KINETICSOF THE
Sept., 1951
[CONTRIBUTION PROM THE
4307
IODINATION OF PHENOL
MARIONEDWARDS PARK LABORATORY OF BRYNMAWRCOLLEGE]
Kinetics of the Iodination of Phenol BY ERNSTBERLINER The kinetics of the iodination of phenol were studied in an attempt to distinguish between the undissociated phenol and the phenoxide ion as the reacting species. The rate constants calculated for the iodination of the phenoxide ion are much greater than those for the iodination of aniline, and the activation energy is smaller. This is in accord with predictions made by the electronic theory of substitution, and supports the assumption that the phenoxide ion is involved rather than phenol. The term (HOX)(H+)appears in aqueous iodination, bromination and chlorination, and the similarity between the ihree reactions is emphasized.
Painter and Soper’ have shown that the kinetics of the iodination of phenol are compatible with a reaction of either undissociated phenol and hypoiodous acid (1) or the phenoxide ion and positive iodine (2). dx/dt E ko(CeHsOH)(HOI) dx/dt = ko(CeH60-)(If)
two reactions. According to Painter and Soper,‘ the observed rate constants for the iodination of phenol can be expressed as the sum of the catalytic constants as shown in (3).
+ kc,t(CeH~,O-)(A~01) (1)
(3)
+ kw.t(CeHsO-)(AcOI)
(2) There is an “uncatalyzed” reaction4 whose rate In addition, there is for both cases a general acid- varies inversely with the hydrogen ion concentracatalyzed reaction of phenoxide ion and hypoiodous tion and an acid-catalyzed reaction whose rate is acid which the authors formulated as indicating re- inversely proportional to the square of the hydroaction between acyl hypoiodite and the phenoxide gen ion concentration. This relationship can be ion. A distinction between these two possibilities rewritten in the form (4). If expressed in this (1 and 2) could not be made on kinetic grounds, but kobsd = ko(HO-) f kcst (HO-)(base) (4) the authors preferred the latter of the two. The form, the equation becomes very similar to that present study was undertaken in order to add furfound for the aniline reaction (5), except that a hyther evidence in favor of a reaction involving the kobsd = ko k c a t (base) ( 5 ) phenoxide ion. Since the kinetics alone do not lead to a single answer, it seemed necessary to in- droxide ion catalysis is superimposed on both the troduce non-kinetic considerations, but even then, catalyzed and uncatalyzed reactions. The cataas will be seen, the results are not unambiguous. lytic form of the phenol reaction (4) is rather unuA possible distinction between the two mechanisms sual; while there are many reactions which show was suggested through the following considera- specific hydroxide ion and general base catalysis15 tions : the electronic theory demands that the order in the above case the hydroxide ions appear in both of activation by different substituents on the ben- terms. This kinetic form indicates the possibility zene ring toward an electrophilic reagent should be that the hydroxide ions serve a specific purpose, -0- > -NH2 > -OH.2 Since the kinetics of the not connected with the substitution process, iodination of aniline have recently been r e p ~ r t e d , ~namely the formation of the phenoxide ion. it should be possible to arrive a t a decision simply In order to obtain all the data required for the by establishing the relative order of the two reac- comparison some of the work of Painter and Soper tions. Because the actual species used in the iodin- had to be repeated under conditions identical with ations is the undissociated phenol, the rate for the those maintained in the aniline reaction. The phenoxide ion reaction must be calculated from the iodinations were carried out a t 25” with an 0.008 known concentration of the phenoxide ion a t every M phenol concentration and a 0.002 M concentra#H. There are, therefore, obtained two sets of rate tion of iodine. The iodide ion concentration, as constants, an observed rate constant (which is pos- in the reactions with aniline, was 0.12 M and the sibly a true rate constant for phenol) and one calcu- total ionic strength was 0.3 (NaC1). Almost all lated for the phenoxide ion, and these have to be reactions were carried out with phosphate buffers compared with the data for aniline. If the rate where both secondary and primary phosphate are constants calculated for the phenoxide ion reac- catalytically active, although the latter much less. tion should be faster than those for the aniline re- Five different buffer ratios were used (pH 5.71action, the evidence can be used as a further sup- 6.61). If the observed rate constants are plotted port for reaction (2), though not as a proof. If against the primary phosphate ion concentration, they are slower a reaction involving the phenoxide lines as shown in Fig. 1 are obtained. The obion is unlikely. If the reacting species is phenol, served rate constants can be expressed by equation the aniline reaction should be faster, except for the (6). If the intercepts at different pH’s are plotted unlikely possibility that phenol reacts faster than ko k’ay)(base’) + >(base”) k* (6) aniline because HOI is a more powerful iodinating kobsd = (H+) -k (H (H+) agent than positive iodine. The Catalytic Constants.-A complication arises against the inverse hydrogen ion concentration, the from the different catalytic characteristics of the line, whose slope is equal to ko, goes through the origin so that the phenol reaction apparently does not (1) €3. S. Painter and F. G. Soper, J . Chcm. Soc., 342 (1947); F. G.
+
Soper and G. F. Smith, ibid., 2757 (1927). (2) C. K. Ingold, Chcm. Reus., 1 6 , 225 (1934); Rec. fyau. chim., I S , 797 (1929). (3) E. Berliner, THISJOURNAL, 72, 4003 (1950).
(4) T h e reaction referred to in the following as the uncatalyzed reaction is actually an inverse hydrogen ion catalyzed reaction. (5) See for instance, L. P. Hammett, “Physical Organic Chemistry,” McGraw-Hill Book Company, Inc., New York, N. Y.,1940, p. 215 8.
430s
Vol. 7 3
ERNSTBERLINER
in compounds such as p-aminophenol are very doubtful and the results fortuitous.6 In order to obtain the rate constants for the iodination of the phenoxide ion one can use, for each experimental run, expression (8), where K is the kobad
(cd&oH)(Iz) = kcsnso-(CeH@-)(I2) and kCsHsO'
kobd
=-
Of+) (8)
K
dissociation constant of phenol. The observed rate constants for the phenoxide ion reaction can be expressed by an equation of the type (9). This rekobad
0 0.02 0.06 0.10 0.16 Concn. KHzPO,. Fig. 1.-The influence of buffers on the iodination of phenol. The lines correspond, from left to right, t o the following pH's: 6.61, 6.31, 6.13, 6 01, 5.71.
= ko
kat
(base)
(9)
action has the same catalytic characteristics as the aniline reaction, and the respective data can be compared, because in both reactions the catalytic constants are independent of the hydrogen ion concentration. From Fig. 3, in which the rate constants for the phenoxide ion reaction are plotted against the concentration of primary phosphate,
have a water reaction. It seemed preferable to plot k o b s ( H + ) instead of kobs (7). In that case all kobsd
(H') = ko f k ' d b a s e ' ) -k
k"mt
(base") (7)
the lines converge in one point (Fig. 2). The intercept is equal to ko and from the slopes of all the lines the two catalytic constants can be calculated by the method used for aniline. The catalytic constants thus obtained are ko = 3.34 f 0.06 X io-'; kNa,Hpo, = 22.5 x kKHIm, 0.7 x The corresponding constants for aniline are K O = 0.123; kNarHPO4 = 29.3; kKHIPO4 0.3. These two sets of data are clearly not comparable, because the aniline reaction lacks the inverse hydrogen ion catalysis (or direct hydroxide ion catalysis) in all terms. A comparison of the numerical data in terms of reactivities would therefore be quite meaningless and misleading. This shows the need for exercising caution in evaluating relative reactivities, even when the experimental conditions are the same. For instance, relative activating powers deduced from the direction of substitution
0
0.02
0.04 0.06 0.08 0.10 0.12 Concn. KH2PO4. Fig. 3.-The influence of buffers on the iodination of phenol: a plot of Rc,E,oagainst the concentration of KH?POI.
the rate constants are found to be ko = 3.06 f 0.05 x lo3; kNa*HPO, = 20.8 x lo4; kKH,p04 0.6 x lo4. Table I shows the observed rate constants and those calculated assuming equation (10). kobad
= 3 06 X IOa
+ 20.8 X 104[Na2HP0,]+ 0.6 X 10' [KHzPOII (10)
A comparison with the aniline reaction shows that the uncatalyzed reaction of phenoxide ion is 25,000 times faster than the uncatalyzed reaction of anih e , and the catalytic constant kNa,HPOI is 7,100 times greater than the corresponding constant for aniline. Not only is the assumption of phenoxide ion as the reacting species in agreement with the data, but the numerical differences are reasonable on chemical grounds. The charged phenoxide ion and the uncharged aniline should differ in reactivity by an a preciable factor. The nergies of Activation.-Another comparison is possible through the energies of activation, but here again two sets of data are obtained. A clear-cut decision cannot be made and it can only be decided whether or not the data are in agreement with the assumption of the phenoxide
E
o
0.06 0.08 0.10 0.12 Concn. KH2PO*. Fig. 2.-The influence of buffers on the iodination of phenol: a plot of kobsd. ( H f ) against thc concentration of KH2P04. 0.02
0.04
(6) W. Theilacker, B e y . , 71, 2065 (1938); W. Fuchs, Monolsh., 86, 331 (1917)
KINETICSOF THE IODINATION OF PHENOL
Sept., 1951 TABLE I THERATESOF
IODINATION OF PHENOL AT VARYING CONCENTR ATIONS~ kohid
KH;-
Concn. NatHPO4
pH Po4 6.13 0.024 0.008 .03 .01 .0375 .0125 ,015 ,045 ,0525 .0175 .02 .06 6.01 0.032 0.008 .048 .012 .064 ,016 .08 .02 6.01 0.032 ,048 .064 .08 5.61 0.05 .06
0.008
.lo
,012 .016 .02 0.005 ,006 ,008 .01
Concn. HAC
Concn. NaAc
.08
(H') kCsHsO(H+) x 107 ~ c ~ H ~xo io-; X lo7 (calcd.)* X (cslcd.) 5.31 5.33 4.89 4.87 5.80 5.32 5.32 5.80 5.86 5.87 6.41 6.39 6.42 6.45 7.00 7.03 7.64 7.56 6.94 7.01 7.55 7.58 8.23 8.26 4.93 4.92 5.38 5.36 5.87 5.84 6.38 6.40 6.77 6.77 7.39 7.38 8.40 8.35 7.66 7 70 kobsd
kohsd
0.719 .783 ,862 ,944 1.02 1.11 0.550 .655 ,755 .854
T = 35' 2.27 22.2 2.74 26.8 3.22 31.5 3.64 35.6 0.843 20.7 ,903 22.2 1.02 25.0 1.12 27.5
14.8 17.9 21.0 23.7 13.8 14.8 16.7 18.3
22.4d 26.9 31.4 35.9 20.6 22.0 25.0 27.9
14.9' 17.9 21.0 24.0 13.8 14.7 16.7 18.7
T = 25' 5.60 0.01 ,012 .014 .016
0.1 .12 .14 .16
1.53 1.70 1.88 2.05
0.665 .736 ,815 .888
1.54' 1.71 1.87 2.04
All rate constants are in liters moles-' min.-'. a Calculated from kohad (H+) = 3.34 x 10-7 22.5 x 10-6 [ N ~ ~ H P O I ] 0.7 X [KHsPOd]. Calculated from k,-.n.o- = 3.06 X 108 20.8 X lo4 lNazHPOal 0.6 X 10;-"[KHtP01]. * Calculated from kobsd 1.33 X 91.7 X 10" [NazHPOJ 5.4 X'YO+'6~$H~POJ. e Calculated from &cas0- = 8.88 X 103 60.7 X lo4 [NarHPOd] 3.7 X lo4 (KHzPOp]. 'Calculated from K C ~ H ~ O = 0.71 X lo4 8.3 X lo4 [CHsCOONa].
+
+
+
+
+
+
+
ature are those of Lundeq8 and from his data the K a t 35' (1.50 X 10-lo) was obtained by extrapolation. In order to use a consistent set of data Lunden's value of 1.09 X 10-lo was used for K a t 25'. With this value for K I , ko for the phenoxide ion reaction a t 35" is 8.88 f 0.06 X lo3 and the activation energy is 19.4 f 0.4 kcal. This is less by 4.5 kcal. than the value for aniline and constitutes again a reasonable difference between the phenoxide ion reaction and that of aniline. The values for the activation energies of the catalyzed reactions could not be found with the same accuracy as those for the uncatalyzed reactions. This is partly due to the uncertainty with which the catalytic constants for primary phosphate are determined,9 because they only constitute a few per cent. of the constants for secondary phosphate (the former being a much weaker base), and partly because the phenol reaction could not be studied over a wide enough range of $H on account of the great variations in reactivity. The data for secondary phosphate are recorded in Table I1 together with other relevant data. The energy of activation for the catalyzed reaction of phenoxide ion ( E N a a P O , ) is less than that for aniline, as expected, but the difference is rather small and not very significant. The difference is decidedly smaller than that for the uncatalyzed reactions for reasons which are not a t all clear. Included in the table are also some data obtained in acetate buffers. Here again the uncatalyzed phenoxide ion reaction proceeds with a lower activation energy than the uncatalyzed aniline reaction. The energies of activation as given in Table I1 are not true activation energies for the iodinations; they are composite values and include the heats of reaction involved in the formation of the iodinating species as well as the heats of dissociation of the buffers. However, a comparison of the activation energies of the phenoxide ion and the aniline reaction is meaningful if, as is assumed here, the iodinating species for the two reactions are identical. The Iodinating Agent,-The kinetic data so far presented make it seem probable that the iodination of phenol actually takes place on the phenoxide ion. The following argument is based on another analogy with the aniline reaction. The kinetics of the iodination of aniline are in agreement with a reaction between aniline and I + and a general base-catalyzed reaction between the same two species (11).3 The function of the
BUFFER value for
KI 0.12 M , p = 0.3,T = 26' Concn.
4309
+
+
ion reaction.' From the plot of kobd.(H+)a t 35' against primary phosphate a t five different buffer ratios (PH 6.01-5.61), similar to the one represented by Fig. 2 and equation (7) KO is calculated From this value and the to be 1.33 f 0.01 X value a t 25" the experimental energy of activation for the observed uncatalyzed reaction is 25.2 kcal. Aniline was iodinated a t 35' using four different phosphate buffer ratios (PH 7.23-5.61). At 35" ko for aniline is 0.455 f 0.005 from which in conjunction with k~ a t 25' the activation energy of the uncatalyzed aniline reaction is calculated to be 23.9 f 0.5 kcal. To calculate the energy of activation of the phenoxide reaction it is necessary to know the dissociation constant of phenol a t two different temperatures. The only data in the liter(7) There is a point to he considered when the reaction is conducted at two temperntures. The hydrogen ion concentration of a buffer is practicdly independent of the temprrature ( D . I. IIitchsock and A. C. Taylor, THISJ O U R N A L , 60, 2710 (1938); D. A. MacInnes, D. Belcher and T. Shedlovsky, ibid., Bo, 1094 (1938))but since Kw has a strong temperature dependence the hydroxide ion concentration changes and is almost doubled between 25 and 35'. The two equations (8) and (4) are therefore not equivalent any more, and the energy of activation for the observed resction is quite different (almost by a factor of two) depending on whether a hydroxide or inverse hydrogen ion catalysis is assumed to take place. Since in the present case both equilibria in question, namely, that of hypoiodous acid and that of phenol, involve stoichiometrically the hydrogen ions rather than the hydroxide ions the expression containing hydrogen ions was used in calculatinp E lor the observed reaction. This difficulty is absent in the phenoxide ion reaction.
dx/dt = ka( CaHsNH2)(I +)
+
kcat( CsHbNH*)(I +)(base)
(11)
base could be visualized as being concerned with the removal of the proton from the benzene ring. However, good arguments have been advanced tn the effect that the proton loss in aromatic substitution is kinetically insignificant and takes place after the rate-determining step.I0 If this also applies to iodination the base may not have t h i s function and a second possibility, kinetically indistin(8) H. Lunden. 2.physik. Chem , 70,249 (1910). (9) The recorded values for the catalytic constants of primary phosphate are very approximate (10) L. Melander, A d a Chem. Scand., S, 95 (1949); ArkivfGr K o n i , 3, 213 (1950); E. D.Hughes, C. K. Ingold and R. I. Reed, J . Chem. Soc., 2400 (1950).
4310
Vol 73
ERNSTBERLINER TABLE I1 THECATALYTIC CONSTANTS AND ENERGIES OF ACTIVATION OR koa6
Eo (kcal.)
ko35
C6HsOHb 3 . 3 4 f 0 . 0 6 X 10-7 CsHsO 3.06 f 0 . 0 5 X IO* CaHsrjH~ 0.123 f 0.002
1 . 3 3 f 0 . 0 1 X 10-6 8 . 8 8 i 0.06 X 103 0.455 f 0 . 0 0 5
THE
IODINATION OF PHENOL AND ANILINE'
kZSNe~HPOd
k3sNalHPOd ENazHPOi
ck~"
2 5 . 2 i 0 . 5 2 2 . 5 X 10-8 9 1 . 7 X 10-5 2 5 . 6 1 9 . 4 rt 0 . 4 2 0 . 8 X l o 4 6 0 . 7 X 10' 1 9 . 5 0.71 X 10' 23.9 i 0 . 5 29.3 88.8 2 0 . 2 0.121
ckn35
2 . 0 8 X 10: 0.435
19.6 23.3
a All rate constants are in liters moles-' minutes-'. These are the values for the observed reaction. The rate constants for the "uncatalyzed" and the catalyzed reactions are inversely proportional t o the hydrogen ion concentration. See equation ( 7 ) and footnote 4. In acetate buffers.
guishable from the first, involves a hydrogen ion or general acid-catalyzed reaction of aniline and hypoiodous acid3 (12). The complex of hypoiodous
C1) + , I 7 and the difficulty to distinguish between the two possibilities in some cases has recently been emphasized.I8 In the present work the reactions are first order with respect to aromatic compound, dx/dt = kn( Ca&NHa)( HOI)(H +) + probably because the equilibria leading to the formakot( GHsNH*)(HOI)(acid) (12) tion of hypoiodous acid as well as the breaking of acid and an acid can be represented as a hydrated the iodine-oxygen bond are easier energetically I+ or as the conjugate acid of hypoiodous acid than the breaking of the chlorine-oxygen bond, and (H20I) +, and in either case the function of the acid the former must be comparable with the energy would be to weaken the oxygen-iodine link either required for substitution. If chlorination proa t the moment of or prior to the attack on the ben- ceeds through C1+, bromination and certainly iodizene ring. This possibility seems a rather plaus- nation should be much more likely to proceed ible one, but when applied to the iodination of through the positive halogen itself,19 but the kinephenol it fits, kinetically, only a reaction involving tics of iodination (as of bromination) do not allow the phenoxide ion (13); the undissociated phenol that conclusion, although they do not rule it out. dxjdt = kn(C&O-)(HOI)(H+) + The iodination is at the present time best pictured Kcat(CeH,O- )(HOI )(HA ) (13 ) as involving either a preliminary formation of I+ cannot react with HOI by a reaction involving acid from HOI and H+ (or a general acid), and an attack catalysis.ll The second term in (13) is identical by the equilibrium I + on the aromatic compound, with that of Painter and Soper, except that the or an attack by the conjugate acid of HOI. The striking similarity between the kinetics of latter assumed the direct formation of an iodinating species between hypoiodous acid and the buffer chlorination, bromination and iodination speaks acid, which in the present interpretation does not very much in favor of mechanism (13) for which the phenoxide ion has to be assumed to be the reseem necessary. There have recently been described a number of acting species. That substitution takes place on halogenation reactions in water which show the phenoxide ion has recently been postulated even for kinetic form of reactions (12) and (13). Wilson nitration,20where the concentration of the phenoxand Soper12 have observed acid catalysis in the ide ion is much smaller than in the above experibromination of o-nitroanisole and of benzene with ments. I t was also quite convincingly established hypobromous acid, and Derbyshire and Waters13 for chlorination with hypochlorous acidz1 and for have noted the same in the bromination of w-tolu- diazo coupling.22 Experimental enesulfonic acid. The appearance of the term (HOBr)(H+), noted earlier by Schilov and NaniMaterials.-Two samples of phenol were tested. One aev, l4 has independently been interpreted by was obtained by hydrolysis of twice recrystallized phenylbenzoate (Eastman Kodak Co., m.p. 68.348.9') and two these authors as indicating bromination by either distillations, b.p. 180.2-180.4°(753 mm.). The other was Br+ or the conjugate acid of hypobromous acid. Merck Reagent Grade Phenol which was twice distilled a t As in the iodination, kinetically no distinction be- atmospheric pressure (b.p. 181.4' (763-764 mm.)); it was tween the two can be made. Furthermore, De La then heated for four hours at 50' over D r i d t e and for 30 minutes over calcium chloride, and distilled in vacuo. Mare, Hughes and Vernon16found that the kinetics Since both samples gave the same kinetic results in test runs, of chlorination with hypochlorous acid followed a the second sample was used in subsequent runs. The anisimilar expression. Under their experimental con- line was the sample used in previous experiments* and was ditions the formation of C1+ is more surely estab- redistilled in vacuo before use. All other reagents were best commercial reagent grade chemicals and were used without lished, because the reactions are of zeroth order further purification. with respect to the aromatic compound (if the latKinetic Runs.-All stock solutions were prepared as deter is sufficiently reactive) and the kinetically im- scribed before except that the phenol was made up so as to portant step, according to these authors, involves give a 0.008 M solution rather than 0.004 M . At the lower phenol concentration the rate constants had a tendency to the formation of C1+, similar to zeroth order nitra- rise during the later parts of the reaction which may have tion.I6 Under favorable conditions these authors been due to some precipitation of the product. These diffialso visualize a second order reaction with (H20- culties were not encountered with the higher concentrations (11) I t can, however, react in a hydrogen ion or acid-catalyzed reaction with hypoiodite ion, although this is unlikely on chemical grounds (12) W J Wilson and F G Soper, J Ckcm. Soc , 3376 (1949) (13) D H. Derbyshire and W A Waters, abrd , 564, 574 (1950) (14) E A Schilov and N P Naniaev. Compl r e i d acud Scr R s s , a 4 , 8 9 0 (1939) A , 3 4 , 4 0 6 2 (i94oj1 (15) De La Mare E D Hughes and C A Vernon, Rcscurch 3, 192
[c
u
(1950) (16) E I) Hughec C 11 I n g o l d
.tii
