1120 JONES AND LEWIS: CXXV1.-Studies in Catalysis. Part XIV. The Mechanism of the Inversion of Sucrose. By CATHERINE MARGARET JONES a.nd WILLIAM CUDMORE MCCULLAGH LEWIS. IN comexion with the investigations caxried out in this l a h a t t o r y on the oheanioal reactivity of mollelcular and iolnia species from the point of view ot the radiation hypothe& it was found necessary, in one selatioln of the work t o make a series of determinations of vedooitly-coinstant,s in the case od a felw familiar reiactioas suoh as the inversioln olf SUCTOBB with the object of studying molre particu-larly the influence olf change od elnvironment on the rate of the reaction. The principle olf mass actioln in it8 usual formulation, affords no1 information of an a prio1.i kina regarding the effectt of efiviromnentr.It is helped that tlhe radidion hypolthesis may be of assistanm in this colnnexioln (oompare Lewis Sc?;emth 1919, 25). As B preliminary it is nenemary tol p"sess as o l a r a view IM pwsible of the acttaal material mechanism of the selected reac-tion ifi olrder to! be able t o proceed with solmei degree of oonfidenm to the furthelr problems invollved in altelrations oif material aad tapretture eliivironment. The inversion of sucrolse is a readion particularly suitlabla for the purpose in vielw in so far as preoision of measuremeat is coacerned. Furt(her this reiactdon has been frequently illveatigated but ih spite of this the inversion proces STUDIES IN CL4TALYSIS. PART XIV. 1121 obvio;usly requires further inv&iga.tioln more particularly as regards ths way in which the hydrogen ion entelrs into the proms.The p r a i n t mssarch was underta'lcen prima'rily for this purpw. As will be seen the expeIrhnent~a,l melasurelmelnts co'nsist oa the olne ha,nd of deteirminatioas od relaction velocitiee and on the olther of determinations olr" the a#verage (geomeltric mva.n) activities of the hydrogen iojns olbst'ained by the ele,c.t,ro~met~ric melthold. In agree ment with t'he ooncllusion a;lrela3dy relacheid by Ha*rned ( J . Amer. Chem. Soc. 1918 40 1461) in c80anexion with othe'r reactions cahalysed by ions it has been found that tlhe a,ctivitly od the hydrogen ioa as defineid by G. N. Leiwis is the detelrmining fa.ct.or for the rake of the reaotion. This colnclusioa is of p a t importanm, as it not only includes t o a certain exte'nt the phenomenon known aa the sollvmt displa,oement effect b'ut adsol a.s polinbd o u t by Harned appea4rs t,o exclude the neatmity of ascribing any catalytio influence t o tlhe undisso,ciated molleloule olf the acid.In order to obta'in data which could be mplolyed for various kinds of oaJcula-tion care has been t,akeln t.0 deltexmine the amounts of all aon-stituents watees included in knolwn volumes od the va'riorus sollu tions. AS will be sholwn later the most probable material mechanism of the inversion rea,&ioln is that expresseld by RH'+H,O + dextrose + lmdow. Thah is t'he rea.ation is a true bimollecular onel bNet,wean a moleoule of wafer aad a complelx ion formeld by tfhe a,ddition of hydrogen ion to the sucrose moleloule.E x P E R I M E N T A L. Rtmctiom Velocity Measurements. The rat& of inversion olf sucrolse was oarried out in the ordinalry mannelr with the usual pretcantiolns a t 20° 30° 40° and 50° in the presence of N / 10-sulphurio acid the initJa1 concnent8ration oif SUCTOSCY being va8riield over a wide range namely from 10 to 70 per cent. in order to b,ring into prominence1 any elffelct produced by the displacement of t'he soclvent,, water the volume of the va.rious d u -tdons being ma,inta.ined const,a;nt,. All de!telrminatqiolns were carried o u t in duplica,tei and satiaf act80ry mimolecular velocity-constiants weire obt,aineId t,hronghoiut the entire1 ra,nge examined. To! save space the meran valuels only olf tIhe olbselrveld vedwity-coastants, relfelrred to bhe ba,w e and in seconds -1 as well1 a.s tbel initia'l ooncentrations of sucrolse aad watelr are1 given in the following t*a blel 1122 JONES AND LEWIS: TABLE I.CataJyst 0.1 AT-H,SO,. Grams of sucrose in 100 C.C. of solution. 10 20 30 40 60 60 70 Gram-molecules of water ( M = 18) per litre. 51.95 44.99 41.62 38.09 34.69 30.94 48-45 Unimolecular velocity -constant. at 20". at 30". at 40". at 50". 4-43 1-83 6-73 2.29 4-79 1.97 7.37 2-56 6.2 1 2.12 8.04 2.81 5.64 2.29 8.80 3.08 5.95 2-45 9.53 -6.22 2.58 10.22 -6-29 2-66 10.92 3.94 A - x 106 x 105 x 106 x 104 It will b obsesveld that in all cases displacement of the1 water by the sucrose1 causes a definito increlase in the velocity-constant a further illustration of the anti-catalytic elff ect of water already observed in other reaotiolns for example hydrolysis of esters.If the reaction is a true1 birnolecular one it is necessary to divide the velocity-constants by the oolrresponding ooncentration of water in olrdelr to obtain comparable quantitieies in so far as this stoicheiol-metric colrrelctioii is concerned. Whether the obselrved velocity-constants require to be thus divideld or not it is possible to obtain values for the critical inurementl 3 by means of the equation d log k/dT=E/RT2 from the velocity-constank a t diffelrent temperatures. It was thus found that with an increasing initial colnmntratJon of sucrose the value1 of B rises shadily over all three ranges of temperature invatJgated namely 20° to 30° 30° to 40°, and 40° to 50°.That is the critical inarment (or tlhe temperature coefficient) apparently risee whilst the velolcity-constant itself also inoreases. As this is in direclt contradiction t O the conclusion already drawn namely that the gre'ater the1 oritical increment the smaller ceteris parribus the1 velocity-constant it was concluded that the catalytio oonditions were nolt cotmparabk a t m y pair of temperatures as the contentl of sucrose was varied. The behaviour, in fact coald be accounteld folr on the1 assumptiolns that ( a ) the catalytic influence od the acid a t any given temperature! increlases with increasing concentration of sucrosel and ( b ) tlhat this inorease in aatalytio act4ivity is greater the higher the temperature.This conclusioln did not appear to1 be1 a very probable one on the basis of tbe degree of detctrolytic dissooiatioa of t3he acid f o r in the oase of sulphurio acid the1 catalytio effelct ascribed to the undis-sociated molecule is approximately the same as that of the hydrogen ion and in general the1 ratio1 otf the1 catalytic effect ascribed to the molecule1 to that o€ the ion decreases as the temperature increases. Owing to tlhel large viscosity effects producleld by a variation in th STUDIES IN CATALYSIS. PART XIV. 1123 clonaeintratioln of suorow eleotrical conductivity measuraents could not give direct information even assuming that the original q u a -t'ion olf Arrheiiius wa,s valid an assumpt,ion which ha.s recently beeln called in question.On this account it was decideld to delt,elrmine the a>ctivity or thermodynamic cornaentra.tJoa of the hydrogeln ions (striatly speaking the geomet'ria mean of the activity of hydrion and HSO,') by xmans of dwtromotivel-folrce medsurments. Activity of Hydrogem I o n ilt Apzce~ows Solutims of Sucrose. The cdl employed was of the following type: I H,SO + sucrose i saturated i normal calomel :!:::% I 0.1 N J KC1 electrode The use of a saturated solution of potassium chlolride w a middle liquid t o eliminate colntact potential differelnces has been fre-quently recommended (compare Falee and Vosburgh J. Amer. C'hem. Solc. 1918 40 1291). The absolute value of the normal calomel electrode wits taken to be +0*56 volt at 18'. Using T. W. Richards' value for the temperaturecmfficienb of this electrode, namely 0*0006 vollt per degree the polbntial difference of the calomel electrode at 20° was taken to be 0.5612 volt a t 40° 0.5732 vollt.The value of rH the potential of the hydrogen ellectrode in the sulphuria acid-sucrose solutions was obtained by means of the folllowing eiquations : rH = 0.277 + 0.058 log, H' a t 20° rH = 0.296 + 0.062 log, H' at 40°, where the eleotrdytio potential of hydrogen has bwn assumed to vary directly its the absolute temperature. The absolute value may not be correct on this basis but the quantity involved is a constant at any given temperature the relative valuea of rII at two different temperaturee are comparable. Table I1 contains the E.M.F. data obtained a t 20° the last column giving the average aotivity of the hydrogen ioln expressed in gram-molleaulm per litre.Table I11 contains similar data for 40°. In all oases diwolved oxygeln was removed from the sumose solutions t o prevent any ohange in the suarme in contact with the platiniwd platinum eledrder. an 1124 JONES AND LEWIS: TABLE 11. Temperaltus.e 20°. 0*1N-H+30,. Grams of sucrose in 100 c.c of solution. 0 10 20 30 40 50 60 70 Gram-molecGes of sucrose per litre. 0 0.292 0.585 0.877 1.169 1.460 1.755 2.047 Gram-molecules of water per litre. 55.55 51.95 45-45 44.99 41-62 38.09 34.59 30.94 (M = 18) E.M.P. of cell observed, in volts. 0-3555 0.3520 0.3485 0-3450 0.3410 0.3380 0-3345 0.33 13 P.d.of hydrogen electrode, 0-2057 0.2092 0.2127 0.2170 0.2202 0,2282 0.2267 0.2299 r n . Activity of hyfk.ogen ion in gram-molecules per litre. 0-060 0.068 0.078 0.0895 0.105 0.118 0.139 0.1 62 TABLE 111. Tempera;turq 40Q. O.lN-H,SO, Grams Of sucrose in 100 C.C. of solution. 0 10 20 30 40 50 60 70 E.M.P. of cell in volts. 0.3580 0.3545 0.3522 0.3460 0.3370 0.3280 I -P.d. of hydrogen electrode. 0.2 152 0-2 187 0.2210 0.2272 0.2362 0.2452 --Activity of H' in gram-molecules per litre. 0.050 0.056 0.062 0.078 0.09 1 * 0.109 0-130* 0.152 (Values marked with an asterisk are interpolated from the curve.) It will be observeld that the aotivity od the hydrolgen ion is 1-s atr 40° than it is a t 20°.This would be elxpsated for the cmcen-tratiom of the ion since the dissauiation of the aoid is amlmpanied by a n evolution of helat. The first conclusion to be drawn from the1 above datla is that the aativity of the hydrogen ion inareasee apprecbbly at both tempera-turels with an inoream in the aoncentration of the sucrose and furthw as shown latelr in the figure the rate of increlase in aotivity is greater a t 40° than it is a t 20°. The belaring of these rwults on the velocity-constants wiIl bs considered in the next sedioln. The wlclolnd oondusion is that the aativity of an ion is related to its colncent8ratlion in a very secondary manner. This has already been sholwlz to1 be the case by several Anmrican investigators.The &ect is weill marked in the prewnt instance. Thus when the suc1"olse colntelnt is 70 per cent. tIhe thermodynamio colncentration of the hydrogen ion is 0,162 at 20° and 0.152 at 40° although the STUDIES IN CATALYSIS. PART XIV. 1125 maximum aattaal conoatration 09 hydrogen ioin oannot exmeld 0.10. The third conclusioin is that the displacement of the solvent by a noin-ellelctrolyte sucrom producm effects entirely analogous t o those! prolduoeld by addition of nelutral salts the1 addition olf the sucrose and cionselquent eilimination of water causing an increlase in the activity of hydrolgen ion from two- to three-fold. It is p m polsed t o investigate this asp& olf the displacememtl e$fect (compare Griffith and Lewis T. 1916 109 67) in furthelr researches in this laboratory.The1 activity oif ions in presenae of a noin-eleictrollyta appears tlol have bem investigatetd in a single instance only namely, by Harned ( J . Amer. Chem. Soc. 1915 37 2467)) who deter-mined the activity of the hydrogea ioln from hydrochloric add in the presence of mannitol witholut olbselrving ho.uelvelr any marked effect over the concentsation range employed. THE PROBABLE MECHANISM OF THE INVERSION PROCESS. It was found that the direotly obseirved velocity-oonsbanb (table I) a t 20° when divided by the concentration of the water and also^ by /tthe corresponding activity of the1 hydrogen ioin gave a quantity which was a constlant within tihe limit of the expan-mental error. Similarly the data atl 40° gave a colnstant independent of the sucrose or water content tqhe numerical value beling od cosrse grelater than that a t 20°.These coastants repre-sent bimoleioular velocity-oolnstantt3 relduced to unit activity of hydrogen ions. The va,lues are given in tables IV and V. TABLE I V . Temperature; 20°. O*liV-H,SO,. Grams of sucrose in 100 C.C. of solution. 0 10 20 30 40 60 60 70 Gram-molecules of water (M=18) per litre. 55.55 51.95 48.45 44.99 4 1.62 38-09 34-59 30.94 Uniniolecular velocity-constant observed. x 108. 4*14* 4.43 4.79 5.21 5.54 5.95 6-22 6-29 Unimolecular velocity-constant divided by water c oncen t.ra ti on. x 108. 7.46* 8.53 9-85 1 I -58 13.31 15.61 17.97 20.33 Unimolecular velocity-constant x lo6 divided by water and also by activity of H ions.1.24 1-25 1.27 1.29 1.27 1.32 1.29 1-25 (Values marked with VOL. CXVII. Mean = 1 . 2 7 ~ lo-'. an asterisk are obtained by extrttpolation.) u'i 1126 JONES AND LJEWlS: TABLE V. Tempemture 40°. O.lN-H,SO,. Grams of sucrose in 100 C.C. of solution. 0 10 20 30 40 50 60 70 Grain-molecules of water per litre. 55-55 51.95 48.45 44.99 41.62 38.09 34.59 30.94 Unimolecular velocity-c on.. t ant observed. 5*98* 6-73 7.37 8.04 8.80 9.53 10.22 10.92 x 105. Unimolecular Unimolecular velocity- velocity-constant constant divided by dividcd by concontration. by H' activity. 1*08* 2.16 1-30 2.31 1.52 2*46? 1.79 2-30 2.1 1 2.31 2.50 2-30 2-96 2.27 3.53 2.33 water water and also x 106.x 105. -Mean = 2.305 x 10-6. (Values marked with an asterisk are obtained by extrapolation.) The two1 mean values 1-27 x 10-6 at 20° and 2.305 x at. of the 40° repre,sent t.he true bimo1eoula.r velolcJty-cons.t8ant k rea,cltioln whelm Icbi is de,fined by where uni is tlis olbserveid unimo1e.cula.r ve!locity-const,ant, [I-%,O] t,he oonoelntration of the water (M= IS) and [H'] is the act.ivit,y o!f t'he hydrogen ion. The vaduee of kbi as t.hus defineid a,rel in-dependelnt olf t'he colnoelntlration of sucrolse. or water and also of the activity olf the aoid ca.ta.lystl. Thei valuels od kbi depend o'nly on the temperature. Sinaa t,he change in the olbserved velocity-coastantl with t'hhe com-position of the solutioln is elntirelly acrcojunted f olr after t,he int.ro8-ductioln od the ne8mssary stoicheiornet,ria colrrelotion f olr wa,telr by t,he aotivity of tlhel hydrogen ioln it f 01lo~wwe that t$he undissociated molecule od the aoid possesses8 no1 ca8talytio effelct in this oasel.This osndusion has alre'ady bem drawn by Harned (Zoic. c i t . ) in olther easels in the prelmnoe. o'f neubral salts. It is hopeld that furthe'r invwt.igatbon with othes acids will se.tt.le the1 que'stioln a.s t'ol wheltheir tlhis csndusioa is a ge,neral one folr aque'olus sollut'io.ns. I n the case of gaseous syst,ems catadysis b,y means of acid is sca.rcedy knofx-n. Del Hemptinne. (Zeitsch. physiknl. Chem. 1892 13 561) stmates that a positive uata.lyt,io eEelc t' is producleid by liydrogeln ohlo'rids gas in the hydroslysis of gaselous ee.teirs.I n t'hheisel ciroumst.anoels, a.nd probably hhere9 olre in nom-ionising sollvent.s thel undissociated molelculel of the acid is the ca,ta,lysing individual. Th0 fadl that const,ant values for kbi as definsd abovel a,re k i = kuni /[H,O] [H'] STUDIES IN CATALYSIS. PART XIV. 1127 obt&ed a t a given temperature re'movw the diffiaultly already ment~ioineid in connexio,n with t'hhe apparent varia,tio.n o:€ t,he critica,l increment with compositJon of the1 syst.em. When the catalytlo condit'ions are1 re'ally made the] same a.t two difielre,nt temperatures, a single value for E thel critical increment per gram-molecule od sucrolse invelrted is ob,t'a.ined this value being indetpenclent od t,he cofnce,ntrafiomn oif sucrose olr water in the system.The vadue oif E doulat'ed from kbi ah Zoo a,nd 40° is 26,390 calories. The accuracy od this value depe'nds on the reht'ive a,ccura,cy with which t,he a.ct,ivities of the hydrogeln ions atl 20° a,nd 40° ha,vvei be,ea deter-mined. This numetrical value represents the sum of t,hel orittical increments of the reactmanta which will bma shown later t,o be t,he complelx ion (sucrose R') and wa,ter. The! constanoy olf kbi may be' employeld as a orit8e1rioa in ahtempt-ing to dete;rminel t,he meohaaism o,f t8h0 inversion pro'wss by m.ea,ns of stoicheiomet'rio equations f osr t'hesel elqua.t4iosiis must be such as t.01 lela,d fina,lly to1 an elxpression of t,lw folrm kbi= k,i /[HZ01 [H'].Apart from this consideraiion many adtermtive moldes oif expressing the inve.rsioi-1 pro'oess suggest themsellves. We shall considelr five oif the moistl probable of t'hese and sholw tlha.t olnly olne of them is in agrelelmeint with the critlerion. The lat,itade od choice a,risee from the doub,t as to whether (a') the hydrolgea ion which re.a,at's is hydrated oc not (6) whethetr the suoro~e whioh rea.cts is hydrated olr noh. The pro'b81em of hydrahioln is olnei od the least sat.isfact,ory a,speds of the thelory of solutdons the widetncei in most ca,sers beJng conflioting. Possibility l.-Assumel tha,t practically a31 t8he suctrose is hydrakd and tha.tl the! same1 is t,ruel for the' hydrogea ions butl thajt it is t'he non-hydra,t'ed sucrose which.reacts with a hydra,te,d hydrogein ion (H,O,H'). Applying t'het law of mass action t,ol the hydration of the sucrose R + H2Q = (R,H,O) we ham X = [R,H,Q] / [R] [H2Q], whe're X is t,he elquilibrium oonstant 09 hydratdoa and R stlands for non-hydra,ted sucrowl. The inversion process assumed to be correlot ia R + (H20,H*) -+ dextrose + lzevulose. Hence ra'te; of invers'ion = kbi [R] [M20,H']. Butl ra.te of inversioln = kuni &serv&[(R,H@)], since pra,cticc,zlly all t,he sucrose has beeln assumeid t,o bs hydratad. Hence, kbi = kuni [(R,W,O)]/[R] [f120,H'] =Auui . K . [H20]/[H20,1-3[']. This me,chanis;m requires thelref o're t,hat4 the product of the olbseirveld vdooity-oonstant by the wa8ter colnaentra4tlon divided by the activity of the hydrogen ions sholuld be it colnstlant indelpendelnt of t,he initial concent,rat.ion 04 sucrose.This diffelrs from t,he true n u 1128 JONES AND LEWIS: oriterbn and in fa.& the right-hand side of the quation is not a colnstaat. Possibility 2 .-Assume( t,ha5t pra.atica811y all the sucrme is hydrateid butl tha,t the hydrolgea ioln is practically all unhydra,M, and that t'he inversion prooees is (R,H,,O)+H' + dextroBe + lavuloee. The right-hand side of this ecxpression is not oonst,aat. Henm, pmsibility 2 is incolrrect. PossibiZity 3 .-Assume) tlha8t praotiaadly all the suaroae is hydrated and tha.t t,he hydrogeln i o a is also1 practlically all hydrated but t$at it is the noln-hydrated hydrogen i o a which relaots with hydratad suc!rolsle as in possibility 2.We have nolw to allow for Ki t.he equilibrium colnsta,nnt of hydra(t.io1n off the ioas, This assumpt,ioa leads finally t o kbi = k,I,i .&[RzO] . Ki . /[H,O,R'], which is indis,binguishab,le frolm possibilitly 1 and is thelref ore i ncolrr ect. Po~si~biJity 4 .-Assume thatl the] sucrose1 a.nd the1 hydrogeln ions amre practdca,lly all hydratsd a.nd that it is tshel hydra(ted folrm of ea.ch which relact8s a(oco1rding t,oi the e;qua,tian (R,P-I,Q) + (H,O,H') -+ delxt'rose + lawuloael. Then kbi = kmi /[H,O,H'] which beling equivalentl to pmsibility 2, is incomelot. Po.ssibiZity 5.-Assume t'hat t'be sucrose as well as the1 hydrogen iolns is pra,ctioally all noa-hydra.te:d and thamtl the first stage is a,dditioa t801 form a co!mplsx ion t,hus R -+ 33' =RHO which is f ollloweld by the true inversion reactioa, RHO + H,O 4 dext'rmel + lzvulmei.Wo then have, ra.tel of inversion =kUAi [R] alsoI ratel olf invelrsion= k . [RH'] [H20] wheaoe kbi = kuni [R]/[RH'] [HzO]. Nolw R,, the elquilib,rium colnstantl olf the complex-ion foirmation is given by H,=[RIR']/[R] [H'] whelnce khi= kuni / K c . [€I,Q] [H']. This is in agrejelmelnt with t,hel critelrion the1 right-ha,nd side being a. coinsta,ntl, since it only diffe,rrs frolm the! criterion in conntaining the colnst'ant, X,. It ma,y be1 sholwn furtheir ttha,t if the above1 mechanism is assumetd with. the1 modifioatioa tha;t t'hei sucrose is marly all hydrate,d an expression is obtained which is not in agrmmelnt with the experime,nt,al re8sult,. Also( if polssibility 5 is eunploqye,d with the mofdifica.tioa thatl t'he ions arel pra#ctically all hydrateid an equally incorrelcb expreasioln is obtained.Finally if possibsility 5 is etmplolyeld with tjhei moIdifi&ion that bolth suorose aad the ions a#re practic3a.lly all hydrated an incorrect expreaion is obtIaineld. Helnnael possibility 1 is incolrred. This leads fina811y t.0 the conclusion tha,t kbi = kuni / [H '1. Ri= [R~O,H']/[H'] [HZQ] STUDIES IN CATALYSTS. PART XIV. 1129 It is reasolnable ta coIndude therefore that polssibility 5 as i t stands is the only satisfactolry meahanism. The two stloicheiometric equations which exprem t'hei inversion prolcess are thelref ore : and Itl is obviolus tlhat this melohanism accolunts for the1 faat thatl the hydrogeln ion is neIcmsary in olrdsr that inversion may proceeld with measurable speed.There1 is also some evidence od a prelliminary nature f o r the existleinm of an addition complex of sucrose and hydrogen ion which itl is hopeid may be communicated in detail later. The mechanism of the1 inversion process to which we have been leld involves etquatiolns (1) and (Z) and also! the assumptioln that the hydrogen ions and thel sucrose molwules are not appreciably hydrated in aqueous solution. This relpresents the simpled set of conditions but the r e u l t is somelwhat surprising. Sucrose is usually regarded as being helavily hydrated in sollution. As already pointed out howeiVelr such an assumptioln would no6 lead to the obselrved constanoy of the quantity elxprelssed in the final clollumns of ta$blels IV and V.R + H' = RH' (practically instantaneloas) . . (1) RH' + H,O = dextrolsel+ lzvulowi (measurable) . . (2) Hydvoigem-ioa Activity as a Fwnctiom of the Compositiom of the1 S0~2~tiOYnl. Rosanoff ( J . Amer. Chem. SO:^. 1913 35 173) has suggested that the cat'alytia elffect of a solvemt may be1 eocpressed as an expo,neatia,l funct'ion 04 the1 concentration of the solvent. A similar rehtdon has bean etmployecd by Wilsoln (ibid. 1920 42 715) to a,ocoant folr t.hel effeat of nelutral sa.lts on the1 hydrogen ioln as delter-mined by eleot,rometlric me.asurements. Wilso'n appears tlo relgard t,he increase in macenttra,tion of t'hei hydroge'n ion in the prelselnce of ne>utra,l salts as rela,l. The e,ffect howevelr involvee the a,ctivity, and nolt nemsarily the c,oncentlratdon of the ion.The ana'logy of the1 e,ffed produced by the1 displaoemelnt od water by a'ddition oif suorose to thah prolduced by a> neutral sa'lt has already been inent.icmed. I n the aondit,ioas olbtaining in the1 present' investliga;fmion we deal with t<hs aimultaneloius influeince of sucrose and wa'ter since by a,ddition od one mlnstitne.nt a cert,ain amoant of tihe1 other is elliminateld. Let us suppow thab the activity of the hydrogen ion can be expressed by [H'] = A . e(bR+b'W) . . . . . . . (3 1130 JONES AND LEWIS: whelre R and Jlr are the concentratiolns of suclrme and watelr respectivelly in gram-molelcules per litre b and b' are the1 oatalytia environmental constants characteristic olf suorose and wabr b is a positivel whilst bl is a negative[ quantity.When there1 is no sucrose present W becomes IV, (namely 55.55 gram-molecules per litrel) and H* belcomels TI*(). We can thus repwrite equation (3) in the1 form : Hence, (34 H' = H',e4 ~ R - ~ ' ( W O - W \ . . . . . . loge N' =loge H.0 + bR - b'(W9- W). . . . (4) In the present case itl happens thatl a straight line is obtained when the aoncentration of sucroae is plotteld against the1 oonmn-tration of the water. Hence we can write TV=W,+pR where1 p is a negative constant having the value -12.1 the sucrose and water being expressed in gram-moleculee per litre. Equation (4) can thereforel be writtetn in the present case in the form or alternatively, l ~ ~ ~ E € ' = ~ o g ~ H ' ~ + (b+b'~3)R .. . . (5) logeH' = 10geH.O + (b/p + b')( W - Wo) . . . (6) Equatioas (5) and (6) are both linear thatl is they represent the1 obseirveld belhaviour of the1 activity of the hydrogen ion as a function of the composition of the mixture as is shown in the figure (graphs C and 0). Heam eiquatlioln (3) is justified in so far a t lelast as it satisfies the above condition in the present case. From the1 experimental data the follolwing mean vadues are1 obtained: b - 12.lb' = 0.47 a t ZOO, b - 1 2 W = 0 * 5 6 , 40'. Itl will be olbselrved that equations (5) and (6) being elquivaleint do noh permit of a calclulation of b and bl separately. Furthelr experi-melntal da,ta invollving a diEerent functional relationship beltween the aonoeintration of the sucrose and the water are required.I n equations (3) to (6) we have employed the most general type of folrmula in which a possible catalytic elf€ect has even been ascribed to1 the reactant solute sucrose. In this connexion there is an important coinsidelration which has t o be taken into1 amount. T he1 obselrved unim ole cula r v el ocit y-con st an ts obtained at various temperatures are1 found tot remain constant within tlhe limit; of expeirianental error throughout the entire course of the reactioln. This is the case1 even when the initial concentration of sucrose is as high as 70 per wnt. as is shown by the following results of duplicate expelrimeuzts which give the observeid velody-aonstant ST(JD1ES IN CATALYSlS. PART XN. 1131 obtaineid with 70 per cent. suerow at 40° in the presenoei of N / l O -sulphuric acid.The vellociity-coust'aiit is calculated to1 the base e . -.-I 3 5 4 0 4- 5 5 0 5 5 B O 5 5 5 0 4 5 4 - 0 3 3 3 3 Water concentration moles per litre. (k = okerved veloci% constant) \ water concentration 1 TABLE VI. Experiment 2. I Experiment 1. Time in seconds. 0 780 1600 2100 3060 3900 5160 6420 7500 Unimolecular velocity constant x lo6. 10.82 11-03 10.98 11-00 11.03 11-07 11.11 10.98 (55 per cent. of the sucrose decomposed.) -Time in seconds. 0 1200 1950 2460 3060 3960 4440 6360 6900 Unimolecular velocity constant x 106. 10-81 10.81 10.83 10.77 10.89 10.86 10.96 10.97 (53 per cent. of tho sucrose decomposed.) 1132 STUDIES IN CATALYSIS.PART XIV. The facb tha.t the velocity-coastant does noit alter indioates tha,t the a,ctivity od the hydropn ion is likelwise sensibly constant. As, howelver the conoent,ration olf the sucrose ha.s diminished b,y more than onel ha81f it follo8ws tlhat the eaviroamental influenca of tlhe sucroBel if any ass melasurelcl bsy the1 quant,ity b must blel t,he same! as that of t'hel de'xtrosei a,nd lzvuloleel formeld a's a relsult od t'he invelrsioln. If these sub~st8ance8s a,rel inelrt it woiuld foillow that b = 0 and tha.tl thetreifolre the1 elnvironmelntad mtadytia teirm b', cha,raotelrisbic of wa,t,eIr is -0.039 ab 20' aad - 0.046 a't 40'. The nega<t,ivel value indica,t,ea of colursel t'ha4t the wa.ter is al negative cata,lyst for t'he procelss.The question ot t,ha albsoilutle value olf b will be deIdt with. in a la,te.r clommunica.tion. Rehurning t'ot thei graphs shown in the1 figurel it1 will b'e olbserveld that witb diminution in tshel co8ncent,ra'tioIn of wat'er the aurvels folr the hydrogeln-ion a&ivity art8 20' a,nd 40' a.re1 noit paradlel but coavorgent. The same relation is exhibiteid by the mrvea for the velocity-aolnstantl shown in tbe same diagram graphs A a.nd B. When t'he two1 wfta of curvea adel oombineld coastant va'lues folr k,, as dehed in the prejcelding sectioa are1 olbt8a8ineld. The relahive pmitioas8 olf t,he two sets of ourves a,fford the most direct emidenm yet obhaineld fo'r the1 colnclusion that the activity of t'he ion ooImplet,elly delterminels thel velocity of the relaction.8ummury. (1) The1 vellolciity od inswrsioln od sucrolsa in the1 presence of N/10-slulphuria amaid has bteen dete'rmineld a8t 20° 30° 40° and 50° the compoeitioa od the sollut>ion beling altetreld by gra,dual displa4cmelnt of t'he wa,ter bmy the sucrose. The ~e~lolcity-constanta vary with the initial colmpolsit8ion olf the mixtare. (2) Thei a4velrage activitks of t,he hydrogen ion ha4ve beein deltelr-mineld e~lec1t~rome~trica.lly in tbs varioas mixtures referred t,oi in (1) at 20° and 40°. It is sholwn tha,tl the1 a,ltera,tioln in the vellodty-const'ant ca,n be a,ccountad for completely by (a) allowing folr the st,ob&eiornet,ririo colrreiction f oc t,he wa.tes preseat and ( b ) by allow-ing for the cha*nge in the act'ivity of the hydrogen ions. (3) I n a,greeimetnt with (2) the inversion process is shown t o be bimollecula,r. Iti consists osf tswo proaes6els (i) union olf the non-hydra,teld hydrolgeln iojn with an noIn-hydratd molleiauls of sucrwel, thus R + IP' =RHO this proiceas being pmctioally inst8antaneoas, and (ii) the actual inversion reIachion RH' + H,O = delxtroset hvulme. It is a,lsol concluded that. tlhei hydrogen iolns a,nd the sucrose moleIcule8 a,re not sensibly hydrate'd in a.quwus so1ut.ion. (4) I n agrememt wit'h Roeanoff's suggestion regarding the mod THE PREPARATION OF GUANID'ENE ETC. 1133 of eixpressing the influence of solvent cdalysts the aotivity of the hydrogen i o n is folund to be an eixponential functioln of the coaaentration of sumose and water praeat. (5) It is shown thatl the e'nvironmeatal cahalytio influence of a molleculel of SUCTO~S~ refelrreld tlol in (4) is idetntical in magnitude with that exerted by one molleculej of dextrose tolgether with one molwule of laevuloae. MUSPRATT LABORATORY OF PHYSICAL AND ELECTRO-CHEMISTRY, UNIVERSITY OF LIVERPOOL. [Received August 4th 1920.