Infrared and UV-vis spectra of phenoxonium cations produced during the oxidation of phenols with structures similar to vitamin e

Article abstract of DOI:10.1016/j.electacta.2010.07.096

Several phenols with structures similar to vitamin E were oxidised and the intermediate species produced were characterised by in situ infrared and UV-vis spectroscopies. The Fourier transform infrared (FTIR) measurements were performed by chemically oxidising the phenols with 2 mol equiv. of NO ⁺SbF₆^- in CH₃CN and recording the spectra between 1900 and 1300 cm^-1 with an attenuated total reflectance (ATR) probe utilising a fiber conduit and a diamond composite sensor. The compounds that formed long-lived phenoxonium cations displayed two IR absorbances at 1665 (±15) cm^-1 and one at 1600 (±10) cm^-1 associated with the carbonyl, symmetric ring stretch and asymmetric ring stretch modes. The para-quinones are one of the long-term products of oxidation of the phenols, and displayed solution phase IR absorbances at 1650 (±10) cm^-1. In situ electrochemical UV-vis experiments performed during the oxidation of the phenols led to the detection of bands due to the phenoxonium cations at 295 (±5) and 440 (±15) nm and due to the para-quinones at 260 (±10) nm. The concentration of the substrate and the water content of the solvent had a major effect on the yields of the intermediates and products that were produced during the oxidation reactions.

Full text of DOI:10.1016/j.electacta.2010.07.096

Electrochimica Acta 55 (2010) 8863–8869
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Infrared and UV–vis spectra of phenoxonium cations produced during the
oxidation of phenols with structures similar to vitamin E
Shanshan Chen, Hong Mei Peng, Richard D. Webster∗
Division of Chemistry and Biological Chemistry, School of Physical and Mathematical Sciences, Nanyang Technological University, 21 Nanyang Link,
Singapore 637371, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 8 May 2010
Received in revised form 30 July 2010
Accepted 30 July 2010
Available online 10 August 2010
Several phenols with structures similar to vitamin E were oxidised and the intermediate species produced
were characterised by in situ infrared and UV–vis spectroscopies. The Fourier transform infrared (FTIR)
measurements were performed by chemically oxidising the phenols with 2 mol equiv. of NO+SbF6 in
−
CH3CN and recording the spectra between 1900 and 1300 cm−1 with an attenuated total reflectance (ATR)
probe utilising a fiber conduit and a diamond composite sensor. The compounds that formed long-lived
phenoxonium cations displayed two IR absorbances at 1665 ( 15) cm−1 and one at 1600 ( 10) cm−1
associated with the carbonyl, symmetric ring stretch and asymmetric ring stretch modes. The para-
quinones are one of the long-term products of oxidation of the phenols, and displayed solution phase
IR absorbances at 1650 ( 10) cm−1. In situ electrochemical UV–vis experiments performed during the
oxidation of the phenols led to the detection of bands due to the phenoxonium cations at 295 ( 5) and
440 ( 15) nm and due to the para-quinones at 260 ( 10) nm. The concentration of the substrate and the
water content of the solvent had a major effect on the yields of the intermediates and products that were
produced during the oxidation reactions.
Keywords:
Electrochemical oxidation
Phenoxonium
Nitrosonium cation
Reactive intermediate
FTIR
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
and the phenol can be regenerated in 100% yield [8,9]. The phenox-
onium cations of the other forms of the tocopherols have lifetimes
Phenolic compounds have an important function in many bio-
logical processes; either in a complementary fashion such as acting
as antioxidants, or in a negative role such as acting as endocrine
disruptive chemicals [1]. In both scenarios the oxidative behaviour
of the compounds is important; thus electrochemical studies have
been used extensively to study and characterise the oxidation reac-
tions [2–5]. The initial electrochemical oxidation of phenols can be
considered to occur through a series of electron transfer and pro-
ton transfer reactions that can be represented in a “square-scheme”
mechanism, with the lifetimes of the oxidised compounds varying
considerably depending on the structure of the phenol [4,5].
The tocopherols are a series of naturally occurring phenolic com-
pounds consisting of several structurally related forms (␣, ␤, ␥, ␦)
that differ in the degree of methylation of the aromatic ring. It has
been found that ␣-tocopherol (the most methylated tocopherol),
undergoes an electrochemical ECE oxidation process (−2e−/−H+) to
form a phenoxonium cation (PO+), which is surprisingly long-lived
in acetonitrile and dichloromethane solutions (Scheme 1) [6–16].
The reaction is chemically reversible on the milliseconds and hours
time-scales, providing the moisture content of the solvent is low,
considerably less than the phenoxonium cation of ␣-tocopherol
[10]. In addition to ␣-tocopherol, there are presently only a few
examples of compounds that are known to form long-lived phenox-
onium cations [13,17–21], so it is interesting that one derived from
a natural compound (␣-tocopherol) is long-lived, which has led to
speculation that it may have some biological importance [10,12].
A recent electrochemical study examined a number of phe-
nolic compounds (POH) with structures similar to ␣-tocopherol
and identified several compounds whose associated phenoxonium
cationssurvivedforatleastseveral minutes in acetonitrile solutions
at room temperature [13]. It was proposed (based on processes
detected during voltammetric and electrolysis experiments) that
the phenoxonium cations (PO+) reacted with water in the solvent to
form para-quinone hemiketals (Q(OH)), which in some cases could
also be electrochemically reduced back to the parent phenols, or
reacted further to form the para-quinones (Q) (Scheme 1).
The aim of this study is to provide spectroscopic characterisa-
tion of the phenoxonium cations, that can only be usually produced
as very short-lived species during flash photolysis experiments in
aqueous solutions [22–29]. Under transient conditions the iden-
tity of intermediate species can be difficult to determine, thus
characteristic spectroscopic data from compounds with relatively
long lifetimes can be useful in identifying compounds under con-
ditions where they are likely to be considerably more reactive
∗
Corresponding author. Tel.: +65 6316 8793; fax: +65 6791 1961.
E-mail address: webster@ntu.edu.sg (R.D. Webster).
0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.07.096

8864
S. Chen et al. / Electrochimica Acta 55 (2010) 8863–8869
were controlled with a Thermo Electron Neslab RTE 740 circulating
bath and a thermostatted cell holder. Solution phase background
subtracted in situ FTIR experiments were conducted with a Mettler
Toledo iC10 spectrometer utilising a diamond composite ATR fiber
conduit probe [35,36].
3. Results and discussion
3.1. Electrochemical mechanism
Scheme 1 shows the general mechanism for the electrochemical
oxidation in CH3CN or CH2Cl2 of the class of phenols (POH) used
in this study that have a five- or six-membered ring containing
an ether group, with the ether oxygen in the para-position to the
hydroxyl group [13]. In CH3CN containing Bu4NPF6 as the support-
ing electrolyte, the electrochemical mechanism is similar for all the
compounds and involves the initial one-electron oxidation of POH
at approximately +0.5 V vs. Fc/Fc+ (where Fc = ferrocene) to form
the radical cation (POH+•). POH+• rapidly dissociates via the loss
of the hydroxyl proton to form the phenoxyl neutral radical (PO•),
which is then immediately further oxidised at the electrode sur-
face by one-electron to form the phenoxonium cation (PO+) [13]. A
homogeneous disproportion mechanism may alternatively account
for the second electron transfer step [16]. Although the expression
“phenoxonium” is used to describe the cations, molecular orbital
calculations and NMR experiments on the model phenoxonium
cation of ␣-tocopherol (compound 1b) indicated that most of the
positive charge is located on the carbon atom adjacent to the ether
oxygen atom, rather than on the oxygen atom [9].
The lifetime of PO+ depends on the substituents in the R1–R5
positions (Scheme 1), as well as the temperature and solvent com-
position [13]. Some phenoxonium cations have been found to be
long-lived (for at least several minutes) and can be reversibly
reduced back to the starting materials. A typical cyclic voltam-
mogram of a compound (1a) that forms a phenoxonium ion that
Scheme 1. General oxidation mechanism for the phenols. R1, R2, R3, R4 = H or Me.
R5 = H, Me, CH2OH, OMe, OEt, CO2Et, or CHCH2. n = 1 or 2 [13].
(such as in biological conditions or in aqueous environments). A
range of phenols were selected based on previous electrochemical
experiments which indicated they formed long-lived phenoxo-
nium cations upon oxidation in acetonitrile (Scheme 2, compounds
1a, 2a, 3a, 4a, 6a, 7a) [6–16]. For comparison purposes, the long-
term para-quinone oxidation products of two of the phenols (1d
and 7d) were examined, in addition to one compound (5a) that
was thought to form the para-quinone hemiketal upon oxidation
[13].
2. Experimental
˚
HPLC grade CH3CN (Tedia) was dried over 3 A molecular sieves
and Bu4NPF6 was prepared and purified by standard methods
[30]. NOSbF6 (99%) used for chemical oxidation experiments
was obtained from Sigma–Aldrich. The synthesis of the phenols
and quinones used for analysis has been described previously
(Scheme 2) [13,31].
Cyclic voltammetric (CV) experiments were conducted with a
computer controlled Eco Chemie Autolab PGSTAT 100 with an ADC
fast scan generator. Working electrodes were 1 mm diameter pla-
nar Pt disks, used in conjunction with a Pt auxiliary electrode and
an Ag wire reference electrode connected to the test solution via
a salt bridge containing 0.5 M Bu4NPF6 in CH3CN. Accurate poten-
tials were obtained using ferrocene as an internal standard that
was added to the test solution at the end of the measurements.
Bulk oxidation experiments were performed in a divided controlled
potential electrolysis (CPE) cell separated with a porosity no. 5
(1.0–1.7 ␮m) sintered glass frit [32].
In situ UV–vis spectra were obtained with a Perkin-Elmer
Lambda 750 spectrophotometer in an optically semi-transparent
thin layer electrochemical (OSTLE) cell (pathlength = 0.05 cm) using
a Pt mesh working electrode [33,34]. UV–vis experiments at 253 K
Scheme 2. Structures of starting materials.

S. Chen et al. / Electrochimica Acta 55 (2010) 8863–8869
8865
ples of ␣-tocopherol attached to an electrode surface indicated that
the phenol is immediately converted to the quinone in high yield
when the oxidation occurs in aqueous solutions [16], but in non-
aqueous solvents the yields of the quinones (from the phenols) are
low [31].
The phenoxonium cations can also be prepared in two one-
electron steps, if the reaction is commenced with solutions
containing the phenolate anions [3,7,8]. Preparation of the pheno-
late starting materials requires initial reaction of the phenols with
an organic soluble base, such as a tetraalkylammonium hydrox-
ide, which are commercially available as concentrated aqueous
solutions and difficult to dry [7]. Therefore, because of the higher
amount of water that is present in solutions of the phenolate anions
(due to the addition of the base), the phenoxonium cations (which
are reactive with water) were prepared directly from the phenol
starting materials.
3.2. Infrared spectroscopy
In this study, the phenoxonium cation reaction intermediates
were prepared by chemical oxidation of the phenolic starting mate-
rials in preference over electrochemical oxidation. It has not been
conclusively proven that the identical oxidation pathway occurs
for chemical oxidation of the phenols compared to heterogeneous
oxidation at solid electrodes [8,10,12], although it is known that
chemical oxidation of ␣-tocopherol with 2 mol equiv. of NO+ pro-
duces the phenoxonium cation in 100% yield [9,11]. Fig. 2 shows
in situ background subtracted ATR–FTIR spectroscopic data that
were obtained by chemically oxidising a range of phenols in CH3CN
Fig. 1. Representative cyclic voltammograms recorded at a scan rate of 100 mV s−1
in CH3CN containing 0.2 M Bu4NPF6 for 5 mM of the phenols and related com-
pounds at 22 ( 2) ◦C. (—) Chemically reversible −2e−/−H+ transformation of phenol
(1a) into phenoxonium ion. (. . .) Reduction (+2e−/+2H+/−H2O) of para-quinone
hemiketal (5c) back to phenol. (- - -) Chemically reversible +1e− transformation of
para-quinone (1d) into semiquinone. Data modified from reference [13].
survives for at least the voltammetric time-scale (a few seconds) is
shown in Fig. 1 (solid line) where a forward oxidation peak (Epox) is
observed at approximately +0.5 V vs. Fc/Fc+ and a reverse reduction
peak (Epred) is detected at approximately +0.2 V vs. Fc/Fc+. Both the
forward and reverse processes correspond to the transfer of two-
electrons and one proton. The wide separation between the Epox and
Epred peaks (ꢀEpp) is due to the electron transfer reactions occur-
ring sequentially in two one-electron steps at different potentials,
with a proton transfer reaction interspersed between the electron
transfers (an ECE mechanism) [8–16]. The voltammograms in Fig. 1
are representative of a general reaction scheme of phenols; how-
ever, not all of the processes may be detectable for all compounds.
For example, the phenoxonium cations of the majority of phenols
are not long-lived, therefore, the peak for the reduction of the phe-
noxonium back to the phenol at +0.2–+0.4 V is seldom observed
(especially at slow scan rates).
Most PO+ compounds react quickly with trace water (or other
nucleophiles) in the solvent to form the para-quinone hemiketals
(Q(OH)) as a major product, which can convert into the para-
quinones (Q), although the reaction is concentration dependant
and also involves the formation of other unidentified products
(Scheme 1) [13]. Evidence for the para-quinone hemiketal inter-
mediates came from the observation that during electrolysis
experiments new species were produced with reductive peaks at
approximately −0.4 V vs. Fc/Fc+ for some of the compounds (such
as 5c) (Fig. 1, dotted line). The reduction potential of the interme-
diate occurred between that observed for the phenoxonium ion
(solid line) and that observed for the stable para-quinone (dashed
line). Applying a potential more negative than −0.4 V vs. Fc/Fc+
allowed the intermediate compound be reduced back to the start-
ing material indicating that it was unlikely to be a ring-opened
structure such as the para-quinone (which cannot be reduced back
to the starting materials under the mild conditions used for elec-
trochemistry), and supporting the formation of the para-quinone
hemiketal. The para-quinone hemiketals rearrange slowly in solu-
tion into the para-quinones which can be reduced at approximately
−1 V vs. Fc/Fc+ in a one-electron chemically reversible process (such
as for 1d) (Fig. 1, dashed line). Experiments on solid state sam-
Fig. 2. Background subtracted in situ ATR–FTIR spectra obtained by oxidising 0.1 M
of the phenols in CH3CN with 2 mol equiv. of NOSbF6 22 at ( 2) ◦C. The spectra
represent 256 scans recorded with 4 cm−1 resolution. (. . .) Starting phenols. (—)
Immediately after oxidation, except for 1d, which is a pure compound.

8866
S. Chen et al. / Electrochimica Acta 55 (2010) 8863–8869
with 2 mol equiv. of NOSbF6. The dotted lines in Fig. 2 are the ini-
tial spectra of the phenols and the solid lines are the spectra of
their associated phenoxonium cations. The phenols in Fig. 2 were
chosen based on previous controlled potential electrolysis exper-
iments that indicated that their associated phenoxonium cations
had lifetimes of at least several minutes [13]. CH3CN is the pre-
ferred solvent for the oxidation experiments because it affords the
greatest lifetimes of the phenoxonium ions and because it has few
bands in the infrared region that interfere with the FTIR detection of
the phenoxonium cations. NO+ was used as the oxidant because the
chemical oxidation occurs much more rapidly than electrochemi-
cal oxidation, because its reduction potential is sufficiently positive
to oxidise many phenols and because it is cleanly reduced to NO(g)
which is easily removed from solution [9].
The ATR probe used for the in situ FTIR experiments has a usable
range between approximately 1900–650 cm−1 [35,36], although it
is the carbonyl region (1800–1600 cm−1) that is the most useful in
characterising the phenoxonium ions [8–10]. The sensitivity lim-
its of the ATR–FTIR probe require relatively high concentrations of
substrates, therefore; all experiments were conducted with 0.1 M
concentrations of the phenols. There is presently no evidence to
indicate that the short term oxidation mechanism of the phenols to
form the phenoxonium cations is concentration dependant [8–16].
The infrared spectra of the oxidised forms of the phenols in Fig. 2
appear similar and show three characteristic bands between 1681
and 1604 cm−1 (the highest wavenumber band often appearing
as a shoulder). Compounds 1b and 2b mainly differ in that they
contain a 6-membered heterocyclic ring while compounds 3b and
4b contain a 5-membered heterocyclic ring. Previous molecular
orbital calculations on compound 1b indicated that the structure
was indeed similar to that of a quinone (as drawn in Fig. 2), con-
sisting of a carbonyl group and a cyclic diene [9]. Therefore, the
infrared region between 1800 and 1500 cm−1 would expect to show
absorbances due to a carbonyl stretch, a symmetric diene ring
stretch and an asymmetric diene ring stretch. Nanosecond time
resolved resonance Raman spectroscopy in CH3CN/H2O solutions of
a 4-acetoxy-4-aryl-2,5-cyclohexadienone led to the detection of a
phenoxonium cation, which displayed a C O stretch (1635 cm−1), a
symmetric diene ring stretch (1593 cm−1) and an asymmetric diene
ring stretch (1525 cm−1) [28].
Molecular orbital calculations on compound 1b predict that two
of the bands are associated with the symmetric and asymmetric
ring stretching modes and one band is associated with the carbonyl
stretching mode, although in each case other coupled vibrations
within the molecules contribute to the modes [9]. The absorbance
at close to 1600 cm−1 is predicted to be associated with the asym-
metric ring stretching mode [9]. The two bands that occur at 1663
(which appears as a shoulder) and 1650 cm−1 (in the spectrum of
1b) are predicted to be due to the carbonyl stretch and symmetri-
cal ring stretch, respectively [9]. The molecular orbital calculations
predict the carbonyl band to be at the higher wavenumber, which, if
correct, means that it is weaker in intensity than the symmetric ring
stretching mode. Therefore, in the FTIR spectra of all the phenoxo-
nium cations in Fig. 2, the absorbance at the highest wavenumber
(which appears as a shoulder in the spectrum of 3b at 1681 cm−1), is
likely to be associated with the carbonyl stretching mode. The spec-
trum of the quinone of the ␣-tocopherol model compound (1d) is
provided in Fig. 2 as a comparison to the phenoxonium cations, and
shows one strong band in the carbonyl region at 1645 cm−1, possi-
bly because the C O stretch and symmetric ring stretching modes
are overlapping.
Fig. 3. Background subtracted in situ ATR–FTIR spectra obtained by oxidising 0.1 M
5a in CH3CN with 2 mol equiv. of NOSbF6 22 at ( 2) ◦C. The spectra represent 256
scans recorded with 4 cm−1 resolution (. . .) 5a. (—) The spectrum of 5b was obtained
immediately after oxidation and the spectrum of 5c was obtained from 5b after the
addition of 0.1 M of water.
ing modes of the methyl groups on the quaternary carbon [9]. The
C–O stretching modes in ethers normally occur between 1300 and
1000 cm−1 [37]. Nevertheless, an X-ray crystallographic study on
compound 1b showed that the quinoid like carbon–oxygen ether
bond in the phenoxonium cation is shorter than a normal C–O bond
[11], which could account for the shift in the absorbance of the C–O
bond to higher wavenumbers in the cation.
The lifetime of the phenoxonium cations is partly determined
by how reactive they are with nucleophiles, such as water, in the
solvent [10]. In an ultra-dry environment it has been found that the
phenoxonium cation of the model compound of ␣-tocopherol (1b)
can survive for at least several weeks at −27 ◦C [9,11]. Since water
is always present in organic solvents, the phenoxonium cations
may be longer lived at higher concentrations where the effects of
trace water are less, providing they do not react in a dimerisation
or related processes. The water concentration of the CH3CN inside
the reaction vessel was calculated to be 25 ( 5) mM, based on the
published experimental procedure where values of |E1 − E2| for a
quinone are correlated with the amount of water present (E1 and E2
are the first and second reduction potentials of vitamin K1) [38,39].
The high humidity of the local environment combined with the
cell used for the in situ spectroscopic analysis make it difficult to
completely exclude the water from the solvent.
Fig 3 shows infrared spectra that were obtained before and
after the oxidation of compound 5a with 2 mol equiv. of NO+. The
spectrum obtained immediately after the oxidation of 5a showed
absorbances in the carbonyl region typical of the phenoxonium
cation at 1660 (shoulder), 1651 and 1605 cm−1, and with a strong
absorbance close to 1500 cm−1. The phenoxonium cation of com-
pound 5a (5b) is known to be more reactive with water than
compounds 1b–4b [13], thus when an equal molar amount of
water was added to the solution, the spectrum of 5b immediately
diminished in intensity to most likely form 5c (the para-quinone
hemiketal) (Fig. 3). There are several reports on the existence of
para-quinone hemiketals related to the tocopherols, although the
compounds are difficult to obtain in pure form and rapidly convert
in the presence of acid to the quinones [31,40–44]. The infrared
stretching frequency of the carbonyl group of the para-quinone
hemiketals of the tocopherols have been assigned between 1680
and 1630 cm−1 [40–44], although there is some ambiguity whether
The spectra of all the phenoxonium cations in Fig. 2 also show
a strong absorbance at close to 1500 cm−1, which is the strongest
absorbance between 1900 and 1300 cm−1. Molecular orbital cal-
culations on compound 1b predicted several absorbances close to
1500 cm−1 that coincide with the C–O stretch coupled with bend-

S. Chen et al. / Electrochimica Acta 55 (2010) 8863–8869
8867
dation of the phenolic compounds shown in Fig. 4 are consistent
with the formation of the phenoxonium cations [8–10].
The geometry and shape of platinum mesh electrode inside the
flat UV–vis cell is not conducive to obtaining clearly interpretable
voltammetric peaks, as is achievable for cyclic voltammetry exper-
iments at planar disk electrodes. Therefore, the applied oxidation
potential of the Pt mesh electrode in the OSTLE cell was set by
slowly increasing the potential until a positive current of a few ␮A
was first observed, and then applying a further +100 mV past this
point, which produced an initial positive current of approximately
100 ␮A. The oxidation reaction was run until the current decreased
to <1 ␮A, or until the present UV–vis spectrum was identical to the
previous UV–vis spectrum. In order to regenerate the starting mate-
rial, a reducing potential of 0 V was applied to the Pt mesh electrode.
It was not possible to collect accurate coulometry data during the
UV–vis measurements because only the solution inside the flat por-
tion of the cell underwent exhaustive oxidation, while the bulk of
the solution was located outside the flat portion of the cell. Previous
bulk controlled potential coulometry experiments on the phenols
in a conventional two compartment electrolysis cell [32] confirmed
that the oxidation reaction occurs via two-electrons per molecule
[13].
At 25 ◦C the starting phenols showed an absorption band at
approximately 295 nm with ε ≈ 3000 L cm−1 mol−1 [45,46]. As the
temperature decreased the band narrowed and appeared to split
into two closely spaced absorbances that could not be completely
resolved at −40 ◦C [8]. During the in situ oxidation of the phe-
nols at −20 ◦C, an intense band was observed at 293–296 nm
(ε ≈ 11,000 L cm−1 mol−1) and a less intense very broad band at
432–452 nm (ε ≈ 700 L cm−1 mol−1) (the ε-values supersede the
previously reported values [10,15]). The spectra are very simi-
lar to those obtained of the phenoxonium cations of the ␣- and
␤-tocopherols [10]. In each case the spectrum of the starting mate-
rial could be completely regenerated by applying a 0 V reducing
potential to the electrode, indicating that the oxidised compounds
were long-lived. Laser flash photolysis of a 4-acetoxy-4-aryl-2,5-
cyclohexadienone in O2 saturated CH3CN/H2O solutions led to the
detection of two transient species with absorption bands at 460 and
360 nm [26,28]. The band at 460 nm was assigned to the phenoxo-
nium cation (similar to this work), and the band at 360 nm assigned
to a phenoxyl free radical [28]. Under the slower time-scale exper-
iments in this work, any phenoxyl radicals produced are unlikely
to be sufficiently long-lived to detect, thus the bands at approxi-
mately 300 and 450 nm are both associated with the phenoxonium
cations.
Fig. 5 shows the UV–vis spectra obtained during the electroly-
sis of compounds 4a, 6a and 7a. In this instance, the phenoxonium
cations (4b, 6b and 7b), formed by electrochemical oxidation were
known to be more reactive than compounds 1b, 2b and 3b [13].
The initial oxidation of the phenols did produce the phenoxonium
cations, as evidenced by the initial increase in intensity of the bands
at approximately 300 and 436–452 nm. However, as the electrol-
ysis progressed, these bands diminished in intensity again and a
strong band (or pair of bands) was detected at 254–264 nm. The
band detected at 254–264 nm is very similar to that observed in the
UV–vis spectra of the quinones shown in Fig. 6. Therefore, it appears
that in the low analyte concentration UV–vis measurements, there
is a greater tendency for the oxidised compounds to react to form
the quinones (4d, 6d and 7dꢀ) as one of the final products, which
accounts for the higher apparent reactivity of 4b in the UV–vis mea-
surements compared to in the infrared measurements (in addition,
the UV–vis measurement were performed over a longer time-scale
allowing more time for a hydrolysis reaction). It was found that the
spectra of the starting phenols could not be regenerated by apply-
ing a reducing potential after the initial oxidation of compounds
4a, 6a and 7a.
Fig. 4. In situ electrochemical UV–vis spectra obtained in CH3CN containing 0.2 M
Bu4NPF6 during the oxidation of approximately 1 mM of the phenols at −20 ◦C. (. . .)
Starting phenols. (—) Intermediate scans. (—) Final scan.
this may be a strong ring stretching mode, as is observed in the
phenoxonium cations.
The band that was present in the spectrum of 5b at 1651 cm−1
appears to shift to 1647 cm−1 in the spectrum of 5c. The band
at 1647 cm−1 for 5c could be due to either a carbonyl stretch or
ring stretching mode (or an overlap of both). Although the band
at 1647 cm−1 occurs at a similar wavenumber to that observed for
a quinone, such as 1d in Fig. 2, it is unlikely that 5b converts so
rapidly and quantitatively to the quinone in the presence of an
equal molar amount of water. Normally the conversion of phe-
nols to quinones in non-aqueous solvents is a relatively low yield
reaction with numerous side products [13,31]. The broad band that
occurs at approximately 1720 cm−1 in the spectrum of 5c is due to
water.
3.3. In situ electrochemical UV–vis spectroscopy
Fig. 4 shows in situ electrochemical UV–vis experiments that
were conducted on compounds 1a, 2a and 3a which are known to
form relatively long-lived phenoxonium cations [13]. The spectro-
scopic experiments were performed over a time-scale of around
30 min and needed to be performed at low concentrations (<1 mM)
to avoid overloading the instrument detector. There is a disparity in
the concentration of the phenolic compounds used in the infrared
and UV–vis spectroscopic experiments because of the infrared
experiments requiring very high concentrations in order to obtain
acceptable signal-to-noise ratios. The low concentration of analytes
used for the UV–vis measurements leads to there being a substan-
tially higher ratio of water present that favours the formation of the
para-quinone hemiketals or para-quinones over the phenoxonium
cations. Nevertheless, the UV–vis spectra obtained during the oxi-

8868
S. Chen et al. / Electrochimica Acta 55 (2010) 8863–8869
and UV–vis spectra of the phenoxonium cations, hemiketals and
quinones associated with all phenolic compounds.
4. Conclusions
Infrared spectroscopic measurements made during the oxida-
tion of several phenols led to the detection of a range of products
depending on the lifetimes of the oxidised compounds. The com-
pounds that formed long-lived phenoxonium cations showed two
infrared absorbances between 1685 and 1650 cm−1 (sometimes
overlapping into one band) due to the carbonyl stretching and
symmetric ring stretching modes and a band at approximately
1600 cm−1 due to the asymmetric ring stretching mode. The
phenoxonium cations also showed a very intense band at approx-
imately 1500 cm−1, which is possibly associated with a C–O single
bond mode(s). The solution phase spectra of the ring-opened
quinones showed a single strong infrared band in the carbonyl
region at approximately 1650 cm−1. The UV–vis spectra of phenox-
onium cations showed intense bands at 293–296 nm and relatively
less intense bands at 434–460 nm, compared to the long-term para-
quinone oxidation products that showed bands at 256–264 nm.
Acknowledgment
This work was supported by a Singapore Government Ministry
of Education research grant (T208B1222).
References
[1] Z. Rappoport (Ed.), The Chemistry of Phenols, Wiley, New York, 2003.
[2] K. Yoshida, Electrooxidation in Organic Chemistry: the Role of Cation Radicals
as Synthetic Intermediates, Wiley, New York, 1983, p. 174.
[3] O. Hammerich, B. Svensmark, in: H. Lund, M.M. Baizer (Eds.), Organic Electro-
chemistry, third ed., Marcel Dekker, New York, 1991 (Chapter 16).
[4] A. Rieker, R. Beisswenger, K. Regier, Tetrahedron 47 (1991) 645.
[5] H. Eickhoff, G. Jung, A. Rieker, Tetrahedron 57 (2001) 353.
[6] V.D. Parker, J. Am. Chem. Soc. 91 (1969) 5380.
Fig. 5. In situ electrochemical UV–vis spectra obtained in CH3CN containing 0.2 M
Bu4NPF6 during the oxidation of approximately 1 mM of the phenols at −20 ◦C. (· · ·)
Starting phenols. (—) Intermediate scans. (—) Final scan.
[7] R.D. Webster, Electrochem. Commun. 1 (1999) 581.
[8] L.L. Williams, R.D. Webster, J. Am. Chem. Soc. 126 (2004) 12441.
[9] S.B. Lee, C.Y. Lin, P.M.W. Gill, R.D. Webster, J. Org. Chem. 70 (2005) 10466.
[10] G.J. Wilson, C.Y. Lin, R.D. Webster, J. Phys. Chem. B 110 (2006) 11540.
[11] S.B. Lee, A.C. Willis, R.D. Webster, J. Am. Chem. Soc. 128 (2006) 9332.
[12] R.D. Webster, Acc. Chem. Res. 40 (2007) 251.
[13] H.M. Peng, R.D. Webster, J. Org. Chem. 73 (2008) 2169.
[14] W.W. Yao, H.M. Peng, R.D. Webster, J. Phys. Chem. B 112 (2008) 6847.
[15] H.M. Peng, B.F. Choules, W.W. Yao, Z. Zhang, R.D. Webster, P.M.W. Gill, J. Phys.
Chem. B 112 (2008) 10367.
[16] W.W. Yao, H.M. Peng, R.D. Webster, J. Phys. Chem. C 113 (2009) 21805.
[17] K. Dimroth, W. Umbach, H. Thomas, Chem. Ber. 100 (1967) 132.
[18] B. Speiser, A. Rieker, J. Chem. Res. (S) (1977) 314.
[19] B. Speiser, A. Rieker, J. Electroanal. Chem. 102 (1979) 373.
[20] B. Speiser, A. Rieker, J. Electroanal. Chem. 110 (1980) 231.
[21] A. Vigalok, B. Rybtchinski, Y. Gozin, T.S. Koblenz, Y. Ben-David, H. Rozenberg,
D. Milstein, J. Am. Chem. Soc. 125 (2003) 15692.
[22] M. Novak, S.A. Glover, J. Am. Chem. Soc. 126 (2004) 7748.
[23] M. Novak, S.A. Glover, J. Am. Chem. Soc. 127 (2005) 8090.
[24] S.A. Glover, M. Novak, Can. J. Chem. 83 (2005) 1372.
[25] M. Novak, M.J. Poturalski, W.L. Johnson, M.P. Jones, Y. Wang, S.A. Glover, J. Org.
Chem. 71 (2006) 3778.
[26] Y.-T. Wang, J. Wang, M.S. Platz, M. Novak, J. Am. Chem. Soc. 129 (2007) 14566.
[27] M. Novak, A.M. Brinster, J.N. Dickhoff, J.M. Erb, M.P. Jones, S.H. Leopold, A.T.
Vollman, Y.-T. Wang, S.A. Glover, J. Org. Chem. 72 (2007) 9954.
[28] Y.-T. Wang, K.J. Jin, S.H. Leopold, J. Wang, H.-L. Peng, M.S. Platz, J. Xue, D.L.
Phillips, S.A. Glover, M. Novak, J. Am. Chem. Soc. 130 (2008) 16021.
[29] Y.-T. Wang, M. Novak, J. Org. Chem. 74 (2009) 7697.
Fig. 6. UV–vis spectra of approximately 1 mM of the quinones obtained in CH3CN
at 22 ( 2) ◦C.
[30] A.J. Fry, W.E. Britton, in: P.T. Kissinger, W.R. Heineman (Eds.), Laboratory Tech-
niques in Electroanalytical Chemistry, Marcel Dekker, New York, 1984 (Chapter
13).
[31] A. Patel, T. Netscher, L. Gille, K. Mereiter, T. Rosenau, Tetrahedron 63 (2007)
5312.
[32] R.D. Webster, A.M. Bond, T. Schmidt, J. Chem. Soc. [Perkin] 2 (1995) 1365.
[33] R.D. Webster, G.A. Heath, A.M. Bond, J. Chem. Soc. [Dalton] (2001) 3189.
[34] R.D. Webster, G.A. Heath, Phys. Chem. Chem. Phys. 3 (2001) 2588.
[35] R.D. Webster, J. Chem. Soc. [Perkin] 2 (2002) 1882.
The FTIR spectra obtained during the chemical oxidation of 6a
and 7a at high concentrations with NOSbF6 were complicated and
showed a number of bands in the carbonyl region that were weak
in intensity suggesting that a number of products were formed (in
addition to the phenoxonium cations or quinones) and were diffi-
cult to interpret. Therefore, although Scheme 1 and Fig. 1 represent
a general reaction pathway, it is likely that in many cases other
reactions can occur making it difficult to fully characterise the FTIR
[36] R.D. Webster, Electrochem. Commun. 5 (2003) 6.
[37] D.L. Pavia, G.M. Lampman, G.S. Kriz, Introduction to Spectroscopy, Thomson
Learning, Singapore, 2001, p. 48.

S. Chen et al. / Electrochimica Acta 55 (2010) 8863–8869
8869
[38] Y. Hui, E.L.K. Chng, C.Y.L. Chng, H.L. Poh, R.D. Webster, J. Am. Chem. Soc. 131
[42] A. Nilsson, U. Palmquist, T. Pettersson, A. Ronlán, J. Chem. Soc. [Perkin] 1 (1978)
696.
(2009) 1523.
[39] Y. Hui, E.L.K. Chng, L.P.-L. Chua, W.Z. Liu, R.D. Webster, Anal. Chem. 82 (2010)
1928.
[43] K. Omura, J. Org. Chem. 54 (1989) 1987.
[44] A. Pelter, S.M.A. Elgendy, J. Chem. Soc. [Perkin] 1 (1993) 1891.
[45] M. Sow, G. Durocher, J. Photochem. Photobiol. A 54 (1990) 349.
[46] K.R. Naqvi, T.B. Melø, T. Jávorfi, S. González-Pérez, J.B. Arellano, Phys. Chem.
Chem. Phys. 11 (2009) 6401.
[40] W. Dürckheimer, L.A. Cohen, Biochem. Biophys. Res. Commun.
262.
9 (1962)
[41] R.T. Borchardt, L.A. Cohen, J. Am. Chem. Soc. 95 (1973) 8308.