Mechanistic Study on Electrochemical Reduction of Calix[4]quinone in Acetonitrile Containing Water

Article abstract of DOI:10.1021/jp049864n

The mechanism of electrochemical reduction of calix[4]quinones (CQs) is of vital importance, as their reduction products, calix[4]hydroquinones (CQH ₈s), are known to self-assemble organic nanotubes, which in turn self-assemble novel metal nanostructures by a self-synthetic process of electrochemical reduction of solvated metals. We therefore conducted detailed studies of electrochemical reduction of CQ in rigorously dried acetonitrile (CH₃CN) and in CH₃CN containing varied amounts of water as well as perchloric acid. CQ undergoes a series of reversible one electron reductions in dry CH₃CN leading to anion radicals of each of p-benzoquinone units followed by formation of dianions. However, a series of following chemical-electrochemical reactions with protons takes place after the initial electron transfer in CH₃CN containing water. The final product of this series of reactions is identified as CQH₈ via an eight electron, eight proton reaction. The intermediate species are identified, and the reaction mechanism is elucidated using transient electrochemical, electrochemical quartz crystal microbalance and spectroelectrochemical measurements. The present results could be utilized to fine control the self-assembling reaction of nanotubes by electrochemical processes and to design electrochemically controllable nanomechanical devices.

Full text of DOI:10.1021/jp049864n

J. Phys. Chem. B 2004, 108, 4927-4936
4927
Mechanistic Study on Electrochemical Reduction of Calix[4]quinone in Acetonitrile
Containing Water§
Young-Ok Kim,† Young Mee Jung,† Seung Bin Kim,† Byung Hee Hong,‡ Kwang S. Kim,*,‡ and
Su-Moon Park*,†
Department of Chemistry and Center for Integrated Molecular Systems, and National CreatiVe Research
InitiatiVe Center for Superfunctional Materials, Pohang UniVersity of Science and Technology,
Pohang 790-784, Korea (ROK)
ReceiVed: January 11, 2004; In Final Form: February 18, 2004
The mechanism of electrochemical reduction of calix[4]quinones (CQs) is of vital importance, as their reduction
products, calix[4]hydroquinones (CQH8s), are known to self-assemble organic nanotubes, which in turn self-
assemble novel metal nanostructures by a self-synthetic process of electrochemical reduction of solvated
metals. We therefore conducted detailed studies of electrochemical reduction of CQ in rigorously dried
acetonitrile (CH3CN) and in CH3CN containing varied amounts of water as well as perchloric acid. CQ
undergoes a series of reversible one electron reductions in dry CH3CN leading to anion radicals of each of
p-benzoquinone units followed by formation of dianions. However, a series of following chemical-
electrochemical reactions with protons takes place after the initial electron transfer in CH3CN containing
water. The final product of this series of reactions is identified as CQH8 via an eight electron, eight proton
reaction. The intermediate species are identified, and the reaction mechanism is elucidated using transient
electrochemical, electrochemical quartz crystal microbalance and spectroelectrochemical measurements. The
present results could be utilized to fine control the self-assembling reaction of nanotubes by electrochemical
processes and to design electrochemically controllable nanomechanical devices.
Introduction
rigorously dried nonprotic solvents.5 However, the mechanism
of p-BQ reduction may be summarized as follows in aqueous
Calixarenes have received much attention due to their ability
to bind ions or molecules, acting as a receptor molecule for the
host-guest complexes.1 Electroactive calixarenes could be more
attractive because their electroactive groups including quinone2
or nitroaromatic moieties3 provide chemists with opportunities
to control the binding ability as well as the chemical selectivity.
The electrochemical control of binding properties for guest
molecules may lead to applications otherwise unavailable in
chemical and biochemical reaction systems.
solutions:6
Q + e- h Q-•
(1)
(2)
(3)
(4)
Q
-• + H+ h QH•
QH• + e- h QH-
QH- + H+ h QH2
It is known that CQH8, one of the redox active calixarenes
with four hydroquinone moieties, forms self-assembled organic
nanotube bundles.4 Ultrathin silver wires of a diameter of 0.4
nm were shown to grow inside the CQH8 nanotubes under
ambient aqueous conditions when silver nitrate is added to the
aqueous solution containing the nanotubes. Metal, semiconduc-
tor, or conductive polymer nanowires have been studied
extensively due to their possible uses in manufacturing nanoscale
devices.
Electrochemical reduction of CQ can be thought of as four
consecutive reductions of p-benzoquione (p-BQ) units as CQ
is made of four p-BQs connected in parallel by four -CH2-
groups in between. p-BQ undergoes reversible two one-electron
transfer reductions to form anion radicals and dianions in
where Q represents p-BQ. The sequence of the reactions is thus
described as two rounds of an electron-transfer coupled with a
chemical reaction, i.e., the ECEC or EHEH mechanism, unless
p-BQ is reduced further to its dianion, Q2-, in the absence of
protons. When the acidity of the medium is high, reactions 2
and 4 can be so fast that the whole reaction may be described
as a concerted two-electron, two-proton reaction
Q + 2H+ + 2e- h QH2
(5)
where QH2 is 1,4-dihydroquinone. This reaction is the basis of
what is known as a “quinone/hydroquinone electrode” for pH
measurements.
Now the electrochemistry of CQ in rigorously dried non-
aqueous solutions would most likely be expected to follow the
reactions summarized above. In other words, CQ would undergo
four consecutive reversible one-electron reductions to form anion
radicals of each p-BQ unit, followed by the formation of
dianions in dry nonprotic solvents. In aqueous solutions,
* To whom correspondence should be addressed. E-mail: smpark@
postech.edu. Phone: +82-54-279-2102. Fax: +82-54-279-3399.
† Department of Chemistry and Center for Integrated Molecular Systems.
‡ National Creative Research Initiative Center for Superfunctional
Materials.
§ Dedicated to the occasion of Professor Dong Han Kim’s retirement.
10.1021/jp049864n CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/20/2004

4928 J. Phys. Chem. B, Vol. 108, No. 15, 2004
Kim et al.
however, the sequence summarized above as reactions 1 through
4 or 5 would be observed depending on the acidity of the
medium.
Recently, electrochemical studies of CQ in nonaquous media
including CH3CN,7 N,N-dimethylformamide (DMF),8 and dichlo-
romethane (CH2Cl2)9 have been reported. All of these experi-
ments were conducted under an argon or nitrogen atmosphere
in rigorously dried solvents to remove residual water, and indeed
its electrochemistry in these media was described as the
formation of anion radicals of four p-BQ units before the
formation of dianions. Removal of water ensures the stability
of intermediate species such as anion radicals and/or dianions.
However, the electrochemistry of CQ in aqueous media should
be important just as well because CQH8 would be formed as a
final product upon reducing CQ in aqueous media. Further,
CQH8 is known to form organic nanotubes in aqueous media
under appropriate conditions.4b Water must play an important
role in making organic nanotube bundles by four hydrogen
bonds between OH groups of hydroquinone moieties and water
molecules. For this reason, our present study addresses elec-
trochemical reduction of CQ to CQH8 in solutions containing
water, primarily aiming at elucidation of the reaction mechanism
involved in electrochemical reduction of CQ to CQH8 employing
transient electrochemical, electrochemical quartz crystal mi-
crobalance (EQCM), and spectroelectrochemical techniques, and
its relevance to the formation of CQH8 nanotubes upon CQ
reduction is discussed. Thus, we conducted experiments in
rigorously dried CH3CN first as a control experiment, followed
by a series of experiments, in which varied amounts of water
and proton sources were used.
Figure 1. CVs for reduction of 1 mM CQ at a GC electrode in
rigorously dried CH3CN containing 0.10 M TBAPF6 as a supporting
electrolyte. The scan rate was 100 mV/s.
where ∆f is the change in frequency, ∆m is the mass change,
and Cf is the sensitivity constant. The Cf value was obtained to
be 6.68 ng/Hz/cm2 from calibration experiments using silver
deposition from a silver nitrate solution.
In situ UV-visible absorption spectra were taken with an
Oriel InstaSpec IV spectrometer with a charge-coupled device
(CCD) array detector, which was configured in a near normal
incidence reflectance mode using a bifurcated quartz optical
fiber.12,13 A Xenone lamp (75 W) was used as a light source.
The wavelength of the spectrograph was calibrated using a small
mercury lamp.
Experimental Method
CQ and CQH8 were synthesized and purified according to
the procedure described in the literature.10 CH3CN (Aldrich,
99.8%, anhydrous) was purified by three successive vacuum
distillations over P2O5 to remove organic impurities and water.
Tetrabutylammonium hexafluorophosphate (TBAPF6, Aldrich,
g 99.0%) was recrystallized twice from absolute ethanol and
dried overnight under reduced pressure at 100 °C. After oxygen
was removed, the solvent and the supporting electrolyte were
moved to a glovebox under an argon atmosphere. CQ was dried
at 100 °C under vacuum overnight. Acetonitrile was dried by
three freeze-pump-thaw cycles under vacuum. In some
experiments, where the water-CH3CN mixed solvent was used,
twice-distilled deionized water was used. In these cases, the
solution was bubbled with argon or nitrogen to remove oxygen.
The supporting electrolyte was mostly 0.10 M LiClO4 (Aldrich,
99.99%) unless otherwise stated.
Electrochemical experiments were performed using an EG&G
PAR model 273-A potentiostat/galvanostat. Glassy carbon and
gold disk (area ) 0.20 cm2) electrodes used as a working
electrode were polished successively with 1.0, 0.3, and 0.05
µm alumina slurries (Fisher) and cleaned by ultrasonication in
doubly distilled, deionized water before drying for use. The
platinum spiral wire and Ag/AgCl (or Ag wire) were used as
counter and reference electrodes, respectively.
Results and Discussion
Figure 1 shows typical cyclic voltammograms (CVs) of CQ
in rigorously dried CH3CN in an argon atmosphere. Similar
results have been reported on electrochemical reduction of CQ
in nonaqueous solutions such as CH3CN,7 DMF,8 and CH2Cl2.9
The CVs shown in Figure 1 are very similar to that described
by Gomez-Kaifer et al.7 except that the reduction waves are a
little more reversible and better resolved in our case. These
investigators attributed each CV wave to reduction of each p-BQ
unit to form an anion radical. Thus, the first four waves shown
in Figure 1 must be responsible for the consecutive formation
of anion radicals of four p-BQ units, and the two following
less reversible ones after the initial four waves should be further
reduction of two of those anion radicals to corresponding
dianions. The dianions are more reactive and the reversibility
of the last two waves are significantly reduced compared to the
first four waves. The difference in peak potentials between the
second and third waves are significantly larger than those
between the first and second as well as the third and fourth
waves, indicating that the first two electron transfers take place
at the first and third p-BQ units. The third electron transfer,
however, occurs at the third p-BQ molecule, which is located
between the two p-BQ anion radicals, resulting in a large
resistance to the incoming electron. This explains why the
potential differences between the first and second waves, as well
as the third and fourth waves, are smaller than that between the
second and third waves. After cycling the potential a few times,
the working electrode was passivated due perhaps to the low
solubility of reduced products formed from following chemical
reactions of anion radicals and dianions of CQ (vide infra).
The EQCM experiments were carried out using an EG&G-
Seiko model 917 quartz crystal analyzer (QCA) along with an
EG&G PAR model 273 A potentiostat/galvanostat. A 9 MHz
AT-cut quartz crystal coated with gold was used as a working
electrode. A decrease in frequency at this electrode corresponds
to an increase in mass according to the Sauerbrey equation11
∆f ) -Cf∆m
(6)

Electrochemical Reduction of Calix[4]quinone
J. Phys. Chem. B, Vol. 108, No. 15, 2004 4929
the differences in absorbance arising from the electrochemical
reaction during the potential cycling. A relatively broad absor-
bance peak at around 436 nm with its shoulders ranging from
350 to 500 nm grows as CQ is reduced until another broad band
begins to appear at bout 775 nm at a more negative potential.
It is not clear, however, whether other absorption bands emerge
as the potential becomes negative. For this reason, we ran two-
dimensional (2D) correlation analysis to resolve and identify
the absorption bands.14 In the 2D-analysis method, synchronous
and asynchronous spectra are obtained by applying a correlation
analysis on a series of spectra obtained using an external
perturbation such as changing potentials; synchronous spectra
represent the overall similarity between two separate spectral
intensity variations, whereas asynchronous spectra show the
measure of dissimilarity of the spectral intensity variations.14c
Because of the wide range of applications of this technique, it
has become one of the standard analytical techniques for
interpreting various types of spectroscopic data. The details are
described elsewhere,14 and no further description is given here.
The synchronous 2D correlation spectrum obtained during the
reoxidation of reduced CQ is shown in Figure 2b as a contour
map. The intensity of a synchronous 2D correlation spectrum
represents the simultaneous or coincidental changes of spectral
intensity variations measured at two different wavelengths, ν1
and ν2. A power spectrum extracted along the diagonal line of
the synchronous spectrum is also shown at the top of Figure
2b. Three autopeaks located at the diagonal line are clearly
observed at 389, 436, and 775 nm, which are not readily detected
in 1D spectra (Figure 2a). All cross-peaks located at off-diagonal
line have positive signs indicating that these absorption peaks
increase together. There are three autopeaks at 389, 436, and
775 nm, which also show up as cross-peaks as well. These cross-
peaks show positive signs indicating that the absorbance peaks
increase together.
The asynchronous 2D correlation spectrum shown in Figure
2c shows that the corresponding bands are occurring at different
rates. An asynchronous cross-peak develops only if the intensi-
ties of two spectral features change out of phase (i.e., delayed
or accelerated) with each other. The sign of an asynchronous
cross-peak becomes positive if the intensity change at ν1 occurs
predominantly before ν2 in the sequential order of perturbation.
It becomes negative, on the other hand, if the change occurs
after ν2. This rule, however, is reversed if the synchronous peak
is negative.14 The analysis of the asynchronous spectrum
indicates that the absorption bands emerge in the order of 436
f 389 f 775 f 550 nm. Further examination of the spectra
singled out from those displayed in Figure 2a (not shown)
reveals that the absorption peak at 436 nm increases monotoni-
cally until peaks at 389, 775, and 550 nm begin to show up
when the potential becomes more negative than about -1.0 V,
where dianions are produced. This leads to a conclusion that
the band at 436 nm results from the first reduction product,
i.e., anion radicals of p-BQ units. We thus assign the band at
436 nm to the anion radical, that at 775 nm to the free radical
generated by protonation of the anion radical, i.e., QH•, and
that at 550 nm to QH- from the similarity of their wavelengths
to those observed during electrolysis of p-BQ in DMSO in the
presence and absence of water.15,16 However, the band at 389
nm has not been reported in the literature, which we assign to
the dianion, Q2-, because it is observed only below -1.0 V.
Here the QH• band shows up after the dianion is generated
because of the slow kinetics of the protonation reaction of Q-•
due to the trace amounts of protons available in the CH3CN
solution. The species QH- would most likely be produced by
Figure 2. (a) In situ UV-vis absorption spectra recorded for reduction
of 1.0 mM CQ by scanning between 0.2 and -1.6 V at a GC electrode
in CH3CN with 0.1 M TBAPF6 at a scan rate of 100 mV/s. (b)
Synchronous and (c) asynchronous 2D correlation contour maps of CQ.
The dotted line indicates regions of negative correlation
To show that the four reduction waves do indeed result from
the generation of anion radicals of p-BQ units, we carried out
in situ spectroelectrochemical experiments. Figure 2 shows a
series of in situ UV-vis absorbance spectra simultaneously
obtained during reduction of 1.0 mM CQ by scanning the
potential from +0.20 to -1.60 V at a reflective glassy carbon
electrode in dry CH3CN with 0.10 M TBAPF6 used as a
supporting electrolyte. Since the reference absorbance spectrum
was taken at 0.20 V before the potential sweep started under
otherwise identical conditions, the spectra obtained represent

4930 J. Phys. Chem. B, Vol. 108, No. 15, 2004
Kim et al.
Figure 4. CVs recorded at a GC electrode for reduction of 0.50 mM
CQ in a 0.10 M LiClO4 solution in ACN containing 0.083 M H2O at
a scan rate of 50 mV/s.
water present (more will be discussed below). Second, the CV
peak shows shifts in potential upon addition of proton donors
such as HClO4. This is an indication that the proton causes the
following chemical reaction such as reactions 2 or 4 to the
electrogenerated species.17 Third, the current density is much
larger in the presence of water than in its absence, indicating
that more than just one electron transfer is taking place within
a single CV peak. From the fact that the current density is
increased drastically when water is added, we surmise that the
series of reactions summarized by reactions 1-4 are indeed
occurring within the single CV peak. Finally, the peak separation
between the cathodic and anodic peaks is significantly different
depending on whether there is an acid or not. The peak shift
depends on the acidity of the proton donor (see curves 2 and
3). This is demonstrated clearly in CVs recorded in CH3CN
with different acid dissociation constants (Figure 3b). The peak
separation is 212 mV without HClO4 but 558 mV with the acid
present. This indicates that the reversal (anodic) peak in the
presence of acid is due to the reoxidation of the final product,
perhaps CQH8, whereas that with water alone is due to the
reoxidation of intermediate species, which would not be the
immediate reduction product, CQ4-•. This was confirmed by
recording a CV for an authentic CQH8 under identical experi-
mental conditions (vide infra).
What appears to be a single CV peak when water is present
is resolved when the amount of water is reduced or the scan
rate is changed. Figure 4 shows CVs recorded in the presence
of 0.083 M H2O, in which the first and second CV peaks are
clearly resolved due to the slow rate of protonation such that
two electron transfer steps are resolved. Yet the current density
is significantly larger than that shown in Figure 1. The peak
separation between peaks A1 and C1 is only 137 mV, which
becomes 245 mV when the potential is scanned down to -1.0
V. This indicates that the relatively slow chemical reaction
proceeds following the electron transfer continuously to the final
product, and eventually an intermediate species close to the final
product is reoxidized at peak C2, if it is not the final product.
Note also that the anodic peaks in both Figures 3 and 4 show
characteristics of strong adsorption or precipitation. Thus, the
product, regardless of whether it is an intermediate or final, is
strongly adsorbed or precipitated on the electrode surface. More
will be addressed on this below. The difference in CV shapes
Figure 3. (a) CVs for reduction of 1.3 mM CQ with 2.2 M H2O (1,
‚‚‚‚), with 2.3 M H2O and 62 mM HClO4 (2, s), and with 62 mM
HClO4 and 0.14 M H2O (3, ---) in CH3CN with 0.1 M LiClO4. The
scan rate was 50 mV/s. (b) CVs for reduction of 1.3 mM CQ in an
acetonitrile solution of 2.3 M H2O and 0.1 M LiClO4 with 62 mM of:
(1) perchloric acid, (2) benzoic acid (pKa ) 4.19), (3) acetic acid (pKa
) 4.75), and (4) phenol (pKa ) 9.98). The working electrode was a
GC disk for all of these measurements, and the scan rate was 50 mV/s
and pKa values are the dissociation constants in pure water at 20 °C.
protonation of Q2- as well. Our spectroelectrochemical experi-
ments provide clear evidence that the first four CV waves result
from consecutive reduction of p-BQ upon potential sweep.
Since we obtained spectral information on intermediate
species from the results in dry CH3CN and yet the electrochemi-
cal reduction of CQ in aqueous media is of our primary interest,
the next series of experiments were carried out in the presence
of water. Because of the low solubility of CQ in water, we used
acetonitrile as a major solvent with varied amounts of water
and perchloric acid (HClO4) as a proton donor.
Figure 3 shows CVs recorded for a 1.0 mM CQ solution in
CH3CN with 0.10 M LiClO4 used as an electrolyte under various
experimental conditions. Figure 3a shows CVs recorded in CH3-
CN with (1) 2.20 M water only, (2) 2.34 M H2O and 0.062 M
HClO4, and (3) 0.062 M HClO4 and 0.14 M H2O that came
from HClO4. A few comments can be made on these CVs. First,
CVs shown here are significantly different from those obtained
in dry CH3CN in that a single peak is observed as opposed to
a number of peaks observed in dry CH3CN (Figure 1). Also,
the CV behavior shown by CQ in the presence of water is
entirely different from that of p-BQ in nonaqueous media with

Electrochemical Reduction of Calix[4]quinone
J. Phys. Chem. B, Vol. 108, No. 15, 2004 4931
Figure 6. Correlation between Ep and log[H2O] for the reduction of
0.50 mM CQ in an acetonitrile solution of 62 mM HClO4 and 0.10 M
LiClO4.
must be chelating the CQ molecules as a result of an equilibrium
reaction
CQ + pH2O h CQ(H2O)p
which has a formation constant
(7)
[CQ(H2O)p]
[CQ][H2O]p
Kf )
(8)
Based on eq 7, one can derive an equation
RT
nF
mRT
nF
ln[H+] +
0
Ep ) Ep
-
ln(Kf) +
Figure 5. CVs recorded at a GC electrode for 0.90 mM CQ in a 0.10
M TBAPF6 solution in ACN: (a) without and (b) with 2.20 M H2O
added. The scan rate was 50 mV/s.
RT
pRT
nF
ln(CQ(H2O)p] -
ln[H2O] (9)
nF
for an electron-transfer reaction
CQ + mH+ + ne- h CQHm
shown in Figure 3a1 and that in Figure 5b indicates that the
products, whether they are intermediate or final, are more soluble
in a solution containing TBAPF6 than in that having LiClO4.
In other words, a larger organic supporting electrolyte such as
TBAPF6 acts as a better surfactant for organic products than
their more hydrophilic counterpart such as LiClO4.
(10)
Here Ep0 is the peak potential for reaction 10 when no water is
present in a 62 mM HClO4 solution in CH3CN, and Ep is the
peak potential with water present. Equation 9 explains the
decrease in peak potential for an increase in [H2O] assuming
that the addition of water does not change the proton concentra-
tion. The assumption is not unreasonable as HClO4 is a strong
acid and its ability to protonate the solvent would not be affected
by the concentration of water. From the slope of the plot, we
obtain a p/n value of 4.0 with a Kf value of 1.8 × 1020. Thus,
eight water molecules would be incorporated into each p-BQ
unit.
To find if water molecules incorporated are charged, effects
of the ionic strength have been studied by varying the
concentration of LiClO4 and the result is shown in Figure 7.
As can be seen, the current for CQ reduction decreases as the
ionic strength increases for a solution containing 0.50 mM CQ,
62 mM HClO4, and varied amounts of LiClO4. The decrease in
current upon increase in ionic strength indicates that the
electroactive species undergoing the reaction is an ionic species
as its activity coefficient is affected by the ionic strength. Since
an experiment cannot be run at the zero ionic strength, no
standard state is available in this set of experiments, and there
is no way of knowing the actual activity coefficient. Although
we have no experimental information as to the charge of the
To see if indeed the ECEC reaction occurs within the single
CV peak, we now compare the CVs recorded in CH3CN without
(a) and with (b) water under otherwise identical experimental
conditions in Figure 5. The CVs show that the wave is more
drawn out in this partially aqueous solution, and the ratio of
the charges passed in the same potential range of +0.20 to
-0.771 V, where the second peak hits its minimum just before
its increase for the third peak in the absence of water, with and
without water is about 2.1 for each electron transferred,
indicating that the ECEC reactions must be occurring in a single
CV peak. Note that TBAPF6 was used as a supporting electrolyte
to record CVs shown in Figures 1 and 5, while LiClO4 was
used for the rest of the experiments. This is because LiClO4
was not used in rigorously dried solvents due to the difficulty
of removing hydrated water, whereas TBAPF6 had a solubility
problem in a medium with a high water content.
To see how the water molecules interact with a CQ molecule,
we carried out the water concentration dependency of CQ
reduction in CH3CN containing 62 mM HClO4, and the result
is shown in Figure 6. Since the peak potential shifts in a negative
direction as the water concentration increases, water molecules

4932 J. Phys. Chem. B, Vol. 108, No. 15, 2004
Kim et al.
Figure 7. Plot of ip vs [LiClO4] for reduction of 0.50 mM CQ in an
acetonitrile solution containing 62 mM HClO4 and 2.30 M H2O.
ionic species and, thus, on whether four water molecules are
all or partially protonated, our experiments allow us to conclude
that protonated water molecules are acting as ligands for each
p-BQ during the reduction. This is partially supported by a recent
theoretical work18 wherein Q2- shows a significant charge
transfer to the hydrating water molecules. Since this charge
transfer increases with an increasing number of water molecules,
it is expected that Q2- would give a full electron to neighboring
water molecules in the presence of eight water molecules,
resulting in Q--water or Q2--protonated water in the presence
of proton sources.
From the results described thus far, the electrochemistry of
CQ in dry CH3CN can be regarded as an extension of that of
p-BQ as can be seen from results shown in Figures 1 and 2,
whereas its electrochemistry in the presence of water is entirely
different from that of p-BQ in aqueous media. p-BQ undergoes
a reversible one-electron transfer to form an anion radical upon
reduction in nonaqueous media containing a large amount of
water and even in unbuffered aqueous solutions above pH
2.5.15,16 Its electrochemistry in buffered aqueous media or in
unbuffered solutions of pH lower than 2.5 is described as a
chemically reversible, but electrochemically irreversible, two
electron reaction according to reaction 5,5,19,20 which is the basis
for the so-called quinone electrode used for pH measurements.
No intermediate species has been detected for reaction 5 on an
electrochemical time scale including CV and/or chronoamper-
ometry. However, we saw clearly that the electrochemistry of
CQ in dried ACN is entirely different from that with water as
seen from Figures 1 and 3. When the medium is strongly acidic,
we observe a reaction similar to reaction 5, i.e.
Figure 8. (a) Three consecutively recorded CVs at a Au-coated QCA
electrode in a CH3CN solution of 2.3 M H2O, 62 mM HClO4, 0.10 M
LiClO4, and 1.2 mM CQ; (b) Frequency shifts concurrently recorded
with the CVs in (a). The scan rate was 50 mV/s.
Figure 8 shows CVs and frequency changes concurrently
recorded during CQ reduction in a CH3CN solution containing
62 mM HClO4 and 2.3 M H2O for the first three consecutive
cycles at a gold coated quartz crystal electrode. Here CQ is
reduced to CQH8 according to reaction 11. The CV shown in
Figure 8 is essentially identical to that shown in Figure 3a2
except that there is a large ohmic voltage drop (I‚R drop) in
the cell used for EQCM measurements due to its nonideal cell
configuration compared to a normal cell used for Figure 3.
Because of the I‚R drop, the CV shown in Figure 8 is
significantly sloped, and the peak separation became sizably
larger than that shown in Figure 3a2. Also, the CV recorded
from CQH8 was essentially the reverse of that shown in Figure
8, indicating that reaction 11 is chemically reversible but
electrochemically irreversible. This is another evidence that
CQH8 is produced when CQ is reduced in an organic medium
containing a proton source. While CQ is being reduced to CQH8,
the accumulation of the product, CQH8, on the electrode surface
is shown by decreases in the frequency. The decrease of ∼400
Hz/cycle is relatively small perhaps because much of CQH8
electrogenerated is dissolved as soon as it is produced. The
amount precipitated is only about 15∼20% of that theoretically
estimated from reaction 11. This is also shown by smaller anodic
waves in comparison to cathodic waves at both the glassy carbon
and gold electrodes [Figure 3a curves 2 and 3, and Figure 3b
curve 1) and the gold electrode (Figure 8a)]. Even after the
CQ + 8H+ + 8e- h CQH8
(11)
On the other hand, the intermediate species, which suggest that
the overall reaction consists of the elementary reactions given
as (1)-(4), are detected in the CH3CN solutions with just water
alone or in mildly acidic solutions. Thus, the rate of the reaction
sequence is slow enough for the intermediate species to be
observed (Figure 4). One reason for the higher stability of the
intermediate species generated during CQ reduction is because
the reaction products such as QH-•, QH•, and QH- freeze in
the solid phase on the electrode surface as soon as they are
produced. To show that this is the case, a series of microgravi-
metric experiments have been conducted using an EQCM
technique.

Electrochemical Reduction of Calix[4]quinone
J. Phys. Chem. B, Vol. 108, No. 15, 2004 4933
that the separation of the two anodic peaks has to do with the
scan rate, indicating that different intermediate species may be
detected depending on how much time has been allowed after
the first electron transfer. In the present case, the electrochemical
annealing of the reaction products by cycling repeatedly
produces different states of the precipitate such as different
crystalline forms or hydration numbers, which would cause the
potential shifts.
Two features noticed from Figure 9b are (1) much larger
amounts of insoluble products are precipitated initially on the
surface than those shown in Figure 8b upon CQ reduction, but
(2) a large fraction (∼80%) of the initial precipitate is redis-
solved upon reoxidation as can be seen from the large frequency
increase. This observation suggests that the intermediate species
are more insoluble than the final product, CQH8.
In efforts to see how the rate of precipitation/dissolution
corresponds to its current counterpart, a derivative signal (d∆f/
dt) obtained for the frequency change during the first cycle is
shown in Figure 9c along with the concurrently recorded CV.
In an ideal situation, where the electrogenerated products
precipitate immediately on the electrode surface, the two signals
should be identical.11,21 In our case, the two signals are quite
different, indicating that the electron transfer reaction does not
straightforwardly lead to precipitation. Also, it is seen that the
precipitation is delayed more during the anodic wave. That the
d∆f/dt signal shows a slight delay behind the CV rise during
the cathodic scan indicates that the intermediate species gener-
ated upon first electron transfer, most likely QH•, precipitates
almost immediately, and the intermediate thus precipitated
undergoes the following chemical and electron transfer reactions.
In other words, the second and following electron transfers as
well as chemical reactions take place inside the solid phase on
the electrode after the first reduction product is rapidly adsorbed.
We thus believe that further reactions to QH- and QH2 proceed
within the solid phase. This must be the reason the overpotentials
for the second and third steps are reduced so that only one broad
peak appears. Also, this is why the limiting currents are reduced
drastically to only about 1/7 ∼ 1/6 of the peak current because
the current is now controlled by the diffusion of CQ molecules
and protons through the newly formed solid phase.
Upon reversing the potential, the dissolution of electrogen-
erated products upon anodic scan occurs only after nearly all
of the anodic current has decayed. We assign the anodic peak
at a more negative potential to the reoxidation of QH- to QH•
and the second anodic peak following it to the oxidation of QH•
back to CQ. Depending on how long the products have been
electrochemically annealed, the compactness of the solid phase
and the physical and chemical transformations of the resulting
precipitate, would differ, resulting in further potential shifts
(Figure 9a). Part of the precipitate dissolves back into solution
when the electrogenerated products are oxidized back to CQ.
However, there still are some intermediate species and perhaps
CQH8 left on the surface because reoxidation of the reduced
products requires the diffusion of protons toward the solution,
which limits the reaction, and thus, part of CQH8 remains on
the surface. This explains why the dissolution of the precipitate
from the electrode surface does not take place until all the anodic
current has decayed in Figure 9c.
Figure 9. (a) Series of CVs at a gold-coated QCA electrode for three
consecutive potential cycles in a CH3CN solution of 2.2 M H2O, 0.1
M LiClO4, and 1.2 mM CQ; (b) Frequency shifts concurrently recorded
with the CVs in (a); (c) CV and derivative signals of the frequency
shift with respect to time vs voltage obtained from the first potential
cycle shown in (a) and (b) above. The broken line indicates the -d∆f/
dt signal while the solid line is the CVs. The scan rate was 50 mV/s.
anodic peak is past, no further weight decrease is observed,
indicating that the dissolution is not caused by reoxidation of
CQH8.
Figure 9 shows the first three consecutive CVs (a), corre-
sponding frequency changes (b), and the derivative signal of
the frequency shift with respect to time (c), recorded in a CH3-
CN solution containing only 2.2 M H2O. As the potential cycling
is continued, both the cathodic and anodic peak currents decrease
and the second anodic peak shifts in a more positive direction.
The CVs shown here are different from those shown previously
in that two anodic peaks are resolved instead of the one observed
at glassy carbon electrodes (Figure 3). Even at the GC electrode,
well resolved CVs were often obtained (Figure 4). We found
As already pointed out, an important difference between the
electrochemical behaviors of p-BQ and that of its cyclic
oligomer, CQ, in wet nonaqueous solutions arises from the fact
that the intermediate species of p-BQ, i.e., QH• and QH-, are
soluble, whereas the corresponding intermediate species of CQ

4934 J. Phys. Chem. B, Vol. 108, No. 15, 2004
Kim et al.
Figure 10. (a) UV-vis absorption spectra obtained in CH3CN
containing 2.2 M H2O with authentic samples of 2.1 × 10-5 M CQ
(solid line) and 3.3 × 10-5 M CQH8 (dotted line), respectively; and
(b) UV-vis absorption spectrum obtained after electrolysis of 1.0 mM
CQ at 0 V for 90 min in the CH3CN solution containing 2.2 M H2O,
62 mM HClO4, and 0.1 M LiClO4. The working electrode was a Pt
wire with Ag/AgCl and Pt gauze used as reference and counter
electrodes, respectively.
reduction are not. This explains why the electrochemistry of
CQ is not a simple extension of that of p-BQ.
To detect the intermediate species, we ran in situ spectro-
electrochemical experiments. Figure 10 shows (a) the absorption
spectra of authentic samples of CQ and CQH8 in a 2.2 M H2O
solution in CH3CN and (b) the in situ absorption spectrum
obtained during the electrolysis of 1 mM CQ in a wet CH3CN
solution with HClO4 at a reflective glassy carbon working
electrode. It is seen clearly that the only product upon
electrochemical reduction of CQ in wet CH3CN containing
HClO4 is CQH8 as was expected; the spectrum shown in Figure
10b is just made of the mixture of CQ and CQH8 in solution.
The CQH8 peak absorbing at 296 nm is overlaid on the shoulder
of the absorption peak of CQ at 247 nm.
Figure 11. (a) In situ UV-vis absorption spectra recorded at a
reflective glassy carbon electrode during potential cyc1es between 0.5
and -1.0 V in CH3CN with 2.2 M H2O, 0.10 M LiClO4, and 1.2 mM
CQ. The scan rate was 50 mV/s; (b) Synchronous and (c) asynchronous
2D correlation contour maps of derived from spectra shown in (a). The
dotted line indicates regions of negative correlation.
(Figure 11, parts b and c). There are a number of absorption
bands with their intensities varying during the potential sweep,
which are difficult to identify or resolve. For this reason, we
ran 2D analyses, and the 2D synchronous and asynchronous
correlation spectra are shown in Figure 11, parts b and c. The
synchronous spectrum identifies four auto peaks at 282, 391,
452, and 788 nm. The signs of cross-peaks of the synchronous
Figure 11a shows a series of in situ absorption spectra
recorded during a potential sweep in a wet CH3CN solution of
1.0 mM CQ and 2-D correlation spectra obtained from them

Electrochemical Reduction of Calix[4]quinone
J. Phys. Chem. B, Vol. 108, No. 15, 2004 4935
spectrum are all positive, indicating that they all grow during
the negative potential scan. Analysis of the asynchronous 2D
correlation spectrum indicates that the band at 595 nm occurs
at about the same time as that at 282 and 788 nm after the peaks
at 391 and 452 nm show up first. Thus, the peaks at 282, 595,
and 788 start to emerge at about the same time after the
absorption peak at 391 nm has shown up first, followed by that
at 452.
concerted eight-electron reduction to the final product, CQH8,
is obtained similar to reaction 5 as would be expected from the
electrochemistry of its monomeric counterpart, p-BQ. It was
also found that the CQ reduction potential shows an acidity
dependency, indicating that the proton accelerates the following
chemical reaction after the electron transfer. The water con-
centration dependency of the CQ reduction potential and the
ionic strength dependency of the reduction current in the
presence of HClO4 led us to conclude that protonated water
molecules interact with the CQ molecule by forming a complex
whose formation constant is 1.8 × 1020.
All of the absorption bands identified here were also present
in the spectra obtained in dry CH3CN except for the band at
282 nm, though all of the bands are red or blue shifted compared
to those observed in dry CH3CN shown in Figure 2 and those
shown in Figure 10. We believe relatively large spectral shifts
observed here are due primarily to poor spectral resolution (∼0.4
nm at 546 nm) as a result of a relatively wide slit width used
(25 µm) for a higher spectral sensitivity and also heavy spectral
overlaps of broad absorption peaks resulting from the generation
of large amounts of intermediate species, most of which absorb
light in a broad spectral region. Accepting this, we assign the
282 nm peak to CQH8, 391 nm to CQ2-, 452 nm to CQ-•, 595
nm to CHQ-, and 788 nm to CHQ• in reference to our earlier
assignments for the bands in spectra obtained in dry CH3CN.
The sequence of the band emergence is also very revealing.
The appearance of the 391 nm band (CQ2-) first in the series
must be due to the fact that the last protonation step to produce
the CQH8 is the slowest rate-limiting step of the whole CQ
reduction reactions in the presence of water. The emergence of
the 452 nm band (Q-•) next in the series indicates that the
protonation of the anion radical is the second slowest step and
begins to be populated a little after the electrolysis begins.
Finally the fact that CHQ• and CHQ- appear last in the series
along with CQH8 indicates that further reduction of CHQ• to
CHQ- and its protonation to produce the final product CQH8
is relatively fast. Only after all of these series of reactions
proceed, the final product, CQH8, is produced.
Our assignments here for the results obtained from the two
extremes of the spectroelectrochemical experiments conducted
without water (Figure 2) and with water (Figure 11) are
consistent with each other. These results show indeed that the
reaction is taking place via the elementary reaction steps
summarized by reactions 1 through 4. The fact that the sequence
of absorption band emergence is different in dry and wet CH3-
CN indicates that the reaction rates are modulated by the
presence of a reactant such as proton as well as its concentration.
Although the spectroelectrochemical experiments do not provide
information on reaction kinetics, it certainly gives clues to the
identification of the rate-limiting step, which should be a starting
point for computer simulation of the currents to evaluate the
related rate constants.
The results obtained from the EQCM and spectroelectro-
chemical experiments show that CQ reduction in CH3CN
containing just 2.2 M water undergoes a series of ECEC
reactions to eventually produce the final product, CQH8. The
EQCM measurements indicate that what appears to be a single
CV peak actually consists of a series of electron transfer and
following chemical reactions. The final product, CQH8, and all
of the intermediate species produced during the electrolysis of
CQ in dry CH3CN are detected in the spectra recorded during
the electrolysis in wet CH3CN.
It has been reported that CQH8 is known to form organic
nanotubes via four hydrogen bonds formed between OH groups
of hydroquinone (HQ) moieties and water molecules.4 It is for
this reason that the electrochemical reduction of CQ is important
as a good understanding of its mechanism would allow the
various parameters for CQ reduction and thus the tube formation
to be straightforwardly controlled. In fact, our present results
suggest that CQ is already in the form, which would lead to
the tubular structure upon reduction. Our preliminary results
show that relatively large nanotubes are formed upon CQ
reduction, and studies to fine-tune this reaction are currently
under way in our laboratory. In addition, the exploitation of
structural changes from CQ (partial cone shape) to CQH8 (cone
shape) or vice versa could be utilized to design a novel
electrochemically controllable nanomechanical devices, like
those of quino-cyclophane systems.22
Acknowledgment. This work was supported by a grant from
the Korea Science and Engineering Foundation (KOSEF)
through the Center for Integrated Molecular Systems, Pohang
University of Science and Technology. The graduate stipends
for Y.O.K. were provided through the BK 21 program by the
Ministry of Education of Korea.
References and Notes
(1) Gutsche, C. D. Calixarenes; The Royal Society of Chemistry:
Cambridge, U.K., 1989.
(2) Reddy, P. A.; Gutsche, C. D. J. Org. Chem. 1993, 58, 3245.
(3) (a) Kelderman, E.; Derhaeg, L.; Heesink, G. J. T.; Verboom, W.;
Engbersen, J. F. J.; van Hulst, N. F.; Persoons A.; Reinhoudt, D. N. Angew.
Chem., Int. Ed. Engl. 1992, 31, 1075. (b) Verboom, W.; Durie, A.; Egberink,
R. J. M.; Asfari Z.; Reinhoudt, D. N. J. Org. Chem. 1992, 57, 1313.
(4) (a) Hong, B. H.; Bae, S. C.; Lee, C.-W.; Jeong, S.; Kim, K. S.
Science 2001, 294, 348. (b) Hong, B. H.; Lee, J. Y.; Lee, C.-W.; Kim, J.
C.; Kim, K. S. J. Am. Chem. Soc. 2001, 123, 10748. (c) Kim, K. S.; Suh,
S. B.; Kim, J. C.; Hong, B. H.; Lee, E. C.; Yun, S.; Tarakeshwar, P.; Lee,
J. Y.; Kim, Y.; Ihm, H.; Kim, H. G.; Lee, J. W.; Kim, J. K.; Lee, H. M.;
Kim, D.; Cui, C.; Youn, S. J.; Chung, H. Y.; Choi, H. S.; Lee, C.-W.; Cho,
S. J.; Jeong, S.; Cho, J.-H. J. Am. Chem. Soc. 2002, 124, 14268.
(5) Patai, S.; Rappoport, Z. The Chemistry of the Quinonoid Com-
pounds; Wiley: New York, 1974.
(6) (a) Turner, W. E.; Elving, P. T. J. Electrochem. Soc. 1965, 112,
1215. (b) Eggins, B. R.; Chambers, J. Q. J. Electrochem. Soc. 1970, 117,
186. (c) Koshy, V. J.; Sawayambunathan, V.; Periasamy, N. J. Electrochem.
Soc. 1980, 127, 2761.
Conclusions
The electrochemical reduction of CQ in CH3CN with and
without water and/or HClO4 has been investigated using
electrochemical, EQCM, and spectroelectrochemical techniques.
CQ undergoes consecutive four one-electron transfer reactions,
followed by further reduction to dianions in rigorously dried
CH3CN. This was demonstrated by spectra recorded during the
potential sweep, in which the spectrum of one species increases
until after the fifth and sixth CV peaks are observed, where the
formation of the most reactive species, dianions, is detected.
The results obtained in dry CH3CN provided the basis to the
interpretation of the electrochemistry of CQ in CH3CN contain-
ing water and/or HClO4. When CQ is electrochemically reduced
in CH3CN in the presence of a strong acid such as HClO4, a
(7) Gomez-Kaifer, M.; Reddy, P. A.; Gutsche, C. D.; Echegoyen, L.
J. Am. Chem. Soc. 1994, 116, 3580.

4936 J. Phys. Chem. B, Vol. 108, No. 15, 2004
Kim et al.
(8) Suga, K.; Fujihara, M.; Morita, Y.; Agawa, T. J. Chem. Soc.,
Faraday Trans. 1991, 87, 1575.
(9) Casnati, A.; Comelli, E.; Fabbi, M.; Bocchi, V.; Mori, G.; Ugozzoli,
F.; Manotti-Lanfredi, A. M.; Pochini, A.; Ungaro, R. Recl. TraV. Chim.
Pays-Bas 1993, 112, 384.
(10) Morita, Y.; Agawa, T.; Nomura, E.; Taniguchi, H. J. Org. Chem.
1992, 57, 3658.
(11) Rubinstein, I. Physical Electrochemistry; Marcel Dekker: New
York, 1995.
(12) Pyun, C.-H.; Park, S.-M. Anal. Chem. 1986, 58, 251.
(13) (a) Zhang, C.; Park, S.-M. Anal. Chem. 1988, 60, 1639. (b) Zhang,
C.; Park, S.-M. Bull. Korean Chem. Soc. 1989, 10, 302.
(14) (a) Noda, I. J. Am. Chem. Soc. 1989, 111, 8116. (b) Noda, I. Appl.
Spectrosc. 1990, 44, 550. (c) Noda, I.; Dowrey, A. E.; Marcott, C.; Story,
G. M.; Ozaki, Y. Appl. Spectrosc. 2000, 54, 236A.
(15) Shim, Y.-B.; Park, S.-M. J. Electroanal. Chem. 1997, 425, 201.
(16) Pyun, C.-H.; Park, S.-M. J. Electrochem. Soc. 1985, 132, 2426.
(17) Gupta, N.; Linschitz, H. J. Am. Chem. Soc. 1997, 119, 6384.
(18) Manojkumar, T. K.; Choi, H. S.; Tarakeshwar, P.; Kim, K. S. J.
Chem. Phys. 2003, 118, 8681.
(19) Bard, A. J.; Lund, H. Encyclopedia of Electrochemistry of the
Elements; Marcel Dekker: New York, 1978; Vol. XII.
(20) Cantor, C. R. Energy Transduction in Biological Membranes;
Springer-Verlag: New York, 1990.
(21) Lee, H. J.; Cui, S.-Y.; Park, S.-M. J. Electrochem. Soc. 2001, 148,
D139.
(22) Kim, H. G.; Lee, C.-W.; Yun, S.; Hong, B. H.; Kim, Y.-O.; Kim,
D.; Ihm, H.; Lee, J. W.; Lee, E. C.; Tarakeshwar, P.; Park, S.-M.; Kim, K.
S. Org. Lett. 2002, 4, 3971.