VELOClTY OF FORMATION OF ACETOXIME. 85IX.-The InJEuence of Acids and Alkalis O ~ L theVelocity o f Formation o f Acetoxime.By ERNEST BARRETT and ARTHUR LAPWORTH.IN a paper which appeared in the July number of this Journal(Trans., 1907, 91, 1133), it was shown by one of us that acidsaccelerate oxime formation, and that certain oximes are formed withgreat ease even in presence of concentrated hydrochloric acid. At theconclusion of the paper, a subsequent communication on measure-ments of the velocity of reaction between hydroxylamine and acetoneand aldehyde in presence of both bases and acids was promised. Themeasurements referred to had been made some time before the firstpaper was submitted to the Society ; the work had been going on forsome eighteen months, and the further investigation vas in progress.I n the September number of the Amel-ican Chemical Joumal, 86 BARRETT AND LAPWORTH: THE INFLUENCE OF AClDS ANDpaper by Acree and Johnson appeared (Amer.Chenz,. J., 1907, 38,258), which dealt with many chapters of catalysis, and includedmeasurements on the equilibrium point of the reversible reactionbetween acetaldoxime, acetone, and hydroxylamine hydrochloride, thevelocity of hydrolysis of acetoxime with hydrochloric acid, and thevelocity of formation of acetoxime from acetone and free hydroxyl-amine (Zoc. cit., p. 308 et 8eq.).Towards the end of November, Prof. Acree wrote stating that hehad only recently seen Lapworth's paper for the first time, and drewour attention to his work with Johnson, expressing the hope that wemight see our way to abandon the investigation to him, In thecircumstances, we have decided to submit to the Society a brief accountof the more important observations we had made before last July,when the first communication of the proposed series appeared, and theobservations me select are those which appear to add something to thediscoveries of Acree and Johnson and other workers, but they do notsolve the question of the mechanism of the reaction.On the otherhand, we have agreed to abandon the further study of the velocity offormation and hydrolysis of oximes, at least for the present, in favourof the American authors.The work was begun some two years ago, after hearing fromStewart, the last worker on the velocity of oxime formation (Trans,,1905, 87, 410)) that he did not propose to extend his researches inthis direction.The measurements were made much in the mannerdescribed by Stewart, but were carried out at 0' with solutions inwhich the concentrations of acetone and hydroxylamine were 3/40 innearly all cases. Instead of sodium phosphate, sodium acetate wasused during the titrations, for it was desired to study the formationof aldoximes, in which case sodium phosphate was found to be useless.I n brief, our results supplement those of other workers in thefollowing points. The acceleration of oxime formation by alkalis,first noticed by Auwers (Ber., 1889, 22, 605), we find to be veryconsiderable and nearly proportional to the concentration of alkali.The acceleration of the reaction between hydroxylamine and acetoneby hydrochloric acid is rather less marked at first, but risea withincreasing proportions of acid very rapidly to a maximum, in whichthere is little or no suggestion of lag, to a point where there ispresent about 0.6 to 0.6 equivalent of acid, the velocity here beingmore than fifty times as great as with free hydroxylamine alone.It then falls almost as rapidly to the point where there is presentrather more than one equivalent of hydrochloric acid, after which thevariation of velocity with amount of acid present is small (seediagram, p.92).With acetaldehyde, more dificulty mas experienced in estimatinALKALIS ON THE VELOCITY OF FORMATION OF ACETOXIME. 87the greater velocity of reaction, and much higher dilutions werenecessary.The curve obtained on plotting our preliminary resultswas much like that for acetone, but the point of maximum velocitywas reached with a somewhat larger proportion of acid.The acceleration of oxime formation by addition of alkali, and theapproximate proportionality of the velocity to the concentration ofalkali present, suggests at once that the hydroxylamine behaves aa aweak acid and reacts with the carbonyl compound in much the samemanner as does hydrocyanic acid (compare Trans., 1903, 83, 99’7;1904, 85, 1214 and 1355, &c.),Thus, to take the simplest possible ionic view, the hydroxylaminemay be supposed to yield the ions H* and *NH*OH, the latter ofwhich, as a weak ion, forms a complex with the carbonyl compound :R2C = 0 + *NH*OH f+ R2C<NH.0H,+ -- 0-this complex ion formation being relatively very slow.The idea that ammonia in analogous cases reacts as an acid andin the parts H and NH, or NH,*OH has been suggested by Knorr(Bey., 1899, 32, 731), and revived in a slightly different form byLowry (British Assoc.Report, 1904,The question of the mechanism of the acceleration of oxirne formationby acids is certainly much less straightforward. From the fact thatan equilibrium is attained when acids act on acetoxime, Acree andJohnson conclude that the reaction involves the union of the hydroxyl-ammonium ion with the ketone as neutral constituent :+Dynamic Isomerism,” p. 11).+ +(CH,),CO + NH,*OH -+ (CH,),CO*NH,*OH.This is highly improbable; firstly, because of the already highelectro-affinity of the hydroxylammonium ion, and secondly, becauseall the evidence hitherto goes t o show that only negative groups becomeattached to the carbon atom of the carbonyl group.A more likely suggestion is that the hydroxylammonium ion is notdirectly concerned, but that the acetone forms complex ions withhydrogen ions which are present as the result of hydrolysis of thehydroxylamine salt : + 4 OH(CH3),C0 + 11.+-+ (CH,),C<+(the oxonium ion (CH,)2C:O<H being possibly formed at an inter-mediate step), and this positive ion then attacks the free hydroxyl-amine, forming a substituted hydroxylammouium ion 88 BARRETT AND LAPWORTH: THE INELUENCE OF ACIDS ANDAcree and john son'^ view would predict a regular rise in thevelocity from the state where free hydroxylamine is present to thatwhere there is one equivalent of acid, and after this a regular, butvery slight, fall.On the view now suggested, addition of acid to free hydroxylaminewould at first produce little effect, and afterwards a rapidly increasingacceleration for a time, leading to a curve at first distinctly conmvetowards the line A in the diagram, and near and beyond the pointwhere an equivalent of acid is present a maximum followed by a slightdecreasing velocity as with Acree and Johnson's proposition.It is not worth while at present to enlarge on these views.Neitherexplains the curious variation of the velocity between the points Aand B on the diagram.It seems hardly possible that this can beelucidated by any view as to the mechanism of oxime formation fromhydroxylamine and acetone. A supposition which would lead to acurve attaining a maximum value between these points would be oneassuming that the measured change involved the interaction of freehydroxylamine, hydroxylammonium ions, and acetone, but this would beroughly of the form y =.(a - x), and would show a rapid rise near A , aslow change near the maximum, and an increasingly rapid fall to B.The curve rather appears to suggest that its form may be due to apeculiarity either of acetone or of hydroxylamine itself. Is itpossible, for example, that hydroxylamine gives a salt, (NH,*OH),,HCl,yielding a basg, (NH,*OH),, by hydrolysis, and transformed byexcess of hydrochloric acid into NH,*OH,HCl 0 The tendency of theoxygenated derivatives of ammonia to form more complex aggregatesmay be recalled, and a salt of the formula quoted is known to beproduced readily enough in the solid form, but whether this exists toany large extent as such, or as its ions, in aqueous solution, does notappear to be known.A fact which seems to militate against suchan explanation is that the condition for maximum velocity withacetaldehyde does not appear to coincide with that with acetone, thepoint lying somewhat nearerethe line B.EXPERIMENTAL.In the following series of experiments with acetone, the initialconcentrations of hydroxylamine and acetone were, in all cases, N/40.The solutions in which the reaction was studied were prepared bymixing equal volumes of N/10 solution of acetone and hydroxylaminehydrochloride to the latter of which had previously bean added vary-ing quantities of sodium hydroxide or hydrochloric acid, it being soarranged that immediately after admixture there should be exactly1 gram-molecule each of ketone and total hydroxylamine per 40 litresOF solution, and all operations were ci-trried out a t 0'ALKALIS ON THE VEL~CITY OF FORMATION OF ACETOXIME. 89In the tables given, the amounts of hydrochloric acid or sodiumhydroxide present are also stated in gram-molecules per 40 litres, andit is to be understood that the stated amount of hydrochloric acid ineach case includes both the free acid and that combined with thehydroxylamine; similarly, in experiments I1 to IV, the amount ofsodium hydroxide given refers to the excess of the latter used overthat required to convert all the hydroxylamine present into free base.In the majority of the experiments, therefore, sodium chloride wasnecessarily present, but we have found that the effect of this salt washardly perceptible, and did not produce any effect on the velocity ofreaction sufficient to alter in any way the general conclusions towhich the investigation lends.After many preliminary trials, the method adopted for estimatingthe amount of oxime formed was to determine the quantity ofhydroxylamine which remained by oxidising it with excess of standardiodine solution as other workers have done, but having found that inalkaline media, such as sodium phosphate or bicarbonate, concordantresults could not be obtained in presence of aldehydes, we were finallyled to carry out the titrations in presence of sodium acetate.Thissalt, if highly purified, serves a similar purpose, and in its presencehydroxylamine may be determined with a very fair degree ofaccuracy, providing that much free acid is not present with thehy drox y lamine.In each case, the time which elapsed between the moment ofadmixture and that of withdrawing an aliquot portion of the solutionfor titration is given in minutes. The number indicating the amountof oxime formed represents the number of gram-molecules presentper 4000 litres, the maximum possible at the end point being, ofcourse, equal to 100, and was arrived at by subtracting the quantityof hydroxylamine unchanged in 40 litres from 100.The last numberin each case represents the quantity of oxime formed at the endpoint.I. No acid or alkali present (that is, the solution was preparedso as to contain 1 gram-molecule each of hydroxylamine, hydro-chlcride, and acetone per 40 litres, the free base being liberated byaddition of 1 gram-molecule of sodium hydroxide per 40 litres) :t= 2 5 10 20 30 -Oxime Somed= 5-8 17'0 29'2 43.7 52.9 99'5SERIES l.--In Presence of Alkali.11. 0.20 gram-molecule of free sodium hydroxide present. Inexperiments I1 to IV a quantity of acid sutlicient to neutralise th90alkali present was added to the sodium acetate before the titration ofhydroxylamine :BARRETT AND LAPWOKl'H: THE INFLUENCE OF ACIDS ANDt = 2 5 10 21 36Oxime formed= 33.9 52-2 69'6 83 '0 89.3111. 0.25 gram-molecule of sodium hydroxide present :t = 2 5 10 20 -Oxime formed= 49.5 68.3 81.6 90.9 98-8IV.1-00 gram-molecule of sodium hydroxide present :t= 1 2 5 10 20( a ) Oxime formed= - 64.3 82.7 92.3 97.1(6) ,, ,, = 51.4 66 -3 84.4 92'6 97.9SERIES 2.--In Presence of Acid.V, 0.066 gram-molecule of hydrogen chloride present :t= 3 5 10 20 30Oxime formed= 7.3 13.4 28.7 52'2 68.6VI. 0.25 gram-molecule of hydrogen chloride present :t= 2 5 10 20 30Oxime formed= 19.3 43.9 73.8 81.7 83-5VII. 0.30 gram-molecule of hydrogen chloride present :t= 1 2 5 10 15Oximeformed= 21.7 34.5 67'4 81'4 85 -0VIII.0.40 gram-molecule of hydrogen chloride present :t = 1 2 5 11 20Oxime formed= 19.1 38.0 64'7 72.6 77.6IX. 0.50 gram-molecule of hydrogen chloride present :t= 2 5 10 20 30Oxime formed= 49.8 54.5 65'2 68.9 70-4X. 0.60 gram-molecule of hydrogen chloride present :Oxime formed= 31'6 44.2 54.9 68.5 68'5 69'4XI. 0.75 gram-molecule of hydrogen chloride present :t.= 3 5 10 20 30Oxime formed= 31.3 46.5 50.4 58.7 61.1XII. 1 *00 gram-molecules of hydrogen chloride presentt= 1 2 5 18 25 36-99.5-10000-98.185.9-85 *8I77.6-71 *O-69.4-61 '9(that isto say, in this experiment, initially only acetone and hydroxylaminehydrochloride were present) :- t= 2 5 10 15 37Oxirne formed= 16.3 28.6 40'1 46.1 57 -1 60-ALKALIS ON THE VELOCITY OF FORMATlON OF ACETOXIME.91XILI. 1.50 gram-molecules of hydrogen chloride present ;t = 2 5 10 20 35 -Oxime formed= 3*6 7.7 21.5 32 -1 45'1 58.3XIV. 1 -5 12 gram-molecules of hydrogen chloride present :t = 1 2 5 10 25Oxime formed = 1 *6 4.7 12.6 23.5 32.7 58.3XV. 1064 gram-molecules of hydrogen chloride present :Oxime famed= 4 ' 1 13.4 18.5 28-0 35.3 55.0XVI. 2.024 gram-molecules of hydrogen chloride present :Oxime formed= 1.6 2.0 (2) 12.2 23.8 31-2 50.6-t = 2 5 10 20 30 -t = 1 2 5 11 25 -I n the last cited and in other experiments in which more than2 molecular proportions of acid were present, the disturbing influenceof the excess of acid on the titration of the hydroxylamine makesitself felt.I n all cases, the quantity of hydroxylamine found wasgreater than that really present, so that the amount of oxime calcu-lated on the same basis as in the preceding experiments appears lessthan is actually the case. Nevertheless, the results showed clearlyenough that., even when as much as 40 molecular proportions ofhydrogen chloride are present, oxime formation takes place withconsiderable velocity, and this does not appear to vary much withinvery wide limits. It had been our intention to examine thisregion, using sodium phosphate instead of acetate as the mediumduring titration.Fairly concordant numbers obtained on repeating a considerablenumber of the above experiments, indicate that the results may beregarded as correct within 2-3 units.We have also carried out experiments on similar lines, using acidsother than hydrochloric acid, but the results were without muchfurther significance.The most noteworthy points revealed by a glance a t the numbers inthe above tables are, first, that there is a minimum velocity at, or verynear, the point where only acetone and hydroxylamine are present(experiment I) ; secondly, that very large acceleration is caused bythe addition of alkalis or acids, the former having proportionately aconsiderably greater influence ; thirdly, that there is a maximumvelocity point between this point and that where the solution containsnothing but acetone and hydroxylamine hydrochloride.This is at aboutthe point attained in experiment JX,and, as will be seen, the amountof oxime formed here in two minutes is larger than in any other caseon the side where acid is present; fourthly, beyond this point rapi92 VELOCITY OF FORMATION OF ACETOXIME.fall in the velocity occurs, but even where an enormous excess of acidis present there is evidence that the velocity of oxime formationremains perhaps larger than when free hydroxylamine alone is present.Owing to the rapid change of the velocity with acidity (and thisvaries during each experiment), it is not possible to obtain velocityconstants, and the errors of time measurement in the first stages of theoxime formation render it difficult to obtain a value for the initialvelocity under any prescribed conditions. However, rememberingthat when much acid is present partial hydrolysis of the oximeoccurs, the reaction therefore being incomplete, a fair idea of therelative velocities with different canditions as to concentration of acidand alkali may be obtained by carefully plotting the results andascertaining the time required for the reaction to proceed half way t othe point a t which change ceases.It is clear, of course, that it wouldbe better to take points corresponding t o one-quarter, one-tenth, orless, but this leads to a magnification of other errors, and it maybe stated that the curves thus obtained are found to be very similar,showing precisely the some peculiarities.The following diagram exhibits the velocity of reaction estimated1.0 0-5Grammol. NaOH. Gmwi-mols.IFC1 per 40 Zitres.by taking the inverse of the time required for the reaction to proceedhalf-way towards completion. The vertical line at A correspondswith conditions when acetone and free hydroxylamine only are present ;to the left of this, alkali, and, to the right, acid is present. Thevertical line at B corresponds with the point where acetone andhydroxylamine hydrochloride only are presentCOLORIMETRIC METHOD FOR THE DETERMINATlON OF IRON. 93The velocities were found by ascertaining, graphically, the timerequired for the formation of one-half the amount of oxime preseut atequilibrium point, and the numbers given are one hundred times thereciprocals of these.The method of exhibiting the results, although rough, at leastaffords a general idea of some of the peculiarities in the formation ofacetoxime. It is difficult to state precisely the points or magnitudesof the maximum or minimum velocities, but greater or less deviationsthan those indicated have not been found in spite of repeated searchwith slightly varying conditions.It is noteworthy that the extension of the curve on the rightappears to be nearly horizontal beyond the limits shown; in otherwords, it seems that with excess of hydrochloric acid the velocityvaries only very slightly with the concentration.By varying either the amount of acetone or hydroxylamine whilekeeping the concentration of the other unaltered, me have ascertainedthat the velocity is nearly proportional to the concentration of each ofthese separately.The nature of the results we obtained in using acetaldehyde insteadof acetone have already been alluded to. We hesitate to give thedetails, because those experiments were of a preliminary character,and, owing to the much higher dilution necessary, eubject to errorswhich we hoped to be able to eliminate on repeating the measure-ments.Our thanks are due to the Research Fund Committee of theChemical Society for a grant, which helped to defray the cost of theinvestigation.GOLDSMITHS’ COLLEGE, NEW CROSS, S. E