pubs.acs.

org/joc

Dual Reactivity of Hydroxy- and Methoxy- Substituted o-Quinone

Methides in Aqueous Solutions: Hydration versus Tautomerization.
Selvanathan Arumugam and Vladimir V. Popik*
Department of Chemistry, University of Georgia, Athens, Georgia 30602, United States

vpopik@chem.uga.edu
Received August 17, 2010

4-Hydroxy-6-methylene-2,4-cyclohexadien-1-one (1) and 4-methoxy-6-methylene-2,4-cyclohexadien-1-one

(2) were generated by efficient (Φ = 0.3) photodehydration of 2-(hydroxymethyl)benzene-1,4-diol (3a) and
2-(hydroxymethyl)-4-methoxyphenol (4a), respectively. o-Quinone methides 1 and 2 can be quantitatively
trapped as Diels-Alder adducts with ethyl vinyl ether or intercepted by good nucleophiles, such as azide ion
(kN3(1) = 3.15
104 M-1 s-1 and kN3(2) = 3.30
104 M-1 s-1). In aqueous solution, o-quinone methide 2
rapidly adds water to regenerate starting material (τH2O(2) = 7.8 ms at 25 °C). This reaction is catalyzed by
specific acid (kHþ(2) = 8.37
103 s-1 M-1) and specific base (kOH-(2) = 1.08
104 s-1 M-1) but shows no
significant general acid/base catalysis. In sharp contrast, o-quinone methide 1 decays (τH2O(1) = 3.3 ms at
25 °C) via two competing pathways: nucleophilic hydration to form starting material 3a and tautomeriza-
tion to produce methyl-p-benzoquinone. The disappearance of 1 shows not only specific acid (kHþ(1) =
3.30
104 s-1 M-1) and specific base catalysis (kOH-(1) = 3.51
104 s-1 M-1) but pronounced catalysis by
general acids and bases as well. The o-quinone methides 1 and 2 were also generated by the photolysis of
2-(ethoxymethyl)benzene-1,4-diol (3b) and 2-(ethoxymethyl)-4-methoxyphenol (4b), as well as from (2,5-
dihydroxy-1-phenyl)methyl- (3c) and (2-hydroxy-5-methoxy-1-phenyl)methyltrimethylammonium iodides
(4c). Short-lived (τ25°C ≈ 20 μs) precursors of o-quinone methides 1 and 2 were detected in the laser flash
photolysis of 3a,b and 4a,b. On the basis of their reactivity, benzoxete structures have been assigned to these
intermediates.

Introduction biological processes.1-3 The chemical behavior of o-QMs

resembles that of R,β-unsaturated ketones. However, the
o-Quinone methides (o-QMs) are very reactive species that
zwitterionic resonance form in the former is additionally
have been implicated as intermediates in many chemical and
stabilized by aromatic conjugation, increasing o-QMs polar-
ity and enhancing their reactivity.1,4 o-QMs react very
(1) Quinone Methides Rokita, S. E., Ed.; ; Wiley Series of Reactive
Intermediates in Chemistry and Biology, Vol. 1; Wiley: Hoboken, NJ, 2009. rapidly with nucleophiles and undergo efficient Diels-Alder
(2) (a) Wan, P.; Brousmiche, D. W.; Chen, C. Z.; Cole, J.; Lukeman, M.; cycloaddition with electron-rich olefins.1,4 It has been
Xu, M. Pure Appl. Chem. 2001, 73, 529. (b) Wan, P.; Barker, B.; Diao, L.; demonstrated that o-QMs are efficient dDNA alkylating
Fischer, M.; Shi, Y.; Yang, C. Can. J. Chem. 1996, 74, 465. (c) Petet, M. G.
Angew. Chem., Int. Ed. 1982, 94, 376. (d) Berson, , J. In The chemistry of the and cross-linking agents,1,5 and are believed to be res-
quinonoid compounds; Patai, S, Rappoport, Z., Eds.; Wiley: Chichester, UK, ponsible for the cytotoxicity of antitumor antibiotics of mitomy-
1988; Vol. 2, Part 1, pp 455-536; (e) Van De Water, R. W.; Pettus, T. R. R. cin C and anthracycline families.6
Tetrahedron 2002, 58, 5367. (f) Fischer, M.; Shi, Y.; Zhao, B.-P.; Snieckus,
V.; Wan, P. Can. J. Chem. 1999, 77, 868. (g) Bulger, P. G.; Bagal, S. K.;
Marquez, R. Nat. Prod. Rep. 2008, 25, 254 and references cited therein. (4) (a) Leo, E. A.; Delgado, J.; Domingo, L. R.; Espinos, A.; Miranda,
(3) (a) Freccero, M. Mini Rev. Org. Chem. 2004, 1, 403. Lignins: Occurrence, M. A.; Tormos, R. J. Org. Chem. 2003, 68, 9643. (b) Chiang, Y.; Kresge,
Formation, Structure and Reactions; Sarkanen, K. V., Ludving, C., Eds.; Wiley: A. J.; Zhu, Y. J. Am. Chem. Soc. 2000, 122, 9854. (c) Diao, L.; Yang, C.; Wan,
New York, 1971. (b) Stowell, J. K.; Widlanski, T. S.; Kutateladze, T. G.; Raines, P. J. Am. Chem. Soc. 1995, 117, 5369. (d) Wang, H.; Wang, Y.; Han, K.-L.;
R. T. J. Org. Chem. 1995, 60, 6930. Peng, X.-J. J. Org. Chem. 2005, 70, 4910.

7338 J. Org. Chem. 2010, 75, 7338–7346 Published on Web 10/06/2010 DOI: 10.1021/jo101613t
r 2010 American Chemical Society
Arumugam and Popik
JOC Article
o-QMs can be efficiently generated by photodehydration
of o-hydroxybenzyl alcohol derivatives.2a,4a-c The enhanced
acidity of phenols in the excited state facilitates intramole-
cular proton transfer (ESIPT)7 of the phenolic proton to the
oxygen atom in benzylic position.4a-c C-O bond heterolysis
can be concerted with ESIPT or the loss of water might occur
in the ground state after proton transfer is complete.8 In
either case, the formation of o-QMs is usually complete
within a nanosecond pulse.4a-c o-QMs also can be generated
by photochemical elimination of ammonia or amines from
o-hydroxybenzylamines.9
The major reaction of o-QMs in aqueous solutions is rapid
addition of water producing o-hydroxybenzyl alcohol deriva-
tives. Efficacy enhancement of o-QM-based antitumor agents,1
as well as development of o-hydroxybenzyl photolabile protect- FIGURE 1. UV spectra of ca. 10-5 M aqueous solutions of 3a
ing groups,10 requires a better understanding of o-QM behavior (solid line) and 4a (dashed line).
in this medium. However, only kinetics of hydration of the
parent o-QM, 6-methylene-2,4-cyclohexadien-1-one, has been
investigated in detail.4b-d Little is known about the influence of
the electronic properties of substituents on the reactivity of
oQMs.4a,5d,11 Our recent studies of o-naphthoquinone methides
have demonstrated that the presence of an additional electron-
rich aromatic ring causes dramatic changes in the mechanism of
formation and reactivity of o-QM.11 These results prompted us
to investigate the effect of electron-donating substituents on the
dynamics of o-quinone methide hydration. In the present report
we discuss the photochemical generation and reactivity of
electron-rich o-quinone methides 1 and 2 (Scheme 1).
SCHEME 1

FIGURE 2. Emission spectra at λex = 266 nm of ca. 10-5 M

aqueous solutions of 3a (dashed line) and 4a (solid line).

from (2,5-dihydroxy-1-phenyl)methyl- (3c) and (2-hydroxy-5-

methoxy-1-phenyl)methyltrimethylammonium iodides (4c).12
Results and Discussion The formation and reactions of o-QMs in aqueous solutions
were monitored with a nanosecond kinetic spectrometer
o-QMs 1 and 2 are conveniently generated by the photolysis equipped with pulsed Nd:YAG laser.12 The product analysis
of 2-(hydroxymethyl)benzene-1,4-diol (3a) and 2-(hydroxy- of photochemical reactions of 3a-c and 4a-c was conducted by
methyl)-4-methoxyphenol (4a); 2-(ethoxymethyl) benzene-1,4- HPLC, using individual compounds isolated from preparative
diol (3b) and 2-(ethoxymethyl)-4-methoxyphenol (4b); as well as scale photolyses as references.
Photophysical Properties and Photochemical Reactivity of
(5) (a) Wang, P.; Song, Y.; Zhang, L. X.; He, H. P.; Zhou, X. Curr. Med.
Chem. 2005, 12, 2893. (b) Freccero, M.; Di Valentin, C.; Sarzi-Amade, M.
o-QM Precursors. UV spectra of 2-(hydroxymethyl)ben-
J. Am. Chem. Soc. 2003, 125, 3544. (c) Weng, X.; Ren, L.; Weng, L.; Huang, zene-1,4-diol (3a) and 2-(hydroxymethyl)-4-methoxyphenol
J.; Zhu, S.; Zhou, X.; Weng, L. Angew. Chem., Int. Ed. 2007, 46, 8020. (d) (4a) are very similar (Figure 1) and red-shifted by 25-30 nm
Weinert, E. E.; Dondi, R.; Colloredo-Melz, S.; Frankenfield, K. N.; Mitchell,
C. H.; Freccero, M.; Rokita, S. E. J. Am. Chem. Soc. 2006, 128, 11940. (e) compared to the spectrum of o-hydroxybenzyl alcohol.13 A
Wang, P.; Liu, R.; Wu, X.; Ma, H.; Cao, X.; Zhou, P.; Zhang, J.; Weng, X.; major absorption band of 3a lies at 295 nm (log ε = 3.91) and
Zhang, X.-L.; Qi, J.; Zhou, X.; Weng, L. J. Am. Chem. Soc. 2003, 125, 1116. of 4a at 300 nm (log ε = 3.97). Both o-QM precursors are
(f) Richter, S. N.; Maggi, S.; Mels, S. C.; Palumbo, M.; Freccero, M. J. Am.
Chem. Soc. 2004, 126, 13973. (g) Colloredo-Mels, S.; Doria, F.; Verga, D.; fluorescent showing emission band with λmax at 333 (3a) and
Freccero, M. J. Org. Chem. 2006, 71, 3889. 336 nm (4a) (Figure 2). The quantum yield of fluorescence is
(6) (a) Han, I.; Russell, D. J.; Kohn, H. J. Org. Chem. 1992, 57, 1799. Li,
V.-S.; Kohn, H. J. Am. Chem. Soc. 1991, 113, 275. (b) Tomasz, M.; Das, A.;
higher for 2-(hydroxymethyl)-4-methoxyphenol (4a, ΦFl =
Tang, K. S.; Ford, M. G. J.; Minnock, A.; Musser, S.; Waring, M. J. J. Am. 0.16 ( 0.01) than for 3a (ΦFl = 0.050 ( 0.002).12,14 This
Chem. Soc. 1998, 120, 11581. (c) Egholm, M.; Koch, T. H. J. Am. Chem. Soc. emission hinders monitoring of laser flash-induced transfor-
1989, 111, 8291. (d) Gaudiano, G.; Frigerio, M.; Bravo, P.; Koch, T. H. J. Am.
Chem. Soc. 1990, 112, 6704. mation of 3a and 4a at shorter wavelengths, but the intensity
(7) (a) Lahiri, S. C. J. Sci. Ind. Res. 1979, 38, 492. Van der Donckt, E. Prog. of fluorescence dies off above 400 nm and thereby allows us
React. Kinet. 1970, 5, 273. (b) Ireland, J. F.; Wyatt, P. A. H. Adv. Phys. Org. Chem.
1976, 12, 131. (c) Shizuka, H.; Tobita, S. Mol. Supramol. Photochem. 2006, 14, 37.
(d) Tolbert, L. M.; Solntsev, K. M. Acc. Chem. Res. 2002, 35, 19. (12) See the Supporting Information.
(8) (a) Wan, P.; Barker, B.; Diao, L.; Fischer, M.; Shi, Y.; Yang, C. Can. J. (13) Kammerer, B.; Kahlich, R.; Biegert, C.; Gleiter, C. H.; Heide, L.
Chem. 1995, 74, 465. (b) Yijian, S.; Wan, P. Can. J. Chem. 2005, 83, 1306. Phytochem. Anal. 2005, 16, 470–478.
(9) Nakatani, K.; Higashida, N.; Saito, I. Tetrahedron Lett. 1997, 38, 5005. (14) (a) Wang, Y.-H.; Zhang, H.-M.; Liu, L.; Liang, Z.-H.; Guo, O.-X.;
(10) (a) Kostikov, A.; Popik, V. V. Org. Lett. 2008, 10, 5277. (b) Tung, C.-H.; Inoue, Y.; Liu, Y.-C. J. Org. Chem. 2002, 67, 2429. (b) Berlman,
Kostikov, A.; Popik, V. V. J. Org. Chem. 2007, 72, 9190. I. B. Handbook of Fluorescence Spectra of Aromatic Molecules; Academic
(11) Arumugam, S.; Popik., V. V. J. Am. Chem. Soc. 2009, 131, 11892. Press: New York, 1971; p 473.

J. Org. Chem. Vol. 75, No. 21, 2010 7339

JOC Article Arumugam and Popik

to monitor the formation and the decay of o-QMs 1 and 2 at SCHEME 3

wavelengths >400 nm.
Prolonged irradiation of an aqueous solution of o-QM
precursor 3a (pH 6.96 ( 0.06) with 300 nm light yielded
methyl-p-benzoquinone 5 as the only photoproduct. The
quantum yield of the formation of 5 is Φ3a = 0.060 ( 0.005.
Since competing hydration of the o-QM 1 regenerates the
starting material 3a and thus is undetectable, we have explored
the aqueous photochemistry of the corresponding ethyl ether
This observation suggests that the secondary photoproducts are
3b. Low conversion (∼15%) photolysis of 3b in aqueous
most likely o-QM oligomers, which are trapped on the column.15
biphosphate buffer solutions at pH 6.96 ( 0.06 yielded diol
o-Hydroxybenzyltrimethylammonium salts are also known
3a as the major product (76((2)%) and 2-methyl-1,4-benzo-
to generate o-QMs upon irradiation.16 To compare the reactiv-
quinone (5) as the minor product (19((2)%) with quantum
ity of o-QM generated from an alternative source, we have
efficiency Φ3b = 0.31 ( 0.01 (Scheme 2, Table 1).
studied the photochemistry of (2,5-dihydroxybenzyl)trimethyl-
SCHEME 2 ammonium iodide (3c) and (2-hydroxy-5-methoxybenzyl)tri-
methylammonium iodide (4c). 300 nm irradiation of the
ammonium salt 4c in aqueous solution cleanly produces the diol
4a. As in the case of ethyl ether 4b, low conversion photolysis of
4c gave a quantitative yield of the product, while at high con-
version the yield dropped to 76% without formation of detect-
able byproduct (Table 1). Irradiation of 3c in aqueous bipho-
sphate buffer solutions at pH 7.0 yielded both diol 3a and
p-quinone 5. The ratio of hydration to ketonization products
was similar to that obtained in the photolysis of the ether 3b
TABLE 1. Photolysis of o-QM precursors 3a-c and 4b,c in Aqueous (Table 1). This observation provides additional support for the
Biphosphate Buffer Solutions (pH 6.96 ( 0.6) mechanism of formation of 3a and 5 from o-QM 1.
o-QM yield of 3a or yield of 5/% yield of 6 or It is interesting to note that the ratio of hydration to
precursor 4a/% (conversion)a,b (conversion)a,b 7/% (conversion)a,c,d ketonization products (3a/5) depends on the acidity of the
3a N/A 96 ( 2 (10%) 94 ( 1 (93%) solution. Thus, diol 3a is the major product of the photolysis
3b 76 ( 2 (15%) 19 ( 2 (15%) 93 ( 1 (99%) of 3b at neutral pH (Table 1) and in aqueous perchloric acid
31 ( 2 (81%) 59 ( 2 (81%) at or below 10-3 M concentration (Table 2). At higher
3c 72 ( 1 (12%) 22 ( 2 (15%) 95 ( 1 (98%) acidities, ketonization to produce p-quinone 5 becomes the
37 ( 3 (75%) 57 ( 2 (81%)
4b 94 ( 3 (16%) -e 92 ( 2 (98%) predominant pathway of o-QM 1 decay. o-QM 2 in aqueous
73 ( 3 (80%) solutions produces only diol 4a irrespective of the media
4c 96 ( 2 (15%) -e 96 ( 1 (99%) acidity. No new products were detected in the photolyses of
76 ( 3 (82%) 3b and 4b in perchloric acid solutions.
a
λirr = 300 nm. bCa. 3
10-4 M in water. cCa. 3
10-4 M solutions of
the substrate; ca. 0.03 M ethyl vinyl ether. dIn 50% CH3CNaq. eNot TABLE 2. Photolysis of 3b and 4b in Aqueous Perchloric Acid Solutions
detected. o-QM yield of 3a or yield of 5/%
precursor [HClO4]/M 4a/% (conversion) (conversion)
At higher conversion, the chemical yield of 5 increases as the 3b 0.001 64 ( 2 (20%) 34 ( 2 (20%)
primary product, diol 3a, is also photoactive. The chemical 3b 0.01 34 ( 2 (20%) 63 ( 2 (20%)
yields of the products upon 300 nm irradiation of 3b are near 3b 0.05 23 ( 2 (22%) 75 ( 2 (22%)
quantitative (∼90%) even at higher conversion (>80%) as the 3b 0.1 15 ( 1 (21%) 80 ( 2 (21%)
4b 0.001 94 ( 3 (15%) not detected
benzoquinone product 5 does not have significant absorption at
4b 0.1 92 ( 3 (11%) not detected
the irradiation wavelength (Table 1).
Photolysis of aqueous solutions of o-QM precursor 4a with
Significant difference in electrophilicity between o-QMs 1 and
300 nm light did not produce any detectable amounts of new
2 becomes evident when these species are generated in aqueous
products due to the rapid and efficient rehydration of QM 2 to
acetate buffer. Photolysis of 3b in aqueous 0.1 M acetate buffer
yield back the starting material. On the other hand, irradiation of
(pH 4.57 ( 0.05) produced the same products as in biphosphate
the aqueous solution of ethyl ether 4b resulted in a rapid con-
buffer solutions, i.e., diol 3a and p-quinone 5. The product ratio,
sumption of the substrates and the formation of 4a (Scheme 3).
however, is reversed and 5 is a major product (Table 3). No
The quantum yield for the photoelimination of ethanol from
detectable amounts of new products were observed. Irradia-
4b at 300 nm is Φ4b=0.28 ( 0.01.12 The chemical yield of 4a pro-
tion of ether 4b under the same conditions, on the other hand,
duced in this reaction is almost quantitative at low conversions,
produced substantial yield of 2-(acetoxymethyl)-4-methoxyphe-
but is somewhat reduced at longer irradiation times, apparently
nol (4d, Scheme 3, Table 3). The latter is apparently formed by
due to secondary photochemical processes (Table 1). However,
the attack of acetate ion on the electron-deficient methide car-
no new photoproducts were detected by HPLC or isolated by
bon atom. Similar reactivity was also observed in the case of the
flash chromatography, even after prolonged irradiation of 4a.
parent o-QM.17

(15) (a) Itoh, T. Prog. Polym. Sci. 2001, 26, 1019. (b) Dolenc, J.; Sket, B.; (16) Modica, E.; Zanaletti, R.; Freccero, M.; Mella, M. J. Org. Chem.
Strlic, M. Tetrahedron Lett. 2002, 43, 5669. 2001, 66, 41.

7340 J. Org. Chem. Vol. 75, No. 21, 2010

Arumugam and Popik
JOC Article

FIGURE 3. Transient spectra obtained at 5 μs (dashed line) and 1 ms (solid line) after the laser pulse in photolysis of ca. 0.1 mM aqueous
solutions of 3a (A) and 4a (B) at pH 7.0.

TABLE 3. Photolysis of o-QM Precursors 3b and 4b in Aqueous sients, which were subsequently identified as o-QMs 1 and 2 (vide
Acetate Ion Buffer Solutions (0.1M buffer, pH 4.57( 0.05) infra), decay at a relatively slower rate (τ = 3-8 ms). Since the
QM yield of 3a or yield of 4d/% yield of 5/% formation and the decay of o-QMs proceed at very different
precursor 4a/% (conversion)a (conversion)a (conversion)a
rates, these processes were recorded in separate experiments with
3b 27 ( 2 (29%) no acetylation 70 ( 2 (29%) use of different time scales and signal amplifications. The experi-
product mental data were fitted separately for each transient to a double
4b 66 ( 2 (13%) 31 ( 2 (13%) not detected
exponential function (Figure 4) and the first order rate constants
a
λirr = 300 nm; ca. 3
10-4 M in water.
were determined from the corresponding decay curve.12
Quinone methides are known to undergo very efficient
hetero-Diels-Alder reactions with electron-rich alkenes.4
Irradiation of 3a and 4a in aqueous acetonitrile in the
presence of 30 mM ethyl vinyl ether resulted in the formation
of adducts 6 and 7 in 90% isolated yields (Scheme 4).12
SCHEME 4

HPLC analysis shows near-quantitative formation of vinyl

ether adducts in photolysis of alternative o-QM precursors 3b,c
and 4b,c at both high and low conversions (Table 1). Products of FIGURE 4. Formation and decay of o-QM 1 in the flash photolysis
o-QM hydration (3a and 4a) or ketonization (5) were not of ca. 0.1 mM aqueous solution of 3a at pH 7.0.
detected in these experiments. The exclusive formation of
Diels-Alder adduct despite the presence of more than a thou- The identity of the longer lived transients was established on
sand-fold excess (33 M versus 0.03 M) of a nucleophilic solvent the basis of their reactivity toward nucleophiles. Thus, decay of
indicates that addition of ethyl vinyl ether to o-QMs is more than the second transients in wholly aqueous solution is relatively
2 orders of magnitude faster than a hydration reaction. Direct slow: kobs(3a)=300 ( 5 s-1 at pH 7.0; and kobs(4a)=126 ( 4 s-1
kinetic measurements of the rate of o-QM 1 and 2 reaction with at pH 7.0. Addition of the azide ion dramatically increases
ethyl vinyl ether were precluded by strong absorbance of 266 nm the rate of this process: kN3(3a) = 3.30
104 M-1 s-1; and
laser pulse by the alkene under experimental conditions ([QM kN3(4a)=3.15
104 M-1 s-1 (Figure 5). On the other hand, the
precursor] = 10-4 M; [vinyl ether] = 0.02-0.1M). lifetime of the first transients generated from 3a and 4a was not
Kinetics of the Formation and Reactions of 4-Hydroxy-6- affected by the presence of azide anion. Azide anion was pre-
methylene-2,4-cyclohexadien-1-one (1) and 4-Methoxy-6- viously demonstrated to increase the decay rate of o-QMs due to
methylene-2,4-cyclohexadien-1-one (2). Rate measurements their pronounced electrophilicity.11,18 Additional support for
were conducted in aqueous solutions at 25 ( 0.1 °C and ca. 10-4 the structural assignments of the second transients came from
M concentration of a substrate. Excitation of diols 3a and 4a the results of the laser flash photolyses of 2,5-dihydroxybenzyl-
with 4 ns 266 nm pulses of a Nd:YAG laser under these con- trimethylammonium iodide 3c and 2-hydroxy-5-methoxyben-
ditions results in the formation of short-lived transients zyltrimethylammonium iodide 4c. Photochemical decompo-
(τ ≈ 20 μs) with λmax ≈ 410 nm, which rapidly decay to yield new sition of 3c and 4c produces the same set of products as the
intermediates with λmax ≈ 420 nm (Figure 3). The latter tran-
(18) Richard, J. P.; Amyes, T. L.; Bei, L.; Stubblefield, V. J. Am. Chem.
(17) Chiang, Y.; Kresge, A. J.; Zhu, Y. J. Am. Chem. Soc. 2001, 123, 8089. Soc. 1990, 112, 9513.

J. Org. Chem. Vol. 75, No. 21, 2010 7341

JOC Article Arumugam and Popik

FIGURE 5. Quenching of o-QM 1 triangles) and 2 (circles) by FIGURE 6. Rate profile for the hydration of o-QM 1 (solid squares)
sodium azide in aqueous solutions at pH 7.0. and 2 (open circles) in aqueous solution at 25 °C.

photolyses of 3a,b and 4a,b, which are produced from o-QMs 1

and 2. However, only a single transient is observed in laser flash
photolysis of ammonium salts 3c or 4c. The spectral properties,
lifetime, and reactivity of these intermediates are identical,
within the uncertainty limits, to that of the second transient
generated from 3a and 4a.
Kinetics of Decay of o-QMs 1 and 2 in Aqueous Perchloric
Acid, Sodium Hydroxide, and Buffer Solutions. Rates of the
decay of o-QM 1 and 2 were determined in dilute aqueous
solutions of perchloric acid and sodium hydroxide, as well as
biphosphate ion, bicarbonate ion, and acetic acid buffers.
The ionic strength of these solutions was kept constant at
0.1 M by adding sodium perchlorate as required. The rates of
decay of the first transient observed in photolyses of 3a and
4a remain constant through the entire pH range. The decay FIGURE 7. Specific acid catalysis observed of o-QMs 1 (triangles)
of second transients, o-QMs 1 and 2, on the other hand, is and 2 (circles) generated from 3c and 4 in aqueous perchloric acid.
catalyzed by by perchloric acid, hydroxide ion, as well as
some buffers (vide infra). Kinetic measurements in buffered transient observed in the photolyses of ammonium salts 3c
solutions were performed in a series of solutions of varying (kHþ = (8.12 ( 0.10)
103 M-1 s-1) and 4c (kHþ = (3.49 (
buffer concentration but constant buffer ratio. The observed 0.07)
104 M-1 s-1) is very similar to that of o-QMs
rates were then extrapolated to a zero buffer concentration. generated from the diols 3a and 4a. This further confirms
Relatively strong buffer catalysis was observed for the decay that the second transients observed in laser flash photolysis
of both o-QMs 1 and 2 in acetate buffer solutions. However, of 3a and 4a are indeed the o-QMs 1 and 2, respectively. As in
only the rate of decay of o-QM 1 shows appreciable buffer the case of the parent o-QM,4b,19 the observed specific acid
catalysis in phosphate and bicarbonate buffer solutions. catalysis can be attributed to the rapid equilibration between
While the buffer catalysis for the decay of QM 2 in phosphate o-QMs and more electrophilic o-hydroxybenzyl cations 1þ
and bicarbonate buffer solution was very weak, the zero- and 2þ (Scheme 5). The specific acid catalysis of the o-QM 2
concentration intercepts for the rate of decay of QM 2 were hydration is somewhat weaker than that observed for the
well-determined. The buffer-independent rate constants, parent benzene-1,2-quinone methide (kHþ = 8.4
105 M-1
together with observed rate constants determined in per- s-1),4b,17 apparently due to the presence of the electron-rich
chloric acid and sodium hydroxide solutions, are shown as methoxy group. Similar reduction of the electrophilicity
the rate profile in Figure 6. of methide carbon by m-methoxy substitution was observed
The rate of hydration of o-QMs 1 and 2 is independent of on addition of various nucleophiles to o-QMs.5d Electron-
the acidity of aqueous solutions in the range from pH 4 to 10 donating abilities of hydroxy and methoxy groups are close
resulting in a broad horizontal region in the rate profile but acid catalysis of o-QM 1 decay is stronger than that of 2.
(Figure 6). The observed rate constants on the plateau are The higher acid sensitivity of 1 indicates that both hydration
kH2O = 300 ( 27 s-1 for o-QM 1 and kH2O = 126 ( 16 s-1 for and ketonization reactions of this o-QM are catalyzed by
o-QM 2. The lower uncatalyzed hydration rate of o-QM 2 in acid. In fact, the 5 to 3a ratio increases at higher hydronium
comparison with the parent o-QM (kH2O = 260 s-1)17 is ion concentration apparently due to stronger acid catalysis
apparently due to its reduced electrophilicty. In the case of of the ketonization reaction (Table 2).
o-QM 1 the lower rate of hydration is apparently augmented The decay of o-QMs 1 and 2 is also strongly catalyzed by
by the ketonization reaction. The decay rates of o-QMs 1 and the hydroxide ion: kOH(1) = (3.51 ( 0.19)
104 M-1 s-1 and
2 rise in a linear proportion to acid concentration: kHþ(1) =
(8.37 ( 0.16)
103 M-1 s-1; kHþ(2) = (3.30 ( 0.01)
104 (19) Keeffe, J. R.; Kresge, A. J. In The Chemistry of Enols; Rappoport, Z.,
M-1 s-1 (Figure 7). The acid catalysis of the decay of the only Ed.; Wiley: New York, 1990; Chapter 7.

7342 J. Org. Chem. Vol. 75, No. 21, 2010

Arumugam and Popik
JOC Article
SCHEME 5

kOH(2)= (1.08 ( 0.02)
104 M-1 s-1. The reactivities of in Scheme 5. Rate measurements were conducted in D2O
o-QMs 1 and 2 toward nucleophilic attack by the hydroxide solutions of DCl in the range of concentrations from 0.001 to
ion are very similar to those of the parent o-QM. However, 0.1 M. The kinetic solvent isotopic effect on the acid-
the rate of hydration of 2 in aqueous sodium hydroxide catalyzed hydration of 2 (kHþ/kDþ = 0.48) is very similar to
solution is somewhat reduced in comparison to that of parent that of the parent o-QM (kHþ/kDþ = 0.42).5b Such an inverse
quinone methide (kOH = (3.0
104 M-1 s-1).17 This is again solvent isotopic effect is attributed to the rapid pre-equilib-
attributed to the reduced electrophilicity of this o-QM due to rium substrate protonation and is due to the fact that acids are
electron-donating substituents in the ring.5d On the other hand, weaker in D2O and the fraction of more reactive protonated
the hydroxide ion catalysis of the decay of 1 is more than three species 2þ grows faster with Dþ concentration than with Hþ.20
times stronger than that of 2. This higher reactivity is due to the The hydration reaction of 2 on the plateau region was slower in
fact that the second pathway open to o-QM 1, i.e., ketonization D2O (kH2O/kD2O = 1.63). The nature of this isotope effect stems
to 5, is also catalyzed by a specific base. from the fact that positively charged O-H bonds, such as those
The rate profiles of Figure 6 are readily understood in in the intermediate 8 (Scheme 5), are looser than uncharged
terms of the reaction Scheme 5. Thus, the major reaction O-H bonds, such as those in a water molecule. Conversion
consuming o-QM 2 in aqueous solutions is hydration to 4a. of H2O into ROH2þ leads to a loosening of the hydrogenic
o-QM 1, on the other hand, reacts via two pathways: hydra- environment of the species involved, and that produces an
tion to form 3a and ketonization to p-quinone 5 (Scheme 5). isotope effect in the normal direction. Since the major reaction
The hydration mechanism of both o-QM 1 and 2 is very of the o-QM 1 within the pH 3-10 range is addition of water to
similar to that of the parent o-QM.4b In the base-catalyzed produce 3a, the observed solvent isotope effect (kH2O/kD2O =
region of the rate profiles hydration apparently occurs via 1.49) is similar to that of 2. In a sharp contrast with 2, the isotope
the rate-limiting attack of the hydroxide ion on the methide effect on acid-catalyzed (pH <3) reactions of 1 is in the normal
carbon of o-QMs 1 and 2. Specific acid catalysis at pH <3 direction (kHþ/kDþ=1.60). This observation can be explained by
should be assigned to the rapid pre-equilibrium protonation the fact that ketonization to methyl-p-quinone (5) becomes an
of the o-QM, which produces the more reactive cation 1þ/2þ increasingly important pathway of the o-QM 1 decay at higher
(Scheme 5). The uncatalyzed portion of the rate profile at pH acidities. The ketonization reaction, which is known to proceed
between 4 and 10 has two possible interpretations. The first is via rate-determining proton transfer, is expected to produce a
reversible protonation of the o-QMs by water, followed by rate- normal kinetic isotope effect (Scheme 5).20,21
determining cation capture by the hydroxide ion so formed. As Buffer Catalysis. The rate of o-QM 2 hydration shows
in the case of parent o-QM, this mechanism can be ruled out on virtually no dependence on the concentration of phosphate
the basis of the fact that it would require an impossibly large and bicarbonate buffer, resembling the behavior of the
value of the rate constant for its rate-determining step.19 The parent o-QM.17 This observation agrees well with the nucleo-
more probable mechanism of the pH-independent hydration is philic hydration mechanism (Scheme 5). However, in the
simple nucleophilic attack of water on the o-QM methylene acetate buffer solutions, the rate of decay of o-QM 2 shows
group, with or without simultaneous proton transfer to avoid a significant buffer catalysis. Rate measurements were per-
zwitterionic intermediate 8 (Scheme 5). formed in four series of solutions of varying acetate buffer
Hydration of o-QM 1 to diol 3a is accompanied by the concentrations but constant buffer ratio in each series.
tautomerization to methyl-p-quinone (5), which also shows Slopes of buffer dilution plots at various buffer ratios
acid and base catalysis. Conversion of o-QM 1 to 5 is, in fact, represent pH-independent catalysis by the buffer compo-
a vinylogous ketonization reaction. Such a process usually nents. As is evident from Figure 8, the efficiency of acetate
proceeds via rate-limiting proton transfer on a δ-carbon of a
conjugated enol and, therefore, is catalyzed by general and (20) Kresge, A. J.; More O’Ferrall, R. A.; Powell, M. F. In Isotopes in
Organic Chemistry; Buncel, E., Lee, C. C., Eds.; Elsevier: New York, 1987;
specific acids.19 Within the pH-independent part of the rate Chapter 4.
profile water takes over as the major protonating species. At (21) (a) Capon, B.; Siddhanta, A. K.; Zucco, C. J. Org. Chem. 1985, 50,
3580. (b) Chiang, Y.; Guo, H.-X.; Kresge, A. J.; Tee, O. S. J. Am. Chem. Soc.
pH >9, the fraction of o-QM 1 existing in more reactive 1996, 118, 3386. (c) Chiang, Y.; Kresge, A. J.; Santaballa, J. A.; Wirz, J.
enolate form 9 grows with increasing pH producing an J. Am. Chem. Soc. 1988, 110, 5506. (d) Pruszynski, P.; Chiang, Y.; Kresge,
apparent hydroxide ion catalysis (Scheme 5). A. J.; Schepp, N. P.; Walsh, P. A. J. Phys. Chem. 1986, 90, 3760. (e) Chiang,
Y.; Kresge, A. J.; Nikolaev, V. A.; Popik, V. V J. Am. Chem. Soc. 1997, 190,
Solvent isotope effects on the rate of o-QMs 1 and 2 11183. (f) Capponi, M.; Gut, I. G.; Hellrung, B.; Persy, G.; Wirz, J. Can. J.
consumption provide additional support for the mechanism Chem. 1999, 77, 605.

J. Org. Chem. Vol. 75, No. 21, 2010 7343

JOC Article Arumugam and Popik

of this mechanism will produce hydroxide ion catalysis and the

second is catalyzed by a general acid. Combination of these to
catalyses is operationally equivalent to a general base catalysis.
Elucidation of the o-QMs Precursor Structures. In agree-
ment with previous reports,4b parent o-QM is formed within
the duration of a laser pulse in the flash photolysis of o-hydroxy-
benzyl alcohol. We observed similar “instant” formation of
o-QMs 1 and 2 in photolyses of ammonium precursors 3c and
4c. In a sharp contrast with these results, laser flash photolysis of
2-(hydroxymethyl)benzene-1,4-diol (3a) and 2-(hydroxy-
methyl)-4-methoxyphenol (4a), as well as their ether analogues
3b and 4b, allowed us to detect a short-lived (τ ≈ 20 μs) kinetic
precursor to o-QMs. In unbuffered aqueous solutions o-QM 1 is
produced with the observed rate k = (4.80 ( 0.20)
104 s-1 and
o-QM 2 with kobs = (5.00 ( 0.45)
104 s-1. The rate of o-QMs
FIGURE 8. Dependence of the acetate buffer catalysis of o-QM 2 formation is independent of pH of the solution and is not
decay on the fraction of free acid in the buffer. affected by the presence of the azide ion. The fluorescence
lifetimes of 3a and 4a (τ ≈ 8 ns) are much shorter than the rise
time of the corresponding o-QMs, indicating that the latter
species are not formed directly from the singlet excited state of
the diol precursors. Saturation of the solution with oxygen or
addition of increased amounts 1,3-cyclohexadiene to the reac-
tion mixture before the photolysis do not quench the first
transient or affect the rate of its decay. This observation allows
us to conclude that triple excited states of 3a and 4a are not
involved in the formation of the respective o-QMs. The zwitter-
ionic structure 8, which can be potentially formed in ESIPT to
benzylic oxygen atom in 3a or 4a (Scheme 6), is an unlikely
candidate for o-QM precursor because reverse proton transfer in
this structure should proceed at least at the diffusion-controlled
rate limit. In fact, pH insensitivity of the first transient allows us
to exclude all ionized forms of 3a and 4a from consideration.
FIGURE 9. Dependence of the buffer catalysis of the decay of SCHEME 6
o-QM 1 on the fraction of free acid in the buffer.

buffer catalysis decreases with increased fraction of free acid.

This observation, as well as the observed formation of
acetylation product 4d (Scheme 3, Table 3), indicate the
nucleophilic nature of the acetate buffer catalysis of o-QM 2
hydration. Similar reactivity was also observed in the case of
the parent o-QM.17 ESIPT from the phenolic oxygen to the aromatic carbon is
In a contrast to o-QM 2, the rate decay of o-QM 1 is more well documented for phenols and naphthols.22 Such carbon
sensitive to the free acid concentration in both phosphate and protonation leading to the unstable keto-form of phenol (such as
acetate buffer solutions (Figure 9). The strong buffer catalysis of 10, Scheme 5, or its isomers) should result in incorporation of
the decay of o-QM 1 observed in all buffers examined can be deuterium in hydration products when photolysis is conducted
explained by the fact that the ketonization reactions are usually in deutrated solvents. GC-MS analysis of the products formed
catalyzed by general acid (Scheme 5).22 Thus, in 0.1 M acetate after prolonged irradiation of diols 3a and 4a in D2O showed no
buffer o-QM 1 does not form any detectable amounts of new deuterium enrichment in the aromatic rings. This observation
products, but the ratio of ketonization to hydration products allows us to rule out participation of 10 or isomeric structures. It
increases dramatically (Table 3). Buffer catalytic coefficients has been shown, on the other hand, that o-quinone methides
can be partitioned into contributions from the acidic and basic might reversibly isomerize to benzoxete derivatives, which are
components by extrapolation of the plots in Figure 9 to fA = 1 stable only at cryogenic temperatures.23 A similar ring-opening
(general acid: kH2PO4- = (1.99 ( 0.02)
103 M-1 s-1; kHOAc = reaction is actually used for the generation of thio-o-quinone
(4.20 ( 0.11)
103 M-1 s-1) and fA = 0 (general base catalysis: methides from benzothietes.24 We have recently reported
kHPO42- = (7.40 ( 0.02)
102 M-1 s-1; kAcO- = (1.44 ( 0.05)

103 M-1 s-1). Ketonization reaction in buffer solutions can also (23) (a) Qiao, G. G.; Lenghaus, K.; Solomon, D. H.; Reisinger, A.;
Bytheway, I.; Wentrup, C. J. Org. Chem. 1998, 63, 9806. (b) Tomioka, H.
occur by ionization of o-QM 2 to the enolate ion 9 followed by Pure Appl. Chem. 1997, 69, 837. (c) Sauter, M.; Adam, W. Acc. Chem. Res.
the proton transfer from the buffer acid (Scheme 5). The first step 1995, 28, 289. (d) Shanks, D.; Frisell, H.; Ottosson, H.; Engman, L. Org.
Biomol. Chem. 2006, 4, 846.
(24) (a) Mayer, A.; Meier, H. Tetrahedron Lett. 1994, 35, 2161. (b) Meier,
(22) (a) Lukeman, M.; Wan, P. J. Am. Chem. Soc. 2003, 125, 1164. (b) H.; Mayer, A.; Groschl, D. Sulfur Rep. 1994, 16, 23 and references cited
Lukeman, M.; Veale, D.; Wan, P.; Munasinghe, V. R. N.; Corrie, J. E. T. therein. (c) Meier, H.; Eckes, H.-L.; Niedermann, H.-P.; Kolshorn, H.
Can. J. Chem. 2004, 82, 240. (c) Basaric, N.; Wan, P. J. Org. Chem. 2006, 71, Angew. Chem., Int. Ed. Engl. 1987, 26, 1046. (d) Kanakarajan, K.; Meier,
2677. H. J. Org. Chem. 1983, 48, 881.

7344 J. Org. Chem. Vol. 75, No. 21, 2010

Arumugam and Popik
JOC Article
SCHEME 7 Solutions of ca. 3
10-4 M of compounds 3b, 3c, 4b, and 4c
were irradiated in water and in the presence of ethyl vinyl ether
(3
10-2 M) in 50% MeCN in water, using a mini-Rayonet
photochemical reactor equipped with eight fluorescent UV
lamps (4 W, 300 nm). Reaction mixtures after photolysis were
analyzed by HPLC and chemical yields were determined from
the calibration plot constructed with known standards of the
pure product. Quantum efficiencies of photochemical reactions
were measured by ferrioxalate actinometry.26 Buffer solutions
for kinetic experiments were prepared by using literature pKa
the detection of naphthoxete precursors of naphthoquinone
values of the buffer acids and activity coefficient recommended
methides.11 We, therefore, believe that oxetanes 11 and 12 are by Bates.27
the likely precursors of o-QMs 1 and 2 (Scheme 7). Kinetic Experiments. Rate measurements were conducted
Since the fluorescence spectra of 3a and 4a are very similar with a nanosecond kinetic spectrometer equipped with a Nd:
to spectra of corresponding phenols,25 we can assume that YAG laser (pulse width = 4 ns) fitted with second and fourth
excitation of these substrates results in the formation of the harmonic generators. Degassed solutions of o-QM precursors
phenolate ions in the excited state. If the o-benzylic position with OD ≈ 0.4 were thermostated at 25 ( 0.05 °C. Since the
is substituted with a good leaving group (e.g., X = NMe3þ, formation and the decay of o-QMs proceed at very different
Scheme 7), the o-QM is formed directly from the phenolate rates, these processes were recorded in separate experiments
ion. However, in case of poor leaving groups (e.g., X = OH with different time scales and signal amplifications. The experi-
mental data were fitted separately for each transient to a double
or OR), oxetane is formed. The latter then opens up in a
exponential function and the first order rate constants were
ground state reaction to give an o-QM. Ring-opening of 11 determined from the corresponding decay curve. Second order
and 12 to give o-QMs 1 and 2 is an electrocyclic reaction and, rate constant of o-QM reactions with azide ion were determined
therefore, is not sensitive to acid/base catalysis or the pre- from the plot of the concentration of the trapping reagents vs the
sence of reactive nucleophiles such as azide ion. observed pseudo-first-order rate constants. The experimental
rate constants are summarized in Tables S1-S20 in the Sup-
Conclusions porting Information.12
Fluorescent Measurements. Fluorescent spectra of 3a and 4a
Two o-quinone methides, 4-hydroxy-6-methylene-2,4-cy- were recorded at λex = 305 nm in doubly deionized water with
clohexadien-1-one (1) and 4-methoxy-6-methylene-2,4-cy- the substrate concentration ca. 1
10-5 M, using Varian steady
clohexadien-1-one (2), are efficiently generated by the state fluorimeter. The excitation source slits and the detector
irradiation of three different types of precursors: substituted slits were set to 2 and 5 nm, respectively. The fluorescence
o-hydroxybenzyl alcohols 3a and 4a; their ethyl ethers 3b and quantum yields were determined with phenol as the standard
4b, as well as (2,5-dihydroxy-1-phenyl)methyl- (3c) and (2- reference.27,28
hydroxy-5-methoxy-1-phenyl)methyltrimethylammonium Materials. Ethyl vinyl ether and sodium were purchased from
Sigma-Aldrich and used as received. 2-(Hydroxymethyl)benzene-
iodides (4c). The reactivity of o-QMs 2 resembles that of
1,4-diol (3a),29 (2-(hydroxymethyl)-4-methoxyphenol (4a),30 (2-
other o-quinone methides albeit with reduced nucleophili- hydroxy-5-methoxy-1-phenyl)methyltrimethylammonium iodide
city. In aqueous solution it undergoes efficient rehydration (4c),5d and 2-(acetoxymethyl)-4-methoxyphenol (4d)31 were pre-
back to 4a and this reaction is catalyzed by both hydroxide pared following the literature procedures.
and hydronium ions. In the case of o-QMs 1, addition of 2-(Ethoxymethyl)benzene-1,4-diol (3b). A solution of 3a
water to give 3a competes with tautomerization to form (70 mg, 1 mmol) in 80% aqueous ethanol (500 mL) was irradiated,
methyl-p-quinone (5). This is a rather unique phenomenon, using a mini-Rayonet photochemical reactor equipped with
as the same carbon atom shows both electrophilic (addition 254 nm lamps for 1 h. Photolysate was extracted with ethyl acetate
of water) and nucleophilic (protonation) reactivity. The then dried over sodium sulfate; solvents were removed under
ketonization reaction shows pronounced general acid cata- vacuum. The residue was separated by chromatography (30%
EtOAc in hexane) to give 50 mg (60%) of 3b as a colorless oil. 1H
lysis. Both o-QMs readily react with other nucleophiles such
NMR δ 6.72-6.74 (m, 1H), 6.65-6.67 (m, 1H), 6.52-6.53 (m,
as azide anion and undergo efficient Diels-Alder cycloaddi- 1H), 4.61 (s, 2H), 3.60 (q, 2H, J = 5.2 Hz), 1.26 (t, 3H, 5.2 Hz). 13C
tion to electron-rich olefins. Photochemical dehydration of NMR δ 149.8, 148.9, 123.4, 117.3, 116.1, 115.1, 71.9, 66.5, 15.2;
3a and 4a to produce o-QMs 1 and 2 proceeds via the GC-MS m/z (%) 168 (10), 138 (100), 137 (40), 122 (60), 121 (25),
formation of a reactive intermediate. On the basis of the 110 (65), 94 (50), 82 (37), 63 (45), 53 (47). FW calcd for C9H12O3þ
reactivity of these transient species the benzoxete structures 168.0786; EI-HRMS found 168.0783.
11 and 12 were assigned to these transients. 2-(Ethoxymethyl)-4-methoxyphenol (4b). Concentrated HCl
(5 mL) was added to a solution of 4a (500 mg, 3.25 mmol) in 95%
Experimental Section aqueous ethanol (20 mL), then the reaction mixture was stirred

General Methods. All organic solvents were dried and freshly (26) Murov, S. L.; Carmichael, I.; Hug, G. L. In Handbook of Photo-
distilled before use. Flash chromatography was performed with chemistry; Marcel Dekker: New York, 1993; p 299.
40-63 μm silica gel. All NMR spectra were recorded in CDCl3 (27) Bates, R. G. Determination of pH Theory and Practice; Wiley: New York,
and referenced to TMS unless otherwise noted. Solutions were 1973; p 49.
(28) Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic
prepared with HPLC grade water, methanol, and acetonitrile. Molecules; Academic Press: New York, 1971; p 473.
Substrate concentration for kinetics experiments was kept at ca. (29) Casiraghi, G.; Casnati, G.; Puglia, G.; Sartori, G. Synthesis 1980, 2,
1
10-4 M for 3a and 4a and 5
10-5 M for 3c and 4c. 124.
(30) Ono, M.; Kawashima, H.; Nonaka, A.; Kawai, T.; Haratake, M.;
Mori, H.; Kung, M. P.; Kung, H. F.; Saji, H.; Nakayama, M. J. Med. Chem.
(25) (a) Wehry, E. L.; Rogers, L. B. J. Am. Chem. Soc. 1965, 87, 4234. 2006, 49, 2725.
(b) Stalin, T.; Devi, R. A.; Rajendiran, N. Spectrochim. Acta 2005, 61A, 2495. (31) Bray, C. D. Synlett 2008, 16, 2500.

J. Org. Chem. Vol. 75, No. 21, 2010 7345

JOC Article Arumugam and Popik

for 3 h at rt and poured onto crushed ice. The aqueous layer was 137 (52), 136 (40), 123 (12), 108 (17), 94 (12), 73 (17), 70 (37), 61
extracted with ethyl acetate and dried over sodium sulfate; (50), 53 (27), 44 (62), 45 (100). FW calcd for C10H16NO2þ
solvents were removed under vacuum. The crude product was 182.1181; EI-HRMS found 182.1188.
purified by column chromatography (40% dichloromethane in 2-Ethoxy-6-hydroxychroman (6). A solution of 3a (42 mg, 0.3
hexane) to give 4b as a yellow oil (366 mg, 62%). 1H NMR δ 7.27 mmol) and ethyl vinyl ether (2.9 mL, 30 mmol) in aqueous
(br s, 1H), 6.72-6.80 (m, 2H), 6.56-6.57 (m, 1H), 4.64 (s, 2H), acetonitrile (1:1, 300 mL) was irradiated, using a mini-Rayonet
3.72 (s, 3H), 3.58 (q, 2H, J = 5.2 Hz), 1.25 (t, 3H, 5.2 Hz). 13C photochemical reactor equipped with 254 nm lamps for 1 h.
NMR δ 153.2, 150.3, 123.3, 117.3, 114.5, 113.9, 72.3, 66.5, 56.0, Photolysate was extracted with ethyl acetate and dried over
15.3. GC-MS m/z (%) 182 (20), 136 (100), 137 (20), 108 (45), 79 sodium sulfate; solvents were removed under vacuum. The
(20), 65 (20). FW calcd for C10H14O3þ 182.0943; EI-HRMS residue was separated by chromatography (20% EtOAc in
found 182.0943. hexane) to give 53 mg (90%) of 6 as a colorless oil. 1H NMR
(2,5-Dihydroxy-1-phenyl)methyltrimethylammonium Iodide (400 MHz, CDCl3) δ 6.69-6.70 (m, 1H), 6.53-6.60 (m, 2H),
(3c). A solution of 2,4-dihydroxybenzaldehyde (900 mg, 6.67 5.20 (t, 1H, J = 2.8 Hz), 3.82-3.88 (m, 1H), 3.59-3-64 (m,
mmol) in methanol (15 mL) was added to a mixture of 1.16 g 1H), 2.88-2.97 (m, 1H), 2.60, 2.55-2.59 (m, 1H), 1.88-2.03 (m,
(14.2 mmol) of dimethylamine hydrochloride, sodium acetate 2H), 1.18 (t, 3H, 7.6 Hz). 13C NMR (100 MHz, CDCl3) δ 149.3,
(0.91 g, 11.1 mmol), and 0.49 g (7.7 mmol) of sodium cyanobor- 146.3, 123.71, 117.8, 115.6, 114.5, 96.9, 63.8, 26.7, 20.9, 15.3. GC
ohydride in methanol (30 mL).The pH of the solution was MS m/z 195 (10), 194 (90), 149 (30), 148 (70), 147 (100), 122 (50),
maintained throughout the reaction in the range 7-8 by the 94 (30), 77 (20), 65 (25), 55 (28), 45 (15). FW calcd for C11H14O3
addition of concentrated HCl. The solution was stirred at room 194.0943, EI-HRMS found 194.0952.
temperature during 24 h. Acetone (50 mL) was added and the 2-Ethoxy-6-methoxychroman (7). Compound 7 was prepared
solution was acidified to pH 2-3 with 6 N HCl solution. in 91% yield following the same procedure that was described
Solvents were removed under vacuum, and the residue was for 6. The spectral properties are consistent with the previously
redissolved in water (15 mL) and washed with ether (4
15 reported values.32 1H NMR (400 MHz, CDCl3) δ 6.69-6.78(m,
mL). NaOH solution was added to the remaining aqueous phase 2H), 6.61-6.62 (m, 1H), 5.23 (t, 1H, J = 2.8 Hz), 3.84-3.90 (m,
until pH reached 8-9. The resulting aqueous solution was 1H), 3.76 (s, 3H), 3.60-3-66 (m, 1H), 2.94-3.02 (m, 1H),
extracted with ether (6
25 mL), the organic layer was dried 2.60-2.66 (m, 2H), 1.90-2.07 (m, 2H), 1.20 (t, 3H, J = 7.6
over MgSO4, and solvent was removed under vacuum. The Hz). 13C NMR of 2 (100 MHz, CDCl3) δ 153.8, 146.3, 123.5,
fractional distillation of amber oil residue produced 0.8 g of pure 117.7, 114.1, 113.5, 97.0, 63.8, 55.9, 26.8, 21.2, 15.4. GC MS m/z
2-(N,N-dimethylaminomethyl)benzene-1,4-diol (71% yield) as 208 (100), 180 (5), 163 (30), 162 (35), 161 (40), 147 (8), 136 (75),
a colorless oil. 1H NMR (400 MHz, CDCl3) δ 6.62-6.50 (m, 108 (27), 91 (12), 73 (12), 65 (12), 55 (7), 43 (5). FW calcd for
3H), 3.54 (s, 2H), 2.28 (s, 6H). 13C NMR (100 MHz, CDCl3) C12H16O3 208.1099, EI-HRMS found 208.1102.
δ 151.1, 149.9, 123.0, 116.1, 115.4, 115.0, 62.6, 43.9. GC-MS
m/z (%) 168 (10), 167 (45), 152 (6), 122 (12), 94 (15), 58 (25), Acknowledgment. The authors thank the National
44 (100). Science Foundation (CHE-0449478) and the donors of the
Methyl iodide (1.5 mL) was added to a stirred solution (0 °C) ACS Petroleum Research Fund (434444-AC4) for the sup-
of 2-(N,N-dimethylaminomethyl)benzene-1,4-diol (450 mg, 2.5 port of this project.
mmol) in acetonitrile (1.5 mL) and the mixture was stirred for 2 h
at rt. Anhydrous ether (25 mL) was added to the reaction Supporting Information Available: Tables of raw kinetic
mixture, and white precipitate separated and was washed with measurements and NMR spectra of newly synthesized com-
ether to give 0.8 g (98%) of quaternary ammonium salt 3c as a pounds. This material is available free of charge via the Internet
white solid. 1H NMR (400 MHz, D2O) δ 6.78-6.86 (m, 3H), at http://pubs.acs.org.
4.29 (s, 2H), 3.63 (s, 3H), 2.99 (s, 9H). 13C NMR (100 MHz,
D2O) δ 150.0, 148.7, 120.6, 119.7, 117.8, 115.4, 64.1, 52.7. GC- (32) Manetsch, R.; Zheng, L.; Reymond, M. T.; Woggon, W.-D.; Reymond,
MS m/z (%) 182 (7), 181 (45), 168 (10), 168 (5), 167 (30), 149 (8), J.-L. Chem.;Eur. J. 2004, 10, 2487.

7346 J. Org. Chem. Vol. 75, No. 21, 2010