol. To Potentiotrol. - 8 -TYPE "B" Uranium Glass This is composition "A" plus 5% of its weight of U-jOg. Si0 2 53.7 per cent Na20 13.3 " " PbO 28.2 " " U 30 8 4.8 " " Batch materials used for the above compositions were: Si0 2 PbO, Na2C03, and U02(N03)2.6H20. The Si0 2 was in the form of a very fine powder and was obtained from Linde. The batch was melted in a cylindrical platinum crucible, 5 centimeters in diameter and 4.5 centimeters in height, at a temperature of 1160 °C for a period of 12 hours after the batoh had become molten. Immediately prior to being poured i t was heated to 1260 °C, then poured onto a previously heated graphite tray and placed in an annealing oven. It was kept at a temperature of 532 °C for 30 minutes, then allowed to cool to room temperature at a rate of 5 °C per minute. (b) Phosphate Glasses Glasses of the following compositions were prepared: v. - 9 -TYPE "C Non Uranium Glass PgOj. 67.0 per cent A 1 2 0 3 4 .0 " » BaO 19.0 " " CaO 10.0 1 1 n TYPE "D" Uranium Glass This is Type nC" plus 5% of i t s weight of U^Og. P o0,- 63.7 per cent A 1 2 0 3 3.9 " " BaO 18.0 " " CaO 9.5 " " U 5 0 8 4.9 " " TYPE "E" Non Uranium Glass This is composition "Dtt but contains PbO instead of U~0 o . PgOpj •• 63.8 per cent A1,0„ 3.8 1 1 " I 3 BaO 18.1 u " CaO 9.5 " " PbO 4 .8 " n - 10 -Batch materials used for the phosphate glasses were: H 3P0 4, A1203.3H20, BaHP04, Ca(H 2P0 4) 2.H 2 0, PbO, and UOg (NO^g.eHgO. In the preparation of these glasses the batch was thoroughly-mixed in a beaker then added to a clay crucible and placed in an ordinary laboratory blast furnace. After the batch had become molten the glass was allowed to remain in the furnace for a period of 1 hour and 30 minutes. It was then poured onto a previously heated graphite tray and placed in an annealing oven, kept at a temperature of 500 °C for a period of 30 minutes, then allowed to cool to room temperature at the rate of 5 °C per minute. When an attempt was made to join the uranium phosphate glass (Type "D") to non uranium phosphate glass (Type "C") i t was found that upon annealing and cooling a fracture developed at the join. To surmount this d i f f i c u l t y i t was decided to use Type "E" instead of Type "C". It was hoped that by adding the same percentage of PbO as of U 0 the co-3 8 efficients of expansion of these two glasses would be similar. However, this also proved fruitless with fractures occurring in the same place. In an effort to make such a join two pieces were placed end to end in a graphite holder, after the ends had previously been polished. Then the holder and glass were placed in a furnace and the temperature raised to 670 °C. - 11 -The glass flowed, hence a join was made and the ensuing sample annealed by allowing the furnace to cool to room temperature i n a period of 15 hours. This did not completely remove a l l the strain, for a l l the samples developed a fracture near this join in the subsequent grinding and polishing procedure. However, only those samples whioh did not fracture a l l the way across the join were used. In some cases these fractures resulted in a complete break between the two glasses, the breaks occurring upon the removal of the samples from the discs after polishing, the samples having been adhered to the disos with wax, GRINDING AND POLISHING EQUIPMENT The machine used i n this experiment was a standard commercial model, the rotating iron disc being 12 centimeters in.diameter. The i n i t i a l fine grinding was performed on this diso, using a fine grade of emery, slurried with water. Polishings were then carried out by using a f e l t pad 10 centimeters in diameter glued onto the diso. In most of the polishings the standard weight (5.5 kilograms) supplied with the machine was used. To regulate the flow of slurry a gravity fed arrangement was constructed as shown in the attaohed photograph I. A pinoh cock controlled the rate of flow. This was measured before and after polishing and averaged. During polishing the sample was held by hand in such a manner as to ensure that the direction of polish was always i n a direction from the active to inactive side. PHOTOGRAPH I (To follow page ll ) - 12 -In the polishings made with radioactive ceric oxide the slurry was not applied by this arrangement but poured onto the disc from a beaker. In this case pressure was applied by hand and not by the standard weight. For these experiments the polishing disc was completely covered by a lucite dome with a pair of heavy duty rubber gloves f i t t e d into two conveniently placed holes cut into its wall. In addition to this dome a small circular cellophane holder was bu i l t around the rotating disc. These precautions were taken to prevent radioactive contamination to workers and surroundings. Photograph II. demonstrates this arrangement. AUTORADIOGRAPHIC MEASUREMENTS (a) Emulsions For alpha measurement NTB - 3 plates were used. An alpha particle entering this emulsion is registered as an individual track upon subsequent development of the plate, thus enabling a minute amount of an alpha emittor to be detected. The alpha tracks registered on the plates were counted on a B & L micro-scope with magnification of 560 X, with an occular disc i n the eye piece. (b) Development of Plates The procedure used for the development of the nuclear track plates is given i n detail below. PHOTOGRAPH II (To follow page 12) - 13 -(1) The plates were developed i n D - 19, diluted with an equal volume of water at a temperature of 4 °C for a period of 30 minutes. Then two volumes of water at a temperature of 22 °C were added and the resulting solution kept at room temperature for a further period of 30 minutes. (2) The plates were then placed i n a 2% acetic acid bath at 4 °C for a period of 30 minutes. Strong nitrogen agitation was applied while the plates were i n this stop bath. (3) Upon removal from the stop bath the plates were placed in the fixing solution, (30$ hypo) main-tained at a temperature of 4 °C for a period of 1 hour. After this time the fixing bath was allowed to come to room temperature and strong nitrogen agitation was again applied. The plates began to clear approximately 5 hours after the commencement of the fixing. The fixing was continued for a period of 15 hours. Following this the plates were washed i n running tap water for 3 or 4 hours. (o) Sensitivity of the Photographic Method 7 Cook and Hudswell were able to prepare thin films of U_0Q - 1 4 -on platinum f o i l or stainless steel by, (1) Defining the area to be covered by the U^Og with a solution of nitro-cellulose in ether-alcohol. (2) Applying uranyl nitrate in pyridine to this area. (3) Heating gently to drive off the pyridine and to destroy the boundary, then they ignited to convert the uranyl salt to U^Og. They were able to determine the amount of U^Og by weighing, finding i t necessary to follow this procedure since losses of uranyl salt or oxide may have occurred during ignition. For the purpose of this experiment i t was not practical to weigh out very small amounts of uranium, therefore i t was..decided to use the same method as Cook and Hudswell, but to omit the ig n i -tion, hence eliminating the necessity of weighing. A minute amount of uranyl nitrate i n pyridine was prepared, - 5 6.132 X 10 grams per ml of solution. Of this solution 0 .5 ml was pipetted onto a previously cleaned pieoe of platinum f o i l . 2 An area of 9 cms was defined on the platinum by painting a border with collodion. The pyridine was allowed to evaporate at room - 15 -temperature. Finally, to expel any excess pyridine, the sample was very gently warmed. At this point an autoradiograph was taken of this area. The track plate was placed directly over this area and clamped. A calculation giving the theoretical number of alpha tracks that this activated area w i l l give per cin^ is given below. Concentration of uranyl nitrate = Quantity of uranyl nitrate used = Area covered = Concentration of uranyl nitrate = Time of exposure of autoradiograph -t i of uranium = Molecular weight of uranyl nitrate = The rate of decay of uranyl nitrate for 3.47 X 10 ° gms per hour is given by, . 6 9 3 - 1 6.02 X 10 J X 3.47 X 10 4.5 X 10 9 X 365 X 24 X 502 = 74 disintegrations per hour But U 2 3 4 is in equilibrium with U 2 3 8 therefore there are two alpha emittors. 6.132 X 10"° gms/ml 3.066 X 10" 5 gms 8.85 cm2 3.47 X 10" 6 gms/cm2 2 hours 4.5 X 10 9 years 502 gms - 16 -Therefore rate of decay, i f this amount is spread evenly over one square cm. 148 disintegrations/hour/cm 2 However, only one half of the alpha particles w i l l activate the film. For a 2 hour exposure this w i l l represent 148 alpha tracks/cm 2 (Theory) Experimental results: Three different areas of this plate were counted. When this plate was observed under the microscope the distribution of alpha tracks appeared to be random and no clumps were noticed. To de-termine the background two blank plates were also counted. Below are the results of these counts. Area Counted Ho. of alpha tracks o Blanks with no uranium 0.4228 cm 59 0.3881 " 53 Total area 0.8109 cm2 Total number of alpha tracks 112 ± 10.6 o Therefore number of alpha tracks/cm 138 ± 13 Sample 0.4101 cm2 114 0.4536 " 128 0.4040 " 111 Total area 1.2677 cm2 Total number of alpha tracks 353 t 18.8 Therefore number of alpha tracks/cm 2 280 £'14.7 Therefore the number of alpha traoks/om from the uranyl nitrate i s 280 138 ± \j(l3)2-h (H.7)' = 142 ± 17 tracks cm If the exposure were increased to a period of 200 hours, —8 3.47 X 10" grams of uranyl nitrate would give this same number of tracks. Thus, the reasonable limit of the sensitivity of this method is reached, for in order to decrease the amount of the nitrate by a factor of 10 this would necessitate prolonging the exposure to 2,000 hours which is impractical. ••8 A similar calculation shows that 1.98 X 10 grams of U_0o 3 8 w i l l give 148 tracks/cm 2 with a 200 hour exposure. I f during the polishing operation 1.98 X 10 8 grams of U^Og were trans-ported from the active side to the inactive side and assuming the glass transported was spread evenly over a square cm of the surface, the depth of this flowed layer would be, Density of glass 2.5 gms/cm3 Composition of glass 4.8% U^ O^ - —8 Therefore number of grams of glass which contain 3.47 X lo" grams of U 30 8 — X 1.98 X 10"8 4.8 -7 = 4.1 X 10 gms of uranium glass Volume of this layer 4.1 X l o " 7 , ^ -7 — — 1,54 X 10 cos. 2.5 - 18 -This would represent a layer of 16.4 A 0. However, i t xvould be possible to detect an increase in track count of 40 tracks/cm^ over the background instead of 148 tracks/cm 2. This would represent a layer of 4.5 A 0. This then is the minimum thickness of a layer which could be detected by this photographic method. (d) Determination of Polishing Compound Transferred to Glass To detect any of the activated polishing agent that may have fused with the glass during•polishing, i t was decided to f i r s t use the Geiger counter, then when positive results were obtained auto-radiographs of the surface were taken. From the following cal-culation i t may readily be seen that the method of detection of betas by the Geiger counter is more sensitive than the method of autoradiography. 20 mcs of activated CeOg were available for this project. 7 1.0 mcs = 3.7 X 10 disintegrations/sec. 8 ' 20 mcs = 7.4 X 10 disintegrations/sec. The 20 mcs were contained in 5 grams of the activated Ce02» Therefore 5 grams of activated Ce02= 4.4 X lO"^ disintegrations/min. If the sample is placed 1.25 cms from the window of the tube in the counter approximately ijr of the total count w i l l register. Hence for the purpose of counting 1.1 X 10^ counts per minute were - 19 -available. It was planned to use. activated CeCv) in the ratio of 1:20 with ordinary Ce02. Therefore 10 grams of the mixture would equal I.I X 109 cts/min. It is possible to detect with ease a count of 25 cts/min over the background. Therefore 25 X 10 = 2.3 X l O - 7 grams of 1.1 X i o 9 the mixture are needed to give this count of 25 cts/min over the background. One of the best films for detecting beta particles is No-screen X-ray film. To darken this f i l m sufficiently to detect 7 / 2 8 beta particles 10 disintegrations/cm are needed. The sensitivity of the photographic method w i l l now be considered and compared with the sensitivity obtainable with the Geiger counter. 141 t i . of Ce -~ 21 days 3 X 10 4 mins Nov/ 25 cts/min on the counter is equivalent to approximately 100 disintegrations/min. dx = - "V N dt Therefore N =I00 X 3 X 10* ^ 4 # 3 x 1 Q6 a t o m s o f ^141. .693 From this i t may readily be seen that i t is impossible for the film to be darkened by 10^ beta particles i f the activity is spread uniformly over one square cm. However, i f the total - 20 -count were 1000 ots/min i t is obvious from the above that 4 . 3 X 10^ atoms of Ce"^ would be present. Since one half 7 of the beta particles activate the film approximately 2 X 10' atoms would be available. Therefore i t would require approxim-ately 21 days to activate the film. From previous calculations —6 i t may be seen that 2 .3 X 10~ grams of the mixture would be necessary to give a total disintegration rate of 1000 dis/min. Thus the counter method is more sensitive than this photographic method by a factor of approximately 10. The following calculation illustrates the depth a disturbed layer must be in order that i t may be detected by the counter method. I t is necessary, for the purpose of this calculation, to assume that the glass into which the ceric oxide is incorporated contains 10% of this activated agent. As previously calculated —7 the minimum amount of Ce02 detectable is approximately 2 .3 X 10 grams. -6 Number of grams of glass detectable = 2 .3 X 10 Density = 2.5 grams/cc Volume of glass = 9.2 X 10"7 ccs o Area of the sample polished = 6.2 cm Therefore depth of the layer 9.2 X 10""7 6.2 - 21 -The following equation gives the relation between the depth of this layer and percentage of-incorporated Ce0 2. d X % = C Where d — depth of layer in Angstrom units. c — constant, in this case equal to 150. RESULTS A series of blank autoradiography were taken at random whenever a new batch of photographic plates were opened and used. Listed below are the results of these blanks from which i t was possible to determine the background count. In addition to these blanks autoradiographs of glass, ground from the inactive to the active side, were also taken. In these cases i t was most unreason-able to suppose that there would be any probability of glass being flowed onto the inactive side. This series also acted as a check i n determining whether there happened to be any radioactivity i n the material used to prepare the glass, excluding uranium, of course. - 22 -BLANKS Sample No* Area Counted No. of Alpha Tracks 1 0.377 cm2 52 2 0.325 " 42 3 0.390 n 54 4 0.372 " 51 5 0.342 " 44 Total area 1.806 cm2 Total number of alpha tracks 243 — 15.6 Therefore number of alpha tracks/om2 135 - 8.6 GROUND GLASS AUTORADIOGRAPHS Sample No. Area Counted No. of Alpha Tracks 1 0.403 om2 58 2 ' 0.390 " 53 3 0.387 " 52 4 0.396 ,f 53 5 0.345 " 46 Total area 1.921 cm2 Total number of alpha tracks 262 ± 16.2 Therefore number of alpha traoks/cm2 137 — 8.5 In the results tabulated below, the method used the number of alpha tracks per square centimeter was to record to mark off - 23 -a small square (approximately 0.5 cm2) on the resulting auto-radiograph, about 0.2 cms from the boundary between the active and inactive side. The number of alpha tracks in this area was then counted, use being made of a microscope with magnifi-cation of 560 X. For any positive results (by positive results i t is implied that uranium glass was flowed over to the non uranium glass) one half of the total of the alpha particles from the uranium would register as distinct tracks in the photographic emulsion. I t was possible to arrive at this factor of one half because the glass surface was clamped tightly to the plate. There would have been no correction necessary for absorption of alpha tracks by the glass. Any amount flowed would have been so small that the absorption of alpha particles by this flowed glass would have proved negligible. The following series of polishings had the undermentioned standard conditions :-SERIES A (1) Type of glass lead glass (Type A & Type B) (2) Finely ground with fine emery (3) Composition of slurry 25 gms of Ce02/l000 mis of water (4) Rate of flow of slurry 40 mls/min (5) Standard weight of 5.5 kilograms (6) Time of exposure of autoradiograph 100 hours - 24 -Sample Ho. Time of Polish No. of Tracks IA 5 sees 130 2A 10 » 132 3A 20 it 128 4A 3 0 n 134 5A 1 min 132 6A 2 mins 131 7A 4 t: 136 8A 8 it 132 9A 10 tt 132 In the above series the time of polishing was increased continually from five seconds to ten minutes, thus, i f a flow did take place, i t would have been possible to determine at which time during the polishing operation this flow was most pronounced, and i f i t varied with the degree of polish. p By comparing the number of tracks recorded per cur on the p above autoradiographs with the number of tracks per cm on the blanks, i t is at once obvious that there was no flow during this set of polishings. I t was then decided to increase the amount of polishing agent i n an effort to determine whether this variable had any effect on flow. Hence, the amount of polishing agent i n the next set was increased from 25 grams to 50 grams of CeOg/lOOO mis of water. Other than this change in composition of the slurry, this series WB" had the same standard conditions as the previous s eri es. - 25 -SERIES B Sample No, Time of Polish No. of Tracks per cm' IB 5 sees 140 2B 10 " 138 3B 20 " 134 4B 30 " 135 -5B 1 min 131 6B 2 mins 136 7B 4 " 130 8B 8 " 132 9B 20 " 143 There was no indication of flow, even with twice the rate of addition of ceric oxide. If the manner of recording results is considered, i t is possible to reason that there may have been flow in the previous polishings, but this flow was over such a minute distance that i t was impossible to detect i t because no glass would have flowed a l l the way from the boundary to the area counted, i.e., a dis-tance of 0.2 cms. In an effort to answer this question counts were made as close as possible to the boundary for a l l plates in the "A" and "B" series. This was done by making a single sweep of the autoradiograph, the edge of the f i e l d being 0.218 nuns from the outermost position of the boundary between the inactive and active sides. This dividing line could readily be seen on the plate, i t was very distinct both visually and under the microsoope. - 26 -Sample No. Area Counted Uo. of Tracks IA 0.015 cm2' 1 2A 0.033 n 3 3A 0.028 " 4 4A 0.040 " 5 5A 0.034 tt 2 6A 0.042 " 6 7A 0.041 " 7 8A 0.043 " 6 9A 0.031 " 4 13 0.011 " 2 2B 0.026 *.* 3 3B • 0.012 " 4 4B 0.020 " 2 5B 0.019 " 2 6B 0.026 " 3 7B 0.020 n 4 8B 0.021 " 3 9B 0.014 " 3 Total area counted 0.476 cm Total number of alpha tracks 64 ± 8 Therefore number of alpha tracks/cm 2 138± 17 Again no evidence of any increase in track count is apparent, thus indicating that i f flow does take place i t is over a distance of less than 0*22 mms. After negative results were obtained using ceric oxide as the polishing compound i t was decided to use rouge in order to deter-mine whether this type of agent had any effect on a flow process. Because in the previous series the amount of agent appeared to have no effect i n a flow i t was deemed unnecessary to vary the amount of the polishing compound. The following set of polishings had the undermentioned standard conditions. - 27 -SERIES '0 (1) Type of glass lead glass (Type A & Type B) (2) Finely ground using fine emery (3) Composition of slurry 50 gms of Fe20y^l000 mis of water (4) Rate of flow of slurry 40 mls/min (5) Standard weight of 5.5 kilograms (6) Time of exposure of autoradiograph 100 hours Sample No. Time of Polish No. of Tracks per cm' 1C 5 sees 129 2C 20 " 136 3C 50 " 135 4C 100 " 133 5C 200 " 136 Again there was no evidence of any flow. A phosphate glass was then polished with the thought that its lower softening point would be more conducive to flow. Again i t was deemed unnecessary to vary the amount of the agent supplied for reasons already given. The following series had the under-mentioned standard conditions :-SERIES D (1) Type of glass phosphate glass (Type D & Type E) (2) Finely ground using fine emery (3) Composition of slurry 50 gms of Fe„0.,/1000 mis of water - 28 -(4) Rate of flow of slurry 40 mls/min (5) Standard weight of 5.5 kilograms (6) Time of exposure of autoradiograph 100 hours Sample No. Time of Polish No. of Tracks per cm' ID 5 sees 137 2D 15 " 130 3D 50 " 141 4D 120 " 136 5D 240 " 129 Again i t is readily seen that there was no increase in the background count indicating no flow. Hence this l e f t the alterna-tive of using ceric oxide instead of rouge as the polishing agent, therefore, in the following series the only change i n the standard conditions of polishing was that ceric oxide was used in place of rouge. The composition of the slurry being 50 grams of ceric oxide per 1000 mis of water. Sample No. Time of Polish No. of tracks per cm' IE 5 sees 133 2E 15 " 138 3E 50 " 135 4E 120 " 150 5E 240 " 129 Again there was no evidence of any flow taking place. In the f i n a l set of polishings both types of glass were polished with ceric oxide and rouge. The procedure of polishing was altered in that force was applied by hand and not by means of - 29 -the standard weight. The method employed was that of polishing for a period of twenty minutes using a slurry composed of 50 grams of the agent i n 1000 mis of water, the rate of flow of slurry during this wet polish being 40 mis per minute. At the end of 20 minutes the table was allowed to run dry and there were no further additions of the polishing compound. This operation of the dry polish was continued for a period of 10 minutes, following this an autoradiograph with a 100 hour exposure was taken of the resulting polished surface. SERIES F Lead Glass Sample Ho. Type of Agent Ho. of alpha tracks/cm IF Ceric oxide 133 2F Rouge 131 Phosphate Glass 3F Ceric oxide 129 4F Rouge 129 There was no indication of flow even when the table was allowed to run dry. Dry polishing was continued for a period of 10 minutes after 20 minutes of wet polishing. Counts were again made as close as possible to the boundary in a similar manner to that done in series "A" and !,B". The outermost point of the boundary being 0.218 mms from the edge of - 30 -the f i e l d and again a single sweep taken of the autoradiograph. Sample No. Area Counted No. of alpha tracks IF 0.033 cm2 5 2F 0.021 " 2 3F 0.045 " 7 Total area counted 0.099 cm2 Total number of alpha tracks 14 ± 3.7 Therefore number of alpha tracks/cm 2 141 ^37.4 Again there appears to be no evidence of a flow, however, the- probable error is great because of the limited number of plates. Therefore these results are not conclusive in themselves and serve only as an indication. The following procedure was carried out when the glasses were polished with radioactive ceric oxide. (1) 2 mcs of irradiated ceric oxide were mixed with non-irradiated ceric oxide to give 10 grams of the mixture. This was then slurried in 200 mis of water. (2) Approximately 5 mis of this slurry was rubbed into the f e l t and polishing carried on for 15 seconds. At the end of this time the table was stopped and a further 5 mis of the slurry added, this - 31 -procedure being repeated every 15 seconds. Pressure was applied by hand. Picture (2) represents this technique. However, the slurry in the beaker was poured through a funnel in the lucite dome. When polished, the samples were thoroughly washed and scrubbed with a nylon brush in running water, alcohol, a soap solution and f i n a l l y more water. Below is an example of the procedure followed in recording the counting rate of a l l samples. A standard sample was f i r s t counted to determine whether the plateau of the tube had changed. Following this the background was taken and f i n a l l y the sample counted. Example In one determination of the background, 294 counts were recorded in 5 minutes and the sample recorded 406 in the same period of time. Therefore background rate is ' 294 ± 17.0 5 5 59 i 3.4 cts/min Total rate of sample is 406 ± 20.1 5 5 81 ± 4.02 cts/min - 32 -Therefore net rate is 81 - 59 ± \K3.4) + (4.02) 2 = 22 ± 5.2 — 22± 5 cts/min Sample No. I Phosphate Glass Type "C" 6.2 om This sample was polished 2jg minutes then washed as i n the above procedure. Net rate 92 ± 7 cts/min After this treatment the polished glass was subjected to re-washing and re-scrubbing as before and in addition was allowed to stand overnight in alcohol. The oount taken 48 hours later was The decrease in activity was completely accounted for by the decay of the active cerium. Hence this treatment of re-washing and re-scrubbing did not remove any activated CeO>,. An autoradiograph (l) of the sample was taken with a 72 hour exposure. From a study of the radiograph i t was seen that the activity accumulated on the surface i n spots and not i n a uniform manner. The spots on the f i l m corresponded exactly with minute pits i n the polished surface of the glass. Net rate 8 0 ± 6 cts/min AUTORADIOGRAPHY (to follow page 32) Autoradiograph III 4 The sample was then placed in a solution of 6 N HC1 and KI for twenty seconds at 70 °C. This solution readily dis-solves Ce0 2. The count taken of the sample before and after being immersed i n this solution is given below. Count before being placed in the acid solution 59 — 6 cts/min " after " " " " " " 13 -± 5 " " An autoradiograph with an exposure of 72- hours failed to show any activity on the polished surface. Only a faint blacken-ing of the film was observed from activity lodged in the roughly ground sides of the sample. Sample Mo. 2 Phosphate Glass Type "C" 6.4 cm2 This sample was polished and treated, in the usual manner, total polishing time being 8 minutes. Net rate 22 ± 5 cts/min An autoradiograph taken of this surface gave only a few scattered spots similar to radiograph No.l, but they were less dense. This lack of density made i t impossible for a print to be made of this ..film. Again the spots corresponded exactly with minute pits i n the polished surface of the glass. Sample No. 3 Soda-lead-silioate Glass 3.0 cm2 This sample was polished 6 minutes and treated in the - 34 -usual manner. Net rate 384 ± 9 cts/min The resulting autoradiograph (2) taken with an exposure of 24 hours again showed the activity to be accumulated in spots on the surface which corresponded with minute pits in the polished surface of the glass. This sample was polished for a further 15 minutes, giving a total polishing time of 21 minutes. Net rate 242 + 9 cts/min Autoradiograph (3) had an exposure of 44 hours. Again i t is seen that the activity was accumulated in spots. Scratches v i s i b l e to the naked eye appeared i n this sample when the polishing f e l t was changed five minutes before the end of the polishing. They were attributed to particles on the new f e l t . In the resulting autoradiograph (3) darkening appeared on the f i l m directly above these scratches, indicating the presence of CeOg. To improve the polish of the sample, polishing was carried on for an additional 21 minutes. Net rate 12 ± 2 cts/min The autoradiograph of this sample showed no darkening whatsoever. - 35 -DISCUSSION The method used to investigate the problem of determining whether glass flows during polishing was primarily one of being able to detect any radioactive glass flowed over to a non radioactive glass. As previously shown the photographic method used enabled the detection of a layer 4.5 A 0 i n thickness. However, i f during the polishing operation particles of glass are removed and later re-attached these particles would leave clumps of alpha tracks on the resulting autoradiograph. For example, i f 6 alpha tracks were left in a clump this would represent :--8 / o 1.98 X 10 grams of U^Og give 148 alpha tracks/cmr on an auto-radiograph with a 200 hour exposure. Therefore 6 R - . X 1.98 X 10"° 148 -8.04 X l O - 1 0 grams of U^Og are required to give such a clump. Composition 4.8% U,0o j o 100 _io Therefore X 8.04 X 10 4.8 —8 - 1.67 X 10 grams of uranium glass w i l l give such a clump. Density 2,5 gms/co - 36 -Therefore volume occupied by this weight of glass i s 1.67 X 10"8 2.5 6.7 X 10~9 ccs Hence a clump of 6 alpha tracks observed in the resulting -9 autoradiograph would have represented a chip of 6.7 X 10 ccs of uranium glass, had this chip been removed from the active side and later incorporated on the inactive side. Assuming such a chip to be spherical the diameter would be 0.019 mms. Such clumps were noticed i n a few radiographs. However, upon re-washing and re-scrubbing these particular samples the clumps were not reproduced i n the ensuing autoradiographs. Because the particles were so easily removed i t was concluded that they were merely trapped i n the surface and not due to any flow. If during polishing a flow did take place i t was reasoned that the sharp boundary would have disappeared leaving instead a gradient. In a l l the autoradiographs taken there was no evidence whatsoever of this phenomenon, the boundary remained sharp and distinct. - 37 - • From the results of the experiments conducted using radioactive glass, i t was concluded that the most conspicuous feature during polishing was that of removal of glass, rather than that of a smoothing or flowing action resulting from the softening point of the glass being reached. This statement i s modified to include the type of glass polished and the con-ditions of polishing. I t may also be pointed out that even when the table was allowed to run dry there was s t i l l no evidence of any flow taking place. Thus i f flow occurs, i t is over a distance of less than 0.22 mms even when the polishing is done with a dry f e l t under .3 kg/cm2 for 10 minutes. The results of the experiments conducted with activated CeOg indicate that some of this polishing agent remains on the polished surface i n spots after polishing is ceased. If ceric oxide i s entrapped i n the crevices of the surface this would account for the spottiness of the resulting autoradio-graphs. I t would also account for the decrease i n activity with longer polishing times for with prolonged polishing a finer polish i s naturally attained, thus fewer crevices are available to entrap the ceric oxide. The pits and lines i n the surface corresponded exactly with the dark areas on the autoradiographs, as could be seen with the naked eye. Further evidence whioh suggests that the CeOg i s merely entrapped in the surface was - 38 -gained from the experiments in which the ceric oxide was so readily dissolved by the acid solution. I f the Ce0 2 were fused with the glass i t would be expected that the rate of attack would have been much slower. The acid solution w i l l attack the glass, therefore this evidence is not at a l l con-clusive and is presented as a suggestion. If fusion of the CeCv, with the surface glass took place i t was reasoned that there would probably have been an increase i n the amount of the agent incorporated with the glass as the polishing time was prolonged. However, there was no increase in the amount of ceric oxide on the surface with prolonged polishing times but rather a decrease. This mechanism, there-fore, does not appear to be the one involved. To account for the impossibility of removing CeO^ from the surface by washing i t was possible to reason that during polishing crevices were flowed over, thus securely entrapping this polishing agent. I f this were the case i t was expected that a considerable amount of activity would have been present on the surface after prolonged polishing. Such was not the case, hence i t was again reasoned that the most conspicuous feature of the polishing operation was that of glass removal. - 39 -FUTURE WORK (1) Use radioactive rouge to determine whether this agent fuses with the glass during the polishing operation. (2) Further work might be carried out on the study of the flowing over of scratches. A possible method of attacking this problem would be that of scratching a well polished surface then rubbing radioactive ceric oxide or rouge into the scratches. After this treatment the glass could be polished i n an attempt to flow over these scratches. A failure to remove any of the radioactive material by washing would indicate that the scratches had been flowed over. By using a glass that is not readily attacked by a solution which would dissolve the radioactive agent i t would be possible to determine the depth of the glass flowed over. The rate of attack of this solution on the glass would have to be determined. (3) The work done on phosphate glasses might be checked to de-termine whether the fractures interfered with a flow. A study of the coefficients of expansion of this glass would f i r s t be necessary i n order that a good join be made. Y/hen a join is made in a flame the boundary is more distinct than i f it-' is made i n a furnace.