fssmm ofwwaw of A O f r o m silica c n Hr Otoggo toj&wp- Bomor A fESSIS Brtsostoi to tko OKOtaftt* SokooL of lioktgan Slot# Collogo Of AgrlttllWt «pA AfffeioO 8«loa«t In Farttoi f&flUaftfti of Fttnlfimonto t m tlio Stggt* of Boot#* of ffeiiooofijjr Cfaomlstsy Bo|>aa?tm»aat loot M m t m * Hiokigwt ProQuest Number: 10008259 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest. ProQuest 10008259 Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code Microform Edition © ProQuest LLC. ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 dr/S~ H c. m s HEAT ®F WUfBra OF ACfinm silica f$*ft fth»#tnre ©f sill#® om has bean rather definitely Shewn by falls and Firth U*F*sy»*Oh#w* *||# S4l fltssH* and fenee {y*fhys#4,3>lfS 7(1931}) expressed the opinion that vapors sdaevhed on an adsorbent wore held in e rt under an enormous pressor©* lank and Ooolidg© calculated th a t ik t neat developed in excess o f the heat o f condensation was &a* t© oompreeSion*' larfciaa and Msg lndleated that the m$f%m entity tf m adsorbent is the seat ©f great energy ant esAssisted a tela* if erer SQOOO atmosphere* pressure holding adsorbed aeleewie* m a snrfae® of ©arboa# Coolidge IJ>aia.Ohom*See.* ti flt&i ?) later toeited that his high iralm© fm eewpfeesiea was tee high ant Instead ef stating that net heat ©f ddeevptlen was due t© ©©mpreesloa dooited m a eeyrelatlen with peieayifs thsory ant deetrootton of *e»faee* .£atrials aat Crimm (3»Aa*Chem«See# *jg:*8i44 (ifSi I oaXehuted the area «sp*g*d hy l g* ©f elliea gel fre® heat of wetting mssirtiyemftat* m the theory that the gel exhibited a water enrfate »hd the heat evelved developed from destrmetien tf the total nevfaee m m g y of this water earfate* *Patriot! and frtiier ifm&tim*Ohs** IIffetU determined the heat ©f wetting ©f slllea gel hy water sapor at satmraUon at @%* ant reeslte obtained were In agreement with the # Three different methods for preparing eilioa gel with a definite water content were'followed in this investigation* first*, following a procedure of van lemelenfs* the gel was suspended in vacuum las regards air) in m atmosphere of water vapor from a sulfuric aeid solution* The pressure' of the water vapor varied inversely as the concentration of the acid solution and directly as Its temperature* Second* replacing the sulfuric acid with phosphorus pentOKide and using an electric heating element around the gel* the temperature and water vapor pressure of the gel could then he increase! with heat while the phosphorus peatoniie retained at S5°€* its original vapor pressure* third* In ft vacuum apparatus hull! in this laboratory the gel waft heated to various temperatures and evacuation of the liberated water vapor was carried m to a pressure of *0001 m* of mercury* 8 Wjt %ht«8.three methods II was possible to prepare gel willi wall* contents from gel* al will* la 1*4$ ami reproduce these In the firs! method the feet that lb# tape? pteemrave. af an atmosphere eaa If controlled by using eulfurle acid seiutieme was utilised* using approximately a#8 g. of gel aa received and werblng la quadruplicate through eleven different atmospheres of controlled vapor pressure spaaed from «eto mm# to saturation through font temperature ranges the graph*, fig' 1* was obtained* Theea adsorptlon-de sorption reversible serves indicating m hysteresis could be duplicated sat were duplicated three years 'later which might indicate that the gel tit act deteriorate upon standing* The gel in weighing bottles was suspended in vacuum as regards air over a definite percent by weight of sulfuric acid solution* in ordinary stoefc bottle was used and the top sealed with a rubber stopper and paraffin* the whole wee then submerged in « constant water {oil for the two higher temperatures) bath and allowed to remain for four days* three days were found sufficient in seme of the oases. At the end of this time the gel had adsorbed sufficient water from the vapor to give a weight which was constant at that vapor pressure and temperature* Its e m vapor pressure under those conditions was then at equilibrium with that of the acid solution ever which the gel was suspended and the water on the gel was evenly distributed over the surface of the gel as regards total surface energy exposed. 25 10 fo of adsorbed water on 1 g. of equilibrated gel * 20 c/o 80 40 60 by weight of sulfuric acid. 100 Fig, 1,- adsorption isotherms: No, 1, 15°C.; No, 2, 25°C.; No. 3, 4 0 ° G ,; No. 4, 60°G. 9 By this method gels could be reproduced at any desired water content between the limits of 4 .0 te 86#8$# Ko matter what the water eentent of the gel as received this method gave expected result# within its range# the second method of varying the gel water eontent was by means of heating the gel in vacuum over phosphorous Pentoxide the latter held at 25°0. the result was to drive off more water from the gel by increasing its water vapor pressure thus distilling water from a substance with a high vapor pressure te that of a low vapor pressure namely the phosphorous pentoxide# the gel was suspended in a three liter round bottom single netted flash fitted with a rubber stepper* through the rubber stepper pertruded a stepeott# thermocouple and two lead wires for the heating element wound around the container for the gel* The thermocouple was placed at the very surface of the gel {% g* )* The phosphorous pentoxide was placed at the bottom of the flash* The entire flash was then submerged in a £5*6# constant water bath 1. #006*6# The gel prepared in this manner was tested for % m m metallic elements by a medium quarts spaat©graph with results which showed it retaar&ably free from mettalie impurities. In the third method vacuum technique was followed in that the gel was heated at a final pressure ©f #0001 mm# of mercury and then sealed off* This procedure required a longer time than either of the other two methods and gave results In agreement with those obtained with the gel prepared over phosphorous pentoxide# 10 Hat Mmmwmmt*#* the gel* sealed 4** evacuated bulbs 2*S-3t0 m« 4m diameter, of Jm&m water oentent, m o placed i» tit* oradle prepared far It and immersed in the water of the calorimeter# A volume of ltd oil* of water was mood for the determiaetlems* fhe Beehaamm thermometer* stirrer, moating elements and thermopile wore also entered In the If# ml* of water* the assembly was then ready to to brought to etuilibyiem temperature with the copper sulfate bath, around the enter teabeiner of the silvered eu$* and the temperature of the air hath in anion the entire apparatus m o held* #hi® efmillhr|.mn stage retired from four to six hm m * It m e test to begin ell measurements at 88*0* instead of attempting to end at an equilibrium stage at that temperature* the reaeon obviously was associated with possibilities of heat mfcange*, it m o found that the heat transfer from the silvered eup to the outside was very email in# to the air insulation surround lag the m $ if the copper sulfate hath was held to within »0£*0, of the temperature of the silvered cap* dart m o always baton to he sure no heat entered the system from an o u ts id e source, fhe two Beefcmean thermometera and thermopile gave two theoho on the variations of temperature between the inner calorimeter and sapper sulfate bath* the small metallic stirrer rotating at about 100 r*p*m* gar® no measureable heat of stirring* As a preliminary test of the assembled calorimeter a heat equivalent was tahen and n the® ill# whole again returned to 2S®a* id come te equilibrium #e» the gel rmi When equilibrium at 25°c* Bad resulted the bulb was broken at lie tip by the screw and usually the bulb was completely shattered throwing the gel freely late the water* fh# rise la temperature of the Inner calorimeter waa followed with the same increase la temperature of the copper sulfate hath* the entire apparatus m s then oooled to 2S°C* until equilibrium became evident which required one to two hoars# the heat equivalent m s then determined and the average of four trial®.recorded* , The electrical energy meed for determining the heat equivalent was evaluated by measuring the voltage drop across the heavy copper conductors leading to the heating element also measuring the voltage drop necrose a standard resistance* Between one and ten readings for each of these voltage drops were made during a run and the average taken for oalsulatlcns# fhe most' constant source of current used was that obtained from two $ v* storage batteries connected In parallel* Before using the current the batteries were discharged at a like rate for ten to thirty minutes or until constant voltage readings across a known resistance were obtained* After fully charging It was found expedient to discharge the parallel batteries at *1 amp for ten to twelve hours* thus hastening their equilibrium discharging condition* the heat equivalent was calculated according to Hfcalories5 * At/d*l® or IIt/4*lf* fhe temperature rise 12 #f the electrical heat equivalent was an exact reproduction a* the temperature rise at *500 to *005°0* caused by the gel* Bleeu#slon*«* According te Adams {Physics ant Chemistry ef Surface® p* 500) the heat or energy liberafet by wetting a silica gel surface 1# due te a destruction of total surface energy* She equation f # w„x * f &wal shows that this total surfas# energy is composed of free surface energy and latent energy of the surface molecules themselves* f is total energy* w8l is the worh of adhesion or free energy of wetting* f is absolute temperature# and 1 &%l is a heat term or potential *l w * ww' energy associated with the surfs## molecules# Available Cate Indicates that most of the energy liberated by wetting suet be due to the latter term* this would indicate the! activated silica gel with its enormous area and honeycombed shell presents a surface of silica molecules which hate tremendous potential sMer&r* Zsigmondy, tan Bemmelen# and Anderson worfced on the static control of water on fresh silica gel by means of sulfuric acid but evidently, warning in the presence of air* they did not allow ample time for the adsporbios to reach, equilibrium for they found, a decided hysteresis during the adsorption* desorption runs* la our data no noticeable hysteresis was observed unless its magnitude was below *0Q1 g* in 5*500 g* At times the gel required a longer Interval to attain equilibrium* 13 fable X 4§16 *wn* X/M at temperatures 13 *8 40 40*0 #04? #046 *046 *046 91*4 #044 *046 #64? #04? 86.6 #060 *049 .048 #04$ fi*l *064 *066 •064 *063 net; *o?o #06$ *06? *068 61*9 oo« *106 *102 *099 63.1 *2?4 *2f0 *268 #266' 44.4 •til *646 *834 #386 8»*t #166 *661 *888 *826 19*0 *649 *84? *848 *388 06*0 *863 *368 *861 *341 fable I give© the result© eg sqwiliferati&g siliea gel ever a©lt selaKsft# at 28* BS* 40 aat.600O* fhs serve* #f fig* I fallow the same general shaft f m sash isetherm aid IMS.©ato a ©3?©$$lug off of *4*s*ptieft capacity at ikighs* temperatures* tf' the vaper » m « w t v© # weight aold surfs' at 88*8# la soaptttst to tut above s o t of the same temperature the two shew great similarity in ©tiwimre «ai ftrseaet tho shape of the vapor prestore • % /mmm* Hotting the l/® raises -agolost vapor pressors (InternatisaaX Opitisel fables#,. JJ# 303} value© at 280G* give# a straight lias graph (fig*3) shioh ssulfi ladleate that a# litsal (Surface Ohsiaisbry* pt 88} has stated# "fh# transition fro® a equilibrated gel 70 on 1 g. of 40 of adsorbed water 30 10 20 > 10 15 Vapor pressure in mm, of mercury. Fig. 2.- Adsorption of water vapor by silica gel at 25°C. u Xayor to a polymoleoular adeorfcod layer* ia aaoh a M & that it# f*Q9 mvt&m vomit possess the fro# smrfao# of th# iifmlt im Mils, i# not ahrmpt*" fig* a i»t#.a Xog**X#g Changing mam (fig* 3} aseording to FrowtSlloli*# ©©©r&laatee glvo® a emrve with thro# straight sections to whteh reforene© will to moto lator* lit lotting tho applioatllity of fr#madXl©hf#* Fatriofc*© tad fltetgg*# Adsorption Fetation© wo will ©htorwt only th# eomfral'straight lino aootioo of' th# l#g*l#g graph of th# M/M * m m & prossmr# © m m # at tho varioa® temperaimros* Flgo* 4,# 3* and 6 giro th© gonartl ©hap® of tht omtfos and #*#!* II th# ©lopes and constants of the lines. Th® variation* fftfel© II Fx#w#ftli#fe*« Id® M* m w* ■V# l i Wi ff i? *MT *? § ,w .828 .888 k ?0.66 38.23 18.88 4.74 Fttyitfe** 1/n .7*4 .778 .816 .880 k 18.2 16*1 18.6 10.6 ®reggft l/n .768 k’lo 80*8 .388 21.8 over th# toaipofatmr© rang® in ihiswt figaro® ar# lot# la ?atrte&,« sat @r«Mf# than la Framed, ioh*® with the added rooalt that Fatriofc*©* as ®g*s X*g»l*g lðerat tmrv## all fall v®tf oloso together* At long a® Fatriofc*# and Cn#®g## Adenapt ion Sfnatioaa* however, do not give a® they 1000 Logarithm of X/M in mg, of water 1QQQQ 100 10 Logarithm of pressure in mm. of mercury. Fig. 3.- Freu n d l i c h ’s adsorption isotherm at 25°0. 1000 in mg, of water 10000 60000 Logarithm of X/M oo 100 10 100 Logarithm of pressure in mm. of mercury. Fig, 4,- F r e u n d l i c h ’s adsorption isotherms: No. 1, 15°G.; No. 2, 25°C.; No. 3, 40°C.; No. 4, 60°C. -water Logarithm of V/M in ml# of 1000 100 10 10 100 Logarithm of P s / P 0 . Pig. 5.- Patriot's adsorption isotherms: No. 1, 15°G.; No. 2, 25° C . ; No. 3, 40°C.; No. 4, 60°C. Logarithm of V/M in ml. of v/ater ♦ 10 -4 Logarithm of (P/P0T) x 10 Fig. 6.- Gregg's adgorption i s o thgrms: 100 15 m m intended « means of calculating adsorption values at one tesserature knowing them at another, wa will deal only with Freuncllleh’B simple relationship# Curves from figs* 7* 7® and ?fc art results paralleling parte of the experiments of Bartell and May {J#Bhye*Chea* * M . 478 (193$)}« fhi curves of fig* 7 are gulte identioal with their # water V* fdC* of activation curve# In hath fig*# 7a and 7b the activation temperature was S60°C* and although Bartell and M ay did net work at that temperature their results still indicate a difference which might he explained by noting that they worked with gels containing leas than 4$ water whleh is our value for gel of greatest activity# Obviously therefore their results are reasonable for they are probably only removing adsorbed gases and seme adsorbed water in the 80 minutes ef heating at 300°C# without destroying any of the internal surface by this short activation treatment* A longer heating period or higher temperature would result la a liberation of sufficient water to cause loss of activity# Bus to the unreversibility of the activity of the gel containing less than 4$ water Bartell and M®y*s results cannot be expected to be exactly parallel to this work* for maximum activity they ccacluded that 3QQ°0* for 30 minutes was beet while we found W @0* for many hours gave a gel ©f highest activity* They also state that water content of a gel is Intimately connected with activity which agrees with our results but do not seem to place as much importance on the gel on water % of 125 250 Temperature °C 375 500 Fig, 7,- Relation of water content to temperature of activation A, heating over PgOgj B, heating and evacuating. 30.2 29.8 Heat of wetting in cal./g. of gel 30.6 29 29.0 25 50 100 75 125 Time of activation in hours. 150 Fig. 7a.- Relation of heat of Tjetting to time of activation. 175 15 water on gel 10 00 50 100 150 Time of activation in hours at 260°0 250 Fig* 7b,- The relation of water content to time of activation 16 activation treatment* la covering the range of water Qontent of the gel from 0$ to 26$, by weight* commercial concentrated sulfuric aeia would prepare a gel only as low as 4*7$ water eontent* to go still further the email vaeuum eleetric furnaoe was constructed which produced the series of gels from 4*7$ d am to 1*4$ water content* Talcing 4*7$ gel or gel OfUillhrattd over concentrated sulfuric acid and heating it gradually from 2$ to $00®0* produced the curve of fig* 7# The noticeable fact, as indicated by this curve and later in the heat of wetting measurements, is that a change of condition of the water on the gel occurs at a water content of 4$* A water content below the 4$ point produced a gel of lower activity as will be explained later* The transition of color which accompanied the removal of water was interesting* With the application of heat the gel became colored ranging from light tan to almost blaelu The darkest color occurred during the preparation of the 4$ water content gel ant that gel of lower water content* The color did not in all cases indicate a more active gel although when colored the gel was very active as measured by heat of wetting* Another observance was that if the gel remained at the same temperature long enough the color would disappear, leaving at that time a gel which was the most active at that particular temperature and water content. 17 A laager time was necessary for fading of the color at the lower tcsipevatmres* This color change could he due to an unsaturatien of molecular forces (ailhert £• Lewis, 3* Am* Chem# Soe*t 38, 784 (1914))# e difference of refractire index Cue water present on some surfaces}, or a light interference phenomenon due to a condition of the inner surfaces with respect to each ether* Whatever condition does exist is readjusted hy continual application of heat* It If sealed in a vacuum bulb while colored the color will remain for years,if no heat is applied beyond room temperature,without loss of activity# On the theory that the heat of wetting value will give an accurate means of evaluating activity determinations of the heat evolved from wetting’gel® of different water content from 1*4$ to 36*0$ have been made, fig* 8 and table III give the actual experimental values obtained at 25°C* 18 $ Water o#i m C a l,^g , $ Water Oel on n&xd e' 36#80 0*10 (*08) 7*10 £3.60 (.80) 34*80 0*80 (#04) 8*85 86.60 (.30) 28*50 8.85 (*06) 4*70 88.00 (.20) 25*50 3*78 (.05) 4*40 88.80 (.80) 84*78 4*28 (*08) 4*18 30.00 (.80) 80*88 7*50 (#00) 18*@Q 7*80 (#08) 4.00 4*00 4.00 30.60* 31.40* 81.80* 17*10 10.80 (#10) 3*50 89.10 (.20) 18*7© 15*86 {*05) 8.18 88,90 (.10) 12.05 14*10 (#08) 8*18 87.10 (.10) 7*60 23#fQ (.80 ) 1*90 86,66 (.10) 1*83 26,30 (.06) 1,42 84.60 (.06) 0,00 00,00 60 50 Cal./g. of gel 40 fo of water on gel. Fig. 8.- Eeats of wetting of water on silica gel: •, Ray and Ganguly, -o- Bartell and Almy; Patrick and Greider; A Bartell and Fu; <|>, Patrick and Grimm;* Stenzel; « Nasif; o 9 Swing and Bauer. 19 fhe values are the mean from at least six trials with the maximum deviations in brackets* In the region of greatest activity the value© obtained were variant due to a possibility that in this highly aetive condition email changes in the gel structure and manner of activation will cause a great difference In activity# fbe point "38 calories* is an Interpolated one for value® above 30 calories were difficult to duplicate* Below 4$ water content the gel activity was reduced probably by destruction of its Interior active surface, While for 0$ water (fused gel) the particles seemed to be negative to wetting although the transparent particles were wetted* "Referring to fig# 8 that portion of the curve on the right side of the 4$ gel was a perfectly reversible procedure but that portion of the curve on the left of the most active point* 4$, was not reversible* that is* once the gel had been activated to a water content of less than 4$f the high activity of 38 eal#/ g# of gel could not again be reached# In other word®, once the water below 0 had been removed it could not be replaced on the gel with a corresponding increase of activity to 38 cal#/g of gel resulting* Fitting the egiuatlon of a circle to the experimental points of the curve of fig# 8 the extended line crosses the $ water axis at 3V#f$ which will be taken as the total amount of water finable epaee of the silica gel. Bay and Oanguly 20 *»*• 37*8$ water content as the saturation limit for their «il* la an earlier experiment by Ifcenaaa (unpublished) conducted ia this laboratory on this gel a volume of *333 c*c« was determined as the volume of water taken m at saturation* If we subtract from *377 {considering the total water on the gel as averaging a density value of 1*00) the value of .040 giving .387 *«*«« then our value for the volume of water on the gel at saturation eemj®ree favorably with the former value of .338 e.e. fhe heat of wetting of silica gel of varying amounts of water have been evaluated by Patrick* Bay and Ganguly* Patrick and Grimm* Patrick and Greider* Bartell and Ahoy, and Bartell and Fu with values ranging from 17 to 3® eal./g* for their most active gel. the values for the heat of wetting by these authors are also given on fig* 3 as accurately as their water contents will permit* With these various values indicating either a difference in water content of gel* activation* or gel structure investigators are tuite evident to differ in results obtained unites a knowledge of the entire history ©f each gel sample indicating paralleling work is sufficient proof of the dupiieaMlity of results# Bartell and Fa (Colloid Sym* Mono** 7*188 (1787)) have calculated the specific surface of silica from adhesion tension data using an etuatien'ef the Gibbs Hemholtss type and obtained a value of 4*8 sc 10s cm2* They used for their value ©f change of latent energy of the molecules *.1311 zx ©ygn/em8 which i* the value for water therefore neglecting the potential energy associated with the silica surface molecules, They also obtained a value of S4*31 ml*/ g* ef hydrated (4$ water eontent) silica for the heat of wetting of water and assumed that the surface exposed was predominately silica molecules ©resting an area of 7.5 x 10* cm2 calculated by the equation employed hy Patrick and Grimm* Also, Bartell and Fu from an equation similar to one presented by Harkins and Fwing and from interfaelal tension data calculated an area of 4*5 x 10* em2 for a dehydrated gel* This value of 34*31 cal./g. of gel is distinctly above that obtained by Bartell and Almy but the activation history of the three types of gels is also different as Is their water content* A gel from the data of Bartell and Fu with a water content of 4$ develops the highest heat of wetting which Indicates that their gel also has a high activity point at 4$ water content. Fig* 9is. an attempt to compare our heats of wetting values with the net heats of adsorption of other investigators* The results of the experiments of Lamb and Coolidge agreed with their empirical equation and fig* 10 indicates that heat of wetting value© will also determine a straight line except for the most active gel* The total heat of adsorption in this ease was determined from table If where the change in cal*/g. of gel for *01 g* of water from 4$ to 37.7$ is tabulated* 300 Cal./g. of gel due to water adsorbed 240 °]0 of water on gel,. Fig. 9.~ Heat o'f adsorption of water vapor on silica gel: o, Ray and Ganguly; e, Patrick and Greider; •, Ewing and Bauer. Fos. 1, 2 and 3 are the heats of condensation of water vapor at 0 ° C . } 25°0. and 30°G. respectively. 100 (H) ogarithm of oal./g. of water adsorbed 1000 10 Logarithm of $ 1° water on gel (X) 100 lr lc Fig. 10.~ Appl Application Coolidgee Empirical ica t ion of the Lamb and Coolid g .898. 74 8a8 n i 9 l m E quation of H = raX , is 9.74 and is f H Equat zz fable XV fatey on del Oal./g* del *4 Cel. Are. mtu w,t«, 3 6 7 t 9 10 U 12 13 U 13 1# 1? 10 10 00 21 22 23 24 23 3# it 28 30 m 31 33 33 34. 35 37*7 32*00 27,25 33*33 23.60 33*00 30*30 10*00 17.00 13.13 14*90 13*33 IE,*30 11*43 10*40 9*43 a*40 7*70 3.90 4*13 3*40 4.70 4*10 ■ 3,50 2.93 2,50 2,05 1,60 1*20 *90 *40 .35 #15 0*00 4.78 8*00 1*48 1*60 1*50 1*50 1*40 1.43 1*30 1.25 1.15 1*03 1*05 #95 ,95 *90 .80 .75 *75 *70 *40 ,60 *53 *43 *46 .46 *40 .30 *30 .36 *20 *16 8.80 1*04 1*44 •26 1*30 •M *94 #19 .73 .20 .65 ' *16 *40 te heat of eondenaatlon of the total amber ,01 g* of water to the total' change In eal*/g* of gel oaueea by this aaoont of water starting at 4$ gel am& 32 eel*/g* of gel will giro the total heat of aasorptlon* 83 Xu fig* 11 an attempt Ismade to apply aa aquation to the heat of wetting data in order to he able to determine the heat of watting obtainable from a gel of known water content* with referenoe to thegel used in this experiment* fhe graph is semi log with the ©el*/ g* of gel as the unit scale and the % water content of the gel ae the log scale* fhe central portion fall® on a deoldedly straight line but the extremities deviate from the equation given directly beneath the figure, and one of the deviations Is at a water content of 8$* In an experiment on determination of adsorbed water density ©f varying amount® of water* Swing and Spurway {3*Am*Cheat*See* *5£* 433# (19305) indicated a break in the $ water adsorbed vs* density curve at a water content of approximately which agrees with our first dleeoatinuity in figs. 3 and 11* la their data the break comes at the 4$ point but in the preparation of their gel 4$ water was with the gel before the addition of their first 4$ water* It Is noteworthy that the simple relationship® observed by lay and Ganguly as to the ratio of mole of water to mols of silica were observed in these results* it the point of greatest activity { 4$ gel) the rati® is 3 mols of water to 13 mols of silica while at the saturation point the ratio becomes 30 mols of water to 16 mols of silica* In order to explain the heat effect observed upon 32 28 Cal./g. of gel 24 16 12 10 fo 100 water on gel. Fig. 11.- Heat evolved by gel of definite water content. A n equation for the line is h = m log X/ M ♦ K where m is -34.7 and K is 88.071. £4 isiatrsiag a gal la water as fta® to destruction at a surface ||» area smst be evaluated# As the areas calculated for allloa g«l *» ailea hare varied froa 4.S x 10*ob.2 to IS x 10sea>.2 defending a* tea theory, aethed, liquid far wetting or taa&lllan of gal as©a aaf sis# the accuracy w© will refer I# fig#* 3 sat 13, fay 4at® on which to has© our assumptions few stts© fha straight Haas drawn through tha three straight parti#**® of fig* $ Inticate a change In the gal condition or gal water condition* Mteris© the carr© of fig* 11 shews two changes In nativity at approximately the same points that ware Indicated In fig* 3* fhe first point# that of gel of 3$ water eantent* wilt serve as ©widene© to assnme that when gel has water on it that the serfsoe of the gel is ©overed by s nenentlecnlar layer of water molecules* ' In this assumption XangBulr*# theory of m##s#le#sler adsorption over a plain surface. Is accepted and * further asstsaptlon that th© gal will in part exhibit such a plain surface* Areas severed by a given amount of water are calculate& from V BB aqml& 4*04 x 10 # langmulr#3 value for the area covered by one molecule of water of 7*4 x to*1*#***# and that the adsorbed water has a density of water* If 0 water is the lowest vein# before the gel begins to last its activity this might IMieste that the first 0 water Is m integral part of the silica gel* then the next 0 water is that water covering the silica gel surface with a ®cnomolecnlar« 28 Ittycr* when calculated this area hacomas 1 x 106om*2, Mtrloll has slated that the- internal area is much less than a a 4 x 10 cm* and that the Initial stages of adsorption of mater consist of the formation of a moaomoleoular layer over the surface* Table If lists the + eel*/ g* of gel/ *01 g* of water adaorhed by the gel* In this data, takes .from fig* 5, are grouped the above value® and with their averages per *01 g* of water for changes of *04 g« of water adsorbed* The differences between the storages is greatest between the first and eceond grouping® Mile the average® for the- next four groupings -and the second group decrease in quit# regular value* Mother assumption must be. made to account for the enormous energy liberated from such a small area namely* that the surface of the gel is dry at the 4$ point and that to, destroy the total energy associated with the surface by ■welting with water is the sett of wet of the energy* *# completely wet a surface, so that further water will only destroy a water surface, requires at least five- or six layers of water molecules* Selecting six as the number of layers* the weight of water per gram, of gel needed will bo *24 g* Increasing by #04 «# and the end of the second straight portion of M e curve® of fig* 5 and fig# 11 is reached at *28 g* or 25$ water* the curve® seem to Indicate the 26 break at *32 g* but a statement later will explain the reason far the previous decision* la the equation % » WglA - f{aw8l/df)A, where "A" is the area# we are gives the value of 108*0 ergs/ ea*2 for the adhesion tension of water against allies hut are not given the footer ftw^/os.. As long as "u" is obtained experimentally and "A1* by assumption «wel/d* can be calculated becoming approximately equal to *4*00 trgs/em*E* this would indicate the high eetlvity or potential energy of the surface molecules that m s predicted earlier* fe determine the heat evolved when a water surface 1® destroyed involves only the total surface energy of a bulk liquid of water*