Generation and reactivity of the radical cations of coniferyl alcohol and isoeugenol in solution

Article abstract of DOI:10.1139/v03-075

Nanosecond laser flash photolysis of coniferyl alcohol and isoeugenol in acetonitrile leads to the formation of transient species that are identified as the corresponding radical cations. These radical cations decay with rate constants of ca. 1 × 10⁶ s^-1 in dry acetonitrile. Both radical cations react rapidly with hydroxylic solvents like water and alcohols to give 4-vinylphenoxyl radicals, indicating that these reagents behave as bases rather than nucleophiles. In addition, anionic reagents (acetate, cyanide, and chloride) react rapidly with the radical cations with second-order rate constants that are close to diffusion controlled. The main products generated in the presence of the anionic reagents are again the 4-vinylphenoxyl radicals, suggesting that these reagents also behave as bases. The lifetime of the radical cations in acidic acetonitrile was found to increase dramatically due to a shift in the radical cation - vinyl phenoxyl radical acid-base equilibrium to the side of the radical cation. An estimate of the pK_a of the radical cation in acetonitrile of 4.0 was obtained from the data.

Full text of DOI:10.1139/v03-075

799
Generation and reactivity of the radical cations of
coniferyl alcohol and isoeugenol in solution
N.P. Schepp and Y. Rodríguez-Evora
Abstract: Nanosecond laser flash photolysis of coniferyl alcohol and isoeugenol in acetonitrile leads to the formation
of transient species that are identified as the corresponding radical cations. These radical cations decay with rate con-
stants of ca. 1 × 106 s–1 in dry acetonitrile. Both radical cations react rapidly with hydroxylic solvents like water and
alcohols to give 4-vinylphenoxyl radicals, indicating that these reagents behave as bases rather than nucleophiles. In ad-
dition, anionic reagents (acetate, cyanide, and chloride) react rapidly with the radical cations with second-order rate
constants that are close to diffusion controlled. The main products generated in the presence of the anionic reagents are
again the 4-vinylphenoxyl radicals, suggesting that these reagents also behave as bases. The lifetime of the radical cat-
ions in acidic acetonitrile was found to increase dramatically due to a shift in the radical cation – vinyl phenoxyl radi-
cal acid–base equilibrium to the side of the radical cation. An estimate of the pKa of the radical cation in acetonitrile
of 4.0 was obtained from the data.
Key words: radical cations, laser flash photolysis, lignan, vinylphenols.
Résumé : La photolyse éclair au laser, avec des impulsions de l’ordre de la nanoseconde, de l’alcool du coniféryle et
de l’isoeugénol dans l’acétonitrile conduit à la formation d’espèces intermédiaires qui sont en fait les cations radicalai-
res correspondants. Ces cations radicalaires se désintègrent avec une constante de vitesse de l’ordre de 1 × 106 s–1 dans
l’acétonitrile. Les deux cations radicalaires réagissent rapidement avec les solvants hydroxylés tels que l’eau et les al-
cools pour donner des radicaux 4-vinylphénoxyles, indiquant ainsi que ces réactifs se comportent plutôt comme des ba-
ses que comme des nucléophiles. De plus, les réactifs anioniques (acétate, cyanure et chlorure) réagissent rapidement
avec les cations radicalaires avec des constantes de vitesse d’ordre deux qui sont très proches de celles d’une diffusion
contrôlée. Les produits majoritaires générés en présence des réactifs anioniques sont encore des radicaux 4-vinylphén-
oxyles, ces résultats suggèrent que ces réactifs se comportent également comme des bases. On a trouvé que le temps
vie de ces cations radicalaires, dans l’acétonitrile en milieu acide, augmente dramatiquement à cause d’un déplacement
de l’équilibre acide–base, cation radicalaire – radical vinylphénoxyle, du côté du cation radicalaire. A partir des don-
nées on a pu évaluer à 4,0 le pKa du cation radicalaire dans l’acétonitrile.
Mots clés : cations radicalaires, photolyse éclair au laser, lignan, vinylphénols.
[Traduit par la Rédaction] Schepp and Rodríguez-Evora 806
Introduction
wood. While specific details concerning the biosynthesis of
these compounds are not well understood, it is generally ac-
cepted that initial oxidation induced by an oxidizing enzyme
such as a peroxidase or a laccase to give a resonance-
stabilized 4-vinylphenoxyl radical is a key step. Enzyme-
mediated dimerization of these 4-vinylphenoxyl radicals
then leads to production of the lignan-type materials (2, 3).
One pathway for the formation of the 4-vinylphenoxyl
radicals involves enzyme-induced, one-electron oxidation of
the 4-vinylphenol substrate to its corresponding radical cat-
ion which then undergoes rapid deprotonation, Scheme 1.
Naturally occurring compounds (including 4-hydroxy-
cinnamates like syrginic acid and 1-(4-hydroxyphenyl)-
propen-3-ols like coniferyl alcohol) that contain a 4-
vinylphenol moiety are important materials in a number of
biological systems. In particular, these kinds of organic mol-
ecules are building blocks for the biosynthesis of natural
products such as plant lignans (1), which often have impor-
tant medicinal properties, and lignin, a major constituent of
While the chemistry of 4-vinylphenoxyl radicals has been
studied using several methods, including direct detection
methods such as flash photolysis and ESR spectroscopy (4–
6), the radical cations of the 4-vinylphenol substrates have
not yet been investigated. In the present work, we focus on
the chemistry of two 4-vinylphenol radical cations, namely
the coniferyl alcohol and isoeugenol radical cations,
Scheme 2.
Received 17 February 2003. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on 30 June 2003.
This paper is dedicated to Professor Don Arnold in
appreciation of his many contributions to radical ion
chemistry.
N.P. Schepp1 and Y. Rodríguez-Evora. Department of
Chemistry, Dalhousie University, Halifax, NS B3H 4J3,
Canada.
Our principle goals were to obtain rate constants for the
reactions of these radical cations, both with solvent and with
1Corresponding author (e-mail: norman.schepp@dal.ca).
Can. J. Chem. 81: 799–806 (2003)
doi: 10.1139/V03-075
© 2003 NRC Canada

800
Can. J. Chem. Vol. 81, 2003
Scheme 1.
tions of this solution using acetonitrile. To minimize the
possibility of acid-catalyzed decomposition of coniferyl al-
cohol or isoeugenol, these materials were added to the acidic
sample solutions just prior to performing the laser experi-
ments. All kinetics experiments were carried out at room
temperature (22 ± 1°C). For measurement of absorption
spectra, a flow cell system was used to ensure that only
fresh solution was subjected to each laser pulse.
Scheme 2.
Results and discussion
Laser irradiation in acetonitrile
added bases and nucleophiles, and to examine the effect of
acidity on their reactivity. We were also interested in deter-
mining if these radicals cations are sufficiently long-lived to
participate in bimolecular reactions. If so, it may be possible
that such species play a more significant role in the bio-
synthesis of lignans by participating directly in the dimeri-
zation process.
Laser irradiation of coniferyl alcohol in dry acetonitrile
leads to the formation of distinct absorption bands at 380
and 580 nm, Fig. 1. These two bands are formed promptly
within the laser pulse, and both decay in a first-order manner
with identical rate constants of 8 × 105 s–1 (Fig. 1, inset).
The similar kinetic behaviour of these two bands indicates
that both bands belong to a single transient species.
The absorption spectrum is very similar to the known
spectra of substituted styrene radical cations that have two
strong bands near 360 and 580 nm (8). In addition, the decay
of the transient is insensitive to oxygen, but is greatly accel-
erated in the presence of hydroxylic solvents like water and
methanol and in the presence of anionic reagents like bro-
mide and chloride. These observations provide strong evi-
dence for the identification of the transient species as the
radical cation of coniferyl alcohol produced by laser-induced
photoionization of coniferyl alcohol, eq. [1].
Experimental
Materials
Coniferyl alcohol and isoeugenol were purchased
(Aldrich) and used as received. Acetonitrile was spectro-
scopic grade (Omnisolve, BDH). All other materials were
the best available commercial grades and used without fur-
ther purification.
Laser flash photolysis
The nanosecond laser flash photolysis system at
Dalhousie University has been described previously (7). A
Lambda Physiks excimer laser (308 nm, XeCl, 10 ns/pulse,
<100 mJ/pulse) was used as the excitation source for most
experiments. In experiments using chloranil or di-tert-
butylperoxide, a Continuum Nd:YAG NY-61 laser (355 nm;
8 ns/pulse; <35 mJ/pulse) was used as the excitation source.
Solutions for kinetics experiments were prepared by addi-
tion of small aliquots (ca. 10 µL) of stock solutions of
coniferyl alcohol or isoeugenol to 2.0 or 3.0 mL of
acetonitrile in 7 × 7 mm2 laser cells made out of Suprasil
quartz tubing. The absorption of the 4-vinylphenols in the
laser cells was approximately 0.4 at 308 nm, corresponding
to concentrations of ca. 0.1 mM. Most laser experiments
were carried out using aerated samples. When oxygen-free
solutions were required, the samples were bubbled with a
slow stream of dry nitrogen 20 min prior to laser irradiation.
Measurements of quenching rate constants were made by
adding small amounts (1–20 µL) of a stock solution with a
known concentration of the quencher (n-Bu4NCl, n-
Bu4NCN, or n-Bu4NO2CCH3) dissolved in acetonitrile. So-
lutions for the experiments carried out under acidic condi-
tions were prepared by making a 0.010 M HClO4 stock
solution in acetonitrile and then making appropriate dilu-
[1]
Evidence to support this assignment comes from results
obtained under photoinduced electron-transfer conditions.
Selective laser irradiation (355 nm) of chloranil in the pres-
ence of 1.0 mM coniferyl alcohol in acetonitrile leads ini-
tially to the formation of the triplet of chloranil with an
absorption maximum at 500 nm (9). This band decays rap-
idly in the presence of coniferyl alcohol, Fig. 2 (inset), and
leads to the formation of absorption bands at 380 and
580 nm, Fig. 2, that are closely similar to those produced
upon direct laser irradiation of coniferyl alcohol. Absorption
due to the presence of the chloranil radical anion at 450 nm
(9) is also observed, with the rate constant for the growth of
this band being identical to that for the decay of the triplet
and for the growth of the bands at 380 and 580 nm. These
results are consistent with initial formation of the chloranil
triplet with absorption at 500 nm, which then undergoes an
electron-transfer reaction with coniferyl alcohol to produce
© 2003 NRC Canada

Schepp and Rodríguez-Evora
801
Fig. 1. Transient absorption spectra generated (᭺) 200 ns, (᭹)
920 ns, (ٗ) 3.2 µs, and (᭿) 6.4 µs after 308 nm laser irradiation
of coniferyl alcohol in air-saturated acetonitrile. Inset shows
time-resolved changes in optical density at 380, 580, and 340 nm
upon 308 nm irradiation of coniferyl alcohol in air-saturated
acetonitrile.
Fig. 2. Transient absorption spectra generated (᭹) 1 µs, (ٗ)
2 µs, and (᭿) 7 µs after 355 nm laser irradiation of chloranil in
air-saturated acetonitrile with 0.001 M coniferyl alcohol. Inset
shows time-resolved changes in optical density at 500 and
580 nm upon 355 nm laser irradiation of chloranil in air-
saturated acetonitrile with 0.001 M coniferyl alcohol.
the 4-vinylphenol radical cation and the radical anion of
chloranil, eq. [2].
presence or absence of oxygen, nor is its decay influenced
by the addition of protic reagents like water or methanol or
anionic reagents like chloride or cyanide.
[2]
As observed for methoxy-substituted phenol radical cat-
ions, which rapidly deprotonate to give simple phenoxyl rad-
icals (10), the main reaction of the 4-vinylphenol radical
cation is likely to be deprotonation to give in this case a 4-
vinylphenoxyl radical, eq. [3].
[3]
It is therefore reasonable to suggest that the 340 nm ab-
sorption band generated upon decay of the coniferyl alcohol
radical cation is the 2-methoxy-4-(3-hydroxy-1-propenyl)
phenoxyl radical. Evidence supporting this suggestion was
obtained upon 355 nm laser irradiation of a solution of di-
tert-butylperoxide in the presence of small amounts of
coniferyl alcohol. This is expected to lead to the formation
of the phenoxyl radical by initial photochemical formation
of the tert-butoxy radical, followed by rapid hydrogen ab-
straction from the phenol hydroxy group of the coniferyl al-
cohol. As shown in Fig. 3, a transient with an absorption
maximum at 340 nm identical to that produced upon reac-
tion of the coniferyl alcohol radical cation is observed.
These results therefore provide good evidence that the main
reaction of the radical cation is indeed deprotonation to the
4-vinylphenoxyl radical, eq. [3].
As shown in Fig. 1, a strong absorption band centered at
340 nm is observed upon 308 nm irradiation of coniferyl al-
cohol in addition to the radical cation absorption at 380 and
580 nm. However, unlike the radical cation absorption,
which appears promptly after the laser pulse, the 340 nm
band grows in as a function of time. The observed rate con-
stant of 8 × 105 s–1 for the formation of this band is identical
to that for the decay of the 380 and 580 nm bands, Fig. 1
(inset), indicating the transient responsible for the absorption
at 340 nm is a product of the reaction of the radical cation.
The decay of the 340 nm transient is not affected by the
A long-lived species with an absorption maximum at
350 nm has been observed previously upon irradiation of
coniferyl alcohol in aqueous or alcoholic solvents (6, 11,
12). This transient has been assigned as the quinone methide
generated by a mechanism that is thought to involve
© 2003 NRC Canada

802
Can. J. Chem. Vol. 81, 2003
Fig. 3. Transient absorption spectra generated (᭹) 480 ns, (᭺)
1.9 µs, and (᭿) 13 µs after 355 nm laser irradiation of di-tert-
butylperoxide in air-saturated acetonitrile with 0.001 M coniferyl
alcohol. Inset shows time-resolved changes in optical density at
340 nm.
Fig. 4. Transient absorption spectra generated (᭹) 400 ns, (ٗ)
1.8 µs, and (᭿) 3.0 µs after 308 nm laser irradiation of isoeu-
genol in air-saturated acetonitrile. Inset shows time-resolved
changes in optical density at 380, 580, and 340 nm observed
upon 308 nm irradiation of isoeugenol in air-saturated aceto-
nitrile.
photoinduced protonation of the β-OH group, followed by
elimination of water and deprotonation (6) (12), eq. [4].
Effect of water and alcohols on decay of coniferyl
alcohol radical cation in acetonitrile
Throughout the course of our experiments into the reactiv-
ity of the coniferyl alcohol and isoeugenol radical cations in
neat acetonitrile, rate constants measured for the decay of
the radical cations varied somewhat between 7 × 105 and 1 ×
106 s–1. This variability depended mainly on the quality of
the acetonitrile used: in acetonitrile from a freshly opened
bottle, the radical cation was longer-lived, while the radical
cation in acetonitrile that had previously been exposed to air
was somewhat shorter-lived. The most likely source of this
variability was the presence or absence of water in the
acetonitrile, with acetonitrile from a freshly opened bottle
having a lower water content compared with that from an al-
ready opened bottle. The rate constants of 7 × 105 and 9 ×
105 s–1 were the slowest rate constants measured for the
coniferyl alcohol and the isoeugenol radical cations, respec-
tively, in acetonitrile dried with activated molecular sieves.
To quantitatively determine the influence of water content,
the decay of the radical cation was measured as a function of
water concentration in acetonitrile. As shown in Fig. 5 (in-
set), even small quantities of water greatly accelerate the rate
constant for the decay, such that in the presence of 0.034 M
water, the rate constant more than triples from 0.8 × 106 to
3.0 × 106 s–1. The relationship between the observed rate
constants2 for the decay of the coniferyl alcohol radical cat-
ion and water concentration is linear. Least-squares analysis
leads to the second-order rate constant for the reaction with
water of 6.5 × 107 M–1 s–1.
[4]
However, the absorption band at 340 nm generated in our
experiments decays completely within 1 ms after the laser
pulse, which is consistent with the reactivity of other
phenoxyl radicals (13). This reactivity is considerably differ-
ent from that observed for the quinone methide in aqueous
solution at room temperature, where its lifetime is several
minutes (12). Furthermore, our results clearly show that the
340 nm transient is a product of the reaction of the radical
cation; it is difficult to assign a reasonable mechanism for
the conversion of the radical cation to the quinone methide,
while conversion to the phenoxyl radical by deprotonation is
a sensible reaction pathway.
Similar results to those described above were observed
upon direct 308 nm irradiation of isoeugenol in acetonitrile,
Fig. 4, and upon 355 nm irradiation of chloranil in the pres-
ence of isoeugenol. In both cases, the isoeugenol radical cat-
ion was observed with absorption bands at 380 and 580 nm.
The radical cation decayed in a first-order manner in dry
acetonitrile with a rate constant of 9 × 105 s–1. As the radical
cation decayed, a new absorption band at 340 nm was ob-
served. This new absorption band is assigned to the corre-
sponding 4-vinylphenoxyl radical on the basis of its
similarity to the absorption band observed upon 355 nm irra-
diation of di-tert-butylperoxide in the presence of
isoeugenol.
The decay of the coniferyl alcohol radical cation also in-
creased rapidly with increasing concentrations of methanol,
ethanol, 2-propanol, and tert-butyl alcohol. In each case, a
linear relationship was obtained between the observed rate
constants and alcohol concentration. Least-squares analysis
2 Tables of rate data (Tables S1-S10) have been deposited and may be purchased from the Depository of Unpublished Data, Document Deliv-
ery, CISTI, National Research Council Canada, Ottawa, ON K1A 0S2, Canada (http://www.nrc.ca/cisti/irm/unpub_e.shtml for information
on ordering electronically).
© 2003 NRC Canada

Schepp and Rodríguez-Evora
803
Fig. 5. Transient absorption spectra generated (᭺) 90 ns, (᭹)
460 ns, (ٗ) 1.8 µs, and (᭿) 3.2 µs after 308 nm laser irradiation
of isoeugenol in air-saturated acetonitrile with 0.03 M water. In-
set shows the relationship between the observed rate constants
for the decay of the coniferyl alcohol radical and the concentra-
tion of added water in acetonitrile.
Table 1. Second-order rate constants for the reaction of the
coniferyl alcohol radical cation with water and alcohols in
acetonitrile (22 ± 1°C).
Coniferyl alcohol
Water or alcohol
kNuc (M–1 s–1)
H2O
CH3OH
CH3CH2OH
(CH3)2CHOH
(CH3)3COH
(6.5 ± 0.5) × 107
(3.1 ± 0.1) × 107
(2.0 ± 0.1) × 107
(4.7 ± 0.2) × 107
(4.9 ± 0.2) × 107
vinylphenol radical cations, with the hydroxy group on the
aromatic ring being replaced by a methoxy group. The 1-(4-
methoxyphenyl)propene radical cation thus has no strongly
acidic proton, and the low value for the rate constant for its
reaction with water and alcohols indicates a relatively slow
nucleophilic addition reaction. Rate constants for nucleo-
philic addition to the coniferyl alcohol radical cation should
be similar; the fact that the measured rate constants are
greater is consistent with the availability of a more rapid
deprotonation reaction.
The rate constants for deprotonation from the coniferyl al-
cohol radical cation are of the same order of magnitude for
the deprotonation from methoxy-substituted phenols in ace-
tonitrile. For example, deprotonation of 3,5-dimethoxyphenol
radical cation by water and methanol in acetonitrile takes
place with rate constants of 2.5 × 108 and 8.2 × 107 M–1 s–1
(10), respectively, which are about three-fold greater than
the rate constants for the deprotonation of coniferyl alcohol.
gave second-order rate constants that were similar for each
alcohol, ranging from 2 × 107 to 4 × 107 M–1 s–1. These rate
constants are summarized in Table 1.
In the presence of added water or alcohols, the accelerated
decay of the radical cation continued to lead to the forma-
tion of the 340 nm band assigned to the substituted 4-
vinylphenoxyl radical, Fig. 5, and the rate constants for the
growth of the radical continued to match the rate constants
for the radical cation decay. No evidence was found for the
presence of an oxygen-sensitive phenethyl radical that would
have been formed upon nucleophilic addition of water or al-
cohol to the radical. Thus, water and the alcohols appear to
act as bases rather than nucleophiles in their reaction with
the radical cation, eq. [5].
Effect of anions on decay of coniferyl alcohol and
isoeugenol radical cations in acetonitrile
In acetonitrile, styrene radical cations react with anionic
reagents like halide, cyanide, and acetate ions at diffusion-
controlled rate constants of approximately 1010 M–1 s–1, with
the primary reaction being nucleophilic addition to generate
substituted phenethyl radicals (8). In the present work, the
reactivity of the coniferyl alcohol and isoeugenol radical cat-
ions in the presence of these anions in acetonitrile was also
measured. In each case, the decay of the radical cations did
increase rapidly as a function of increasing concentration of
the anions. The relationships between the observed rate con-
stant for the decay of the radical cations and the concentra-
tions of the anions were all linear. Least-squares analysis
gave second-order rate constants in the 1–2 × 1010 M–1 s–1
range for reaction of acetate and cyanide ions, Table 2, indi-
cating that these two anions react with the radical cations at
the diffusion-controlled limit. The second-order rate constant
for the reaction of the chloride ion with the two radical cat-
ions was similar at about 2 × 1010 M–1 s–1.
Figure 6 shows the absorption spectrum generated upon
quenching of the coniferyl alcohol radical cation in the pres-
ence of 0.1 mM chloride. At this concentration of chloride,
the bands at 380 and 580 nm associated with the radical cat-
ion are almost completely quenched, and in their place is an
absorption band with a maximum at 340 nm. This absorp-
tion band, which is also formed upon reaction of the radical
cation with acetate and cyanide ions, is identical to that of
the 4-vinylphenoxyl radical produced by deprotonation of
[5]
In addition, little variation is observed in the second-order
rate constants for the reaction of the radical cation with the
four alcohols, Table 1. Such a lack of sensitivity to the steric
bulk of the alcohols is expected for a situation in which the
alcohols, which have similar basicities but different
nucleophilicities, act as bases rather than nucleophiles. Fur-
thermore, the rate constants given in Table 1 are consider-
ably larger than the rate constant of 3 × 104 M–1 s–1 for the
reaction of methanol with the 1-(4-methoxyphenyl)propene
radical cation (8). This radical cation is analogous to the 4-
© 2003 NRC Canada

804
Can. J. Chem. Vol. 81, 2003
Fig. 6. Absorption spectra generated (᭹) 900 ns, (ٗ) 3.6 µs, and
(᭿) 6.4 µs after 308 nm irradiation of coniferyl alcohol in
acetonitrile with 0.10 mM n-Bu4NCl.
Fig. 7. Transient absorption spectra generated (᭹) 480 ns, (᭺)
3 µs, and (᭿) 6 µs after 308 nm laser irradiation of isoeugenol
in air-saturated acetonitrile with 0.7 mM HClO4. Inset shows
time-resolved changes in absorption at 580 nm because of the
decay of the isoeugenol radical cation in acetonitrile at different
concentrations of HClO4.
Table 2. Second-order rate constants for the reaction of the
coniferyl alcohol and isoeugenol radical cations with anionic re-
agents (22 ± 1°C).
Coniferyl alcohol+•
Isoeugenol+•
Anion
kNuc (M–1 s–1)
kNuc (M–1 s–1)
taining 0.7 mM HClO4. Clearly, the radical cation is still
generated under these conditions in the same manner as ob-
served for experiments in acetonitrile without added acid.
However, unlike the situation in neat acetonitrile where the
radical cation decayed rapidly, absorption due to the pres-
ence of the radical cation remains essentially unchanged
over the 10 µs time range of the experiment. Thus, the life-
time of the radical cation is dramatically increased in the
presence of low concentrations of the strong acid.
(1.9 ± 0.2) × 1010
(2.0 ± 0.2) × 1010
(2.5 ± 0.4) × 1010
(1.0 ± 0.1) × 1010
(1.4 ± 0.2) × 1010
(1.9 ± 0.3) × 1010
–
CH3CO2
CN–
Cl–
the coniferyl alcohol radical cation in neat acetonitrile or in
acetonitrile containing water. The 340 nm band is also unaf-
fected by the presence of oxygen.
This situation is more clearly illustrated in Fig. 7 (inset),
which shows the effect of perchloric acid concentration on
the lifetime of the radical cation in acetonitrile. In very di-
lute acidic acetonitrile, [HClO4] = 0.01 mM, absorption due
to the radical cation at 580 nm decays in a first-order man-
ner to almost zero absorbance with a rate constant of ap-
proximately 106 s–1. In the presence of 0.1 mM perchloric
acid, the radical cation no longer decays via simple first-
order kinetics. Instead, the radical cation now decays via two
sequential first-order processes. The fast process occurs with
a rate constant of ca. 106 s–1, while the second component is
considerably slower, taking place with a rate constant of ap-
proximately 2 × 104 s–1. As the acid content increases, the
significance of the faster component continues to decrease,
while the second component becomes more dominant. By
the time the acid concentration reaches 0.7 mM perchloric
acid, the fast component is no longer visible, and the radical
cation now decays with a single slow component having a
rate constant of ca. 2 × 104 s–1.
These results indicate that, at increasing acid concentra-
tions, the radical cation is no longer completely converted to
its conjugate base, the 4-vinylphenoxyl radical. Instead, the
radical cation decays until an equilibrium state consisting of
significant amounts of both the radical and the radical cation
is reached. The equilibrium state then decays more slowly
either by reaction of the 4-vinylphenoxyl radical or by
nucleophilic addition of solvent or residual water to the radi-
cal cation.
These observations indicate that the 4-vinylphenoxyl radi-
cal observed in neat acetonitrile and in acetonitrile contain-
ing hydroxylic additives is still generated upon reaction of
the radical cations with anionic nucleophiles. Thus, despite
the demonstrated ability of acetate, cyanide, and chloride to
react rapidly with styrene radical cations by nucleophilic ad-
dition, deprotonation continues to be an important reaction
pathway in the case of the coniferyl radical cation, due to
the presence of its acidic proton.
We considered the possibility that the vinylphenoxyl radi-
cals are generated as secondary products, with the
nucleophile adding to the radical cation to generate a substi-
tuted phenethyl radical that then undergoes a hydrogen atom
abstraction reaction with the pool of unreacted coniferyl al-
cohol. However, bimolecular rate constants for hydrogen
atom transfer from phenols to alkyl radicals are typically
slow (14), and such a reaction at the low concentration of
coniferyl alcohol (ca. 0.15 mM) used in the present study is
unlikely. Thus, the most likely explanation for the observa-
tion that reaction of the radical cation with anionic
nucleophiles leads to 4-vinylphenoxyl formation is that the
anionic reagents like chloride and cyanide are acting as
bases rather than nucleophiles.
Acidity of isoeugenol radical cation in acetonitrile
Figure 7 shows the transient absorption spectra generated
upon 308 nm irradiation of isoeugenol in acetonitrile con-
© 2003 NRC Canada

Schepp and Rodríguez-Evora
805
Fig. 8. Fraction long-lived isoeugenol radical cation measured
from the absorption at 580 nm after equilibration (Abseq) and
immediately after the laser pulse (Absinitial) as a function of acid
content in acetonitrile. The data point (᭿), corresponding to –log
[HClO4] = 7, is for the ratio measured in neat acetonitrile and
was not included in determining the line-of-best-fit.
By measuring the fraction of radical cation present after
the equilibrium state is reached as a function of acid concen-
tration, an estimate of the pKa of the radical cation in
acetonitrile can be determined. According to eq. [6], the ra-
tio of the absorption of the radical cation at 580 nm after the
equilibrium state is reached, Abs 5eq80nm, to the initial absorp-
tion of the radical cation at 580 nm immediately after the la-
ser pulse, Abs initial
concentration.
580nm, will change as a function of acid
580nm
[H+ ]
[H+ ] + Ka
Abs
Abs
eq
[6]
=
580nm
initial
The ratio is plotted as a function of H+ concentration in
Fig. 8. As can be seen, the ratio is small at low acid concen-
trations because of the 4-vinylphenoxyl radical being the
dominant species. As the concentration of acid increases, the
ratio increases as the equilibrium is shifted to the side of the
radical cation. At acid concentrations greater than 0.5 mM,
the ratio reaches a maximum value near one, and no further
increase is observed. Fitting of these data to eq. [6] then
leads to a value for Ka = 1.1 × 10–4 M, indicating that the
radical cation has a pKa ≈ 4 in acetonitrile. Since cationic ac-
ids are typically 7 to 8 orders of magnitude less acidic in
acetonitrile than in water (15), we can estimate a pKa ≈ –3 to
–4 for the acidity of the isoeugenol radical cation in aqueous
solution. This value is consistent with pKa values near –5 es-
timated for other methoxy-substituted phenol radical cations
in aqueous solution (16).
such as the 2-methoxy-4-vinylphenol radical cation are
currently under way to determine if 4-vinylphenol radical
cations lacking a β-alkyl group are capable of undergoing
coupling reactions in competition with deprotonation.
Acknowledgements
Conclusions
We are grateful to the Natural Sciences and Engineering
Research Council of Canada (NSERC) for financial support
of this research. We also thank Prof. Fran Cozens for helpful
discussions.
The results presented here demonstrate that the radical
cations of 4-vinylphenols in a dry polar solvent are short-
lived species, with deprotonation to the corresponding 4-
vinylphenoxyl radical being the major reaction pathway.
However, in an acidic environment, the lifetimes of the radi-
cal cations are dramatically enhanced. Under these condi-
tions, the radical cations should be sufficiently long-lived to
participate in an addition reaction with their neutral counter-
part to give lignan-like dimers in a manner similar to that
previously demonstrated for 4-methoxystyrene radical cat-
ions (8, 17–19). The enhanced lifetime also raises the possi-
bility that the biosynthesis of lignans could, at least in
principle, proceed via a radical-cation-mediated dimerization
process instead of via vinylphenoxyl radical coupling, espe-
cially if the enzyme active site provides a dry, acidic envi-
ronment.
It is worth noting that no quenching of the coniferyl alco-
hol or isoeugenol radical cations by their neutral precursor
was observed in our experiments even under acidic condi-
tions where the radical cations are long-lived. However, this
inability to detect coupling by time-resolved techniques does
not necessarily indicate that radical-cation-mediated dimeri-
zation cannot take place. For example, the decay of 4-
methoxystyrene derivatives such as the 1-(4-methoxy-
phenyl)-propene radical cation with a methyl group attached
to the β-position of the carbon–carbon double bond shows
no dependence on precursor concentration in time-resolved
experiments (20), while product studies clearly demonstrate
radical-cation-mediated dimerization does indeed occur (21).
Further experiments using potentially more reactive species
References
1. M.J. Kato, A. Chu, L.B. Davin, and N.G. Lewis. Phyto-
chemistry, 47, 583 (1998).
2. D.C. Ayres and J.D. Loike. Lignans. Chemical, biological and
clinical properties. Cambridge University Press, Cambridge,
U.K. 1990.
3. L.B. Davin, H.B. Wang, A.L. Crowell, D.L. Bedgar, D.M.
Martin, S. Sarkanen, and N.G. Lewis. Science (Washington,
D.C.), 275, 362 (1997).
4. G.J. Smith and I.J. Miller. Aust. J. Chem. 28, 193 (1975).
5. I.J. Miller and G.J. Smith. Tetrahedron Lett. 2277 (1973).
6. G. Leary. J. Chem. Soc. Perkins Trans. 2, 640 (1972).
7. F.L. Cozens, M. O’Neill, R. Bogdanova, and N.P. Schepp. J.
Am. Chem. Soc. 119, 10 652 (1997).
8. L.J. Johnston and N.P. Schepp. J. Am. Chem. Soc. 115, 6564
(1993).
9. J.J. Andre and G. Weill. Mol. Phys. 15, 97 (1968).
10. T.A. Gadosy, D. Shukla, and L.J. Johnston. J. Phys. Chem. A,
103, 8834 (1999).
11. G. Leary. J. Chem. Soc. Chem. Comm. 688 (1971).
12. K. Radotic, J. Zakrezewska, D. Sladic, and M. Jeremic. Photo-
chem. Photobiol. 65, 284 (1997).
13. E.J. Land and G. Porter. Trans. Faraday Soc. 59, 2016 (1963).
14. C. Evans, J.C. Scaiano, and K.U. Ingold. J. Am. Chem. Soc.
114, 4589 (1992).
15. D. Augustin-Nowacka, M. Makowski, and L. Chmurzynski.
Anal. Chim. Acta, 418, 233 (2000).
© 2003 NRC Canada

806
Can. J. Chem. Vol. 81, 2003
16. F.G. Bordwell and J.-P. Cheng. J. Am. Chem. Soc. 113, 1736
(1991).
19. N.P. Schepp and L.J. Johnston. J. Am. Chem. Soc. 116, 10 330
(1994).
17. S. Kadota, K. Tsubono, K. Makino, M. Takeshita, and T.
Kikuchi. Tetrahedron Lett. 28, 2857 (1987).
20. N.P. Schepp and L.J. Johnston. J. Am. Chem. Soc. 118, 2872
(1996).
18. R.M. Wilson, J.G. Dietz, T.A. Shepherd, D.M. Ho, K.A.
Schnapp, R.C. Elder, J.W. Watkins, L.S. Geraci, and C.F.
Campana. J. Am. Chem. Soc. 111, 1749 (1989).
21. F.D. Lewis and M. Kojima. J. Am. Chem. Soc. 110, 8664
(1988).
© 2003 NRC Canada