' ( A STUDY OF THE GAS EVOLVED BY HEATING SILVER BROMATE BY william Mclennan Morgan thesis For the Degree of Bachelor of Science IN CHEMISTRY COLLEGE OF LIBERAL ARTS AND SCIENCES UNIVERSITY OF ILLINOIS 1921 UNIVERSITY OF ILLINOIS M8JL28* i 9 ri.. THIS IS TO CERTIFY THAT THE THESIS PREPARED UNDER MY SUPERVISION BY Willi am M c L ennan M o r g an ENTITLED A.. ...tM...Saa...E)y.Dly.ad...by...Ee.a.tlng...S.i.l.v.e.r. Bromate IS APPROVED BY ME AS FULFILLING THIS PART OF THE REQUIREMENTS FOR THE DEGREE OF Instructor in Charge Approved : HEAD OF DEPARTMENT OF ’ \t)l\ N\^ A ACKNOWLEDGEMENT The writer wishes, at this time, to express his appreciation for the invaluable assistance of Doctor J. H. Reedy, who suggested the problem and under whose able direction this research was done. TABLE OF CONTENTS I Introduction II Experimental 1. Material Used (a) Preparation of the Silver Bromate (b) Determination of purity 2 Evolution of gas upon heating (a) Methods and Apparatus (b) Analysis of gas samples (c) Spectrum of Lhe gas (d) Analysis of residue (e) Adsorptive power of ignited Silver Bromate 3 Properties of Silver Bromate (a) Behavior towards light (b) Decomposition upon heating (c) Existence of other Silver Oxy-bromine Compounds 4 Silver Bromate as a standard in Iodimetry III Summary TV HI hT 1 r->Vvw Digitized by the Internet Archive in 2015 https://archive.org/details/studyofgasevolveOOmorg LIST OF PLATES Pag© I Apparatus Used to Collect Gas Evolved Upon Heating Silver Bromate 5 II Apparatus Used for Nitrogen Determination 8 it III Plucker Tube Apparatus Used for Spectrum Analysis 11 1 A STUDY OF THE GAS EVOLVED BY HEATING SILVER BROMATE I INTRODUCTION The object of this thesis is to study, in general, the decomposition of silver bromate, but more especially to study the gas evolved by silver bromate when heated. In the work on the transition point of silver bromate, which was carried out by S. J. Gould in the Chemical Laboratory at the University of Illinois, it was noticed that a gas was evolved upon heating silver bromate in contact with a high- boiling paraffin oil. . The evolution of this gas became notice- able slightly above 100 degrees C. Very naturally, it was at first assumed to be carbon dioxide, formed by the oxidation of the oil. At the same time there was a darkening of the oil, suggesting the formation of asphaltic substances and a blackening of the silver salt. Upon removing samples of the silver bromate residues, washing thoroughly with ether, and then with ice water, using small amounts only, no sensible change in the composition of the salt could be detected. This led to an attempt to identify the gas. The test for carbon dioxide was negative, and as only a very small amount of the gas was available at the time, further identification of the gas was not attempted. The study of this phenomena has been resumed and extended so as to include the general chemical behavior of silver brom- ate. There is practically nothing in the available chemical 2 literature which deals with this problem, hence the work has been more interesting because of tne ract that the problem is almost an entirely new one. II EXPERIMENTAL (D Preparation of the Silver Bromate In the present work, the silver oromate was prepared from equivalent amounts oi silver nitrate ana potassium uromate free from potassium oromide. The silver salt was covered with water, waxen was brought to an incipient ooil; the potassium bromate was added in tne form of a hot solution. The mixture was digested with frequent stirrings for two to three hours, ana then thoroughly cooled in ice water. The precipitated silver bromate was filtered out, and the digestion process repeated twice. The final crop of crystals was dried as much as possible on a suction filter and then transferred to a watch glass and heated to 140 degrees C. for an hour in an electric drying oven. The product was pure white, and has been preserved for several months over anhydrous calcium chloride in a desiccator with no darkening effect whatever. An interesting thing might be mentioned in this connection: The decomposition of silver bromate at high temperatures is hastened by contact with certain substances that might be considered to be inert. A quantity of silver bromate crystals were heated on a porous porcelain drying plate in an electric oven for about one hour, the maximum temperature being 131 deg.C. a-— il t : ... . j . . - • ' . • . c/isS 4 . ; . • ■ V . ; . . . • ■ ■ ; • • • , 3 The part in contact with the plate was quite yellow, indicating the formation of silver bromide. This seems to have been due to a catalytic effect of the clay on the decomposition of silver bromate, rather than an actual reduction, and it is important to note that in reduction actions upon silver bromate as upon all silver salts, the product is usually black. When the drying was carried out on a glass plate no decomposition occurs as when the clay plate was used. Determination of the purity of Silver Bromate The purity of the recrystallized silver bromate was deter- mined by converting it into silver bromide by means of hydro - bromic acid, and weighing the residue after evaporation. The reaction: AgBrO^-t 6HBr — ► AgBr + 3H 2 0 + 3Br 2 The hydrobromic acid was carefully redistilled so as to remove all non-volatile matter. The record of this work is: (0 ( 2 ) wt. of dish silver bromate 33*2111 gms 38.8524 tl 11 tl 32.6989 38.3369 u 1! it it .5122 • 5155 Wt'. of dish silver bromate 33. 1068 guts 38.7475 n 11 n 32 . 6989 38.3369 n II : ” " (exp) .4079 .4106 ii 11 " " Uet) .4079 .4106 Purity U)Q/o 100^ t ' . . ■ 4 Analysis of the samples shows that the usual purity of silver bromate prepared In this way ranges from 99*6 to 99 . 8 $, but that on prolonged heating for several hours at 130 to 135 degrees C. a purity of 100$ was obtainable. The deviation from absolute purity might probably be due to the presence of a small amount of silver perbromate, but the most likely assumption is ( 2 ) that the foreign substance is water or air. Stas has reported that silver bromate of 100$ purity could not be prepared, owing to the fact that the last traces of water could not be driven off below the decomposition point of the salt. This supports the assumption that the impurity is water. On the other hand, however, it was later discovered in this research that a one gram sample of silver bromate adsorbs two cubic centimeters of air which would amount to an impurity of about .3$ just as found above. It therefore seems to be a very reasonable assumption that the impurity is adsorbed air. Evolution of G-as upon Heating Methods and Apparatus In order to obtain samples of the gas which Gould noticed was formed upon heating silver bromate in contact with high- boiling paraffin oils, apparatus of the form shown in Figure I was devised. Two-gram samples of silver bromate were introduced into the bulbs. In one tube the salt was covered with fresh conductivity water, so as to exclude as much as possible the presence of dissolved air. A high-boiling paraffin oil was used fbjRUjvj Vn? jo (ouicj (jis tvoivu VjM Hupucj ^ivtt Uimujt * 5 in the second apparatus. To keep out air, an atmosphere of hydrogen was maintained in the set-up. The two tubes were arranged in series, and the bulbs immersed in the same bath of cottonseed oil which was kept at a temperature of about 110 degrees G. by means of a carbon incandescent electric light. The gas was evolved more rapidly in the water-filled tube than . 6 r in the oil filled one, the ratio being about four to one. This was possibly due to greater viscosity and to the fact that the oil may have united chemically with some of the oxygen liber- ated, forming asphaltic substances. The volume of the gas collected in the apparatus containing water was 8.4 cubic cent- imeters in two to three weeks. Analysis of G-as Samples The gas, which was evolved when the silver bromate was heated under water, was analyzed in the following manner: The collected gas was treated with potassium hydroxide to remove any carbondioxide that might possibly be present, then with alkaline pyrogallol to remove the oxygen. As these two absorbents remove carbon dioxide and oxygen quantitatively, the contraction in volume enables the analyst to calculate the percent of the various constituents present. The hydrogen was removed by sparking with a known amount of pure oxygen in a eudiometer. The excess oxygen was later removed by absorption in alkaline pyrogallol. The fact that there was no contraction in volume after the sparking indicates that there was no hydrogen present in the gas. As is the usual procedure in gas analysis the nitrogen was determined by difference. The record of the analysis of two samples of the gas is as follows: . 7 Sample I Original volume 8.4 cc After KOH 8.4 co 2 none After alkaline pynogallol 6 . 9 °2 1.5 cc 17.90$ Oxygen added (pure) 3-3 — Total volume 10.2 Sparked for hydrogen 10.2 h 2 none Nitrogen by difference h 2 6.9 cc 82.10$ Sample II Original volume 5*9 cc After KOH 5-9 co 2 none After alkaline pyrogallol 4.9 o 2 1 .0 cc 16.95$ Oxygen added (pure) 2.3 Total volume 7.2 Sparked for hydrogen 7.2 none Nitrogen Dy difference n 2 4.9 cc 83.05$ These two analyses agree very well for experimental work, hence all assumptions in the remainder of this paper will he based upon them. The test for nitrogen was made qualitatively. The apparatus as shown in Figure II was evacuated and then filled with hydrogen. In the interior of the evacuated tube were several pieces of magnesium ribbon. The gas upon which the test for nitrogen was to oe made, was placed in the tube with the 4 . • . * . . . * . c . . . 8 Jlmwm fbtp fins. Hnutt* hTtinutiniwi n 1 magnesium ribbon, the stopcock closed and the tube heated until the nitrogen present combined with the magnesium. 3Mg -1- Np ► Mg, Up After the reaction was completed, the stopcock was opened and the water from the large dish allowed to enter the bulb. The water was allowed to react with the magnesium nitride, and the resulting ammonia gas absorbed in hydrochloric acid. The I ammonium chloride formed was neutralized with sodium hydroxide and the solution boiled. The odor of ammonia given off during ! i I { -■ - si , ■ 9 the boiling, and the fact that the fumes evolved turned blue litmus red, indicated conclusively that the original gas contained nitrogen. Mg 3 N 2 + 6H 2 0 3Mg(0H) 2 + 2NH 3 2NH,+ 2HC1 2NH.C1 3 4 NH.C1 + NaOH - NaCl + H.O + NH_ 4 2 3 Since all halogen compounds would have been absorbed in the water over which the gas was collected, it is certain that none of them were present in the gas as analyzed, and as nitrogen is always determined by difference in gas analyses it may be considered that nitrogen is present in this gas to the extent of 82.10^. This would indicate that the gas evolved Dy silver Dromate when heated is adsorbed air. Since the amounts of oxygen ana nitrogen in the air are 21 % and 19 % respectively, it is seen from the above data that silver bromate is a better absorbent for nitrogen than for oxygen. Analysis of the gas samples obtained from the silver bromate under oil, while they were all smaller in quantity, gave the same analysis of the gas, showing that the gas came from the silver salt and not from the liquid with which it was covered. . - . . . * * Spectrum of the gas 10 In making a spectrum analysis of the gas, a sample of approximately six cubic centimeters was used. The hydrogen, oxygen, and carbon dioxide were removed as before. The appa- ratus, as shown in Figure III, was evacuated by means of an oil pump. At intervals a small quantity of gas was admitted to the apparatus to sweep out any remaining traces of air that might have been present. When the last of the gas had been admitted it to the tube and a high vacuum obtained, the Plucker tube was sealed off from the rest of the apparatus. Several attempts were made at photographing the spectrum of the gas. This work was done by L. F. Yntema in the Chemical Laboratory of the University if Illinois. The attempts along this line were in the main part unsuccessful, although suffi- cient information was obtained to establish the fact that the gas was air. In an effort to obtain additional data, a spectro- scopic examination was made. The spectrum of the gas as viewed through the spectroscope was not very distinct. Lines were located, however, which indicated the presence of argon and bromine. Two distinct band spectrums for nitrogen were visible. The presence of the gases is easily explained in view of the fact that argon and nitrogen are present in all air, and that bromine is one of the elements present in the original salt from which the gas was obtained. Further spectroscopic examination was considered unnecessary, in view of the fact that all evidence pointed to the conclusion that the gas was air. * . ' . . - . . . * . . 11 hum ‘“vu ftmwvj? V^ed : 0L «?|triVNt j\MMY5l3 It is supposed that the air was occluded during the pre- cipitation of the crystals, or else adsorbed later. The latter supposition is favored by the fact that the silver bromate was always precipitated from hot solutions. Analysis of residue In order to determine the composition of the residue remaining in the tube after the evolution of the gas, the follow- ing experiment was performed: The residue, after being allowed to dry In a desiccator over calcium chloride, was weighed. It was then treated with 33$ hydrobromic acid, and evaporated to dryness on a steam bath. Following this evaporation, the residue was placed in an electric drying oven, and heated for two hours at a temperature of 110 degrees C. The silver bromate in the residue was con- verted into silver bromide in this process, the loss in weight being equal to the oxygen given off by the silver bromate. A comparison of the weight of the oxygen evolved by the residue on treatment with hydrobromic acid, checks within experimental error with the theoretical amount of oxygen that would have been evolved by an equal amount of silver bromate under the same conditions, considering the salt to be pure. This proves that the residue is nothing other than silver bromate, and that the gas which is evolved upon heating silver bromate does not result from any decomposition of the salt* Adsorptive power of ignited Silver Bromate To obtain some information relative to the adsorptive power of silver bromate, a study of the adsorptive power of a two gram sample was made. Original wt. of dish 24.2605 grams Wt. of silver bromate 2.0000 Total wt. 26.2805 The dish containing the silver bromate was heated in an electric drying oven for ninety-six hours at a temperature of . . • 13 approximately 110 degrees C. The loss in weight of .0030 grams was considered to be the weight of the adsorbed air driven off from the salt. The sample was now removed from the §ven and placed in a desiccator over anhydrous calcium chloride. The gained weight very slowly, and it was assumed that this slow increase in weight was due to a slow adsorption of air. Wt. after heating for 96 hours Wt. after standing for 144 hours Wt. after standing for 288 hours Wt. after standing for 432 hours Original weight 26.2775 grams 26.2780 26.2785 26.2793 26.2805 This shows that there is a slow adsorption of air which will eventually continue until the salt has as much adsorbed air in it as it originally had. Properties of Silver Bromate Behavior towards light In studying the behavior of silver bromate towards light, d some pure silver bromate prepared as outline previously was used. Pure silver bromate appears to be wholly unaffected by ( 2 ) light when dry. The wet salt was not found to be so stable. A portion of the pure salt was placed in a clean test tube, covered with water, and the top of the tube sealed off. For this work pure conductivity water was used. After standing 14 several weeks the side next to the window showed a slatey gray color, while the rest was only slightly affected. To see if there was any change in volume during this darkening effect, due to the formation of some insoluble gas (such as oxygen), some wet silver bromate was sealed in a tube with a manometer containing some 1 nujol' colored with azobenzene to facilitate reading. After standing for several weeks there was no displacement of the levels in the arms as shown by a blank apparatus containing only water placed alongside. In a further study of the stability of silver bromate, some wet crystals were heated. A marked darkening effect was noticed. This does not occur with the dry salt which is stable up to its melting point when pure. This evidence leads to the conclusion that dry silver bromate is stable towards light and heat, but in the presence of water darkens slowly at low temperatures, and rapidly at high temperatures. Decomposition upon heating On heating in a bath of Wood's metal, the purified silver bromate was found to have a sharp melting point at 308 to310 degrees C. At that temperature the salt melts to a clear color- fa) less liquid. 'Stas' has reported that silver bromate when heated to 150 degrees C. will decompose. This was found to be erroneous by experiment with 100/6 pure silver bromate. . • , . . \ ... . . . 15 After obtaining a definite melting point for silver bromate, a study of the decomposition was made. It was found that any impurity catalyzes the decomposition of the salt into a terrific explosion. Using a sample of silver bromate weighing about .02 grams, this decomposition was demonstrated when a minute quantity of ferric oxide was placed in the tube. Similar results were obtained when a drop of 'nujol', a drop of water, or a small piece of dirt from the laboratory desk were placed in the tube with the silver salt. All these explosions took place at temperatures ranging from 260 to 295 degrees C. Very impure silver bromate can be decomposed explosively at temperatures as low as 155 to 150 degrees. The explosions resulting from several of these decompo- sitions were so violent that the Wood's metal of the bath was blown up against the ceiling. The writer has found that all precautions must be taken for the safety of the experimenter when this decomposition is being studied. Upon decomposition, silver bromate goes to pieces yielding, very probably, the following substances! Silver Silver bromide Bromine Oxygen. These substances are probably the results of the reaction, but the statement cannot be verified because the explosive nature of the reaction makes it impossible to collect the — ■ - — . , .. v l * V ' . v. • - . ... V . — 16 products. However, as a distinct odor of bromine is noticed after the explosion, it seems very probable that silver bromate decomposes in some such manner as outlined above. Existence of other Silver Oxy-bromine Compounds The deviation of the composition of silver bromate from 100^ purity and the formation of a gas upon heating, might be due to the presence of small quantities of some other silver oxy-bromine compounds, such as, silver perbromate, or silver ( 3 ) hypobromite. With regard to the postulate of the existence of silver perbromate, the following remarks should be made: The actual existence of perbromic acid or any of its salts ( 4 ) is very much to be doubted. 'Kammerer' reported that per- bromic acid was formed by the action of bromine on dilute perchloric acid solutions, and that the usual series of salts were formed by this acid. These results have not been confirmed by other investigators. The same uncertalnity exists with regard to the existence of silver bromite and silver hypobromite. With regard to the latter it should be said that their presence in silver bromate would make the results with the hydrobromic acid analysis too high, and not too low, as was actually found. Furthermore, reasoning from analogy with the corresponding silver oxy-chlorine compounds, they would be more soluble than the silver bromate, and their presence in a carefully purified substance like the ■ . • . 1 t ■ ■ ' ■ 17 salt used in this work is hardly to be expected. Nevertheless, evidence of the presence of such silver- oxygen-bromine compounds was sought in the following way : Samples of the purified silver bromate, weighing one gram each, were placed in Erlenmeyer flasks of 200 cubic centimeters capacity provided with refluxing condensers. Ground glass joints were used, so as to eliminate contact with reducing materials. The salt was covered with 100 cubic centimeters of distilled water, and the flasks placed in jackets designed to heat them at constant temperatures for considerable lengths of time. The heating was effected by electric lights, and was extended throughout a period of seven days. One lamp gave a temperature of 102 to 108 degrees C., while the other gave 85 degrees. In both flasks, crusts of crystalls collected on the surface of the solution. These were skimmed off, analyzed by the hydrobromic acid treatment, and the following data obtained: Wt. of dish silver bromate 25.3843 grams Wt. of dish 24.2810 Wt. of silver bromate 1.1033 Wt. of dish silver bromide residue 25.1687 Wt. of silver bromide .8877 Silver bromlte equivalent 1.1146 This high value was to be expected, since the crust had a yellow tint indicating the presence of silver bromide. In the flask which was heated to 108 degrees C., a black 18 deposit formed on the bottom. The analysis of this deposit likewise indicated that partial decomposition of the silver br ornate had occurred. Confessedly, this attempt to obtain evidence of a tendency of silver bromate to undergo decomposition, forming products intermediate in composition, was a failure, as none of the products obtained appeared to be definite. Silver Bromate as a Standard in Iodimetry On account of the high purity, and the ready method of determining composition, the possibility of using silver bromate as a standard in iodimetry suggests itself. This was tried out as follows i The purity of a quantity of silver bromate was determined, by converting it into silver bromide by the hydrobromic acid method as outlined previously in this paper, and a value of 99*8/£ obtained. A sample of silver bromate (1.0463 grams) was weighed out, water was added, then an excess of potassium iodide, and the mixture finally boiled for several minutes. After cooling, the solution was made up to 250 cubic centimeters, and 25 cubic centimeter portions were titrated against 0.1 nor- mal sodium thiosulfate in this manner • The 25 cubic centimeter sample was diluted to about 100 cubic centimeters, 5 cubic centimeters of dilute sulfuric acid added, and the liberated iodine titrated in the usual way. The titer of the thiosulfate was found to be .09715 normal. Upon . , * 19 standardizing the thiosulfate by the usual arsenous oxide method, the titer was .09981 normal. The work was repeated and good checks obtained. The fact that the arsenous oxide method gave definitely higher values than the silver bromate method can be explained only on the assumption that the silver bromate had a greater purity than 99*8$, or that the arsenous oxide was not 100$ pure. In order to explain this discrepancy, there are two possibilities : (1) The silver bromate was purer than 99*8$, (2) The arsenous oxide was not 100$ pure, although it was of the highest purity on the market which is usually assumed to be 100$. The work was repeated with the 100$ pure silver bromate prepared above, giving almost identical results as those obtained by using the 99.8$ pure salt. This put the suspicion on the arsenous oxide. This was carefully purified by re crystallization from hydrochloric acid solution, and sublimation at the lowest temperature possible. 4*95 grams of arsenous oxide were weighed out and made up to one liter of sodium arsenite in the usual way, giving a 0.1 normal arsenite solution. This gave values with the sodium thiosulfate that checked very acceptably with the silver bromate values. It would appear then, that silver bromate has two distinct advantages over the arsenous oxide as a standard in iodimetry, viz: ( 1 ) Higher purity (2) More direct procedure in titration. By the latter is meant: 20 The liberated iodine from the reaction between silver brom- ate, potassium iodide, and sulfuric acid, is titrated directly by the sodium thiosulfate; in the arsenous oxide method, the arsenite and thiosulfate solutions are compared by titrating against an iodine solution, involving two titrations in the lat- ter case to one in the former. This would double the chance for error. On the other hand, the available oxygen in silver bromate (six equivalents) is high in comparison with the reducing action of sodium arsenite (two equivalents), so that an error in weigh- ing out the silver salt is slightly more significant in the way of error than in the case of arsenous oxide. This work, however, indicates that on the whole, the silver bromate should have the preference where maximum accuracy is desired. The definiteness of arsenous oxide is brought into question in another way. Upon subliming arsenous oxide, partial decom- position occurs: 2As 2 0 3 * As 4 + 30 2 Since arsenic is volatile there is a chance that the sublimate may contain more or less of the free element which would intro- duce appreciable errors in the reducing value of the oxide. «. ' . . , ■ *■ 1. Analysis of silver bromate, of the highest purity obtainable, when converted into silver bromide by means of pure hydrobromic acid, indicates purities ranging from 99-6% to 100$. 2* Upon heating in contact with an inert medium, like water or a saturated paraffin oil, silver bromate gives off a gas, the amount being approximately 3 to 4 cubic centimeters per gram of the salt used. 3» Analysis of this gas shows it to contain mainly oxygen and nitrogen, the latter being something in excess of the amount in the atmosphere. Spectroscopic examination also indicates the presence of an appreciable amount of argon. 4. Silver bromate, heated to temperatures below its Ignition point, on cooling, gradually increases in weight. 5. Dry silver bromate of high purity is stable toward light and heat, but in the presence of water darkens slowly at low temperatures, and rapidly at high temperatures. 6. Silver bromate when pure, melts at 308 to 310 degrees C. without decomposition. Many substances catalyze the decomposition, which may take place with explosive violence even at temperatures as low as 150 degrees C. 7. No evidence of the existance of silver perbromate, or any other silver oxy-bromlne compounds was obtained. 8. Silver bromate may be used as a standard in iodimetry, seeming to possess certain advantages over arsenous trioxide. V BIBLIOGRAPHY 22 1. Abegg Vol.IIpt.I pp 712 2. Mem. de l'acad. de Belgique 3. Abegg Vol.II pt.I pp 712 »» it it n ii « 373 4. Jour. Pract. Chemie ( 1863) £0 190 Michael and Conn Jour. Amer. Chem.Soc. 1901 2£, 89