Jou rnal of the American Chemical Society / 101:lO 1May 9,1979
2698
(1968). (13) The absence of a significant deuterium kinetic isotope effect for the [Rh~(C0)4CI*]-catalyzedrearrangement of 2 (Le., kza = kZb= 8.6 f 0.4 X 10-3 M-1 s-1 ‘In C6D6 at 50 “c)is consistent with this interpretation. (14) The equilibrium constant for the substitution reaction, (I/2)[Rh(NOR)Cll2 -k COD F= (1/2)[Rh(COD)C1]2 NOR, is approximately 1(J. Halpern and M. Takagi. unpublished results). (15) L. A. Paquette, R. S. Beckley, and W. E. Farnharn, J. Am. Chem. Soc., 87, 1089 (1975). (16) E. W. Abel, M. A. Bennett, and G. Wilkinson, J. Chem. SOC., 3178 (1959). (17) A. C. Cope, S. Moon, C. H. Park, and G. L. Woo, J. Am. Chem. Soc.. 84,
+
4865 (1962). (18) (a) M. Sakai, A. Diaz, and S. Winsteln, J. Am. Chem. Soc., 92,4452 (1970); (b) M. Sakai, personal communication. (19) T. A. Antkowiak, D. C. Sanders, G. E. Trimitsis, J. E. Press, and H. Schechter, J. Am. Chem. Soc., 94, 5366 (1972). (20) (a) E. E. van Tamelen, S. P. Pappas, and K. L. Kirk, J. Am. Chem. Soc.,92, 6092 (1971): (b) R. N. McDonald and C. E. Reinecke, J. Org. Chem.. 32, 1878 (1967); (c)M. J. Goldstein and M. S. Benzon, J. Am. Chem. Soc., 94, 5119 (1972). (21) M. Avram, I. G. Dinulesu, E. Marica, G. Mateescu, E. Sliam, and C. D. Nenitzescu, Chem. Ber., 97, 382 (1964). (22) P. Gassman and J. Marshall, Org. Syn., 48, 25 (1970).
Anomalous Rates of Proton Transfer to and from Nitrogen and Carbon in the Reactions of Amines with 1,l -Dinitro-2,2-diphenylethylene Claude F. Bernasconi* and David J. Carre Contribution from the Thimann Laboratories of the University of California, S a n t a Cruz, California 95064. Received September 5, 1978
Abstract: Two kinetic processes are observed in the reactions of 1 , l-dinitro-2,2-diphenylethylene with piperidine, morpholine,
n-butylamine, and aniline in 50% aqueous dimethyl sulfoxide. The first, relatively fast and reversible, refers to nucleophilic addition of the amine to the 2 position of l,l-dinitro-2,2-diphenylethylene to form a zwitterionic addition complex (T*), which is subsequently deprotonated to the anionic addition complex (T-). In the piperidine and n-butylamine reactions proton transfer is fast and nucleophilic attack is rate determining; in the morpholine and aniline reactions proton transfer is partially rate limiting. The second kinetic process, which is relatively slow, refers to protonation of T- on carbon to form TO, followed by rapid and irreversible cleavage of To into Ph2C=O and -CH(N02)2. The rate constants for the various elementary steps were evaluated. They show a dramatic steric effect not only on the nucleophilic attack step but also on the proton transfer rates. For example, deprotonation by N-methylmorpholine of T* derived from morpholine has a rate constant of 2 X lo3 M-’ s-I instead of the 108-109 M-l s-I expected for a diffusion-controlled reaction. Rates of protonation of T- on carbon by water and several general acids are as expected for similar dinitro compounds, but protonation by H+ is 103-104times faster than expected. This To. high rate is tentatively attributed to an intramolecular proton switch mechanism, Ti
-
This is the first paper in a series in which we will explore mechanistic features and structure-reactivity problems which not only relate to nucleophilic additions to olefins’ but which are relevant to other classes of reactions such as nucleophilic vinylic substitutions,2 ElcB elimination^,^ and proton transfer reactions involving “normal” acids4 as well as carbon acids.5 The present paper deals with the reactions of piperidine, morpholine, n-butylamine, and aniline with 1,l-dinitro-2,2diphenylethylene in 50% aqueous dimethyl sulfoxide (v/v). Two kinetic processes were observed. The first ( l / ~ ) )rela, tively fast and reversible, was studied most thoroughly and is consistent with eq 1: Ph
,NO,
Ph,
+
k,
I
/NO2
I
%NO2
:.-
RR’NH e Ph-C--C
Ph/c=c\N02
k-i
S
“FtR
+
T’ Ph k , p , base k-zp. acid
Ph-C-C
I I
/NO,
:.-
(1)
\NO,
NRR
T-
where k2p and k-2p refer to proton transfer and are defined as: k2p = k2pW k2poH[OH-] k2pA[RR’NH] k2pB[B] (2)
+
+
+
k-2p = k-2pH[H+]
+ k-2pW+ k-2pAH[RR’NH2+]
+ k-2pBH[BH+]
(3)
with k2pW,k2p0H, klpA,and k2pBbeing the rate coefficients for the deprotonation of T* by the solvent, hydroxide ion, amine, and another base (buffer), which sometimes was added to the reaction solution, respectively, and k-zpH, k - ~ ~ ~ , k-ZpAH, and k-2pBH being the rate coefficients for the protonation of T- on nitrogen by the hydronium ion, the solvent, and the conjugate acids of the amine and of B, respectively. The second process ( 1 / ~ 2 )which , is relatively slow and irreversible, refers to the cleavage of T- yielding benzophenone and dinitromethane anion. Equation 1 can be regarded as a model for the initial two steps in the mechanism of base-catalyzed nucleophilic vinylic substitutions by amines$’ in a “real” vinylic substitution reaction the substrate would have a nucleofugic leaving group (e.g., Ph(X)C=C(N02)2) and there would be productforming steps in which the leaving group departs from T- (and possibly from T*). A question of great relevance to the mechanism of base-catalyzed nucleophilicvinylic substitutions is whether proton transfer can always be assumed to be a rapid equilibrium step6a-d (kzp >> k-1 in eq 1) or whether proton transfer may become rate limiting (kzp > k-1). In this case nucleophilic attack is rate determining in the forward direction and eq 4 reduces to:
where Kaf is the acid dissociation constant of T*. Equation 5 describes the behavior of the piperidine reaction (Figure 2 ) , i.e., a series of parallel straight lines with slope = k l and intercept = k-I[H+]/K,. When ki[RR’NH] >> k-l[H+]/Ka*, the intercepts become vanishingly small and the straight lines merge into one as in the reaction of n-butylamine (Figure 3). From the intercepts, when measurable, only the ratio k - l / K , * can be obtained; in order to evaluate k-1 and Ka* separately, experiments at pH values close to or below pK,* were needed. Since at these low pH values the equilibrium of eq 1 lies virtually completely on the side of the reactants (kllk-1 = K1 small and amine is mainly present in protonated form), the experiments were performed as pH-jump experiments, as described above. The ensuing relaxation is essentially an irreversible breakdown of T-, with 1/71 given by:
The results of these pH-jump experiments for the piperidine and n-butylamine reactions are summarized in Table I. From plots (not shown) according to eq 7, k-1 and Ka* were obtained. In the piperidine reactions the ratio k-,/K,* = 1.66 X lo8 is virtually identical with k-l/K,* = 1.60 X IO8 obtained at high pH from the intercepts in Figure 2 , in support of our analysis. 71 =
l/k-i
+ K,’/k-i[H+]
(7)
Journal of the American Chemical Society / 101:10/ May 9,1979
2700 12
-
IO
-
08
-
Table I. Reactions of 1,l-Dinitro-2,2-diphenylethylenewith
Piperidine and n- Butvlamine.I DH-JumD . . ExDeriments“ . Piperidine
.
0 6 -
5.74d 5.75d 6.02d 6.04d 6.40d 6.44d 6.62d 6.70d
77.1 72.8 59.3 59.1 41.8 35.9 27.1 24.9
5.7 1 5.99‘ 6.3 1 e 6.36f 6.68f 7.09f 7.4o-f 8.00/
22.7 X 15.9 X 9.62 X 9.23 X IO-* 5.74 x 10-2 2.89 X 1.78 X IO-2 0.876 X
n-Butylamine
_‘c
tc pH 936 c pH 370 0
001
pH I064
002
0 03
[n-Bu“J,M
Figure 3.
~ 1 - l for
77.1 72.8 59.3 59.1 41.8 35.9 27.1 24.9
the reaction of l,l-dinitro-2,2-diphenylethylenewith
n-butylamine. With n-butylamine the situation is somewhat more complex. Plots (not shown) of 1/ T I vs. hydrogen ion concentration have a small but significant intercept which is inconsistent with eq 6. This intercept can be attributed to competing protonation of T- on carbon as shown in eq 8:
k-,p, acid
I
I
Ph-C-C-NO,
I
“ p = 0.5 M at 20 OC, KCI compensating electrolyte. Error limit f 3 % . Corrected for protonation on carbon, see textd pH maintained with acetic acid buffer, [bufferItot= 0.005 M. e pH maintained with acetic acid buffer, [bufferItot= 0.01 M. f pH maintained with cacodylic acid buffer, [bufferItot= 0.01 M. Table 11. Reaction of 1 ,l-Dinitro-2,2-diphenylethylene with n-
Butylamine; Cacodylic Acid Catalysis, pH-Jump Experiments” [BHCl, PH
buffer ratiob
6.91
3: 1 3: 1 3: 1 3:l 1:l 1:l 1:l 1:l
7.45
H
Ph
(8)
1
22.2 x 10-2 15.4 x 10-2 9.07 x 10-2 8.68 x 10-2 5.19 X 2.34 X 1.23 X loF2 0.326 x 10-2
M 0.0186 0.0375 0.0562 0.075 0.0125 0.025 0.0375 0.05
‘ p = 0.5 M (KCI), [RR’NHItot I 0.05 M. Cacodylic acid. Error limit f 4 % .
N R R NO?
1O2T1-l, S-1
5.55 7.55 9.72 11.4 2.34 3.85 5.33 5.55
[BH]/[B-1.
T” where k-3p
= k-3pH[H+]
+ k-3pW+ k-3pAH[RR’NH2+]
+ k-3pBH[BH+]
(9)
1/ T I then becomes:
-
Supporting evidence for the reaction TTo comes from two observations. ( I ) Data summarized in Table I1 show that for the n-butylamine reaction 1/71increases linearly with cacodylic acid concentration, apparently due to an increase in the k-3pBH[BH+] term of eq 9. In the piperidine reaction where k-i is much larger than in the n-butylamine reaction but k-3p is expected to be similar, the k-)pKa*/(Ka* + [H+]) term in eq 10 is expected to be negligible. In fact, buffer catalysis with acetic acid was hardly detectable, and the plot of 1/71vs. hydrogen ion concentration has no significant intercept. (2) At high pH the reaction TTo,due to the fact that eq 1 is no longer “irreversible”, gives rise to a separate relaxation process which was observed with all amines (72; see under Kinetics of Cleavage Reaction). In evaluating k-1 and Ka* via eq 7 from the data in Table I, we have corrected for the k-3pKa*/(Ka* [H+]) term in the n-butylamine reaction by subtracting the intercept of the
-
+
plot of 1/71 vs. [H+] from all 1/71values which yields 1 / rI(corr). This procedure is not quite rigorous, since k-3p is pH dependent, but in view of the smallness of the correction this is of no consequence. From plots of 1/ T 1 vs. [BH+],a value for k-3pBH for cacodylic acid was also obtained. B. Rate-Limiting Proton Transfer (kzp k - ~ ~as~ ) , will be justified in the Discussion; thus eq 4 becomes: 1- k l ( k 2 p w k2pA[RR’NH])[RR’NH] 71 k-1 k2pA[RR’NH] k-l(k-zPH[H+] k-2pAH[RR’NH21) (11) k-1 k2pA[RR’NH] For very low amine concentrations we have k2pA[RR’NH] > k - I , eq 1 1 reduces to eq 5, and the curves merge into a set of parallel lines with slope = k l . Interestingly, at pH 9.2 (highest pH used) all points appear to lie on a straight line with a slope of k l . Since k2,OH[0H-] > k-lpW),but not necessarily k2pW> k-1; the plateau value corresponds, thermore, except at pH 4.85 where aniline acts as its own within experimental error, to 1/71 calculated according to eq buffer, a buffer was added in small concentration (k2pB[B] k-1, has occurred even at the highest amine concentrations used. This is confirmed by the fact that increasing acetate buffer concentration enhances 1/71 not only at low (Table IV, entry A) but also at high aniline concentrations (Table IV, entry B). This shows that proton transfer is rate limiting, or at least partially so, under all conditions employed. The rate constants for some of the elementary steps were estimated as follows. At the buffer concentrations used in the experiments of Figure 5 (0.02 M total buffer concentration), the kz$[B] and k-2pBH[BH+]terms in eq 17 can, to a first approximation, be neglected; this is borne out by the data in Table IV which show the buffer effect to be small at these concentrations. Hence the intercepts in Figure 5 are approximated by:
+
intercept
-- k- I k-2pH[ H+] k-, + k2pW
+
from which we can obtain k-1k-2,~/(k-l k2pW),Since the evidence presented above shows that proton transfer is significantly rate limiting under all conditions, kzpWcannot be much larger than k-1 and probably we have k2pW5 k-1. Thus assuming kZpWto be negligible in eq 19 and equating the intercepts with k-zpH[H+] will not introduce an error of more than a factor of two in our estimate of k-2pH. The pKaf is estimated from the pKa* of the n-butylamine adduct and assuming pKaAH(aniline) - pKaf(aniline) = pKaAH(n-BuNH2) - pKai(n-BuNH2), as for the piperidine-morpholine pair. This then allows us to obtain kzpW= Kaf k-zpH. The approximation kzpW> k3p which reduces eq 23 to: 1 7 2 (adj) = k-3p = k-3pH[H+]
+
+
+ k-3pW+ k-jPAH[RR’NH2+]
(24)
The data permit calculation of the rate constants k-3pH, k-3pW, and k-3pAH for protonation on carbon by the hydronium ion, the solvent, and by morpholinium ion, respectively. In the aniline reaction, plots (not shown) of l / ~ ( a d j vs. ) cacodylic acid concentration also form a series of parallel straight lines, again demonstrating general acid catalysis. Here k-3p contains an additional term, k-3 BH[BH],for cacodylic acid (eq 9). On the other hand k-3p [H+] and k-3pWcontribute too little to k-3p to be measurable; this is also the case in the n- butylamine and piperidine reactions where only k-3pAH could be determined. The various rate constants for protonation on carbon are summarized in Table IX.
A
Discussion Rate coefficients for the various elementary reactions are summarized in Tables VII-IX. Nucleophilic Attack (k1, k-1). The order of nucleophilic reactivity (kl) is n-BuNH2 > piperidine 2 aniline > morpholine (Table VII). This is a very unusual order; commonly one finds piperidine >> n-BuNH2, and piperidine >> aniline
Bernasconi, C a r d
/ Reactions of Amines with 1,I -Dinitro-2,2-diphenylethylene
Table 111. Reaction of 1,l -Dinitro-2,2-diphenylethylenewith Morpholine; Buffer Catalysis" [ba~e],~ M
PH
Table IV. Reaction of l11-Dinitro-2,2-diphenylethylenewith Aniline: Buffer Catalvsis' DH
A. Acetate at [Aniline] = 0.02M
A. p-Cyanophenoxide at [ M ~ r p h o l i n e=] ~0.2M
9.43
17.7 18.0 18.9 17.7 17.8 17.1 B. p-Cyanophenoxide at [ Morpholine] = 0.01M 8.45 0 4.60 0.005 8.38 0.01 10.2 0.025 12.5 0.05 12.1 C. Dabco at [Morpholine] = 0.01M 8.37 0.0074 6.37 0.0185 8.02 0.037 8.94 0.056 9.22 10.0 0.074 0.925 9.29 D. N-Methylmorpholine at [ M ~ r p h o l i n e= ] ~0.01M 8.44 0.05 5.71 0.10 5.49 5.53 0.15 0.20 5.88 0.25 6.61 0 0.01 0.02 0.04 0.06 0.08
~
~~
8.13d
Free base.
B. Acetate at [AnilineJC= 0.2M 0.05
0.10 0.15 0.20 0.25
4.31 4.84 5.39 6.35 7.13
At 20 OC, ~r= 0.5 M (KCI). Free base. Free amine. pH maintained by Dabco buffer, [bufferItot= 0.01M. e pH maintained by acetate buffer. f Error limit f3%.
Table V. Cleavage Reaction with Morpholine, Piperidine, and nButylamineu
A. Morpholine
8.41
Error limit f3%.
for nucleophilic attack on a variety of ele~trophi1es.l~ The observed order is undoubtedly due to steric hindrance by the two phenyl groups in S ; this is not surprising since amine addition even to unhindered olefins is known to be quite sensitive to steric effects in the n ~ c l e o p h i l e . ' ~ ~ The steric interpretation is consistent with three additional observations. (a) Nucleophilic attack by the same amines on the less hindered P-nitrostyrene, to form 1, follows the normal
8.72X 10.7X 11.0x 10-2 12.9X lod2 13.9X 13.9X
0 0.05 0.10 0.15 0.20 0.25
6.08'
~~~
At 20 "C, ~r= 0.5 M (KCI) Free amine. (I
q-'j s-I
[AcO-1, M
102r'-',' s-I
2703
8.71
9.41
0.01 0.02 0.03 0.04 0.01 0.02 0.03 0.04
0.02 0.04 0.06 0.08 0.01 0.02 0.03 0.04
0.05
0.05
0.06 0.02 0.06 0.10
0.06 0.004 0.012 0.02
0.815 2.30 3.80 5.56 0.78 1.78 2.86 4.19 5.22 6.37 1.02 2.29 3.04
8.70 13.4 16.0 19.8 4.58 6.12 7.46 9.25 10.3 11.5 1.52 2.66 3.36
B. Piperidine
0.02 0.04 0.06 0.08
11.03
0.02 0.04 0.06 0.08
9.92
0.02 0.03 0.04
0.10
0.05
0.05
0.10
0.10 0.15 0.20
H
0.762 1.08 1.42 1.66
0.762 1.08 1.42 1.66
1.33 2.03 2.75 0.72 1.48 2.34 3.18
1.33 2.03 2.75 0.72 1.48 2.34 3.18
C. n-Butylamine
+ 1
reactivity order. (b) Nucleophilic attack on P-nitrostyrene is about as fast as attack on S with n-butylamine and aniline, and about 200-fold faster than on S with the bulkier piperidine and morpholine,' despite the smaller electronic activation. (c) l,l-Dinitro-2-phenylethylene(one phenyl group only) is so much more reactive than S that it hydrolyzes virtually instantaneously in aqueous solution, precluding a kinetic study. The rather large k-IPip/k-ln-BuNH2 ratio of 280 is also a consequence of the steric effect; the analogous ratio for P-nitrostyrene is 29,' I that for 1,3,5-trinitrobenzene (Meisenheimer complexes) is Based on a comparison between piperidine and morpholine which have the same steric requirement, we have calculated P n u c = 0.37(k1) and PlS = -0.61(k-1)l9; the Pnucvalues compare well with those of other electrophiles.' Acidity of Tf (pK,fl. It is quite remarkable that, despite its negative charge, the C(N02)2 moiety is so strongly electron withdrawing that T* is almost 5 pK, units more acidic than
10.65
0.15
0.20
" At 20 "C,
0.15
0.20
= 0.5M (KCI).
its parent RR'NH2+ (Table VII). In comparison, the acidifying effect of the -dHN02 group in 1 lowers the pK, of 1 by 2-2.5 units relative to RR'NH2+,I1 which is similar to the effect of the C6H2(N02)3- group in Meisenheimer complexes formed from amines and 1,3,5-trinitroben~ene.~* Proton Transfer between T* and T-. One of the most interesting findings of this study is that many of the rate constants for proton transfer between T* and T- are substantially lower than expected for normal acids and bases.4 This effect is particularly pronounced in the morpholine reaction; deprotonation of Ti by morpholine, N-methylmorpholine, Dabco, and p-cyanophenoxide ion is thermodynamically fa-
Journal of the American Chemical Society/ 1Ol:lO / May 9, 1979
2104 Table VI. Cleavage Reaction with Anilinea [RR'NH], DH
~
7.46
7.02
7.50
8.07
[RR'NHz'], M
M
[B-I, M
IBHl,b
0.095 0.095 0.095 0.095 0.00 125 0.0025 0.0038 0.005 0.005 0.01 0.015 0.02 0.008 0.016 0.024 0.032
0.105 0.105 0.105 0.105 0.00375 0.0075 0.0114 0.015 0.005 0.01 0.015 0.02 0.0021 0.0042 0.064 0.085
102r2-',(adj),
10272-',
M
S-1
S-
~~~~~~~~
3.08 x 6.17 x 9.25 x 1.23 x 1.70 x 1.70 x 1.70 x 1.70 x 5.62 x 5.62 x 5.62 x 5.62 x 1.51 x 1.51 x 1.51 x 1.51 x
0.05 0.10 0.15 0.20 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10
10-5 10-5 10-5
10-4 10-4 10-4 10-4 10-4 10-5 10-5 10-5 10-5 10-5 10-5 10-5 10-5
1.81 2.72 3.29 3.73 0.843 1.37 1.87 2.42 1.43 2.53 3.74 4.62 0.835 1.48 2.18 2.78
3.76 4.19 4.47 4.73 2.10 3.41 4.66 6.03 2.13 3.77 5.57 6.88 0.944 1.67 2.46 3.14
At 20 OC, p = 0.5 M (KCI). bBH = cacodylic acid Table VII. Rate and Equilibrium Constants for the Reactions of Amines with 1,l-Dinitro-2,2-diphenylethylenein 5096 Me2SO-SOYo Water at 20 piperidine k 1 , M-' S-' k - i , s-' Ki = kiIk-1, M-I pKa* pKaAHd K 1 Ka*'
6.8 100 6.8 X 6.22 11.oo 4.11 X
morpholine
n - butylamine
0.95 2.4 x 103 4 x 10-4 -3.946 8.72 4.58 X lo-*
40 0.36 1.1 x 102 5.9 1 10.60 1.35 x 10-4
aniline -1
-5 x 106 -2 x 10-7 --0.5' 4.25 6.32 x 10-7'
-
0.5, KC1 compensating electrolyte. Estimated as pKaAH 4.78 where 4.78 = pKaAH- pK,* for piperidine. Esimated as pKaAH dpKaAHdetermined potentiometrically. e KIK,* is a measure of the stability of Trelative to S. Note that for piperidine and morpholine they are about the same, which means that for a given amine concentration and pH the same percentage of S is converted into T-, leading to roughly the same OD, as observed.f Independent spectrophotometric measurement yielded K'K,* = 6.43 x 10-7 0
/.L =
- 4.79 where 4.79 = pKaAH- pKa* for n-butylamine.
Table VIII. Rate Coefficients for Proton Transfers between Ti and T- in the Reaction of 1,l-Dinitro-2,2-diphenylethylenewith Morpholine and Aniline in 5Wo Me2SO-50% Water at 20 OC" morpholine (catalyst) kzpW,s-' k ~ ~ ~ / [ H 2 M-' 0 ] , s-' k-2 ', M-I S-' k2,A],M-' s - I k-2 A H M-' S-' k2$, M-I s-l k 4 p B H M-I , s-'
4.2 X IO2 (H20) 1.52 X I O (H20) 4.2 X lo6 (H+) 3.8 X lo4 (morph) 0.63 (morph-H+) 2 X lo3 (N-methylmorph) 1.1 X lo4 (Dabco) 5 x 10' (p-cNCtjH40-) 0.48 (N-methylmorph-H+) 0.15 (Dabco-H+) 8.5 (p-CNC6HdOH)
ApKb
aniline (catalyst)
ApKb
-5.38 -5.38 5.38 4.78 -4.78 3.62 4.86 4.76 -3.62 -4.86 -4.76
-9 X lo6 (H20) -3.3 x 105 ( H ~ o ) -3 X lo6 (H+) -6 X IO6 (PhNH2) -100 (PhKH3+) -7 x 107 ( A ~ o - )
-0.94 -0.94 0.94 4.79 -4.79 6.24
-40 (AcOH)
-6.24
/.L = 0.5 M (KCI). ApK = pK, (acceptor) - pK, (donor), e.g., for k2pA:ApK = pKaAH- pK,*; for k-ZpAH: ApK = pK,* - pKaAH, ctc.; pKdAHand pKa* from Table VI; pKaBHdetermined potentiometrically: 7.56 (N-methylmorpholine), 8.80 (Dabco), 8.70 @-cyanophenol), 5.74 (acetic acid); pK, = 15.9, determined from pH measurements in dilute KOH solution; pK,(H30+) = -log [ H 2 0 ] = -1.44, pK,(H20) = pK, + log [ H 2 0 ] = 17.34; ApK > 0 means thermodynamically favored reaction.
vored by several pK units a n d should approach diffusion control with rate constants in the order of lo8 to lo9 M-' s-l? but in fact the highest r a t e constant is only 5 X lo6 M-' s-l 0cyanophenoxide ion) a n d the one for N-methylmorpholine is as low a s 2 X lo3 M-' s-l. These dramatic rate depressions a r e another manifestation of the great steric bulkiness of Ti which, a s c a n be clearly seen with molecular models, makes the proton very inaccessible. T h e fact that the rate depression becomes stronger with increasing bulkiness of the base (N-methylmorpholine is the bulkiest, p-cyanophenoxide ion t h e least bulky) is further evidence for a steric effect, a n d so is t h e observation that in the case of the
less bulky 1 comparable deprotonation r a t e constants have normal values in the order of lo9 M-' s-l.ll T h e steric effect also retards the protonation of t h e morpholine T- adduct by t h e hydronium ion (k-zpH) by at least three orders of magnitude.22 In t h e aniline reaction t h e rate depressions a r e much less dramatic, apparently because T- is a secondary rather t h a n a tertiary amine which reduces t h e steric effect. For example, for the same ApK, deprotonation of t h e aniline T* adduct by aniline is 150 times (or 12 times) faster than the deprotonation of the morpholine T* adduct by morpholine (orp-cyanophenoxide ion). Based on these results one would also expect t h e
Bernasconi, Carrd
/ Reactions of Amines with 1,l -Dinitro-2,2-diphenylethylene
protonation of T- by the hydronium ion to be substantially faster for aniline compared to morpholine. We find however that k-zpH is about the same for both amines. This is because ApK is quite small in the aniline reaction which depresses k-2,H.4 Steric retardations of proton transfers between normal general acids and bases have been observed in other syst e m ~although , ~ they ~ ~are~usually ~ ~ less~ dramatic.25Reports on steric retardation of the protonation of a normal base by the hydronium ion are practically nonexistent; we have shown recently that the protonation of 2,6-di-tert-butylpyridine by the hydronium ion is about 50 to 70 times slower than that of pyridine.28 Alternative Interpretation for Slow Proton Transfer? A referee has suggested that intramolecular hydrogen bonding between amino nitrogen and nitro oxygen in T* might be a contributing factor in the slowness of the proton transfer. Intramolecular hydrogen bonding is in fact known to strongly reduce proton transfer rates,4,27,29but the following points show that this cannot be a major factor in our system. (1) The deprotonation of the morpholine-Ti becomes progressively slower with increasing bulkiness of the base; this cannot be explained by hydrogen bonding but is consistent with the steric interpretation. (2) Since the aniline-T* is significantly more acidic than the morpholine-T*, hydrogen bonding should be stronger in the aniline case and should lead to a larger rate reduction than in the morpholine case. This is opposite to the observed behavior. (3) Intramolecular hydrogen bonding, if it occurs a t all, should be stronger in 1 than in the T* adducts because the negative charge is delocalized into one nitro group only, making it a better hydrogen bond acceptor. Hence one would expect even more dramatic rate reductions in the deprotonation of l. We found however that the rates are normal" in the Eigen4 sense. (4) The large rate reduction in the reaction T- H+ T* cannot possibly be explained by intramolecular hydrogen bonding since it is a protonation rather than a deprotonation. Estimate of Experimentally Inaccessible Proton Transfer Rates. k2pA and k-zpAH in the piperidine reaction must be virtually identical with those in the morpholine reaction (3.8 X lo4 and 0.63 M-I s-l, respectively), since the steric situation and ApK are the same in both systems. k2pAand k-zPAH in the n-butylamine reaction must be similar to those in the aniline reaction (same ApK) or perhaps slightly higher due to a somewhat smaller steric effect (estimated k2,A = 1-5 X lo7 M-' s-I, k-2,AH = 2-10 X lo2 M-I s-l). k-2pH in the piperidine reaction should be comparable to that for the morpholine reaction (same steric effect) or perhaps slightly higher because of a larger pK (estimated k-2pH = 0.5-2 X lo7 M-' s-l). k-2pH in the n-butylamine reaction will be substantially higher than in the aniline reaction because it is not depressed due to a small ApK, and because of a somewhat reduced steric effect (estimated k-2pH = 10s-109 M-' s-'). k2,0H is not accessible experimentally in any of the reactions. Taking into account the steric effect, we estimate it to be 106-107M-' s-l in the piperidine and morpholine reactions, 0.5-2 X lo9 M-I s-l in the aniline reaction, and 1-3 X lo9 in the n-butylamine reaction. From this we can now see that our original assumptions that, for the morpholine and n-butylamine reactions, k2,0H[OH-] _ 3 X loQ h4-ls-l. (23)G. J. Janz and R. P. T. Tomkins, "Nonaqueous Electrolyte Handbook", Voi. 1, Academic Press, New York, 1972,p 86. (24)(a) J. Hine and J. Mulders, J. Org. Chem., 32, 2200 (1967);(b) M. I. Page and W. P. Jencks, J. Am. Chem. Soc.,94,8828(1972);(c) J. M. Sayer and W. P. Jencks, ibid., 95,5637 (1973). (25)An exception is the extremely slow proton transfer involvingcertain proton cryptates,2Balthou h part of the rate retardation may be due to internal hydrogen bonding. (26)J. Cheney and J. M. Lehn, J. Chem. SOC.,Chem. Commun., 487 (1972). (27)A. J. Kresge, Acc. Chem. Res., 8, 354 (1975). (28)C. F. Bernasconi and D. J. Carre, J. Am. Chem. SOC..following paper in this issue. (29)F. Hibbert and A. Awwal, J. Chem. SOC., Chem. Commun., 995 (1976). (30) The problem is the same as in SNAr reactions and has been extensively discus~ed.~~,~~ (31)C. F. Bernasconi. R. H. de Rossi, and P. Schmid, J. Am. Chem. SOC.,99,
8
4090 (1977). (32)R. P. Bell and R. L. Tranter, Proc. R. SOC. London, Ser. A, 337, 518 (1974). (33)V. N. Dronov and I. V. Tselinsky, Reacts. Spsobnost. Org. Soedin, 7,263 (1970). (34)C. D. Ritchie and R. E. Uschold, J. Am. Chem. SOC.,90,3415 (1968). (35)J.-C. Halle, R. Gaboriaud, and R. Schaal, Bull. SOC. Chim. Fr., 2047 (1970).
Rate of Protonation of 2,6-Di-tert -butylpyridine by the Hydronium Ion. Steric Hindrance to Proton Transfer Claude F. Bernasconi* and David J. Carr6 Contributionfrom the Thimann Laboratories of the University of California, Santa Cruz, California 95064. Received November 3, I978
Abstract: The rate of protonation of 2,6-di-tert-butylpyridineby the hydronium ion was measured by the temperature-jump method in 20% dioxane-80% water (v/v) at 25 O C . The rate constant is 3.7 f 0.8 X lo8 M-' s-I, which is about 50 to 70 times lower than expected for a diffusion-controlled reaction. The rate reduction is attributed to a steric effect and represents one of the first cases where steric hindrance is seen to affect the rate of protonation of a nitrogen base by the hydronium ion. The results support the notion that the abnormally low basicity of 2,6-di-tert-butylpyridineis due to steric hindrance of solvation rather than due to steric compression of the N-H bond in the protonated base.
Most chemists know by now that thermodynamically favored proton transfers between oxygen or nitrogen acids and bases are usually diffusion controlled or nearly so.' In particular, the reaction of an amine with the hydronium ion in aqueous solution (eq 1) has kl values in the order of 1O'O M-I s-l or slightly higher. Table I summarizes some typical rate constants.
I -N:+H'
I
k,
k-t
-N"
I I
(1)
I n a recent study of the nucleophilic addition of amines to I , 1-dinitro-2,2-diphenylethylene, we found that protonation of the morpholine adduct T- on nitrogen by H+ in 50% aqueous dimethyl sulfoxide (v/v) (eq 2) has a kl = 4.2X lo6 M-' s - ' . This ~ is nearly lo4 times lower than typical kl values for tertiary amines in water (Table I) and was attributed to a steric e f f e ~ t . ~ Even though there are numerous precedents of steric hindrance to proton transfer from a general acid to a nitrogen base, or from a protonated amine to hydroxide ion or a general 0002-7863/79/150l-2707$01 .OO/O
T-
' T
b a ~ e , ' ~we. ~could ? ~ not find any report in the literature about steric hindrance to protonation of a nitrogen base by the hydronium ion. It therefore appeared desirable to look for examples other than reaction 2 where this phenomenon would be observable.6 In view of the rather large steric effect of the two tert-butyl groups on the pK, of di-tert-butylpyridine,'.* we expected that steric hindrance might significantly lower kl in reaction 3. We
I
H+ 0 1979 American Chemical Society
