Buncel
132
Accounts of Chemical Research
Catalysis in Strongly Acidic Media and the Wallach Rearrangement Erwin Buncel Department of Chemistry, Queen’s University, Kingston, Ontario, Canada, K7L 3N6 Received September 16,1974
The study of acid-catalyzed reactions is important since the proton transfer process is fundamental to many areas of chemi~try.l-~ The experimental observation that in a number of reacting systems the initial proton transfer can trigger an array of reaction sequences is particularly challenging. In such cases the investigator is faced with unravelling both quantitative aspects of the initial proton transfer step and a fairly complex series of processes involving transient reaction intermediates (proven as well as hypothetical) and unknown transition states. Acid-catalyzed molecular rearrangements have provided fertile territory in the development of concepts which have led to significant advances in our understanding of mechanisms; it is not surprising that they have received extensive study. Of the numerous known acid-catalyzed rearrangements, of special interest are the Beckmann rearrangement of oximes,* the Schmidt rearrangement,5 the nitramine rearrangement,6 the Fisher-Hepp rearrangement,7 the Bamberger rearrangement (eq 1),8 and the benzidine rearrangement (eq 2).9 A common link for these transformations is that they all involve, in a formal sense, electron-deficient nitrogen centers. H
(+
other products)
,OH
/ H2N
Although the main features of most of these rearrangements have likely been established, none is understood in its entirety. Also, novel experimental approaches periodically bring new perspective to our views of the mechanisms of the transformations. The benzidine rearrangement, however, is unmatched in defying our capability to resolve satisfactorily what appears at first sight to be a relatively simple process. C. K. I n g ~ l d as , ~one ~ of its chief investigators, was Erwin Buncel was born in Czechoslovakia and educated in England, receiving the Ph.D. at University College, University of London, in 1957. Postdoctoral work at the University of North Carolina and at McMaster University was followed by a period in industry with the American Cyanamid Company. In 1962 he joined Queen’s University as Assistant Professor and is now Professor of Chemistry. Dr. Buncei’s research interests cover a number of aspects of physical organic chemistry, especially catalysis in strongly acidic and strongly basic media, deuterium exchange, and kinetic isotope effects.
prompted to show a cloud haloing the potential energy-reaction coordinates diagram, stating that “a cloud hangs over the high central region of the mechanistic route, wherein lies the transition state.” It is noteworthy that the benzidine rearrangement is an intramolecular process. Also, unlike the majority of acid-catalyzed transformations, it is characterized by the requirement of two proton transfers in one of its mechanistic pathways. It will be shown in this Account that the Wallach rearrangement of azoxyarenes (e.g., eq 3)10 also in0-
volves two proton transfers. However, whereas the benzidine rearrangement normally occurs quite readily in the dilute acid region (e.g., 0.05-1 M HC104), the Wallach rearrangement requires the use of moderately concentrated acid (Le., 60-100% HZS04). The latter characteristic is of particular interest in the present context. The reaction medium in the Wallach rearrangement system can by no means be considered as ideally behaved, whereas the medium in the benzidine rearrangement does approximate such to a close degree. A brief overview of the sulfuric acid system is needed in order that the kinetic studies of the Wallach rearrangement can be placed in proper perspective, particularly with respect to the possibility of ob(1) R. P. Bell, “The Proton in Chemistry,” 2nd ed, Cornell University Press, Ithaca, N.Y., 1973. ( 2 ) W. P. Jencks, “Catalysis in Chemistry and Enzymology,” McGrawHill. New York., N.Y.., 1969. ( 3 ) M. L. Bender, “Mechanisms of Homogeneous Catalysis from Protons to Proteins,” Wiley, New York, N.Y., 1971. (4) B. J. Gregory, R. B. Moodie, and K. Schofield, J . Chem. Soc. B, 338 (1970). (5) P. A. S. Smith in “Molecular Rearrangements,” Part I, P. de Mayo, Ed., Interscience, New York, N.Y., 1963, Chapter 8. (6) (a) D. V. Banthorpe and J. A. Thomas, J . Chem. Soc., 7149, 7158 (1965); (b) W. N. White, D. Lazdins, and H. S. White, J . Am. Chem. Soc., 86, 1517 (1964).
(7) T. D. B. Morgan and D. L. H. Williams, J . Chem. Soc., Perkin Trans. 2, 74 (1972). (8) H. J. Shine, “Aromatic Rearrangements,” Elsevier, Amsterdam, 1967, pp 182-190. (9) (a) C. K. Ingold, Chem. Soc., Spec. Publ., No. 16,118 (1962); (h) M. J. S. Dewar and A. P. Marchand, Annu. Reu. Phys. Chem., 16, 321 (1965); ( c ) H. J. Shine in “Mechanisms of Molecular Migrations,” yol. 2, B. S. Thyagarajan, Ed., Wiley, New York, N.Y., 1969, pp 191-247; (d) D. V. Banthorpe, Top. Carbocycl. Chem., 1, 1 (1969); (e) R. A. Cox and E. Buncel in “The Chemistry of the Hydrazo, Azo and Azoxy Groups,” S. Patai, Ed., Interscience, New York, N.Y., 1975, pp 775-859. (10) E. Buncel in “Mechanisms of Molecular Migrations,” Vol. 1, B. 9. Thyagarajan, Ed., Wiley, New York, N.Y., 1968, pp 61-119.
Vol. 8,1975
The Wallach Rearrangement
servation of general acid catalysis. The detection of general acid catalysis1 in moderately concentrated acid media is characterized by inherent difficulties that are not encountered with studies in weak acids (vide infra).
General Acid Catalysis in the Sulfuric Acid System The sulfuric acid system contains a complex milieu of species in equilibrium.ll The nature and concentration of the species present are dependent on the acid region under consideration (Figure l).llj In the dilute acid region the important species (besides HzO) are H(HzO)~+,HS04-, and, to lesser degree, s04’-. The actual hydronium ion, H30+, reaches maximum concentration at ca. 85% HzS04. In the region 85-100% H2S04, the H30+, HS04- and S042species (as well as HzO) rapidly decrease in concentration. Simultaneously, undissociated H2SO4 increases in concentration and becomes the principal species present. In the 100% HzSO4 region and through the dilute oleum region the acidic species present are H2S04, H2S2O7, SOa, and H3S04+, while the basic species are HS04-, HS207- (and HzS04). A number of interesting relationships have been put forth relating various properties of the sulfuric acid system to species concentrations or activities, including acidity function correlations.11 It will be seen from Figure 1 that evaluation of kinetic data in the concentrated sulfuric acid region could, in principle, afford differentiation between specific catalysis by the hydronium ion and general Br6nsted acid catalysis, the latter involving one or more of the species HzS04, H3S04+, and HzS207. The availability of kinetic data for the Wallach rearrangement over a wide range of acidity has allowed evaluation of rate correlations involving species concentrations, activities, and acidity functions, and has led to some general proposals concerning catalysis in moderately concentrated acid media. These proposals are complementary to other criteria which have been advanced in recent years for catalysis in strongly acid media. l2 The Two-Proton Process in the Wallach Rearrangement of Azoxybenzene Although the transformation of azoxybenzene to p-hydroxyazobenzene was first observed nearly 100 years ago,l3 mechanistic studies date back only about 15 years. Since that time, hypotheses concerning the (11) (a) L. P. Hammett, “Physical Organic Chemistry,” 2nd ed, McGrawHill, New York, N.Y., 1970, Chapter 9 (b) C. H. Rochester, “Acidity Functions,” Academic Press, New York, N.Y., 1970; (c) M. Liler, “Reaction Mechanisms in Sulfuric Acid,” Academic Press, New York, N.Y., 1971; (d) N. C. Den0 and R. W. Taft, J. Am. Chem. SOC.,76,244 (1954); (e) P. A. H. Wyatt, Trans. Faraday Soc., 56, 490 (1960); (f) E. Hogfeldt, Acta Chem. Scand., 14,1597,1627 (1960); (9) E. B. Robertson and H. B. Dunford, J. Am. Chem. Soc., 86,5080 (1964); (h) R. J. Gillespie and E. A. Robinson, in “Non-
Aqueous Solvent Systems,” T. C. Waddington, Ed., Academic Press, London 1965, Chapter 4; (i) C. W. F. Kort and H. Cerfontain, Reel. Trau. Chim. Pays-Bas, 87, 24 (1968); 88, 1298 (1969); (i) R. A. Cox, J. Am. Chem. SOC., 96,1059 (1974). ( 1 2 ) (a) F. A. Long and M. A. Paul, Chem. Reu., 57,935 (1957); (b) J. F. Bunnett, J. Am. Chem. SOC.,83,4956,4968,4973,4978 (1961); (c) J. F. Bunnett and F. P. Olsen, Can. J. Chem., 44,1917 (1966); (d) J. F. Bunnett, R. L. McDonald, and F. P. Olsen, J. Am. Chem. Soc., 96, 2855 (1974); (e) A. J. Kresge, R. A. More O’Ferrall, L. E. Hakka, and V. P. Vitullo, Chem. Commun., 46 (1965); (f) K. Yates and J. C. Riordan, Can. J. Chem., 43, 2328 (1965). (13) 0. Wallach and L. Belli, Chem. Ber., 13,525 (1880).
50%
133 H2SOb. % W/W 80% 90% 9779
7010
60%
100% “Z5207
100%
r i
-6l 3
4
’
I
5
6
1
7
I
8
#
9 -H0
/
(
10
# / 12*
11
I
13
L 14
15
-Figure 1. The species present in sulfuric acid. Concentrations of H+(H20),, HS04-, sod2-, and H3S04+, and activities of HzO, HzS04, H3O+, and HzSz07, in log mole fraction units, plotted against -Ho, for the system HzO-HzS04-H&07 at 2 5 O .
mechanism of the transformation have proliferated. These hypotheses have advocated a number of unusual reaction intermediates, of which the main contenders are given by structures 1-10. Of these 0
1
H
I
OH
H 2
H
7
H
\+/ 3
‘OH,
8
4
9
5
10
species, several are in fact known to be formed in strongly acidic media from azoxybenzene. However, the mere fact of the existence of a given species does not necessarily prove that it actually occurs along the reaction pathway. Our subsequent discussion will consider this point in more detail. Some of these structures are symmetrical in nature while others are not. The apparent involvement of a symmetrical intermediate was indicated by Shemyakin’s observation14 that azoxybenzene specifically labeled on one nitrogen yielded on rearrangement phydroxyazobenzene in which each nitrogen carried (14) M. M. Shemyakin, V. I. Maimind, and B. K. Vaichunaite, Chem Ind. (London),755 (1958).
Buncel
134
Accounts of Chemical Research
Table I Kinetic Data for Rearrangement of Azoxybenzene to 4-Hydroxyazobenzene in Aqueous HzSQ4 at 25"
75.30 6.65 0.967 80.15 7.42 0.994 85.61 8.35 0.999 90.37 9.05 1.ooo 9.82 1.ooo 95.19 1.ooo 97.78 10.35 99.00 10.82 1.ooo 99.59 11.18 1.ooo 1.ooo 99.90 11.64 1.ooo 99.97 11.84 99.99 11.90 1.ooo a Data from ref l l b , h. * Calculated using a zene of -5.15 (ref 19). Pseudo first-order rate mined spectrophotometrically.
Scheme 1
H 7
0.016 0.208 2.17 7.23 20.9 43.8
8
I
H+# a
76.8 227 860 2310 4160 pKa for azoxybenconstants as deter-
5
6
f
HA\b
r
H?---H---A
1*
half the 15N label of the reactant. T o account for this observation, Shemyakin proposed the intermediacy of the N,N-oxide species 1. Gore, on the other hand, pointed that the dicationic structure 5 equally fulfilled the I5N equalization result.16 In related work, studies with [l-14C]azoxybenzene17confirmed \--I the isotopic scrambling result in H2S04, while l80 H tracer studies have shown the rearrangement to be intermolecular,l8 i.e., the OH in the product is solvent derived. A kinetic study of the Wallach rearrangement by Buncel and Lawtonlg provided the first evidence for the rate of rearrangement continues to increase bethe dicationic intermediate 5. Some of the kinetic yond the stage of complete monoprotonation of the data, determined as a function of sulfuric acid consubstrate. The lack of leveling off in the 100%H2SO4 centration, are set forth in Table I.20J1a(Data obregion is also noteworthy. This behavior indicates retained a t 75" for 65-90% H2S04 show similar requirement of a second proton transfer.24 sponse of rate to acidity.) Also given are the extents Since protonation of the substrate in most of these of monoprotonation as derived from the measured media is extensive, any reasonable scheme will start pKa, which was evaluated spectrophot~metrically.~~with the conjugate acid of azoxybenzene (6) as the efAzoxybenzene was shown to approximate the behavfective substrate; this is the case in Scheme I. Conior of a Hammett base, with near unit slope in the log ceptually, the second proton transfer may be utilized CS/CSH+VS. H0 in two ways. In path a an equilibrium proton transfer The striking aspect of the data in Table I is that is followed by rate-determining loss of H20 to yield dication 5 , while in path b rate-determining proton (15) P. H. Gore, Chem. Ind. (London),191 (1959). transfer occurs concertedly with N-0 scission to (16) Gore formulated the dication as a dinitrenium species, CeHsN+yield the same dication. Product formation occurs via N+CsHs, while the structure 5 with the triple bond was preferred by Buncel and Lawton due to the favorable energy factor associated with N-N. attack by H2O or HS04- a t aromatic carbon followed ( 1 7 ) L. C. Behr and E. C. Hendley, J. Org. Chem., 31,2715 (1966). by proton loss and rearomatization. It has been (18) (a) S. Oae, T. Fukumoto, and M. Yamagami, Bull. Chem. SOC.Jpn., shownz5 that if the azoaryl hydrogen sulfate were 34, 1873 (1961); (b) M. M. Shemyakin, T. E. Agadzhanyan, V. I. Maimind, and R. V. Kudryavtsev, Izu. Akad. Nauk SSSR, Ser. Khim., 1339 (1963). formed under the reaction conditions (as is likely (19) E. Buncel and B. T. Lawton, Chem. Ind. (London),1835 (1963); Can. once the bisulfate ion concentration is comparable to J. Chem., 43,862 (1965). or exceeds that of water (Figure l)),it would be rap(20) In >95% HzS04, sulfonation of p-hydroxyazobenzene becomes important and the rate measurements must separate the primary rearrangeidly hydrolyzed to the azophenol. ment from the consecutive sulfonation process.21 As an initial approach toward treating involvement (21) (a) E. Buncel and W. M. J. Strachan, Can. J. Chem., 48, 377 (1970); of the second proton, one can plot log k$ - log ( CSH+/ (b) E. Buncel, W. M. J. Strachan, R. J. Gillespie, and R. Kapoor, Chem. Commun., 765 (1969); (c) E..Buncel and W. M. J. Strachan, Can. J . Chem., (CS CSH+))vs. Ho, which takes into account the ex47, 4011 (1969); (d) W. M. J. Strachan, A. Dolenko, and E. Buncel, ibid., 47, tent of monoprotonation.26 This graph shows strong 3631 (1969); (e) E. Buncel, W. M. S. Strachan, and H. Cerfontain, ibid., 49, 152 (1971). curvature; initially the slope is ca. 1.4 but decreases (22) It is interesting that protonation of pyridine N-oxidesZ3* follows to ca. 0.6 in the 90-95% H2S04 region, while in >95% closely the H A whereas protonation of azoxybenzene is reason-
+
ably well described by the (steeper) Ho function. The reason presumably lies in the differing hydration requirements of the corresponding conjugate acids, relative to the neutral bases, possibly as a result of steric factors.23c-e (23) (a) C. D. Johnson, A. R. Katritzky, and N. Shakir, J . Chem. SOC.B, 1235 (1967); (b) A. R. Katritzky, J. B. Stevens, and K. Yates, Can. J . Chem., 42,1957 (1964); (c) E. M. Arnett, Prog. Phys. Org. Chem., 1, 223 (1963); (d) J. T. Edward, Trans. Roy. SOC.Can., [4] 2, Sect. 111, 313 (1964); (e) G. Scorrano, Acc. Chem. Res., 6,132 (1973).
(24) D. Landini, G. Modena, F. Montanari, and G. Scorrano, J . Am. Chem. S O C . 92,7168 , (19701, observed a second proton transfer in the reduction and racemization of sulfoxides by halide ions in aqueous HCI04 (Ho range, -0.5 to -4.4). (25) E. Buncel and W.M. J. Strachan, Can. J . Chem., 47,911 (1969). (26) (a) W.M. Schubert and H. K. Latourette, J . Am. Chem. SOC.,74, 1829 (1952);(b) W. M. Schubert and R. H. Quacchia, ibid., 85,1278 (1963).
The Wallach Rearrangement
Vol. 8, 1975
HzS04 the slope increases once more to a value of ca. 2.3. Evidently, a simple log rate-Ho relationship does not hold. For the mechanism depicted in Scheme I, path a, eq 4-9 apply. It is assumed in eq 8 that the extent of
+ H+ 2 SH' SH+ + H+ f SHz2+ S
SHZ2' X2'
-
X2+
(fast,preequilibrium) (fast,preequilibrium) (rate-determining)
--+ products
(fast steps)
135
+e m
=0 x
-
0 m
kL
(4)
L a +
(5)
-
(6)
(7)
a
-
m 0
-7 -4
-5
-2
-3
l o g aH2SOI
Or
-I
0
l o g 'H3SOIC
-
kO
+ logfBH,2+fSH+
(9)
fBH+f+
diprotonation is negligible (CSH22+
