Generation of o-quinone α-carbomethoxymethide by photolysis of methyl 2-hydroxyphenyldiazoacetate in aqueous solution

Article abstract of DOI:10.1039/b211444p

Flash photolysis of methyl 2-hydroxyphenyldiazoacetate (8) in dilute aqueous perchloric acid solution and acetic acid and biphosphate ion buffers produced a transient species that was identified as o-quinone α-carbomethoxymethide (9). This structural assignment is based upon solvent isotope effects, the form of buffer catalysis, UV absorption maxima, and the identity of decay rate constants with those determined for the transient obtained by flash photolysis of other, more conventional, quinone methide precursors, namely the benzyl alcohol methyl 2-hydroxy mandelate (10) and its acetate, 2′-acetoxy-2-hydroxyphenylacetate (11).

Full text of DOI:10.1039/b211444p

View Article Online / Journal Homepage / Table of Contents for this issue
Generation of o-quinone a-carbomethoxymethide by photolysis of
methyl 2-hydroxyphenyldiazoacetate in aqueous solutionyz
Y. Chiang, A. J. Kresge* and Y. Zhu
Department of Chemistry, University of Toronto, Toronto, Ontario M5S 3H6, Canada.
E-mail: akresge@chem.utoronto.ca
Received 19th November 2002, Accepted 20th December 2002
First published as an Advance Article on the web 17th January 2003
Flash photolysis of methyl 2-hydroxyphenyldiazoacetate (8) in dilute aqueous perchloric acid solution and
acetic acid and biphosphate ion buffers produced a transient species that was identified as o-quinone
a-carbomethoxymethide (9). This structural assignment is based upon solvent isotope effects, the form of buffer
catalysis, UV absorption maxima, and the identity of decay rate constants with those determined for the
transient obtained by flash photolysis of other, more conventional, quinone methide precursors, namely the
benzyl alcohol methyl 2-hydroxy mandelate (10) and its acetate, 20-acetoxy-2-hydroxyphenylacetate (11).
Irradiation of an a-diazocarbonyl compound, 1, usually
leads to a Wolff rearrangement [eqn. (1)] in which loss of nitro-
gen and migration of the carbonyl group substituent X pro-
duces a ketene, 2.1
We now report that still another reaction route becomes
available when an ortho hydroxyl group is introduced into
the benzene ring of methyl phenyldiazoacetate. Irradiation of
methyl 2-hydroxyphenyldiazoacetate, 8 [eqn. (4)] in aqueous
solution produces methyl 2-hydroxymandelate, 10, the same
final product that would have been obtained if a reaction path
analogous to the carbonylcarbene route of eqn. (3) had been
followed.
ð1Þ
However, when the migratory aptitude of X is poor, another
process [eqn. (2)] takes over, in which loss of nitrogen gener-
ates a carbonylcarbene, 3, that, in aqueous solution, undergoes
conjugate addition of water giving an enol, 4.2
ð4Þ
ð2Þ
Now, however, the reaction intermediate observed is not an
enol but is rather the quinone methide 9. We have demon-
strated that this is so by monitoring this intermediate and
showing that it decays with kinetics characteristic of qui-
none methide hydration, with rate constants identical to those
We have, for example, observed this process in the case of methyl
phenyldiazoacetate, 5, where we were able to substantiate this
reaction route [eqn. (3)] by observing the enol intermediate,
6, and characterizing its ketonization to methyl mandelate, 7.3
obtained using quinone methide
9 generated by more
traditional routes,4 namely photolysis of the corresponding
benzyl alcohol, methyl 2-hydroxymandelate, 10, and its
acetate, methyl 20-acetoxy-2-hydroxyphenylacetate, 11 [eqn.
(5)]. Because these quinone methide hydration reactions are
fast, we used flash photolytic techniques to monitor their
course.
ð3Þ
ð5Þ
There is some precedent for the generation of quinone
methides by irradiation of suitably constituted diazo com-
pounds: photolysis of the ‘‘masked’’ diazo substrate 12 in an
argon matrix at 10 K was shown to give o-quinone methide
14 through the diazo compound 13 [eqn. (6)].5
y Dedicated to Professor Dr Z. R. Grabowski and Professor Dr
J. Wirz on the occasions of their 75th and 60th birthdays.
z Electronic supplementary information (ESI) available. Tables S1–S4
of rate data. See http://www.rsc.org/suppdata/cp/b2/b211444p/
DOI: 10.1039/b211444p
Phys. Chem. Chem. Phys., 2003, 5, 1039–1042
This journal is # The Owner Societies 2003
1039

View Article Online
produced a minor base-line drift, and in these cases a linear
term was added to the exponential function.
Results
Product study. HPLC analyses of spent reaction mixtures
were performed for flash photolyses of methyl 2-hydroxyphe-
nyldiazoacetate (8) carried out in 1 ꢁ 10ꢂ3 M perchloric acid
and sodium hydroxide solutions and in acetic acid ([H+] ¼
3 ꢁ 10ꢂ5 M) and biphosphate ion ([H+] ¼ 2 ꢁ 10ꢂ7 M) buffers.
Methyl 2-hydroxymandelate (10) was found to be by far the
major product formed ( > 95%) in the perchloric acid solution
and in the acetic acid and biphosphate ion buffers, as expected
for the quinone methide generation and hydration route of
eqn. (4). In the sodium hydroxide solution, however, another
process took over: no methyl 2-hydroxymandelate could be
detected and only another, unidentified substance was found.
Rate data obtained in basic solutions, moreover, no longer
conformed to a first-order rate law. This behavior is similar
to our experience with o-quinone methide itself in basic solu-
tion,4b and it can be attributed to further reaction of the qui-
none methide transient with its phenol hydration product,
which, in basic solution, exists as the very reactive phenoxide
ion.11
ð6Þ
Experimental
Materials. Methyl 2-hydroxyphenyldiazoacetate, 8, was
synthesized by using triethylamine to cleave the tosyl hydra-
zone of methyl 2-hydroxyphenylglyoxylate (Bamford–Stevens
reaction).6 This hydrazone was made from tosyl azide and
methyl 2-hydroxyphenylglyoxylate,7 obtained by acid-cata-
lyzed esterification of 2-hydroxyphenylglyoxylic acid, itself
prepared by selenium dioxide oxidation of 2-hydroxy-
acetophenone.8 This synthesis produced methyl 2-hydroxy-
phenyldiazoacetate mixed with its ring-closed counterpart,
3-diazobenzofuran-2-one, but the two substances could be
separated by column chromatography (silica gel with 4 : 1 hex-
anes : diethyl ether as the eluent). The diazoester was obtained
as yellow needles, mp 88–90ꢀ, with the following spectral prop-
1
erties: H NMR (CDCl3): d/ppm ¼ 8.27 (br s, 1H), 7.26–7.17
(m, 1H), 7.05–6.91 (m, 3H), 3.92 (s, H); 13C NMR (CDCl3):
d/ppm ¼ 154.0, 129.4 (2 C), 126.1, 121.1 (2C), 119.5, 111.6,
53.4; HRMS: m/e ¼ 192.0535 (calcd. for C9H8N2O3),
192.0536 (found).
Kinetics. Rates of decay of o-quinone a-carbomethoxy-
methide, 9, were measured in dilute aqueous (H2O and D2O)
solutions of perchloric acid as well as acetic acid and bipho-
sphate ion buffers. The ionic strength of these solutions was
maintained at 0.10 M through the addition of sodium perchlo-
rate as needed. These data are summarized in Tables S1 and S2
of the ESI.z
Methyl 2-hydroxymandelate,9 10, was prepared by acid-cat-
alyzed esterification of 2-hydroxymandelic acid, itself made
by sodium borohydride reduction of 2-hydroxyphenylglyoxylic
acid. As reported,8,9 2-hydroxymandelic acid could not be
obtained completely free of solvent, and it was therefore used
as an ethyl acetate solvate.
The rate measurements in buffers were made in series of
solutions of fixed buffer ratio, and therefore fixed hydronium
ion concentration, but varying total buffer concentrations.
Buffer catalysis was weak, but observed first-order rate con-
stants did increase with increasing buffer concentration. The
data were therefore analyzed by least-squares fitting of the buf-
fer dilution expression shown as [eqn. (7)].
Methyl 20-acetoxy-2-hydroxyphenylacetate, 11, was prepared
by treating methyl 2-hydroxymandelate with equimolar
amounts of acetic anhydride and boron trifluoride diethyl ethe-
rate. The product was purified by column chromatography
(silica gel with 10.1 hexanes : diethyl ether as the eluent) to give
a slightly yellow liquid with the following spectral properties:
1H NMR (CDCl3): d/ppm ¼ 7.31–7.23 (m, 3H), 6.94–6.90
(m, 2H), 6.26 (s, 1H), 3.74 (s, 3H), 2.17 (s, 3H); 13C NMR
(CDCl3); d/ppm ¼ 171.2, 170.8, 155.1, 131.3, 129.9, 120.9,
120.1, 117.3, 71.2, 53.3, 20.9; HRMS: m/e ¼ 224.0685 (calcd.
for C11H12O5) 224.0685 (found).
kobs ¼ kint þ kbuff ½bufferꢄ
ð7Þ
The zero-buffer-concentration intercepts, kint , obtained in this
way, together with the perchloric acid data, were then used to
construct the rate profiles shown in Fig. 1. Hydronium ion
concentrations, written here as [L+], of the buffer solutions
needed for this purpose were obtained by calculation using
All other materials were best available commercial grades.
Product analyses. Product compositions were determined by
HPLC using a Varian Vista 5500 instrument with a NovoPak
C18 reverse-phase column and methanol–water (50 : 50 v/v) as
the eluent. Reaction solutions containing substrate at similar
concentrations as used for kinetic measurements (2–3 ꢁ 10ꢂ5
M) were subjected to a single pulse from our microsecond flash
photolysis system.10 Products were identified by comparing
retention times and UV spectra with those of authentic
samples.
Kinetics. Rate measurements were made using a microse-
cond flash photolysis system that has already been described,10
and reactions were followed by monitoring absorbance decay
at l ¼ 420 nm. Initial substrate concentrations were 2–3 ꢁ
10ꢂ5 M, and the temperature of all reacting solutions was con-
trolled at 25.0 ꢃ 0.05 ꢀC. Observed first-order rate constants
were obtained for the most part by least-squares fitting of an
exponential function; in some of the slower runs, however,
instability of the monitoring light source power supply
Fig. 1 Rate profiles for the decay of o-quinone a-carbomethoxy-
methide, 9, in (O) H2O, and (D) D2O, solution at 25 ꢀC.
1040
Phys. Chem. Chem. Phys., 2003, 5, 1039–1042

View Article Online
thermodynamic acidity constants of the buffer acids from the
literature and activity coefficients recommended by Bates.12
These rate profiles show acid-catalyzed and uncatalyzed
portions; the data were therefore analyzed using the rate law
of [eqn. (8)]. Least squares fitting gave the results
component and is consequently normal (kH/kD > 1) rather
than inverse; that on the ketonization of mandelic acid enol,
for example, is kH+/kD+ ¼ 3.3.15
The form of buffer catalysis is also different for quinone
methide decay from that for enol ketonization. In the quinone
methide reaction, the buffer base acts as a nucleophile, adding
directly to the methide methylene group.4b,d The strength of
the buffer reaction therefore depends on the nucleophilic
strength of the buffer base, and because acetate ion is a weak
nucleophile, buffer catalysis in acetic acid buffers is weak, as
observed here for o-quinone a-carbomethoxy-methide. Enol
ketonization, on the other hand, being a rate-determining
hydron transfer reaction, is catalyzed by buffers with buffer
catalysis increasing in proportion to the acid strength of the
buffer acid. Since acetic acid is a moderately strong general
acid, it is a good catalyst, and enol ketonization is catalyzed
strongly by acetic acid buffers, unlike the weak catalysis
observed here for o-quinone a-carbomethoxymethide. The
absence of general acid catalysis found here is, of course, also
consistent with a quinone methide reaction and not with an
enol ketonization.
þ
þ
kprof ¼ kuc þ kL ½L ꢄ
ð8Þ
kH+ ¼ (3.14 ꢃ 0.05) ꢁ 102 Mꢂ1
s
ꢂ1, kH+/kD+ ¼ 0.391 ꢃ 0.009,
(kuc)H O ¼ (5.40 ꢃ 0.08) ꢁ 10ꢂ1sꢂ1,(kuc)H O/(kuc)D O ¼ 1.22 ꢃ 0.03.
2
2
2
The slopes of the buffer dilution plots, kbuff , were separated
into their general acid, kHA , and general base, kB , components
with the aid of the relationship shown as [eqn. (9)], in which fA
is the fraction of buffer present in the acid form.
kbuff ¼ kB þ ðkHA ꢂ kBÞf A
ð9Þ
Least squares analysis showed that in the acetic acid solutions
the buffer reaction was wholly of the general base type, in both
H2O and D2O, with (kB)H O ¼ (2.27 ꢃ 0.14) ꢁ 101 Mꢂ1 sꢂ1 and
2
(kB)H O/(kB)D O ¼ 0.92 ꢃ 0.08. A similar analysis could not be
2
2
carried out for the biphosphate buffer solutions because mea-
surements were made at only one buffer ratio.
Additional support for identification of the presently
observed intermediate species as a quinone methide rather
than an enol comes from its UV absorbance. Quinone
methides have a cyclohexadienone structure which produces
a relatively long wavelength absorption band. o-Quinone
methide itself, for example, has lmax ffi 400 nm, not unlike
lmax ¼ 420 nm observed for the present intermediate species.
b-Phenylsubstituted enols, such as that which might be formed
from methyl 2-hydroxyphenyldiazoacetate, will have a styrene-
type chromophore with an absorption band at considerably
lower wavelengths; for example, lmax ¼ 268 nm for b,b-
dimethoxystyrene,16 and lmax ¼ 275 for the enol of methyl
mandelate.3
Rates of o-quinone a-carbomethoxymethide, 9, decay were
also measured using methyl 2-hydroxymandelate, 10, and
methyl 20-acetoxy-2-hydroxyphenylacetate, 11, as the photo-
chemical substrates. These measurements were made in dilute
aqueous (H2O) perchloric acid only, and the ionic strength
was maintained at 0.10 M by adding sodium perchlorate as
required. These data are summarized in Tables S3 and S4 of
the ESI.z
Discussion
We have shown in previous studies4b–c that the hydration of
quinone methides in aqueous solution occurs by acid-catalyzed
and uncatalyzed routes. The ketonization of enols, such as
those formed by conjugate addition of water to a-carbonyl-
carbenes in the photolysis of a-carbonyldiazo compounds
illustrated in eqn. (3), also occurs by acid-catalyzed and unca-
talyzed routes. The mechanisms of the two reactions, however,
are quite different, and that gives them distinctive kinetic signa-
tures by which they can easily be distinguished.
Quinone methide hydration catalyzed by the hydronium ion
occurs through rapid pre-equilibrium protonation of the qui-
none carbonyl oxygen atom followed by rate-determining cap-
ture of the carbocation so formed by a water molecule [eqn.
(10)].4b–c
Perhaps the strongest evidence of all supporting assignment
of a quinone methide structure to the intermediate species
observed here upon photolysis of methyl 2-hydroxyphenyldia-
zoacetate, 8, comes from the results obtained by flash photoly-
sis of methyl 2-hydroxymandelate, 10, and its acetate
derivative, methyl 20-acetoxy-2-hydroxyphenylacetate, 11.
Benzyl alcohols such as 10 are probably the most commonly
used substrates for photochemical generation of quinone
methides, and their acetate derivatives, such as 11, are even
better quinone methide precursors.4 It is not surprising, there-
fore, that flash photolysis of the benzyl alcohol 10 and its acet-
ate 11 produced a short-lived transient species with a UV
spectral band at lmax ¼ 420 nm, just like the transient
observed upon flash photolysis of the diazo substrate 8. Fig. 2
ð10Þ
Because the oxygen–hydrogen bonds of the water molecule
formed in the pre-equilibrium step are tighter than those of
the hydronium ion reactant,13 this produces an inverse solvent
isotope effect, just like that, kH+/kD+ ¼ 0.39 observed here.
The corresponding isotope effect on the hydration of the par-
ent o-quinone methide itself is kH+/kD+ ¼ 0.42.4b The hydro-
nium-ion catalyzed ketonization of enols, on the other hand,
occurs by rate-determining hydron transfer from the hydro-
nium ion to the b-carbon atom of the enol.14 The solvent
isotope effect on this process therefore contains a primary
Fig. 2 Comparison of rates of decay of o-quinone a-carbomethoxy-
methide generated from methyl 2-hydroxyphenyldiazoacetate, 8, (—);
methyl 2-hydroxymandelate, 10, (O); and methyl 20-acetoxy-2-hydro-
xyphenylacetate, 11, (D) in aqueous solution at 25 ꢀC.
Phys. Chem. Chem. Phys., 2003, 5, 1039–1042
1041

View Article Online
C. Z. Chen, J. Cole, M. Lukeman and M. Xu, Pure Appl. Chem.,
2001, 73, 529–534; (b) Y. Chiang, A. J. Kresge and Y. Zhu, J. Am.
Chem. Soc., 2001, 123, 8089–8094; (c) Y. Chiang, A. J. Kresge and
Y. Zhu, J. Am. Chem. Soc., 2002, 124, 717–722; (d ) Y. Chiang,
A. J. Kresge and Y. Zhu, J. Am. Chem. Soc., 2002, 124, 6349–
6356; (e) Y. Chiang and A. J. Kresge, Photochem. Photobiol.
Sci., 2002, 1, 57–70.
H. Tomioka and T. Matsushita, Chem. Lett., 1997, 399–400.
M. Regitz and G. Maas, Diazo Compounds. Properties and Synth-
esis, Academic Press, New York, 1986, ch. 9.
W. M. Horspool and G. D. Khandelwal, J. Chem. Soc. C, 1971,
3328–3331.
R. Howe, B. S. Rao and H. Heyneker, J. Chem. Soc. C, 1967,
2510–2514.
A. J. Hoefnagel, J. A. Peters and H. van Bekkum, Recl. Trav.
Chim. Pays-Bas., 1996, 115, 353–356.
shows, moreover, that the rate constants for decay of the tran-
sients obtained from all three substrates have the same numer-
ical values. The line shown in this figure was drawn using
parameters obtained from the least squares analysis of mea-
surements made using diazo substrate 8, whereas the circles
and triangles represent measurements made with the benzyl
alcohol substrate 10 and its acetate 11, respectively. It may
be seen that the data obtained with 10 and 11 fit the line repre-
senting 8 as well as do the 8 data points themselves displayed in
Fig. 1.
5
6
7
8
9
Acknowledgements
10 Y. Chiang, M. Hojatti, J. R. Keeffe, A. J. Kresge, N. P. Schepp
and J. Wirz, J. Am. Chem. Soc., 1987, 109, 4000–4009.
11 P. Wan and D. Hennig, J. Chem. Soc., Chem. Comm., 1987,
939–941.
12 R. G. Bates, Determination of pH Theory and Practice, Wiley,
New York, 1973, pp. 49.
We are grateful to the Natural Sciences and Engineering
Research Council of Canada for financial support of this
work. We would like to thank Professor Dr Jakob Wirz for
teaching us how to do flash photolysis.
13 (a) A. J. Kresge, R. A. More O’Ferrall and M. F. Powell, in
Isotopes in Organic Chemistry, ed. E. Buncel and C. C. Lee,
Elsevier, New York, 1987, vol. 7, ch. 4; (b) J. R. Keeffe and
A. J. Kresge, in Investigation of Rates and Mechanisms of
Reactions, ed., C. F. Bernasconi, Wiley, New York, 1986,
pp. 763–764.
14 J. R. Keeffe and A. J. Kresge, in The Chemistry of Enols, ed.
Z. Rapoport, Wiley, New York, 1990, ch. 7.
15 Y. Chiang, A. J. Kresge, V. V. Popik and N. P. Schepp, J. Am.
Chem. Soc., 1997, 119, 10 203–10 212.
References
1
2
3
4
See, e.g. M. Regitz and G. Maas, Diazo Compounds. Properties
and Synthesis, Academic Press, New York, 1986, pp. 185–195.
Y. Chiang, E. A. Jefferson, A. J. Kresge, V. V. Popik and R.-Q.
Xie, J. Phys. Org. Chem., 1998, 11, 610–613.
Y. Chiang, A. J. Kresge, N. P. Schepp and R.-Q. Xie, J. Org.
Chem., 2000, 65, 1175–1180.
(a) P. Wan, B. Barker, L. Diao, M. Fischer, Y. Shi and C. Yang,
Can. J. Chem., 1996, 74, 465–475; P. Wan, D. W. Brousmiche,
16 T. Okuyama, S. Kawao and T. Fueno, J. Org. Chem., 1981, 46,
4372–4375.
1042
Phys. Chem. Chem. Phys., 2003, 5, 1039–1042