Rearrangement mechanisms for azoxypyridines and azoxypyridine N-oxides in the 100% H2so4 region -The Wallach rearrangement story comes full circle

Article abstract of DOI:10.1139/V09-080

Kinetic studies of the Wallach rearrangements of four azoxypyridines, four azoxypyridine N-oxides, and one azoxypyridine JV-methiodide have been, carried out in the 100% H₂SO₄ acidity region. For all of the ss-isomers in the study th

Full text of DOI:10.1139/V09-080

1127
Rearrangement mechanisms for azoxypyridines
and azoxypyridine N-oxides in the 100% H2SO4
region — The Wallach rearrangement story comes
full circle1
Erwin Buncel, Sam-Rok Keum, Srinivasan Rajagopal, and Robin A. Cox
Abstract: Kinetic studies of the Wallach rearrangements of four azoxypyridines, four azoxypyridine N-oxides, and one
azoxypyridine N-methiodide have been carried out in the 100% H2SO4 acidity region. For all of the b-isomers in the study
the reactions proceeded at a spectrally measurable rate, and the log observed rate constants were found to be linear func-
+
tions of the log H3SO4 concentration, as previously found for azoxybenzene itself, suggesting that the reaction mecha-
nism for these substrates is the same as that previously deduced for axozybenzene, i.e., a general-acid-catalysis A-SE2
process. For the a-azoxypyridines no reaction could be observed at all. The two a-azoxypyridine N-oxides in the study did
react, albeit very slowly, but for these two compounds the log observed rate constants were not linear functions of the log
H3SO4+ concentration, but were instead found to be linear in the H0 acidity function, which is known for the 100% H2SO4
acidity region. It follows that the reaction mechanism for these a-isomers is a different one, presumably an A1 process.
This mechanism was proposed back in 1963 for azoxybenzene, but has never actually been observed for any substrate be-
fore the work reported in this study. Thus, the Wallach rearrangement story can be said to have come full circle.
Key words: azoxypyridines, azoxypyridine N-oxides, Wallach rearrangement, reactivities, mechanisms.
´
´
´
´
´
`
´
´
´
´
Resume : Operant dans la region d’acidite du H2SO4 a 100 %, on a effectue des etudes cinetiques sur les rearrangements
´
de Wallach de quatre azoxypyridines, de quatre N-oxydes d’azoxypyridines et d’un N-methiodure d’azoxypyridine. Pour
`
´
´
´
`
`
´
tous les isomeres b de l’etude, les reactions procedent a des vitesses spectroscopiquement mesurables et on a trouve que le
+
´
´
log des constantes de vitesse observees sont lineaires avec le log de la concentration en H3SO4 , tel qu’on l’avait observe
´
`
ˆ
`
´
´
anterieurement avec l’azoxybenzene lui-meme, ce qui suggere que le mecanisme de la reaction pour ces substrats est le
´ ´
ˆ
´
´
`
`
meme que celui deduit anterieurement pour l’azoxybenzene, c’est-a-dire un processus de catalyse acide generale, A-SE2.
´
´
´
´
Pour les a-azoxypyridines, on n’a observe aucune reaction. Les deux N-oxydes de la a-azoxypyridne reagissent, mais la re-
`
´
action est toutefois tres lente et les log des constantes de vitesses de ces deux produits ne donnent pas de relation lineaire
+
´
´
avec le log de la concentration en H3SO4 ; elles sont toutefois lineaires avec une fonction d’acidite H0 qui est connue pour
´
´
`
´
´
`
la reaction d’acidite du H2SO4 a 100 %. On peut donc en conclure que le mecanisme de la reaction pour les isomeres a
´
´
´ ` ´ ´
´
`
est different, probablement un processus A1. Ce mecanisme avait deja ete propose en 1963 pour l’azoxybenzene, mais il
´ ´
´
´
´
n’avait jamais ete observe avec un substrat avant celui rapporte dans ce travail. On peut donc dire que l’histoire du rear-
´
´
rangement de Wallach a retrouve son point de depart.
´
´
´
´
´
Mots-cles : azoxypyridines, N-oxydes d’azoxypyridines, rearrangement de Wallach, reactivites, mecanismes.
´
[Traduit par la Redaction]
organic chemistry, providing unique insights into reaction
mechanisms. For instance, it was the first reaction for which
a general-acid-catalysis mechanism in strong aqueous acid
media was proved to take place,4,5 the possibility having
been suggested previously.6 Catalysis by the species H2SO4
Introduction
Many of the kinetic and other studies of the Wallach re-
arrangement, discovered in 18802 and shown below for
azoxybenzene 1, have been extensively reviewed, along
with other aromatic rearrangements.3 Over the years, this re-
action has proven to occupy a pivotal position in physical
+
and H3SO4 occurred in concentrated H2SO4, but not by
(H+)aq (or H3O+), which was found to be not a strong enough
Received 17 March 2009. Accepted 22 April 2009. Published on the NRC Research Press Web site at canjchem.nrc.ca on 20 July 2009.
Dedicated to the memory of Yvonne Chiang (Mrs. Jerry Kresge), wife, mother, fine chemist.
E. Buncel,1 S.-R. Keum,2,3 and S. Rajagopal. Department of Chemistry, Queen’s University, Kingston, ON K7L 3N6, Canada.
R.A. Cox.4 16 Guild Hall Drive, Scarborough, ON M1R 3Z8, Canada.
1Corresponding author (e-mail: buncele@chem.queensu.ca).
2Corresponding author (e-mail: keum@korea.ac.kr).
3Present address: Korea University, Department of New Materials Chemistry, Jochiwon 339700, South Korea.
4Corresponding author (e-mail: robin.a.cox@sympatico.ca).
Can. J. Chem. 87: 1127–1134 (2009)
doi:10.1139/V09-080
Published by NRC Research Press

1128
Can. J. Chem. Vol. 87, 2009
acid species to catalyze the reaction.4 This was again shown
to be the case, quite strikingly, when the reaction was found
not to take place at all in HClO4, a stronger acid than H2SO4
in H0 terms, but containing no acidic species other than
H3O+ and other (H+)aq.7 Since that time, other general-
acid-catalysis mechanisms have been found, for instance,
in benzamide hydrolysis,8 and others proposed8 in concen-
trated aqueous acids.
The first kinetic study of the reaction of azoxybenzene,9,10 which we reported as early as 1963,9 quantified the relationship
between substrate protonation (pKBH = –5.1811) and rearrangement, and demonstrated the requirement for a second proton
+
transfer;10 the proposed mechanism involved the formation of a dicationic intermediate, which at the time we wrote as having
the structure 2.9,10 However, recent accurate theoretical calculations on several dicationic intermediates shown to be present in
various Wallach rearrangements12 indicate that the structure is more accurately rendered as 3, with the positive charges being
almost entirely situated on the two rings, not on the nitrogens, which retain their lone pairs.12 Dicationic intermediates can
actually be observed as stable species in suitable superacid media.13
This prompted the question as to whether a tricationic species, for instance 4, could be generated on going to heteroaro-
matic structures. (For clarity only one resonance form is given for 3 and for 4.) We have found, using modern theoretical
calculations, that the charge distributions in di- (or tri-) cationic intermediates like this can have dramatic effects on their
reactivities,11 and we developed a novel method of showing this by using the differences in charge distribution over the two
halves of the species concerned.11 The latter study used a series of azoxypyridines and azoxypyridine N-oxides. One of the
observations that the calculation results helped to explain was the large reactivity difference found between a- and b-isomers
of these molecules,11,14 and this paper deals in detail with their kinetic behaviour; it happens that as well as the reaction rates
being quite different, the reaction mechanisms are different as well. The compounds used in this study, 5–13, are shown be-
low, in the protonated forms, which are the true reactants,11 the parent species being entirely monoprotonated under the reac-
tion conditions used. (Except for 13, the counteranions would be sulfate or bisulfate.) The a and b nomenclature is used here
+
in preference to the formal NNO and ONN nomenclature, simply because of the convenience. The pKBH values for diproto-
nation of those molecules for which these could be measured have been given before.11
Experimental
We have reported the syntheses of 5–12 previously.11,15 The methiodide 13 has also been reported;15 in brief, it was ob-
tained from 6 by methylation, refluxing 6 with a slight excess of methyl iodide for 5 h and recrystallizing the resulting red-
brown crystals from ethanol. The acid solutions, in the 100% H2SO4 range, were prepared from 96% H2SO4 and fuming 30%
oleum, and their concentrations determined accurately by conductivity measurements according to Gillespie and Wasif.16 Re-
action rates were determined spectrophotometrically by the ‘‘direct method’’.10,11
The 90%–100% H2SO4 acidity region is interesting and not particularly complex, the only species present being H3O+
(there is no free water), HSO4 , H3SO4 , and undissociated H2SO4 molecules.4 Accurate H2SO4 activities are available,17 and
with these and the value of (7.8 1.7) ꢀ 10–4 mol2 l–2 given by Liler18 for the best value for the autoprotolysis constant of
H2SO4 at 25 8C (combined from several sources) it is possible to calculate the concentrations (although not the activities) of
–
+
–
the species H3SO4+ and HSO4 between 90% and 100% H2SO4 (at 25 8C). This was done. Above 100% H2SO4 the medium
becomes much more complex, as H2S2O7 and higher sulfur acids are formed, together with the ions resulting from their dis-
sociation, and species concentrations cannot be calculated. However, acidity functions for these media are available;19 the one
used in this work is the one generally called H0, measured at ‘‘room temperature’’ by Gillespie, Peel, and Robinson.20 This
scale was determined using the protonation equilibria of aromatic nitro compounds, which protonate on an oxygen of the
nitro group, and was adjudged suitable to use in this work because of the similarity of the nitro compounds to the azoxy
compounds used here. (The other acidity scale in these media, measured by Frederick and Poulter21 and called HA, uses pro-
tonated uracils and related compounds as indicators and was adjudged to be less suitable.)
Published by NRC Research Press

Buncel et al.
1129
All of the observed pseudo-first-order rate constants, kj, at different sulfuric acid concentrations, some at 50 8C and some
at 65 8C, obtained for 6 and for 8–13 are given as Supplementary data in Tables S1–S7, together with the H0 values used and
the values calculated for the H3SO4+ concentrations. Rate constants at 25 8C were already available for 1,10,22 and no reaction
could be observed for 5 or for 7.
for 10 were very similar. For 10 and 12, results at two tem-
peratures were available; although both temperatures gave
linear plots, the slopes were not the same, but we attach lit-
tle significance to this as the rate data were obtained by two
different workers at different times. There could be other
reasons. The results for the N-methiodide 13 are shown in
Fig. 4, which, as might be expected, exhibits very similar
behaviour to that of 10. It is shown in Fig. 5 that the ob-
served log rate constants for 10 and 12 are in fact not linear
in –H0. The observed slopes, intercepts, etc. for the various
correlations in this paper, many of which are shown in
Figs. 1–4, are collected together in Table 1.
The a-isomers behaved differently, however; in fact, no
reaction at all was observed for the azoxypyridines 5 and 7.
The azoxypyridine N-oxides 9 and 11 did react, but much
more slowly than did the b-isomers under the same condi-
tions. The log rate constants were not linear in the log
Results
Previously, we found that the log observed rate constants
for 1 were a linear function of the log H3SO4+ concentration
at sulfuric acid concentrations between 90% and 100%,4 and
the same was also found to be true, at least under some cir-
cumstances, for hexamethylazoxybenzene5 and some azoxy-
naphthalenes.23 Shown in Fig. 1 are the log rate constants
for azoxybenzene 110,21 plotted against the log H3SO4+ con-
centrations as recalculated for this work; as it can be seen,
the linearity is excellent. For these data log rate constant
plots against –H0 are known to exhibit multiple curvature.4
It therefore seemed reasonable to plot the log rate con-
stant results obtained in this work against the log H3SO4
concentrations in the same way. Dealing with the b-isomers
first: Fig. 2 demonstrates the excellent linearity obtained for
the azoxypyridines 6 and 8, and Fig. 3 that obtained for the
azoxypyridine N-oxide 12 at two temperatures; the results
+
Published by NRC Research Press

1130
Can. J. Chem. Vol. 87, 2009
Fig. 1. Plot of log observed rate constants for 1 at 25 8C against the
Fig. 3. Plot of the log observed rate constants for 12 at 50 8C
(lower) and 65 8C (upper) against the log H3SO4+ concentration.
log H3SO4+ concentration.
Fig. 4. Plot of the log observed rate constants for 13 at 50 8C
Fig. 2. Plot of the log observed rate constants for 6 and 8 at 50 8C
against the log H3SO4+ concentration.
against the log H3SO4+ concentration.
it does not work well in the 90%–100% H2SO4 acidity
range, primarily because an extrapolation to pure water is
required, which is unrealistic here. It does not work at all
above 99.5% H2SO4, as the X-function cannot be defined in
this region,17 and clearly extrapolating to pure water from
oleum solutions, which contain no water at all, is not rea-
sonable. Similarly, the Bunnett–Olsen24 and Bunnett25 meth-
ods, both of which require the water activity, cannot be used
in media, which contain no free water either.
+
H3SO4 concentration, as is illustrated in Fig. 6, however,
when the log rate constants for 9 and 11 were instead plot-
ted against the –H0 acidity function for this medium,20 ex-
cellent linearity resulted, as can be seen from Fig. 7. The
slopes, etc. of these plots are also given in Table 1.
Discussion
The excess acidity (X-function) method, developed for the
study of the mechanisms of reactions in acid media,17 has
proved itself to be an excellent tool for this task.17 However,
Therefore, in this acidity region it is more reasonable to
ascertain whether the log rate constants can achieve linearity
Published by NRC Research Press

Buncel et al.
1131
Fig. 5. Plot of the log observed rate constants for 10 and 12 at
50 8C against –H0, illustrating the lack of linearity in these plots for
the b-isomers.
The reaction mechanism shown in Scheme 2 is in fact
analogous to one proposed for azoxybenzene 1 itself as far
back as 1963,9,10 but never actually observed in practice for
any substrate until now. A second protonation of the monop-
rotonated substrate 9, pKSH+ = –10.49,11 occurs first. Follow-
ing this, the (third) proton transfer to the azoxy OH group,
giving 14, is a pre-equilibrium process, rather than being
concerted with water loss and di- (actually tri-) cationic in-
termediate formation as in Scheme 1, and the actual rate-de-
termining step is loss of water to give the intermediate as
shown in Scheme 2. (In Scheme 2, no attempt is made to
show the full electron flow from 14 to the intermediate, as
this would look confusing.) The actual concentration of the
triprotonated precursor must be vanishingly small, even at
these acidities, and the formation of the intermediate from it
is an uphill process, as we found when we calculated the
charge distributions in these species,11 and so the reaction is
very slow. Once the intermediate is formed it will react with
almost anything nucleophilic very quickly, probably a bisul-
fate ion at these acidities, as shown. The difference from the
b-isomers is that, for these, intermediate formation is down-
hill in energy,11 and so it can be concerted with the third
proton transfer.
The information in Table 1 is useful for internal compari-
sons of the reactions of 1 and 5–13, but very few other cor-
relations of this type are available for other reactions, and so
other comparisons are difficult; for instance, the similar re-
action found in the hydrolysis of benzamides at high acidity
when plotted against known species activities or concentra-
tions, or against a reliable acidity function.26 The fact of lin-
earity is what is actually important. The slopes and
intercepts of these lines may not mean much, as there is lit-
tle data with which to compare them, but at least some
meaning can be attached to the relative values. As men-
tioned above, 90%–100% aqueous sulfuric acid is not a par-
ticularly complex medium, only four species being present,
+ 8
involves (H+)aq as the catalyzing acid, not H3SO4 . We can
make no reliable use of the information obtained at different
temperatures for two reasons: (i) because the log rate con-
stant data, obtained at 50 8C or at 65 8C, is plotted as func-
+
tions of H3SO4 concentrations calculated at 25 8C (or an
acidity function measured at room temperature); and (ii)
there is no reliable ‘‘acidity’’ at which to calculate them,
since any value will also have variations due to different
species concentrations built into it (for instance, we cannot
extrapolate to ‘‘pure water’’, as we can when using the ex-
cess acidity method17).
H3O+, HSO4 , H3SO4 , and H2SO4, and the concentrations
of all of them can be calculated.
–
+
Much previous work3–5 has led to the conclusion that
azoxybenzene itself reacts with undissociated H2SO4 mole-
cules in sulfuric acid, which is more dilute than 90%, and
+
Having said that, there are some conclusions to draw from
the information in Table 1. The intercept values for 10 and
the corresponding methiodide 13 are about the same, as
would be expected. The b-3-isomers react consistently faster
than do the b-4-isomers, by about 0.3 of a log unit. The b-4-
isomers 8 and 12 react at the same rate, whereas the azoxy-
pyridine 6 reacts slightly faster than does the azoxypyridine
N-oxide 10. Little difference seems apparent between the
axoxypyridines and the azoxypyridine N-oxides. Most of the
differences can be explained by considering the charge sepa-
ration in the various reaction intermediates, and we have al-
ready done this.11 Little difference is apparent between the
intercepts for the a-isomers 9 and 11. Some comments can
be made about the slopes listed in Table 1: it seems that the
reaction temperature may have a large effect on them, look-
ing at the differences between the 50 8C and 65 8C results for
10 and 12, but unfortunately, they are in opposite directions
for these two; however, the slope for 1 at 25 8C is much
smaller than all of the other slopes. The behaviour of the a-
isomers 9 and 11 seem much the same as one another.
with H3SO4 at higher concentrations. The reaction scheme
has been given recently11 and need not be repeated here.
þ
The linearity against log CH SO also found for all of the b-
2
4
isomers in this study strongly suggests the same mechanism
applies to them. The azoxybenzene mechanism, modified so
as to apply to 6, is given here as Scheme 1; the same
Scheme 1 (with suitable atom arrangements) also applies to
8, 10, 12, and 13.
The a-isomers, however, behave differently; no reaction
could be observed at all for 5 and 7, while for 9 and 11 the
þ
log rate constants are not linear functions of log CH SO , but
2
are instead linear in –H0 (see Figs. 6 and 7). Linearity 4in dif-
ferent acidity functions has long been attributed to the A1
and A-SE2 reaction mechanisms;17,18,26 this is part of the
original Zucker–Hammett hypothesis.27 We have just ruled
out an A-SE2 process for this reaction, leaving an A1 reaction
as the only possibility. A likely reaction scheme for this proc-
ess is given for 9 in Scheme 2. (In Schemes 1 and 2 the pro-
ton in dilute acid media is written as having the form
H+(H2O)n, as this is the most likely situation applying there.8)
Published by NRC Research Press

1132
Can. J. Chem. Vol. 87, 2009
Table 1. Slopes, intercepts, etc. for the observed log rate correlations for 1, 6, and 8–13.
Substrate
T (8C)
Slope
Intercept
r a
Nb
sc
þ
4
Against log CH SO
3
1 (see Fig. 1)
5
6 (see Fig. 2)
7
8 (see Fig. 2)
10
25
50
0.831 0.013 –2.321 0.018 0.997
No reaction
1.755 0.063 –1.521 0.051 0.990
No reaction
1.829 0.041 –2.034 0.028 0.995
1.806 0.035 –1.754 0.019 0.998
25(1) 0.066
18
21
0.099
0.091
50
50
65
50
65
50
13(1) 0.054
10
2.44 0.19
–1.413 0.043 0.954
18
10
0.095
0.072
12 (see Fig. 3)
12 (see Fig. 3)
13 (see Fig. 4)
1.687 0.057 –2.011 0.044 0.995
1.201 0.024 –1.695 0.012 0.995
1.515 0.039 –1.697 0.028 0.997
27(1) 0.041
10
0.065
Against –H0
9 (see Fig. 6)
11 (See Fig. 6)
65
65
2.587 0.070 –35.72 0.83
2.34 0.12 –32.3 1.4
0.994
0.986
18
12
0.11
0.12
Note: Errors given are standard deviations.
aCorrelation coefficient.
bNumber of points. Any rejected by the program used as being off the line, to 95% confidence, are in
parentheses here and in the figures.17
cRoot-mean-square deviation between experimental and fitted points.
Fig. 6. Plot of the log observed rate constants for 9 and 11 at 65 8C
against the log H3SO4+ concentration, illustrating the lack of linear-
ity in these plots for the a-isomers.
Fig. 7. Plot of the log observed rate constants for 9 and 11 at 65 8C
against –H0.
rangement mechanisms resulting from structural changes, as
one traverses the H2SO4 acidity region from dilute to moder-
ately concentrated to highly concentrated and oleum media.
Moreover, analogies between the Wallach and benzidine re-
arrangements are apparent,28 such as two possible mecha-
nisms also being possible in this case, involving mono- and
di-cationic reaction pathways.29
This whole class of azo and azoxy compounds and the
reactions they undergo have contributed much to our under-
standing of physical organic chemistry. For instance, we
have found that there are two different pathways for the
General conclusions
Having delineated the various possible Wallach rearrange-
ment mechanisms in the 100% H2SO4 region, it is of interest
that in more dilute acid media another reaction mechanism
is possible, proceeding via the monoprotonated species,5,23
but only for those azoxy compounds in which the intermedi-
ate species can be stabilized in some way. For instance, by
the retention of full aromaticity in one of the aromatic rings
of several azoxynaphthalenes.23 If formation of one of these
is not possible, the general-acid-catalysis mechanism of 1 is
followed,23 thus completing the spectrum of Wallach rear-
Published by NRC Research Press

Buncel et al.
1133
Scheme 1.
Scheme 2.
acid hydrolysis of the two methoxy groups in 4-[(3,4-dime-
thoxyphenyl)azo]pyridine,30 and an excess acidity analysis
of the acid-catalyzed aryl hydroxylation of phenylazopyri-
dines showed that the reaction was an A2 process with nu-
More generally, studies of the heteroaromatic azoxyarene
series led to the examination of the solvatochromic behav-
iour of hydroxyazo products and their analogs, and the es-
tablishment of the paꢁzo solvent polarity scale.32 More
recently, we have been studying azo moieties as optical
transducers, through light absorption, energy transfer, and
–
cleophilic attack of HSO4 or H2O on the diprotonated
substrate in the slow step.31
Published by NRC Research Press

1134
Can. J. Chem. Vol. 87, 2009
cis/trans isomerism, for molecular photo switching.33 This
continues to be a very fruitful area of research.
A. Can. J. Chem. 2008, 86 (4), 298–304. doi:10.1139/V08-
016.
(12) Cox, R. A.; Fung, D. Y. K.; Csizmadia, I. G.; Buncel, E.
Can. J. Chem. 2003, 81 (6), 535–541. doi:10.1139/v03-049.
(13) Olah, G. A.; Dunne, K.; Kelly, D. P.; Mo, Y. K. J. Am.
Chem. Soc. 1972, 94 (21), 7438–7447. doi:10.1021/
ja00776a029.
(14) Buncel, E.; Keum, S.-R. J. Chem. Soc. Chem. Commun.
1983, 578–579. doi:10.1039/c39830000578.
(15) Buncel, E.; Keum, S.-R.; Cygler, M.; Varughese, K. I.; Birn-
baum, G. I. Can. J. Chem. 1984, 62 (8), 1628–1639. doi:10.
1139/v84-280.
(16) Gillespie, R. J.; Wasif, S. J. Chem. Soc. 1953, 215–221.
(17) Cox, R. A. Adv. Phys. Org. Chem. 2000, 35, 1–66. doi:10.
1016/S0065-3160(00)35011-0.
(18) Liler, M. Reaction mechanisms in sulfuric acid; Academic
Press: New York, 1971; p 79.
Supplementary data
Supplementary data for this article are available on the
journal Web site (canjchem.nrc.ca) or may be purchased
from the Depository of Unpublished Data, Document Deliv-
ery, CISTI, National Research Council Canada, Ottawa, ON
K1A 0R6, Canada. DUD 3970. For more information on ob-
taining material, refer to cisti-icist.nrc-cnrc.gc.ca/cms/
unpub_e.shtml.
Acknowledgments
Financial support of this research through a grant by the
Natural Sciences and Engineering Research Council of Can-
ada to EB, and from Queen’s University via teaching assis-
tantships (to S-RK and SR) is gratefully acknowledged.
(19) Cox, R. A.; Yates, K. Can. J. Chem. 1983, 61 (10), 2225–
2243. doi:10.1139/v83-388.
(20) Gillespie, R. J.; Peel, T. E.; Robinson, E. A. J. Am. Chem.
Soc. 1971, 93 (20), 5083–5087. doi:10.1021/ja00749a021.
(21) Frederick, G. D.; Poulter, C. D. J. Am. Chem. Soc. 1975, 97
(7), 1797–1801. doi:10.1021/ja00840a031.
(22) Buncel, E.; Strachan, W. M. J. Can. J. Chem. 1970, 48 (2),
377–382. doi:10.1139/v70-055.
(23) Cox, R. A.; Dolenko, A. J.; Buncel, E. J. Chem. Soc. Perkin
Trans. 2 1975, 471.
(24) Bunnett, J. F.; Olsen, F. P. Can. J. Chem. 1966, 44 (16),
1917–1931.
(25) Bunnett, J. F. J. Am. Chem. Soc. 1961, 83 (24), 4973–4977.
doi:10.1021/ja01485a021.
(26) Rochester, C. H. Acidity functions; Academic Press: New
York, 1970.
(27) Zucker, L.; Hammett, L. P. J. Am. Chem. Soc. 1939, 61 (10),
2791–2798. doi:10.1021/ja01265a067.
(28) Buncel, E. Can. J. Chem. 2000, 78 (10), 1251–1271. doi:10.
1139/cjc-78-10-1251.
References
(1) Part 28 of the series ‘‘Studies of azo and azoxy dyestuffs’’ of
which Part 27 is ref. 11.
(2) Wallach, O.; Belli, E. Chem. Ber. 1880, 13 (1), 525–527.
(3) (a) Shine, H. J. Aromatic rearrangements; Elsevier: New
York, 1967; (b) Buncel, E. Mechanisms of molecular migra-
tions; Thyagarajan, B. S., Ed. Wiley: New York, 1968;
Vol. 1, pp 61–119; (c) Buncel, E. Acc. Chem. Res. 1975, 8
(4), 132–139. doi:10.1021/ar50088a004.; (d) Cox, R. A.;
Buncel, E. The chemistry of the hydrazo, azo, and azoxy
groups; Patai, S., Ed. Wiley: London, 1975; Vol. 1, pp 775–
859; (e) Cox, R. A.; Buncel, E. The chemistry of the hy-
drazo, azo, and azoxy groups; Patai, S., Ed. Wiley: London,
1997; Vol. 2, pp 569–602.
(4) Cox, R. A. J. Am. Chem. Soc. 1974, 96 (4), 1059–1063.
doi:10.1021/ja00811a018.
(5) Cox, R. A.; Buncel, E. J. Am. Chem. Soc. 1975, 97 (7),
1871–1875. doi:10.1021/ja00840a045.
(6) (a) Kresge, A. J.; Hakka, L. E.; Mylonakis, S. G.; Sato, Y.
Discuss. Faraday Soc. 1965, 39, 75–83. doi:10.1039/
df9653900075.; (b) Kresge, A. J.; Chiang, Y.; Fitzgerald, P.
H.; McDonald, R. S.; Schmid, G. H. J. Am. Chem. Soc.
1971, 93 (19), 4907–4908. doi:10.1021/ja00748a043.
(7) Bolduc, R.; Buncel, E.; Cox, R. A.; Kitzul, R. B. Unpub-
lished results quoted in ref. 2d.
(29) Banthorpe, D. V.; Hughes, E. D.; Ingold, C. K. J. Chem.
Soc. 1964, 2864–2901. doi:10.1039/jr9640002864.
(30) Cox, R. A.; Onyido, I.; Buncel, E. J. Am. Chem. Soc. 1992,
114 (4), 1358–1363. doi:10.1021/ja00030a034.
(31) Cheon, K.-S.; Cox, R. A.; Keum, S.-R.; Buncel, E. J. Chem.
Soc. Perkin Trans. 2 1998, 1231.
(32) Buncel, E.; Rajagopal, S. J. Org. Chem. 1989, 54 (4), 798–
809. doi:10.1021/jo00265a017.
(8) Cox, R. A. Can. J. Chem. 2008, 86 (4), 290–297. doi:10.
1139/V08-015.
(9) Buncel, E.; Lawton, B. T. Chem. Ind. (London) 1963, 1835.
(10) Buncel, E.; Lawton, B. T. Can. J. Chem. 1965, 43 (4), 862–
875. doi:10.1139/v65-113.
(33) (a) Abdallah, D.; Ghani, M. A. A.; Cunningham, M. F.; Kaz-
maier, P. M.; Keoshkerian, B.; Buncel, E. Can. J. Chem.
2004, 82 (9), 1393–1402. doi:10.1139/v04-107.; (b) Ghani,
M. A. A.; Abdallah, D.; Kazmaier, P. M.; Keoshkerian, B.;
Buncel, E. Can. J. Chem. 2004, 82 (9), 1403–1412. doi:10.
1139/v04-106.
(11) Buncel, E.; Keum, S.-R.; Rajagopal, S.; Kiepek, E.; Cox, R.
Published by NRC Research Press