Photochemical and Electrochemical Oxidation Reactions of Surface-Bound Polycyclic Aromatic Hydrocarbons

Article abstract of DOI:10.1021/jp037393j

The oxidation reactions of two PAH, i.e., pyrene and anthracene, attached covalently to silica, indium-doped tin oxide (ITO), and gold surfaces were studied. For both pyrene and anthracene, electrochemical and photochemical oxidation produced first a monohydroxy-PAH followed by the formation of dihydroxy/dione derivatives. Although the products of electrochemical or photochemical transformations were the same, the mechanism of each reaction might be different. In the case of electrochemical oxidation, no effect of oxygen dissolved in the solution on the voltammetric behavior was observed. The photodegradation of pyrene could occur through two paths, i.e., electron transfer and the reaction of pyrene with generated singlet oxygen. Singlet oxygen was not involved in the photochemical transformation of surface bound pyrene and the electron-transfer mechanism was dominant. The reaction of anthracene with water led to the formation of hydroxyanthracene and anthraquinones. The electron-transfer mechanism was dominant as the initial step of anthracene oxidation as well.

Full text of DOI:10.1021/jp037393j

1038
J. Phys. Chem. B 2004, 108, 1038-1045
Photochemical and Electrochemical Oxidation Reactions of Surface-Bound Polycyclic
Aromatic Hydrocarbons
Maciej Mazur† and G. J. Blanchard*,‡
Laboratory of Electrochemistry, Department of Chemistry, UniVersity of Warsaw,
02-093 Warsaw, Pasteura 1, Poland, and Department of Chemistry, Michigan State UniVersity,
East Lansing, Michigan 48823-1322
ReceiVed: NoVember 7, 2003
We have studied the oxidation reactions of two polycyclic aromatic hydrocarbons (PAHs), pyrene and
anthracene, attached covalently to silica, indium-doped tin oxide (ITO), and gold surfaces. Attachment of the
PAHs to the substrates was accomplished by established covalent coupling methods. For both pyrene and
anthracene, we find that electrochemical and photochemical oxidation produces first a monohydroxy-PAH
followed by the formation of dihydroxy/dione derivatives. We have used cyclic voltammetry and steady-
state excitation and emission spectroscopy to identify the products of the surface reactions. The uniformity
of our results for two chromophores and three different types of substrates demonstrates the generality of our
findings and rules out the involvement of surface enhancement phenomena in these processes.
Introduction
we have found that the chromophores are not stable to UV
exposure under ambient atmospheric conditions, and it is
important that we elucidate the chemical and/or physical basis
for this behavior so that it can be avoided when possible.
We report here on the covalent binding of pyrene and
anthracene derivatives to silica, gold, and ITO substrates. Pyrene
is used frequently as a polarity-sensitive probe to study
relaxation dynamics in bulk solvents30 and in a variety of
adlayers.31 Though anthracene has seen much less use as an
optical probe, it is structurally different from pyrene and thus
represents a useful system for comparison. An important
limitation to the use of PAHs as optical probes is their chemical
instability. To this point, it has been unclear whether this
instability is surface enhanced or not, and a significant issue
we consider here is the generality of the propensity of PAHs to
degrade at surfaces. There is some literature precedent for
photooxidation of PAHs. Pyrene and anthracene physisorbed
to various substrates transform into a number of products, mainly
alcohols and ketones, when exposed to oxygen and UV
light.32-36 We are interested in the influence of the substrate
and whether this effect is dominated by the presence of UV
photons and oxygen or requires both to operate. Our data point
to similar photooxidation behavior for both pyrene and an-
thracene bound covalently to conducting, semiconducting, and
insulating substrates. We identify the oxidation products of these
PAHs using emission spectroscopy and cyclic voltammetry. Our
data point to similar chemistry being operative for both PAHs
bound to different surfaces. If there is indeed any surface-
enhancement effect, it arises solely from the proximity of the
bound PAHs to any adventitious water physisorbed to the
surface. The degradation of surface-bound pyrene and an-
thracene requires the presence of both UV light and oxygen.
The design and synthesis of monomolecular assemblies on
solid substrates has drawn much interest in recent years due to
the numerous potential applications of this class of materials in
areas ranging from chemical separations to nonlinear optics and
biosensor devices. The ability to manipulate the chemical
structure of surfaces with molecular depth resolution provides
a means to control properties such as friction, wetting, adhesion,
electron transfer, and corrosion protection.1-5 A testament to
the strength of this research endeavor is the structural diversity
in deposition chemistry that has appeared. Monomolecular layers
can be prepared by a variety of methods, including Langmuir-
Blodgett deposition,6,7 spontaneous self-assembly,8,9 ionic10-13
and covalent binding,14,15 and polyelectrolyte layer deposition.16-23
The most widely studied and, arguably, the best characterized
monomolecular assemblies are alkanethiols on gold.3,5,24,25 Our
interest in monolayer and multilayer structures lies in their utility
for chemical and biosensing applications. Because thiol mono-
layers are limited to forming on metallic substrates such as gold,
silver, copper, and mercury and a few semiconductor sur-
faces,26,27 other approaches to the formation of adlayers on
different substrates are required. Films of silanes,28 fatty acids,8,9
phosphonates,10-13 and acid chlorides15,29 have been demon-
strated on both conductive and insulating substrates, providing
a rich variety of chemistry from which to construct interfaces
useful for sensing.
To effect chemical sensing, it is necessary to incorporate some
means of signal transduction into the interface. Our group has
incorporated optical chromophores into monolayer and multi-
layer assemblies, and we have used steady-state, time-resolved
and nonlinear optical spectroscopies to interrogate these imbed-
ded chromophores. We have found that the identity and
functionality of the incorporated chromophore can have a
significant influence on the information acquired. In some cases
Experimental Section
Chemicals. All chemicals were of the highest quality com-
mercially available: 1-aminopyrene (Aldrich, >98%), 1-amino-
anthracene (Aldrich, 90%), adipoyl chloride (Aldrich, 98%),
4-methylmorpholine (Aldrich, 99%), 1-hydroxypyrene (Aldrich,
* Corresponding author. E-mail: Blanchard@chemistry.msu.edu.
† University of Warsaw.
‡ Michigan State University.
10.1021/jp037393j CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/18/2003

Oxidation Reactions of Surface-Bound PAHs
J. Phys. Chem. B, Vol. 108, No. 3, 2004 1039
98%), 1,6-pyrenedione (Aldrich), pyrene (Aldrich, 99%), 1,4-
anthracenedione (Aldrich), 9,10-anthracenedione (Aldrich), 11-
mercaptoundecanoic acid (MUA) (Aldrich, 95%), triethylamine
(Aldrich, 99.5%), acetic anhydride (CCI, ACS grade), sodium
borohydride (Aldrich, 98%), methylene blue (Fluka, 97%),
acetonitrile (Aldrich, anhydrous, 99.8%), ethyl acetate (Spectrum
Chemical, ACS grade), ethanol (Aldrich, ACS grade), sulfuric
acid (Aldrich, 99.999%), 1-pentanol (Aldrich, 99+%). Aqueous
solutions were prepared from water distilled in-house.
Steady-State Emission Spectroscopy. Emission and excita-
tion spectra were recorded with a Spex Fluorolog 3 spectrometer
using a 3 nm band-pass for excitation and a 3 nm band-pass
for emission collection.
Electrochemical Measurements. Electrochemical measure-
ments were conducted with a PC-controlled Electrochemical
Workstation (CH Instruments Model 650A), using a small
volume three-electrode cell with Pt wire as the counter electrode.
All potentials are quoted versus a Ag/AgCl/1 M KClaq reference
electrode.
Substrate Preparation. Quartz slides (NSG Precision Cells,
Inc., P/N 10040 UV fused silica windows, nonfluorescent) and
gold electrodes (Au, 99.99%; 0.2 cm2) were cleaned by
immersion in piranha solution for ca. 20 min. Caution! Piranha
solution reacts violently with organic matter and should be
handled with extreme care! Indium-doped tin oxide (ITO)
films (Bayview Optics, Dover-Foxcroft, ME) deposited on the
same quartz slides were used for spectroscopic measurements
and ITO films on glass substrates (Delta Technologies) were
used for electrochemical measurements.
Bonding 1-Aminopyrene and 1-Aminoanthracene to Quartz
and ITO. Quartz and ITO substrates were reacted with adipoyl
chloride (0.3 mL) in dry acetonitrile (10 mL), using 4-methyl-
morpholine (0.3 mL) as a Lewis base, under reduced pressure
for 1 h. The reacted substrates were removed from the reaction
vessel, rinsed with dry acetonitrile and ethyl acetate, and dried
under a stream of nitrogen. Then the acid chloride terminal
functionalities were reacted with either 1-aminopyrene or
1-aminoanthracene by exposing substrates with an adipoyl
chloride adlayer to a 30 mM solution of the amino-substituted
PAH in dry acetonitrile for 1 h. The substrates were then
removed from the reaction vessel, washed with dry acetonitrile
and ethyl acetate, and dried under a stream of nitrogen.
Bonding 1-Aminopyrene and 1-Aminoanthracene to Gold.
Monolayers of 11-mercaptoundecanoic acid (MUA) were
prepared by immersing clean gold substrates into 10 mM MUA
in acetonitrile solution for 4 h. Then, the monolayer-modified
gold surface was removed from the solution, rinsed with
acetonitrile and dried. The terminal carboxylic acid groups were
then transformed into interchain anhydrides,37 by placing the
substrate in a solution of acetic anhydride (0.3 mL) and
triethylamine (0.3 mL) in dry acetonitrile (10 mL) for 1 h. The
reacted substrates were removed from the reaction vessel, rinsed
with dry acetonitrile and ethyl acetate, and dried under a stream
of nitrogen. The acid anhydride terminal functionalities were
reacted with either 1-aminopyrene or 1-aminoanthracene by
exposing the anhydride-terminated substrates to a 30 mM
solution of the amino-substituted PAH in dry acetonitrile for 1
h. The substrates were then removed from the reaction vessel,
washed with dry acetonitrile and ethyl acetate, and dried under
a stream of nitrogen.
Figure 1. (a) Excitation and emission spectra of aminopyrene attached
to an adipoyl chloride adlayer on quartz and (b) aminopyrene solution
phase spectra taken in 1-pentanol.
et al.38 for the oxidation of pyrene in dry acetonitrile solu-
tion. 1-Hydroxypyrene, 1,6-pyrenedione, 1,4-anthracenedione,
and 9,10-anthracenedione thin films on gold were prepared by
immersing the substrates into acetonitrile solution (50 mM) of
the compound, then removed from the solution, and allowed to
dry. The modified electrodes were then subjected to electro-
chemical characterization in aqueous solution.
Preparation of 1,6-Dihydroxypyrene. 1,6-Dihydroxypyrene
was prepared by reduction of 2 mL of 10-4 M 1,6-pyrenedione
in 1-pentanol with 2 mL of saturated sodium borohydride in
1-pentanol.
Preparation of Amidoanthracene Acetate. 1-Aminoan-
thracene (0.2 g) was reacted with excess acetic anhydride (15
mL) in dry acetonitrile (15 mL) for ca. 20 h. The reaction
mixture was poured into distilled water, which resulted in
precipitation of amidoanthracene acetate. The product was then
collected by filtration on a separation funnel (80% yield).
Determination of Reactivity with Singlet Oxygen. Sub-
strates with bound PAHs were immersed in a solution of
methylene blue in ethanol (ca. 1 mM) for several seconds, then
removed from solution, allowed to dry, and exposed to 650 nm
light (20 h, ∼7 mW/cm2). The red-irradiated samples were
immersed in ethanol to remove physisorbed methylene blue,
then rinsed with ethanol, dried, and examined for the presence
of PAHs degradation products using emission spectroscopy.
Results and Discussion
The goal of this work is to understand the basis for the
observed degradation of PAHs bound to surfaces. Either UV
light or electrochemical oxidation in the presence of ambient
O2 can play a role in the degradation, with the characteristic
formation of mono- and dihydroxy and dione derivatives of the
PAHs resulting. Because the structures, and spectroscopic and
electrochemical responses, of the two PAHs we report on are
significantly different, we consider our results for surface-bound
pyrene and anthracene species separately.
Preparation of Thin Films of Bipyrene, 1-Hydroxypyrene,
1,6-Pyrenedione, 1,4-Anthracenedione, and 9,10-Anthracene-
dione on Gold. Thin films of bipyrene on gold were prepared
electrochemically according to the procedure developed by Zotti
Pyrene. We have investigated pyrene bound covalently to
quartz by emission spectroscopy. We show in Figure 1a the
excitation and emission spectra of aminopyrene after covalent
bonding to a quartz substrate to form an amidopyrene. The

1040 J. Phys. Chem. B, Vol. 108, No. 3, 2004
Mazur and Blanchard
Figure 3. Dependence of charges under voltammetric peaks for species
II, III, and IV as a function of number of CV scans.
Figure 2. (a) Cyclic voltammograms of 11-(1-amidopyrenyl)-1-mer-
captoundecanoate on gold in aqueous 0.5 M H2SO4. Sweep rate: 0.1
V/s. (b) Dependence of the separation of anodic and cathodic peak
potentials for redox pair III as a function of the logarithm of sweep
rate.
varying sweep rate and plotted the separation of the anodic and
cathodic peak potentials as a function of the logarithm of the
sweep rate (Figure 2b). On the basis of the theory proposed by
Laviron,41 from the slope of the linear portion of the curve we
were able to determine the value of the transfer coefficient R
and the number of electrons involved in the redox reaction. We
obtain reasonable values of R (0 < R < 1) only for a two-
electron reaction. Fitting these data to the theory with n ) 2
yields a value of R ) 0.2.
On the basis of the voltammetric curves recorded on gold in
aqueous solution, we can plot the values of charge exchanged
in redox reactions for the species present on the surface during
subsequent scans. The results are presented in Figure 3. The
concentration (charge) of species II reaches its maximum value
in the second scan, just after formation of radical cations in the
first scan; then it gradually decreases, approaching zero after
multiple cycles. At the same time, the charge associated with
species III reaches the value of 6.2 mC/cm2 in the second scan,
and then increases monotonically to ∼12 µC/cm2. The concen-
tration of species IV reaches 2.8 µC/cm2 and remains ap-
proximately constant beyond five cycles.
We consider next the assignment of the species associated
with these peaks. The degradation of physisorbed pyrene on
various solid substrates has been established32-35 and is usually
explained in terms of the formation of reactive radical cations,
which follow a variety of reaction pathways, including introduc-
tion of hydroxyl and carbonyl groups to the ring system, and
dimerization. These findings provide a frame of reference for
the assignment of the electrochemical features we see for
surface-bound pyrene.
To identify the surface species, we have recorded cyclic
voltammograms of a number of candidate compounds. Although
we study physisorbed and not surface-bound molecules we
believe the electrochemical behavior (formal potentials) in both
cases is similar. We do not report on solution phase electro-
chemical behavior due to very low solubility of the studied
compounds in water. We show in Figure 4a the voltammetric
behavior of 1-hydroxypyrene adsorbed to a gold electrode. In
the first scan we observe an irreversible peak at 0.71 V, which
is followed by the formation of a new product with peaks at
∼0.32 V. In subsequent scans, the peak at 0.71 V diminishes
rapidly, indicating hydroxypyrene is transformed into a new
product. We note that the oxidation peak potential of 1-hy-
excitation spectrum reveals two bands with maxima at 283 and
347 nm, both associated with the surface-bound pyrene.
The emission spectrum shows two bands, with maxima at
385 and 405 nm. The spectra are very similar to those of
1-acetamidopyrene in pentanol39 and are significantly different
from the solution phase spectra of 1-aminopyrene (Figure 1b).
These data indicate that the immobilized pyrene is bound
covalently to the surface through an amide bond. It should be
noted that we do not observe any significant band in the 450-
500 nm range, which would indicate excimer emission. This
finding may be the result of a relatively low surface concentra-
tion of pyrene molecules or their hindered mobility due to
covalent binding to the substrate. The latter case is more likely
because the formation of excimers was observed even at low
surface concentrations of pyrene physisorbed on silica.40
We have also examined the susceptibility of bound pyrene
to electrochemical oxidation using cyclic voltammetry. We show
in Figure 2a a series of voltammetric scans for amidopyrene
bound covalently to gold. In the first scan, when the anodic
potential above ca. 0.8 V is applied to the electrode, we observe
an irreversible peak with a maximum at 0.96 V. For the reverse
scan and in subsequent scans, a number of new redox peaks
appear in voltammograms. These are a pair of peaks at 0.62 V
(designated II in Figure 2a) and two pairs at 0.21 V (IV) and
0.32 V (III). We assign the irreversible peak at 0.96 V to
oxidation of pyrene and the consequent formation of radical
cations. Because radical cations are highly reactive, we do not
observe a reduction peak; the radical cations undergo rapid
reaction to more stable species. We see that the peak pair II is
associated with an intermediate species on the basis of the
observation that the magnitude of these peaks decreases
monotonically in consecutive scans. In contrast, peaks III and
IV increase in magnitude with the number of scans, appearing
to be the final products of pyrene oxidation.
Due to the complexity of this system, a detailed analysis of
the electrochemical data is difficult. Accordingly, we have
analyzed in detail only the redox reaction of the stable species
III in Figure 2a. We recorded a set of voltammetric curves

Oxidation Reactions of Surface-Bound PAHs
J. Phys. Chem. B, Vol. 108, No. 3, 2004 1041
SCHEME 1 : Oxidation of Surface-Bound Pyrene
Figure 4. Cyclic voltammograms of thin layers of (a) 1-hydroxypyrene,
(b) 1,6-pyrenedione, and (c) bipyrene deposited on gold in aqueous
0.5 M H2SO4. The sweep rate was 0.1 V/s for all measurements.
droxypyrene (0.71 V, Figure 4a) is relatively close to the formal
potential of the intermediate species II we see for covalently
bound amidopyrene (0.62 V, Figure 2a). Both species II (Figure
2a) and hydroxypyrene are unstable and are transformed into a
new product, producing peaks at ∼0.32 V. We thus assign
species II (Figure 2a) as a hydroxy derivative of surface-bound
pyrene.
Figure 4b shows a cyclic voltammogram of the 1,6-pyrene-
dione/1,6-dihydroxypyrene redox couple. We note a pair of
voltammetric peaks centered at 0.32 V, with the similarity of
these signals to peaks III (Figure 2a) and products of hydroxy-
pyrene oxidation (Figure 4a) being clear, and we conclude that
the signals at 0.32 V in both Figures 2 and 4a are due to the
1,6-dihydroxy/dione derivative of pyrene.
react with water to form a monohydroxy derivative. The
monohydroxy derivative is further transformed by reaction with
another water molecule to produce more than one isomeric form
of dihydroxypyrene/pyrenedione. We cannot explicitly exclude
the formation of dimeric (bipyrene) or oligomeric species that
are subsequently transformed into hydroxyl/quinone derivatves,
but this possibility seems less probable due to the expected
changes in redox potential for the oligomers and the absence
of experimental data pointing to additional species.
We consider next the origin of peaks IV (Figure 2). It is
possible that these peaks are associated with a dimeric species
or an isomeric pyrenedione/dihydroxypyrene redox couple.
Figure 4c shows voltammetric scans of a physisorbed bipyrene
thin film on gold in aqueous solution. The increase in current
above ca. 0.9 V can be assigned to oxidation of bipyrene. In
subsequent scans the new products produce signals at 0.60, 0.34,
and 0.25 V. On the basis of these data, none of signals II-IV
(Figure 2) can be assigned to bipyrene species. However, the
signals associated with the oxidation products of bipyrene are
qualitatively similar to those for the oxidation products of
surface-bound pyrene, suggesting that the oxidation of bipyrene
in aqueous media produces a monohydroxy derivative (peak
pair at 0.65 V) and a dihydroxy derivative (0.4 V). We also
observe the third pair of peaks at 0.25 V and believe it to be a
counterpart to species IV. Our data indicate that the electro-
chemical behavior of pyrene and bipyrene in aqueous medium
is substantially the same.
The origin of peaks IV in Figure 2a remains to be established.
Moriconi et al.42 have reported that the standard potential of
1,8-pyrenedione/1,8-dihydroxypyrene is cathodically shifted ca.
100 mV with respect to the 1,6-pyrenedione/1,6-dihydroxy-
pyrene couple. On the basis of this information, species IV is
likely the 1,8-dihydroxy/1,8-dione derivative of pyrene.
The electrochemical data suggest the following scenario for
pyrene reactions occurring at the electrode surface (Scheme 1).
The anodic polarization of the electrode above 0.8 V results in
the formation of pyrene radical cations. These radical cations
We can estimate the surface concentration of pyrene mol-
ecules on the basis of its electrochemical reaction, but due to
complexity of the system we need to make some assumptions.
We calculated the values of charge under voltmmetric peaks
for the curves shown in Figure 2a. The QI value represents the
total charge consumed by the oxidation of pyrene, oxidation of
the formed monohydroxy derivative, and oxidation of dihy-
droxypyrenes. Assuming that all the pyrene molecules are
transformed into mono- or dihydroxy derivatives by the second
voltammetric scan, we get the charge associated (formally) with
one- or two-electron oxidation of pyrene by calculating the value
QI - QII - QIII - QIV, where QII, QIII, and QIV are the values
of charge under voltammetric peaks taken from the second scan
and QI is the value of charge under the voltammetric peak taken
from the first scan. Thus, some fraction of the molecules after
one-electron oxidation is transformed into the monohydroxy
derivative and the rest (after two-electron oxidation) is trans-
formed into dihydroxy derivative(s). We need therefore to
estimate the fraction of molecules following the one- and two-
electron processes. We can do this by calculating the values
2QII/(2QII + QIII + QIV) for the molecules experiencing one-
electron oxidation and (QIII + QIV)/(2QII + QIII + QIV) for those
experiencing two-electron oxidation. The quantity [QI - (QII
+ QIII + QIV)][2QII/(2QII + QIII + QIV)]/F gives the surface
concentration of pyrene transformed into the monohydroxy
derivative (in the second voltammetric scan) and [QI - (QII +
QIII + QIV)][(QIII + QIV)/(2QII + QIII + QIV)]/(2F) is the
concentration of the molecules transformed into dihydroxy-

1042 J. Phys. Chem. B, Vol. 108, No. 3, 2004
Mazur and Blanchard
Figure 5. Emission spectra of (a) amidopyrenehexanoate bound to
ITO, (b) products of the electrochemical oxidation of amidopyrene-
hexanoate bound to ITO, (c) products of the photochemical degradation
of amidopyrenehexanoate bound to ITO, (d) 1,6-pyrenedione in
1-pentanol, and (e) 1,6-dihydroxypyrene in 1-pentanol.
Figure 6. (a) Excitation and emission spectra of aminoanthracene
attached to an adipoyl chloride adlayer on quartz. (b) Excitation and
emission spectra of 1-amidoanthracene acetate (solid line) and 1-ami-
noanthracene (dotted line) dissolved in pentanol.
pyrene. The sum of these two terms gives us the total
concentration of pyrene molecules according to the formula:
Γ ) [QI - (QII + QIII + QIV)][2QII + (QIII + QIV)/2]/(2QII +
QIII + QIV)/F ) 2.1 × 10-10 mol/cm2. If we assume that the
full monolayer corresponds to the concentration of alkanethiol
molecules on gold, 7.7 × 10-10 mol/cm2,43 we achieve a surface
coverage of pyrene moieties of ∼27% of a monolayer. Because
the pyrene attachment chemistry involves the use of two MUA
surface-bound molecules, we achieve ∼50% reaction efficiency
for pyrene layer formation.
To study the emission spectra of covalently bound pyrene,
we used ITO-coated silica substrates. The electrochemical
behavior of pyrene on ITO is qualitatively similar to that on
gold. We show in Figure 5 the emission spectra of pyrene bound
covalently to ITO, before (a) and after (b) electrochemical
oxidation in aqueous H2SO4. For non-oxidized pyrene we see
two emission bands at 385 and 405 nm (similar to those seen
on SiOx, Figure 1). Following electrochemical oxidation, these
bands shift slightly to 382 and 404 nm, and two new bands
appear at 372 and 393 nm. We compare these spectra to that of
1,6-dihydroxypyrene (d) and 1,6-pyrenedione (e). The 1,6-
dihydroxypyrene spectrum is characterized by three bands at
372, 382, and 391 nm, and we can find the same bands in the
spectrum of the product of pyrene oxidation (b). The spectrum
of 1,6-pyrenedione has a relatively intense band at 404 nm,
coincident with the band seen in the spectrum of the electro-
chemically oxidized form.
Upon irradiation with UV light (340 nm, 20 h, ∼7 mW/cm2)
and in the presence of atmospheric oxygen, the covalently bound
pyrene is transformed into a species with emission bands at
373, 381, 394, and 405 nm (Figure 5c), bands characteristic of
dihydroxypyrene and pyrenedione. This finding shows that
both electrochemical oxidation and UV photolysis in the
presence of oxygen result in the same products being formed.
The difference in band ratios for electrochemical and photo-
chemical oxidation suggest the presence of additional species
resulting from exposure to UV light (Figure 5b,c). We observed
that the photodegradation of surface-bound pyrene is much
slower under nitrogen than in the presence of oxygen, under-
scoring the central role of oxygen and/or water in the degrada-
tion reactions.
Although the products of electrochemical or photochemical
transformations are the same, the mechanism of each reaction
may be different. In the case of electrochemical oxidation we
did not observe any effect of oxygen dissolved in the solution
on the voltammetric behavior. This finding means that oxygen
does not play a role in surface reactions and the pyrene is simply
oxidized through electron transfer from the molecule to the metal
(or semiconductor). Then the oxidized molecule reacts with
water to produce a monohydroxypyrene.
The photodegradation of pyrene can occur through two paths.
The first is electron transfer between excited pyrene and O2,
resulting in formation of a pyrene radical cation and a superoxide
(O2-). The pyrene radical can then react with water (physisorbed
on the surface) to give the hydroxyl derivative.
The second mechanism by which photodegradation can occur
is the reaction of pyrene with generated singlet oxygen. It is
known that triplet oxygen may quench the excited states of
pyrene, and this quenching process yields singlet oxygen. The
singlet oxygen can react subsequently with pyrene by addition
to the ring system.
To determine which of the mechanisms occur for our systems,
we performed an experiment where we physisorbed methylene
blue (a well-known singlet oxygen sensitizer) onto the substrate
modified by covalent attachment of pyrene and excited the
methylene blue at 650 nm in the presence of oxygen. The singlet
oxygen formed by irradiation of methylene blue could react with
surface-bound pyrene to yield hydroxy and/or carbonyl deriva-
tives. From these experiments, we did not observe (using
emission spectroscopy) any significant loss of pyrene or
presence of its degradation products. We conclude therefore that
singlet oxygen is not involved in the photochemical transforma-
tion of surface bound pyrene and the electron-transfer mecha-
nism is dominant.
Anthracene. We show in Figure 6a the excitation and emis-
sion spectra of anthracene bound covalently to quartz. The
excitation spectrum reveals one relatively broad band centered

Oxidation Reactions of Surface-Bound PAHs
J. Phys. Chem. B, Vol. 108, No. 3, 2004 1043
SCHEME 2: Oxidation of Surface-Bound Anthracene
to the product of the reaction between the anthracene radical
and water. The signal at 0.47 V also decreases in successive
scans, indicating that the oxidation of this product leads to
further reaction. Simultaneously, a new pair of peaks around
0.17 V grow in, as shown in Figure 7a. The formal potential of
this redox couple is within the range of electroactivity of
quinone/hydroquinone systems.
Thus, the reaction of anthracene with water likely leads to
the formation of hydroxyanthracenes and anthraquinones. It is
well-known that the most stable anthraquinones are those having
two carbonyl groups on the same (aromatic) ring.45 The most
likely oxidation products are the (9,10), (1,4), and (1,2) isomers.
We have recorded cyclic voltammograms of two anthraquino-
nes: 1,4-anthracenedione and 9,10-anthracenedione (Figure 7b,
scans ii and iii, respectively). The formal potential of the 1,4-
anthracenedione/1,4-dihydroxyanthracene redox couple (0.14 V)
is very close to that of the oxidation product of surface-bound
anthracene, whereas the potential of the 9,10 isomers is shifted
ca. 300 mV negative. The formal potential of the 1,2 isomer is
known to be 90 mV more positive than that of the 1,4 isomer.42
On the basis of these data we assign the peak at ∼0.17 V to the
1,4 anthracenedione/1,4-dihydroxyanthracene couple.
The chemical nature of the species responsible for the
oxidation peak at 0.47 V remains unclear. One possible
explanation is (similar to that for pyrene) that the substitution
of anthracene with two carbonyl (hydroxyl) groups occurs
sequentially rather than simultaneously; hydroxyanthracene is
formed first and reacts subsequently to form the dihydroxy/
dione species. In this picture, hydroxyanthracene is responsible
for the peak at 0.47 V. The putative reactions of covalently
bound anthracene are summarized in Scheme 2.
Figure 7. (a) Cyclic voltammograms in aqueous 0.5 M H2SO4 of 11-
(1-amidoanthracenyl)-1-mercaptoundecanoate on gold. (b) (i) Steady-
state CV of 11-(1-amidoanthracenyl)-1-mercaptoundecanoate on gold,
(ii) a thin layer of 1,4-anthracenedione deposited on gold, and (iii) a
thin layer of 9,10-anthracenedione deposited on gold.
at 380 nm, with the approximately mirror-image emission spec-
trum centered near 420 nm. These spectra are similar to those
of amidoanthracene acetate recorded in solution (Figure 6b solid
line) and significantly different form those of solution phase
aminoanthracene (Figure 6b dotted line). These data confirm
that the anthracene molecules are attached to the surface through
an amide bond and are not simply physisorbed. The comparison
of the spectrum of surface-bound anthracene (Figure 6a) and
nonsubstituted anthracene44 reveals the lack of vibronic structure
on both the 380 nm absorption band and its corresponding
emission band, centered near 430 nm. We obtain the same result
for amidoanthracene acetate in the solution (Figure 6b) and
conclude that these spectral changes are not the result of surface
binding but are caused by the presence of the amido substituent
on the anthracene ring system. We also do not observe any
emission signals attributable to excimer formation, which is
probably the result of covalent bonding of the chromophore to
the substrate.
As with pyrene, anthracene bound covalently to SiOx or ITO
is not stable when exposed to UV light in the presence of
atmospheric oxygen and water. Sample degradation is not
observed when the sample is placed in the dark and is strongly
attenuated when oxygen and water vapor are purged with
nitrogen. To determine if the surface-bound anthracene moieties
can be oxidized in a controlled manner, we bound aminoan-
thracene to a modified gold surface and examined the electro-
chemical behavior of the resulting interface. We show in Figure
7 a series of voltammetric curves for anthracene attached
covalently to gold. In the first scan we observe an anodic peak
at 0.95 V that is not associated with a cathodic peak when the
polarization is reversed. This irreversible behavior suggests that
anthracene is oxidized and then reacts rapidly to form a new
product. The anthracene radicals generated by electrochemical
oxidation most likely react with water molecules in a manner
analogous to that seen for pyrene. This possibility is consistent
with the electrochemical data, as seen in the second voltam-
metric scan, (Figure 7) where the peak at 0.95 V decreases and
a new oxidation peak appears at 0.47 V. We assign this peak
As with pyrene, we can estimate the surface concentration
of anthracene on the basis of the amount of charge associated
with the oxidation of surface bound species. The charge under
peak I (Figure 7) is associated with oxidation of anthracene and
with subsequent oxidation of newly formed products (peaks II
and III). We calculate QI - QII - QIII ) 12.3 µC/cm2, and
dividing it by nF, we obtain the surface concentration of the
electroactive species. As with pyrene a significant issue is
determining the number of electrons, n, involved in product
formation. When only monohydroxyanthracene is created on
the surface, n ) 1, and when dihydroxyanthracene is obtained,
the n ) 2. We could estimate the fraction of anthracene
molecules transformed into the mono- or dihydroxy derivative
on the basis of QII and QIII values, but we remain unsure if the
hydroxy/quinone derivatives are the only products of anthracene
oxidation. The magnitude of the III peak pair is relatively low,
which may be the result of some other nonelectroactive products
being created that we are not able to detect. We are thus left to
place bounds on the surface concentration of anthracene. For n

1044 J. Phys. Chem. B, Vol. 108, No. 3, 2004
Mazur and Blanchard
oxidation, either by emission spectroscopy (no product signal)
or by electrochemistry (the signals are probably too small to
be detected).
Conclusions
The data we present here for covalently bound pyrene and
anthracene reveal a limitation of PAHs as spectroscopic probes
of surface adlayers. Such measurements must be performed in
oxygen-free and dry environments to prevent the chromophore
from manifesting oxidative degradation. The fact that we can
achieve such consistent results for two chromophores and three
different types of interfaces underscores the generality of the
effects we report and rules out the possibility of any anomalous
surface enhancement phenomena. It is clear that care must be
taken to avoid oxidative degradation in environments character-
ized by large surface-to-volume ratios.
Acknowledgment. We are grateful to the National Science
Foundation for their support of this work through Grant
0090864. M.M. is grateful to the National Science Foundation
and NATO for their support of a postdoctoral fellowship (DGE-
0209459).
Figure 8. Emission spectra of (a) amidoanthracenehexanoate bound
to ITO and (b) products of electrochemical oxidation of amidoan-
thracenehexanoate bound to ITO.
) 1 we obtain Γ1 ) 1.27 × 10-10 mol/cm2 and for n ) 2, Γ2
) 0.64 × 10-10 mol/cm2; thus the surface coverage of
anthracene on the surface (assuming the concentration of thiol
on gold to be 7.7 × 10-10 mol/cm2 43) is not lower than 8.7%
and not higher than 16.5%, slightly less than that seen for pyrene
covalent deposition.
References and Notes
(1) Ulman, A. An Introduction to Ultrathin Organic Films; Academic
Press: New York, 1991.
(2) Dubois, L. H.; Nuzzo, R. G. Annu. ReV. Phys. Chem. 1992, 43,
437.
(3) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides,
G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321.
(4) Bain, C. D.; Whitesides, G. M. Angew. Chem., Int. Ed. Engl. 1989,
28, 506.
We now consider the spectral characteristics of the im-
mobilized chromophore as a function of oxidation. Because the
emission intensity of any chromophore is quenched efficiently
by the presence of a metallic substrate, we attach the molecules
to the surface of ITO. We show in Figure 8 the emission
spectrum of anthracene bound covalently to ITO through an
amide bond. The spectrum (curve a) exhibits a band with a
maximum at 410 nm, similar to the data for anthracene on quartz
(Figure 6a). The electrochemical oxidation of anthracene on ITO
results in disappearance of the emission band (Figure 8b), but
no new bands appear in this spectral window. We obtain the
same result for irradiation of the sample with UV light in the
presence of atmospheric oxygen and water. We have verified
that the fluorescence quantum yield of amidoanthracene acetate
is much higher (ca. 60×) than that for 1,4-anthracenedione. As
mentioned above, it is possible that hydroxyl and quinone
derivatives are not the only products of anthracene oxidation,
but if such products, exist we are not able to detect them either
electrochemically or spectroscopically.
As with pyrene we studied the effect of singlet oxygen on
the chromophore molecules. The singlet oxygen was gen-
erated by excitation (650 nm) of methylene blue physisorbed
on the sample surface. We do not observe any significant
changes in the emission spectrum of the surface-bound an-
thracene as a result of methylene blue irradiation and con-
clude that singlet oxygen is not involved in photochemical
degradation process. Therefore it seems that the electron-transfer
mechanism is dominant as the initial step of anthracene oxidation
as well.
(5) Ulman, A. Chem. ReV. 1996, 96, 1533.
(6) Langmuir, I. J. Am. Chem. Soc. 1917, 39, 1848.
(7) Blodgett, K. B. J. Am. Chem. Soc. 1935, 57, 1007.
(8) Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 45.
(9) Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 52.
(10) Lee, H.; Kepley, L. J.; Hong, H. G.; Akhter, S.; Mallouk, T. E. J.
Phys. Chem. 1988, 92, 2597.
(11) Hong, H. G.; Sackett, D. D.; Mallouk, T. E. Chem. Mater. 1991,
3, 521.
(12) Katz, H. E.; Scheller, G.; Putvinski, T. M.; Schilling, M. L.; Wilson,
W. L.; Chidsey, C. E. D. Science 1991, 254, 1485.
(13) Vermeulen, L. A.; Thompson, M. E. Nature 1992, 358, 656.
(14) Kohli, P.; Blanchard, G. J. Langmuir 2000, 16, 4655.
(15) Major, J. S.; Blanchard, G. J. Chem. Mater. 2002, 14, 2574.
(16) Schlenoff, J. B.; Laurent, D.; Ly, H.; Stepp, J. AdV. Mater. 1998,
10, 347.
(17) Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153.
(18) Schlenoff, J. B.; Laurent, D.; Ly, H.; Stepp, J. Chem. Eng. Technol.
1998, 21, 757.
(19) Schlenoff, J. B.; Ly, H.; Li, M. J. Am. Chem. Soc. 1998, 120,
7626.
(20) Schlenoff, J. B.; Dubas, S. T. Macromolecules 2001, 34, 592.
(21) Caruso, F.; Lichtenfeld, H.; Donath, E.; Mohwald, H. Macromol-
ecules 1999, 32, 2317.
(22) Feiler, A.; Plunkett, M. A.; Rutland, M. W. Langmuir 2003, 19,
4173.
(23) Wang, L. Y.; Schonhoff, M.; Mohwald, H. J. Phys. Chem. B 2002,
106, 9135.
(24) Finklea, H. O. In Electroanalytical Chemistry; Bard, A. J.,
Rubenstein, J., Eds.; Marcel Dekker: New York, 1996; Vol. 19, p 109.
(25) Xia, Y. N.; Whitesides, G. M. Annu. ReV. Mater. Sci. 1998, 28,
153.
(26) Schvartzman, M.; Sidorov, V.; Ritter, D.; Paz, Y. J. Vac. Sci.
Technol. B 2003, 21, 148.
(27) Sheen, C. W.; Shi, J.-X.; Martensson, J.; Parikh, A. N.; Allara, D.
L. J. Am. Chem. Soc. 1992, 114, 1514.
(28) Sagiv, J. J. Am. Chem. Soc. 1980, 102, 92.
(29) Major, J. S.; Blanchard, G. J. Chem. Mater. 2002, 14, 4320.
(30) Karpovich, D. S.; Blanchard, G. J. J. Phys. Chem. 1995, 99, 3951.
(31) Karpovich, D. S.; Blanchard, G. J. Langmuir 1996, 12, 5522.
(32) Fatiadi, A. J. EnViron. Sci. Technol. 1967, 1, 570.
(33) Mao, Y.; Thomas, J. K. Langmuir 1992, 8, 2501.
One of the unusual results obtained from electrochemical data
is formation of 1,4- and not 9,10-dione/hydroxyl derivatives.
We believe that this result is due to substitution of anthracene
with the NHCOOR group, which influences the reactivity of
the ring system. Unfortunately, we are not able to determine
directly which isomers are generated during the photochemical

Oxidation Reactions of Surface-Bound PAHs
J. Phys. Chem. B, Vol. 108, No. 3, 2004 1045
(34) Reyes, C. A.; Medina, M.; Crespo-Hernandez, C.; Cedeno, M. Z.;
Arce, R.; Rosario, O.; Steffenson, D. M.; Ivanov, I. N.; Sigman, M. E.;
Dabestani, R. EnViron. Sci. Technol. 2000, 34, 415.
(35) Sigman, M. E.; Schuler, P. F.; Ghosh, M. M.; Dabestani, R. T.
EnViron. Sci. Technol. 1998, 32, 3980.
(40) Lochmuller, C. H.; Colborn, A. S.; Hunnicutt, M. L.; Harris, J. M.
Anal. Chem. 1983, 55, 1344.
(41) Laviron, E. J. Electroanal. Chem. 1979, 101, 19.
(42) Moriconi, E. J.; Rakoczy, B.; O’Connor, W. F. J. Org. Chem. 1962,
27, 2772.
(43) Kuwabata, S.; Fukuzaki, R.; Nishizawa, M.; Martin, C. R.;
Yoneyama, H. Langmuir 1999, 15, 6807.
(44) Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic
Molecules; Academic Press: New York, 1971.
(45) Boldt, P. Extended quinones. In Chemistry of the Quinoid Com-
pounds; Vol. 2, Part 2, p 1431.
(36) Dabestani, R.; Ellis, K. J.; Sigman, M. E. J. Photochem. Photobiol.
A 1995, 86, 231.
(37) Ostuni, E.; Chapman, R. G.; Holmlin, R. E.; Takayama, S.;
Whitesides, G. M. Langmuir 2001, 17, 5605.
(38) Zotti, G.; Schiavon, G. Synth. Met. 1992, 47, 193.
(39) Kelepouris, L.; Krysinski, P.; Blanchard, G. J. J. Phys. Chem. B
2003, 107, 4100.