2490 'IIZARD: THE HYDROLYSIS OFCCLVI.---The Hyd~olysis of Aniline SccZts McasuredCo 1 o r ime t r ical I y .By HENRY THOMAS TIZARD.IN the preceding paper it wm shown that the concentration ofhydrogen ions in a solution of methyl-orange, tho molecular coloixrof wliicli is G, is given by the equations :C4-25 x 10-4y= 18.8y+(l -y) . . . . , . . . . . . (1)= (1 -9) x conc. He . . . . . . . . . (2)ANILINE SALTS MEASURED COLORIMETRICALLY. 2491where y denotes the fract,ion of methyl-orange present in the formof undissociated indicator acid.These equations have now been used to determine colorimetricallythe degree of hydrolysis of aniline salts.Aniline Hydrochloride.Aniline hydrochloride was prepared by mixing equivalentquantities of pure concentrated hydrochloric acid and aniline,which was purified in the way recommended by Hantzsch andFreese (Ber., 1894, 27, 2966).The salt was recrystallised fromwater slightly acidified with hydrochloric acid. The colour measure-ments were made in exactly the same way as that described in thepreceding paper. The tempera.ture of observation was 25O. One ofthe chief objections to the measurement of hydrolysis by colorimetricobservations seems to be the difficulty of keeping the temperatureconstant. It was found best to keep the solutions until just beforeuse in a thermostat at bhe required temperature, and then totransfer the requisite amount quickly to the tintometer tube.Since with a certain amount of practice accurate colour measurements can be made very rapidly, the temperature alters only slightlyduring the acbual experiment. I n the experiments communicatedin this paper, it certainly did not vary sufficiently to affect thecolour outside the unavoidable error of observation.The dilutions examined ranged from 1 /ZO- to 1 / 1200-normal.I n a, pure aqueous solution of a salt of a weak base, the con-centration of the hydrolysed base is, of course, equal to that of thefree hydrogen ions in the solution. When, however, an indicator,such as methyl-orange, is also present in the solution, it partlycombines with the hydrogen ions, and the concentration of the latterwill therefore be less than that of the hydrolysed base.Thus, if 12:is the fraction of salt hydrolysed, Vl the dilution of the salt, Vzthat of the indicator, and y (as before) the fraction of the latterpresent in the form of undissociated acid :x: - - - conc.aniline= cow H' + --. VIYvzThis correction is of considerable importance when hydrolysis islarge and the dilut,ion of ths indicator not too great. Thus, totake an example from the following ta,l.de, when V/',=200 andV,=ZO,OOO, the molecular colour is found to be 8-63. Fromequations (1) and (2) we get:y=0*429, conc. 4' = 3-19 x 10-42492 TIZARD : THE HYDROLYSIShenceconc. aniline = 3.19 x += 3.40 xOF0.42930,000The difference between the two concentrations is thus 7 percent. It was not taken into account by Veley in researches of asimilar nature, and this, together with the reasons already putforward in the previous paper, may explain the frequent differencebetween his results and those arrived at by other methods.The hydrolysis constant is given by the equations:x conc.He. K, - conc. C,H,*NH2 x c o w H' - XKb conc. C,H,*NH, 1 -x- - -. - -The following table contains the results obtained :20304060801002003004005006008001000120020,00050,00040,00030,00020,00030,00020,00015,00020,00020,00015,00013,33320,00020,00012.611-911 '410.710'49-868.637-846'976-636'255'865 '484-920.6520'6120.5840-5450.5280.4980'4290.3840.3350.3160.2950.2730-2520.2207'946.715-975.094-764-213-192 *652 '141 *961.781'591 -431.201 -652-072 -513-164 '124.386.808'739-2410.611-914-315'615.7(0'133)(0'142)(0-154)(0.176)(0.204)(0.193)0 '2300.2530.2180.2320'2400.2650'2640.223A 2 per cent.error in colour measurement corresponds on anaverage wit.h an error in the hydrolysis constant of 10 per cent.,so that the fluctuations of the constant from the mean value for thelast eight dilutions are within the error of observations, and,moreover, are not materially greater than those found when othermethods of measurement are employed. The actual mean value ofthe constant for these eight dilutions is:Kw = 0.242 x 10-4, Kbwhereas Bredig (Zeitsch. physikal. Chem., 1894, 13, 239) found byconductivity measurements :The two values are thus practically identical.At higher concentrations, however, the constant is very muchsmaller, and appears to increase quite regularly with the dilution.This difference is too great to be accounted for by the incompletANILINE SALTS MEASURED COLORIMGTRICALLY. 2493dissociation of the aniline hydrochloride, and as the behaviour hasnot been observed when other methods are used, it is presumablydue to the formation of a compound with methyl-orange, whichaffects the colour to a small extent.With aniline acetate, however,no such phenomenon was observed; the colour in a N/2O-solutionof this salt is perfectly normal. The swalled neutral salt actiondiscussed by Szyszkowski (Zeitsch. physikal. Chem., 1910, 73, 269)is apparently unconnected with the above phenomenon, for methyl-orange appears to indicate a higher concentration of hydrogen ionin presence of sodium chloride than the solution contains, whereasthe reverse is true with aniline hydrochloride.On the whole, it is perhaps inadvisable to use concentratedsolutions of salts in colour measurements of this kind.Anzline Acetate.This salt has apparently not been prepared in the solid state.Amixture of equivalent weights of pure aniline and acetic acid willnot solidify when cooled to -20°, but as the liquid is extremelyviscous at this temperature, it is probably supercooled. If theequivalent mixture is kept for some time, acetanilide is formed.The solution used in these experiments was therefore made bymixing equivalent quantities of N/5-aniline and N / 5-acetic acid,both of which were made up accurately by weight from the puresubstances.In a solution of a salt of a weak acid and a weak base, thedissociation constants of which are and Kb respectively, let y bethe degree of dissociation of the salt, and x the fraction hydrolysed;then Ostwald's law gives the two equations :hence :K ~ , z = (1 - X) x COUC.H'K b . X = (L - X) x conc. OH ;or :This is the well-known equation for the hydrolysis of such a salt;it signifies that when the salt is completely dissociated ( y = l ) , thedegree of hydrolysis is independent of the dilution. It has not,however, yet been pointed out, so far as I know, that the con-centration of the hydrogen (and hydroxyl) ions in a solution of aweak salt must always be constant, whatever be the dilution, an2494 HYDROLYSIS OF ANILINE SALTS.whether the salt is completely dissociated or not.once from the above equations, forThis follows atconc. H' = ~ Kmx - - 2/R;: = constant.Y(1 - 4A striking proof of this can be obtained colorimetrically.Theaddition of a few drops of the aniline acetate solution to a neutralsolut'ion of methyl-orange causes the same rise in colour as theaddition of a large quantity. The results of actual experiments areshown below:Aniline acetate.20405080100200500x: (For completeColour. Mean. conc. H'. dissociation).- - - 1'921-921.90 -1 '90 1-91 0.229 x 0.5601-901.91 -1 -90- - -- -- -- -- - -The molecular colour is constant within the errors of observationwhen the concentration of aniline acetate varies from X / 2 0 toN/500.From the colour, the concentration of the hydrogen ionis obtained as before, and t-hen the degree of hydrolysis is calculatedfrom the equation:conc. H' __-- x - - -l - x Ka - -'Ka being 1.8 xArrhenius and Walker(Zeitsch. physikal. Chem., 1889, 5, 18) found that for the samesalt the percentage hydrolysed was 55.5, as a mean of the valuesfor-six different dilutions ranging from V = 12.5 to V = 400.The two methods therefore give practically identical results.From trhe hydrolysis of the chloride, we can calculate :In this way we find x=56'0 per cent.= 4.6 x 1O-I'from that of the acetate:= 3.8 x 10-l'.This difference is within the error of hydrolysis measurement byany method.The colorimetric method used in the investigationseems to give too low values for the hydrolysis constant of thehydrochloride, so that Kt, calculated from the hydrolysis of theacetate is probably more accurate. I n any case the mean value :Kh = 4.2 x 10-lOat 25TUTIN : SYNTHESES Ih’ THE EPINEPHRINE SERIES. PART 11. 2495cannot be far from the truth. The number accepted by LundQn(4.6 x 10-10) appears therefore too high.I n conclusion, the results communicated in the present paper showthat, with proper precautions, hydrolysis can be measured colori-metrically by means of methyl-orange, with an accuracy thatcompares favourably with that att.ained by other methods. Thebase should, however, have a dissociation constant less than 10-7.For bases stronger than this, but weaker than ammonia, methyl-redwill probably be found suitable. Since the apparatus and themethod of working are extremely simple, there is no reason why thecolorimetric method should not come into more extended use.Summary.It is shown that the degree of hydrolysis of aniline salts can beaccurately determined by measuring the depth of colour of methyl-orange in the solution, and then calculating the concentration ofhydrogen ions by means of the equations deduced in the precedingpaper.It is also pointed out that the concentration of hydrogen ions ina solution of a salt of a weak acid and a weak base is always thesame, whatever the dilution and degree of dissociation of the salt.The researches communicated in this and the preceding paperwere carried out in the Davy-Faraday Laboratory of the RoyalInstitution. I should like to express here my thanks to themanagers of the laboratory for the facilities they have placed atmy disposal.DAVY-FARADAY LABORATORY,ROYAL INSTITUTION