THE DISSOCIATION EQUILIBRIUM OF KAOLIN BY CHARLES IMSE ROSE THESIS FOR THE DEGREE OF BAGHELOR OF SGIENGE IN CHEMICAL ENGINEERING COLLEGE OF LIBERAL ARTS AND SCIENCES UNIVERSITY OF ILLINOIS 1922 1922 Pt72 UNIVERSITY OF ILLINOIS .May.. 20 , 1982... THIS IS TO CERTIFY THAT THE THESIS PREPARED UNDER MY SUPERVISION BY .....Cha.ri.e8...Jas.g...^p.g.§, entitled .TJae...Dis,s.o.c.ia.ti.oii.,E.quiliU.riun:....Qf Ka.olm IS APPROVED BY ME AS FULFILLING THIS PART OF THE REQUIREMENTS FOR THE degree of B.a.Gii.elo.r....of > ...,S.c.i.eii.c.e. HEAD OF DEPARTMENT OF. .0.6 r^l.G...Eng.ine.e.Mg Digitized by the Internet Archive in 2015 https://archive.org/details/dissociationequiOOrose ACKNOWLEDGMENTS /f ik kHr -fc kik The writer wishes to express his appreciation of the suggestions and advice that Profe,ssor E.W. Washburn has so kindly given. He also wishes to express his thanks for the help rendered by Dr .E.N. Bunting. N X* & - X- *X* X* Xr X" X* X* X* X - X~ / TABLE OF CONTENTS Page I. INTRODUCTION.. 1 II. LITERATURE 2 III . THEORY . . 3 IV. APPARATUS AND OPERATION 6 V. RESULTS AND CALCULATIONS 16 (1) Data 1'roiri Static Method. 19 (2) Results of Dynamic Method. 20 (3) Heat of Dissociation 21 VI. CONCLUSIONS 22 1 THE DISSOCIATION EQUILIBRIUM OF KAOLIN I. INTRODUCTION A considerable amount of work has been done in studying the dehydration phenomena of clays, and part of this work has been concerned particularly with the rate of loss or percentage loss of the combined water of the clay at various temperatures. The loss of weight of clay samples or the loss of chemically combined water after heating the clay to definite temperatures has been determined for a number of them, but practically nothing has been done to ascertain the tension of the water vapor at those temperatures. With this in mind the problem of finding a method or methods of determining the actual dissociation tension for any temperature was undertaken. During the trials a washed North Carolina Kaolin was used which is composed for the most part of the mineral kaolinite, having the formula: A1 0 . 2SiO . 2H 0 . This clay was not 2 3 2 2 analyzed except to determine the average percentage of hygroscopic and combined water. . * 2 . II. LITERATURE U-.H. Brown and E. T .Montgomery 1 have studied this clay among 1 Technologic Paper No- 21. Bureau of Standards. April, 1913 a number of others, and have determined the percentage loss in weight of the combined water after heating to certain definite temperatures. The graph obtained from their results has been used as an indication of the dissociation tension that might be expected since where their curve shows a relatively large loss of weight at a certain temperature one may be led to believe that the vapor tension will show a relatively large increase, or will be relatively large as compared with the vapor tensions at any lower temperature provided the temperature which produces the fastest rate of dehydration has not been exceeded. A thesis along similar lines by Mr.T.K.Chow which is held by this department was occasionally referred to in order to compare values obtained. This particular work has results which are of doubtful worth and no great amount of weight was given to informa- tion obtained from it. Suggestions for the apparatus have been found in E. W. Washburn' i . . . 2 article y and also many have been offered by him. 2 Jr. Amer. Ceram. Soc. 37,309 (1915) 3 III. THEORY Two methods of attaching the problem were suggested, one a static method, and the other a dynamic one. The static method consisted of placing the kaolin in a vessel, evacuating the vessel, heating it in a small electric furnace and then measuring the pressure of the water vapor by means of a mercury manometer sealed to the vessel. ( See Fig.1) The kaolin was dried before placing it in the vessel so that most of the hygroscopic water would be removed. When the apparatus was evacuated the mercury would rise in the long arm of the manometer. Upon heating the kaolin some of the combined water would be liberated, and would develop a pressure dependent upon the temperature to which the clay was heated. This pressure would depress the mercury column a definite amount. By noting the difference in height of the mercury column in the two arms of the U tube, and subtracting this difference from the barometric pressure, the pressure of the water vapor above the clay was obtained. 1’or low pressures a manometer tube was connected to another similar vessel, and by simply noting the difference in height of the mercury columns in the two arms of the manometer, the pressure of the vapor could be ascertained. It was assumed that the water would possess a definite vapor tension above the clay at a definite temperature, and that repeated trials would give checji results. making a sufficient number of trials at various definite temperatures, it was hoped to obtain values to plot a curve showing - « * * ■ ♦ * 4 . the dissociation tension of the clay at various temperatures. The idea of the dynamic procedure was to pa3s a measured volume of air over some kaolin which was to be heated in a suitable apparatus, collect the moisture taken up, and then knowing the weight of water vapor in a definite volume of air, the pressure of this water vapor could be found by the use of the gas equation. ( See Fig. 2. ) It was assumed here also that at any definite temperature the kaolin would have a definite dissociation tension. The air in passing over the clay was supposed to take up moisture till the partial pressure of the water vapor in the air was in equiliorium with the clay. Necessarily the partial pressure of the vapor in the entering air would have to be less than the pressure of the water vapor above the clay if the air were to take up any moisture in reaching equilibrium. The necessity of passing the air over the clay at such a rate that the equilibrium conditions would not be disturbed to a noticeable extent was realized. The rate of passage of this air would have to be determined by experimentation. In order to aid matters it was desired to pass the air through the apparatus in such a manner that the air would pass around small pieces or particles of the kaolin, and not simply pass over a layer of clay spread over the bottom of the apparatus. Since any hygroscopic water would interfere seriously with the determination, this was first removed before any data were recorded. - . I 5 . It will be noted in connection with this procedure that the water taken up by the drying tube® need not come from the cls.y entirely, and, in fact, the greater part of the water was contained in the air in the first place, since this would allow equilibrium conditions to be reached in the furnace in less time. . I ■ IV. APPARATUS AND OPERATION. 6 . The vessel used, in the static method was of hard glass, and the main portion which held the clay was in the shape of a flask with a moderately long neck. To this neck was attached the side- arms for the vacuum pump, and for the manometer to measure the pressure developed. The clay was introduced into the flask through a hole in the top of the neck. After placing it in the flask, the top was then sealed off. The side-arm for the vacuum pump was provided with two ground glass stop-cocxs, these stop-cocks being about four inches apart. The tops and bottoms of them were sealed with wax when clos after evacuating the bulb. The other arm was in the shape of a rough U with the half of the U connected to the flask, the longer. This half was more than long enough to permit mercury to rise in it equal to the barometric height. ( See Fig.l) . This apparatus was placed in a small electric resistance furnace, the furnace being of such size that the side-aims projected over the upper edge of it. The top of the neck, and the open parts at the top of the furnace were covered with suitable material. The furnace was heated by a current which could be varied to suit one’s needs by the 'manipulati on of a rheostat. Different temperatures could thus be obtained. To prevent any condensation of water vapor in the apparatu when tne pressure of the vapor in it was greater than the saturation pressure of water vapor at room temperature, a fine wire was * . -.1 .. : , . . - § 4 # I ' . ' ■: 7 . wrapped about tire exposed arms. This wire extended over that portion of the large U shaped arm not occupied by mercury when pressure would be developed, and also extended up to the first stop- cock on the other arm. The double stop-cock was necessary because this warming had a tendency to soften the stop-cock grease, and permit a slow leakage of air into the apparatus when one was used alone. A small current passed through this wire was suffici- ent to warm all parts as much as necessary. Temperatures were measured by a thermocouple, and also by an ordinary mercury thermometer, since they rarely exceeded 400 degrees Centigrade . The diagram shows a sectional view of the apparatus. The dynamic apparatus consisted of a fused silica tube about one inch in diameter, and about four feet long. This tube was heated by passing a current through a nichrome ribbon wrapped about it. The coils of the ribbon wera insulated from each other by alundum cement, and the entire heated portion of the tube was surrounded with magnesia pipe covering. A wet test meter was used to measure air volumes. The temperature of the air coming from the meter was read by means of a thermometer incorporated in it. ■ ' , 8 . When it 7/as desired to reduce the moisture content of this air to such a degree that the vapor pressure of this moisture would be less than that supposed to exist above the clay, the air was passed through a calcium chloride tube. Since only three feet of the tube was used as a furnace, it was necessary to warm the exposed ends to prevent any condensa- tion of water vapor in them. The reason for this was similar to that making it necessary to warm the exposed parts of the static apparatus. A fine wire was wrapped about the exposed ends, and a small current passed through it. This current was independent of the heating current, and supplied all the heat desired, it was found from experience that a certain amount of heat came from the furnace and xept these ends above room temperature. Issuing frcm the furnace, the moisture laden air was passed through a series of six-inch U tubes filled with calcium chloride and provided with ground glass stop-cocks. Three tubes were used, and they were found to be sufficient to remove all weighable amounts of water frcm the air. It was also found that calcium chloride was the best agent since the small pieces could be placed in the tubes, and would cane close together, but would nevertheless allow many air passages between them, whereas if sulphuric acid had been used, a certain amount of hydrostatic pressure would have been necessary to pull the air through the tubes, and it was desired to keep the system at atmospheric conditions. An attempt was made to saturate small pieces of pumice stone with the acid and then fill the tubes with tn.is, but the acid dripped from the pieces and collected in the bottoms of the tubes, and thus 9 defeated the purpose of this trial. After passing through a stop-cock in the line used to regulate the flow of air, and a guard tube of calcium chloride to prevent any diffusion backwards into the absorption tubes, the air was exhausted either through a water pump, or through an aspirator bottle The stop-cock mentioned was placed in the line at this point in order that all the parts of the apparatus before it would be at atmospheric pressure except for the small pressure of three-twentieths of an inch of water required to operate the meter. By turning the stop-cock the rate of flow could be adjusted, and by watching the hand on the dial of the meter the rate of flow could be measured. The pump was a small ordinary water pump, and operated under a constant head. The details of this mechanism are shown in the diagram. It will be noted that air was pulled through the apparatus and not blown through. The water pump was used when a gas meter was available, as there was no other way of measuring air volumes when it was used. When a meter was not available, another method of finding the air volumes was employed. This method consisted of using an asnirator bottle to pull the air through the apparatus. uy catch- ing the water coming from the bottle, weighing it, and finding its volume, the volume of the air pulled through could be found. This volume obtained represented a volume of saturated air at the temperature of the water. nnowing the temperature of the water, and its vapor pressure at that temperature, the actual volume of air could be found for dry or otner c ondit ions . The barometric pressure was also needed. c - ■ ' * - ' . - < 10 . It has been noted that a guard tube was used to stop any diffusion backward into the absorption tubes. These guard tubes were placed immediately in front of the pump or aspirator bottle as the case might be. The furnace was heated by a current which could be varied as desired. ±>y using several different currents in a preliminary test, the temperatures corresponding to them were found. From these values a, graph was plotted and enabled one to find a current that would give a desired temperature. Direct current was used to avoid variations. The temperature of the furnace was measured by means of a thermocouple. To test the heating of the furnace the hot junction of the couple was moved along the heated portion of the tube, and the results noted. It was found that the furnace was heated uniformly. The apparatus was tested for leaks by evacuating it, and then sealing it while heated. No leaks were observed. All joints were sealed or wired or both, and rubber stoppers and tubing were used to make connections. In later determi nations it was found necessary to intro- duce a capillary in the line between the exit of the furnace and the first absorption tube. This capillary was used to prevent diffusion of water vapor from the heated clay into the tubes with- out having air carry it. Before loading the furnace the clay was made into small pellets composed of two-thirds ordinary clay, and one-third clay which had been dehydrated by heating for one hundred hours in a - r 1 ■ , 11 . high vacuum. These pellets were broken into small pieces prelim- inary to pushing them into the furnace, but were not broken into small particles. They were then pushed into the furnace, and filled the heated portion to within too or three inches of each end of it. This insured the clay being heated uniformly. The pur- pose of this breaking of the pellets was to provide as much surface as possible, and to allow the air to circulate around the small pieces instead of simply over them. Since the pieces were irregular there were many small openings that the air could pass through. Had the furnace been filled with the fine particles it would have been completely stopped up, or if a less amount had been used then a layer would have formed at the bottom of the tube and the air would have passed over the clay. Before collecting any moisture or measuring any air passing through the furnace the apparatus was allowed to come to a constant temperature. After this haa been attained the absorption tuoes were placed in the line. These tubes were previously connected togetner by small pieces of rubber tubing which were wired fast to the side-arms. Then all joints were sealed and in- spected, the stop-cocks on the calcium chloride tuoes were opened* The regulating stop-cock was opened a slight amount, and air was slowly drawn through the apparatus. The opening ol the main s t op - c o c had to be adjusted for each run. The data necessary were the air volume and its temperature when issuing from the meter, the weight of the water taken up by the calcium chloride tubes, the barometric pressure, r.nd the - El Eli- -t I- J| ■ . . - ; 12 . temperature of the furnace. \ Dynamic Apparatus 16 V- RESULTS MU CALCULATIONS The static method showed that the kaolin adsorbs a large amount of air which is given off slowly when the clay is heated, and which seriously interferes with the measurement of the vapor pressure. Its presence was detected in the apparatus by observing that when the manometer tube was at room temperature there was no condensation of water vapor on the inside walls, of the tube though the pressure measured was far in excess of that allowed for saturation pressure of water vapor in air for the same temperature. There were no leaks in the apparatus when these observations were made. The air is held tenaciously by the clay, and is only removed after a long period of heating in the exhausted vessel. When all of the air was removed most of the water of the clay had been removed, and no reliable results could be obtained. Due to factors mentioned above, and also the development of leaks in the apparatus occasionally but at an important time, no results of any value were obtained with the static apparatus with the open manometer tube. I’or low pressures the closed manometer tube apparatus was used, and some check results were obtained over a moderate range of temperatures. These results were obtained after cooling and heating several times, and after all air had been removed from the clay. The mass of clay used with this particular piece of apparatus was considerably less than in the case of the other piece. (See Graph II). I'rom the data obtained from this test another curve was plotted as shown in Graph III. - - . . •• 340 320 t i 300 g •H -P £ 0) o 280 260 CO 0) 0) h W 0) 240 « «» p -Hlfn 0) ft. 200 | e-i 180 220 160 140 Graph II Pressure-Temperature Curve 8 Vapor Pressure in mm. of Mercury, 10 12 14 16 18 Curve from Van’t Hoff equation. 0017 . C018 .0019 Wlm .0071 .007 19 . D ata fr om Stat i c Metho d, Temperature Degrees Pressure of vapor Centigrade in mm. lie r cur y 145 9.0 190 10.5 235 13.5 260 14.0 275 16,0 300 X 7 .0 The dynamic method was not influenced by adsorbed air. This method snowed tnat tne clay was very slow in approaching equilibrium as did the static method. When air was passed through the apparatus slowly, about two-tenths of a cubic foot per twenty-four hours, water vapor from the furnace diffused over into the absorption tubes, and thus made the results high. When air was passed through too quickly the equilibrium was disturbed, and a low value for the pressure was always certain to be the result. This diffusion was prevented in the last part of the work by introducing a short piece of capillary tubing in the line between the end of the furnace and the absorption tubes. Although adsorbed air did not influence the results here, check results could not be obtained although the manipulations carried out in each test at the same temperature were exactly alike* Care was taken to see that any hygroscopic water was re- ! - ; . - ' H ■ 20 . moved before any run was made, and that the clay in the furnace was net dehydrated. When any clay became dehydrated it was removed and replaced by other material . Results of Dynamic Method Temperature Degrees Pressure of Vapor Centigrad e mm. of Llercurv 317 9.39 320 7.96 433 93.26 200 3.69 175 1.47 293 7.06 297 7.04 320 7.49 322 13.18 322 2.80 352 12.93 322 15.61 322 5.7 347 3.14 The volume and temperature of the air were known, and by subtracting the saturation pressure of water vapor in air from the barometric reading, the pressure of the air was found. Using these three quantities in the gas equation, gave the moles of air. The moles of water could be found from the gain in weight of the 21 . absorption tubes. We had then that : ptr n : P : : N : N y air from which the pressure of the n 2° HO H 2° 2 water vapor could be found. H e at of Di ssociation. - The heat of dissociation for one mole of water was found by the use of the formula: log P H R f-1 - ... A— ) V * h ) where the pressures and temperatures used were those obtained from the static method det erminations . p =9.0 1 p = 17 2 T = 145 + 273 = 418° 1 T = 300 + 273 = 573° 2 log 9.0 = -0.5353 e 1,7 1 =0.00239 1 =0.00174 T 2 (1 - 1 ) =-0.00065 T T 2 1 H .0*0353 = p 1.985 ( -0.00065) H = 1940 calories per mole H 0 P 2 H = 1940 = 107.8 calories per gin. HO p 18 2 22 . VI. CONCLUSIONS Since the values obtained by the static method could be reproduced, they may be taken as approaching the correct values. Results from the dynamic method are low in comparison v/ith those obtained by the static procedure, as might be expected. The highest value given by the dynamic method would be the nearest to the correct value, and even then might be considerable short of it . . ■