Strong photon energy dependence of the photocatalytic dissociation rate of methanol on TiO2(110)

Article abstract of DOI:10.1021/ja4114598

Photocatalytic dissociation of methanol (CH₃OH) on a TiO ₂(110) surface has been studied by temperature programmed desorption (TPD) at 355 and 266 nm. Primary dissociation products, CH₂O and H atoms, have been detected. Th

Full text of DOI:10.1021/ja4114598

Article
pubs.acs.org/JACS
Strong Photon Energy Dependence of the Photocatalytic
Dissociation Rate of Methanol on TiO2(110)
Chengbiao Xu,†,⊥,∥ Wenshao Yang,†,∥ Zefeng Ren,§ Dongxu Dai,† Qing Guo,*,† Timothy K. Minton,*,†,‡
and Xueming Yang*,†
†State Key Laboratory of Molecular Reaction Dynamics, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457
Zhongshan Road, Dalian 116023, P. R. China
‡Department of Chemistry and Biochemistry, Montana State University, PO Box 173400, Bozeman, Montana 59717, United States
§International Center for Quantum Materials, Peking University, 5 Yiheyuan Road, Beijing 100871, P. R. China
S
* Supporting Information
ABSTRACT: Photocatalytic dissociation of methanol
(CH3OH) on a TiO2(110) surface has been studied by
temperature programmed desorption (TPD) at 355 and 266
nm. Primary dissociation products, CH2O and H atoms, have
been detected. The dependence of the reactant and product
TPD signals on irradiation time has been measured, allowing
the photocatalytic reaction rate of CH3OH at both wave-
lengths to be directly determined. The initial dissociation rate
of CH3OH at 266 nm is nearly 2 orders of magnitude faster
than that at 355 nm, suggesting that CH3OH photocatalysis is
strongly dependent on photon energy. This experimental
result raises doubt about the widely accepted photocatalysis model on TiO2, which assumes that the excess potential energy of
charge carriers is lost to the lattice via strong coupling with phonon modes by very fast thermalization and the reaction of the
adsorbate is thus only dependent on the number of electron−hole pairs created by photoexcitation.
from photoluminescence spectra of rutile TiO2(110)19,20 with
3.35 eV photon irradiation. In both these studies, a constant
INTRODUCTION
■
Since the discovery of the photocatalytic splitting of water
(H2O) on TiO2 electrodes by Fujishima and Honda in 1972,1
energy of the emitted photons was observed (at and below the
band gap energy3.05 eV), independent of the energy of the
tremendous research efforts have gone into understanding the
exciting electron or photon. Furthermore, experimental
investigations indicate that the photodesorption yield and
translational energy of O2 from an O2-adsorbed TiO2(110)
surface are independent of the excitation photon energy above
3.4 eV, but instead depend on the photon flux.21,22 It has also
been reported that the photooxidation yield of hexane on a
TiO2 powder is dependent on photon flux rather than photon
energy.23 On the other hand, it has been reported that the
photodesorption and photooxidation yields of other organics
on TiO2 are strongly dependent on photon energy.24−26
Actually, the measurement of wavelength dependent yields of
reaction products is not the best approach to understand the
fundamental processes of photocatalysis. Instead of product
yields, a wavelength-dependent rate measurement for a
photocatalytic reaction system would be a more direct probe
of the underlying mechanisms. We have thus investigated the
wavelength dependence of the initial rate of the photocatalytic
dissociation of methanol (CH3OH) on TiO2(110).
fundamental processes and in enhancing the photocatalytic
efficiency of TiO2.2 These efforts have been driven by the
potential benefits to renewable energy, energy storage, and
environmental cleanup.3,4 In particular, TiO2 has attracted
much attention because of its applications in heterogeneous
catalysis, photocatalysis, solar energy devices, etc.5−14
In an ideal photocatalyst, all photon energy invested in the
generation of charge carriers would be available for redox
chemistry. The higher the potential energy of the electron (or
hole) the more reductive (or oxidative) capacity there is.15 But
charge carrier thermalization is rapid. For example, Gundlach
and co-workers used two photon photoemission (2PPE) to
track thermalization following electron injection from two dyes
adsorbed on rutile TiO2(110).16,17 Fast initial decay of the
2PPE signal resulting from thermalization of the injected
electron occurred on the 10 fs time scale. This result suggests
that excess potential energy is lost to the lattice via strong
coupling with phonon modes, thus reducing the potential
advantage gained by the specificity in the absorption event.
Additional evidence for the rapid thermalization of electrons
comes from photoemission spectra of Ag clusters on rutile
TiO2(110)18 as a function of injection electron energy and
Received: November 10, 2013
Published: December 3, 2013
© 2013 American Chemical Society
19039
dx.doi.org/10.1021/ja4114598 | J. Am. Chem. Soc. 2013, 135, 19039−19045

Journal of the American Chemical Society
Article
irradiated surface area was 56% of the total surface area. The laser flux
was thus 9.5 × 1016 photons cm−2 s−1 for 266 nm and 1.3 × 1017
photons cm−2 s−1 for 355 nm. The TiO2 surface temperature rose by
about 5 K with the surface temperature at 110 K, and we found that
the surface damage at both 355 and 266 nm was negligible during the
exposure times used in our experiments (0−5 s at 266 nm and 0−600
s at 355 nm). These times were chosen such that CH3OH dissociation
to CH3O and CH2O were the dominant processes. Secondary
processes, such as methyl formate (HCOOCH3) formation from cross
coupling reactions,49 were observable but not significant. TPD
measurements were carried out with a heating rate of 2 K/s. The
TiO2(110) surface was annealed in vacuum at 850 K for 20 min
between different TPD runs to maintain a flat and clean surface.
CH3OH photocatalysis on TiO2(110) has become a well-
studied model system and is thus an ideal testing ground for
important photocatalysis concepts. Previous studies showed
that molecular hydrogen production from H2O on TiO2 could
be significantly enhanced by adding CH3OH,27 suggesting that
CH3OH plays an important role in photocatalytic H2O
splitting. Because of the importance of this issue, many
experimental and theoretical studies have been carried out on
the CH3OH-TiO2(110) system.28−39 In recent years, CH3OH
photochemistry on the rutile TiO2(110) surface has been
investigated using various techniques, such as 2PPE, scanning
tunneling microscopy (STM), and temperature programmed
desorption (TPD), in order to understand CH3OH photo-
chemistry on TiO2(110) at the molecular level.40−46 Using
2PPE and STM, Zhou et al.40 concluded that photocatalytic
dissociation of CH3OH on TiO2(110) occurs at the Ti5c
adsorption sites. In a recent study,44 we have shown that
photocatalytic dissociation of CH3OH on TiO2(110) occurs in
a stepwise manner in which the O−H dissociation proceeds
first and is then followed by C−H dissociation to form
formaldehyde (CH2O). Henderson and co-workers also
observed the second step, photocatalytic C−H dissociation
from CH3O to form CH2O.45,46 Theoretical calculations for the
ground electronic state show that the first step is nearly
thermoneutral with a small barrier (0.25−0.3 eV),31,36,43,44
suggesting that O−H bond dissociation is facile, while the
second step has a much high barrier. By extending the detailed
studies of CH3OH photocatalysis on TiO2(110) to include
wavelength-dependent rate measurements, we are able to go
beyond the relatively simple description of the CH3OH
decomposition steps and their energetics and actually probe
the fundamental process by which the photon energy initiates
this decomposition.
RESULTS
■
Figure 1A shows typical TPD spectra collected at a mass-to-
charge ratio (m/z) of 31 for 0.5 ML of CH3OH adsorbed on a
EXPERIMENTAL DETAILS
■
Photocatalysis of CH3OH was studied on a vacuum annealed
TiO2(110) surface with a TPD apparatus44,47 in combination with
laser surface irradiation at 266 and 355 nm. The base pressure of the
main chamber was 6 × 10−11 Torr. TPD products were detected with a
quadrupole mass spectrometer (Extrel), which uses a specially
designed electron-impact ionization region that can be maintained at
an extremely high vacuum of 1.5 × 10−12 Torr. The rutile TiO2(110)
crystal (Princeton Scientific Corp.) was cleaned through multiple
cycles of Ar+ sputtering and UHV annealing at 850 K until all surface
contaminations were below the detection limit of Auger electron
spectroscopy and a sharp (1 × 1) LEED pattern was observed. After
this surface cleaning procedure, the crystal color became blue, and an
oxygen vacancy population of about 4%, determined by H2O TPD
measurement,48 was created on surface. Daily cleaning was
accomplished by annealing the crystal at 850 K for 30 min in UHV.
CH3OH with a purity of >99.9% was purchased from Sigma-Aldrich.
Before using, it was purified further by several freeze−pump−thaw
cycles. In this experiment, we dosed the surface at 110 K with a 0.5 ML
(1 ML = 5.2 × 1014 molecules cm−2) coverage of CH3OH using a
home-built, calibrated, molecular beam doser.
The third harmonic (355 nm) or fourth harmonic (266 nm) output
of a diode-pumped, solid state (DPSS), Q-switched 1064 nm laser
(Spectra-Physics) was used to induce photochemistry on the CH3OH
adsorbate. The laser pulse duration was about 12 ns, and the laser
repetition rate was 50 kHz. In order to minimize the temperature
increase of the surface resulting from laser irradiation and to reduce
surface damage by laser irradiation, only 40 mW of laser light at both
wavelengths was used to illuminate the TiO2(110) surface. The laser
beam diameter was 6 mm, and its incident angle was ∼30° with
respect to the normal of the 10 × 10 mm2 TiO2(110) surface, which
was mounted with an angle point at the top (see Figure S1). The
Figure 1. (A) Typical TPD spectra collected at m/z = 31 (CH2OH+)
following different laser irradiation times at 355 nm. CH2OH+ is
formed by dissociative ionization of the desorbed parent CH3OH
molecule in the electron-bombardment ionizer. (B) Typical TPD
spectra collected at m/z = 31 following different laser irradiation times
at 266 nm. Note “a” indicates the appearance of the methyl formate
product at the longest irradiation times.49 The broad peak near 500 K
is attributed to recombinative desorption of some methoxy groups,
resulting from dissociation at vacancy or nonvacancy sites.42
TiO2(110) surface, with different irradiation times (0−600 s),
using 355 nm laser light. As the main feature at 300 K in the
TPD spectra at m/z = 31 is exactly the same as that at m/z = 32
(not shown), this feature is assigned to desorption of
molecularly adsorbed CH3OH. A broad TPD peak around
500 K is also seen, and it is attributed to the recombinative
desorption of dissociated CH3OH on surface vacancy sites.42
The TPD peak area for molecularly desorbed CH3OH
decreases by 16% after 600 s of 355 nm laser irradiation,
suggesting that the CH3OH molecules adsorbed on the five-
coordinated Ti4+ (Ti5c) sites were photocatalytically dissoci-
ated. This observation is consistent with our previous TPD
19040
dx.doi.org/10.1021/ja4114598 | J. Am. Chem. Soc. 2013, 135, 19039−19045

Journal of the American Chemical Society
Article
Figure 2. (A) TPD spectra collected at m/z = 30 (CH2O+) after laser irradiation times of 0, 60, and 600 s at 355 nm. (B) TPD spectra collected at
m/z = 30 after laser irradiation times of 0, 0.5, and 5 s at 266 nm. (C) TPD spectra collected at m/z = 18 (H2O+) after laser irradiation times of 0, 60,
and 600 s at 355 nm. The peak near 270 K arises from the small amount of H2O impurity in the CH3OH sample and the dissociative ionization
signal of molecularly adsorbed CH3OH in the electron-impact ionizer, while the peak near 500 K is the H2O TPD product formed from
recombination of two H-atoms and one O-atom on a BBO row. (D) TPD spectra collected at m/z = 18 after laser irradiation times of 0, 0.5, and 5 s
at 266 nm.
results on the same surface with 400 nm irradiation.44 The
high-temperature side of the CH3OH TPD peak was depleted
preferentially as the laser irradiation time increased. We
previously attributed this preferential depletion to the
combination of the CH3OH TPD peak depletion and the
shift of the CH3OH TPD peak toward lower temperature
resulting from the hydroxyl groups at bridge bonded oxygen
(BBO) sites resulting from CH3OH dissociation.44 A very small
shoulder also appears on the low-temperature edge of the main
TPD peak after 600 s of irradiation, and this shoulder is
assigned to HCOOCH3 from photoinduced cross coupling of
CH3O and CH2O on the surface.49
Typical TPD spectra collected at m/z = 31 following 266 nm
irradiation are shown in Figure 1B. The evolution of the TPD
spectrum with 266 nm irradiation is similar to that with 355 nm
irradiation. Given the ratio of photon fluxes at 355 and 266 nm
irradiation (1.37:1.00) and the ratio of absorption cross
sections at 266 and 355 nm (1.5:1.0), the number of
electron−hole pairs created per unit time is expected to be
about the same with 266 and 355 nm irradiation. Thus, it is
remarkable to see that the CH3OH depletion after just 5 s
irradiation of the 266 nm laser (40 mW) is significantly larger
than that after 600 s irradiation of the 355 nm laser (40 mW).
This result suggests that CH3OH photocatalysis at 266 nm is
much faster than at 355 nm. Similar to the TPD peak following
355 nm irradiation, the peak following 266 nm irradiation
shows a small shoulder on the low-temperature edge, which is
again assigned to desorption of the secondary HCOOCH3
product.
TPD spectra, a prominent TPD peak near 265 K can be clearly
seen. This peak is assigned to the CH2O product, which
desorbs around 270 K on a clean TiO2(110) surface.50 Note
here that hydroxyls on BBO sites would reduce the molecular
absorption energy on Ti5c sites, as in the case of CH3OH. With
irradiation at a given wavelength (266 and 355 nm), the time
scale on which the CH2O product increases is similar to that of
CH3OH depletion, indicating that the initial main product of
photocatalytic CH3OH dissociation is CH2O. The photo-
catalysis of CH3OH at 400 nm on the same surface also yields
CH2O as the main initial product. Thus, the photocatalytic
dissociation of CH3OH on TiO2(110) at 355 and 266 nm is
assumed to proceed through the same two-step dissociation
mechanism that was demonstrated for 400 nm CH3OH
photocatalysis:44
hv
CH3OH(Ti5c) → CH3O(Ti5c) + HBBO
(1)
hv
CH3O(Ti5c) → CH2O(Ti5c) + HBBO
(2)
where HBBO refers to an H atom that is adsorbed on a BBO site.
Because the increase in surface temperature during laser
irradiation is quite small (only about 5 K), both dissociation
steps must be photoinduced and not thermally activated.
CH2O may be desorbed by light;49 thus, monitoring the
CH2O product yield is not the best approach to measure the
product formation rate, as the CH2O production rate would
not be linked directly with the CH3OH depletion rate. Another
major product from CH3OH dissociation on TiO2(110) is
atomic hydrogen on the BBO sites, resulting from steps (1) and
(2). These products do not desorb by light, but upon heating,
two H atoms on BBO rows will combine with an oxygen atom
In addition to the TPD spectra at m/z = 31, we have
collected TPD spectra at m/z = 30 (CH2O+) at several different
irradiation times with both 355 and 266 nm laser light.
Representative TPD spectra are shown in Figure 2A,B. In these
19041
dx.doi.org/10.1021/ja4114598 | J. Am. Chem. Soc. 2013, 135, 19039−19045

Journal of the American Chemical Society
Article
on the BBO row to form a H2O molecule that desorbs around
500 K, leaving behind an oxygen vacancy (BBOv),44
2HBBO + BBO → H2O + BBOv
(3)
Monitoring H-atom production through the desorption of
H2O is the preferred way to measure the product formation
rate as CH3OH undergoes photocatalytic dissociation. TPD
spectra at m/z = 18 have therefore been measured at seven
irradiation times between 0 and 600 s using 355 nm light and at
six irradiation times between 0 and 5 s using 266 nm light.
Figure 2C shows three TPD spectra at m/z = 18 after laser
irradiation times of 0, 60 s, and 600 s with 355 nm light, and
Figure 2D shows three TPD spectra at m/z = 18 after laser
irradiation times of 0, 0.5, and 5 s with 266 nm light. With these
data, we can determine the formation rate of 2HBBO through
the H2O TPD signal and therefore the CH3OH depletion rate.
The calibrated CH3OH TPD signal at m/z = 31 (peak at 300
K) and the H2O TPD signal at m/z = 18 (peak at 500 K) at
different irradiation times are plotted in Figure 3. The
Figure 4. Laser irradiation time dependence of the H2O (from BBO)
TPD yield following 355 nm (left panel) and 266 nm (right panel)
irradiation of a 0.5 ML CH3OH covered TiO2(110) surface. The solid
squares and circles are the experimental data (calibrated), while the
solid lines are exponential functions that appear to follow the data very
well. The open squares indicate the rise times at 90% of the asymptotic
values of the fits for both 355 and 266 nm photocatalysis.
yields. As a result of the accumulation of H atoms on BBO sites
during CH3OH dissociation, as shown in eqs 1 and 2, the
quantum yield of the reaction will be much lower for longer
irradiation times. The quantum yields were thus calculated from
the TPD spectra after short irradiation times where the yield of
the H2O product had a quasi-linear relationship with laser
irradiation time. The times chosen for 355 and 266 nm were 60
and 0.5 s, respectively. The resulting quantum yields were 3.3 ×
10−6 for 355 nm and 5.5 × 10−4 for 266 nm, suggesting that the
initial dissociation rate of CH3OH at 266 nm is more than 2
orders of magnitude faster than that at 355 nm, in rough
agreement with the simple rise-time analysis.
The large difference between the photocatalytic dissociation
rates at the two wavelengths is surprising, because it is
inconsistent with the prediction from the widely accepted
photocatalysis model that charge carriers in TiO2 rapidly
thermalize to their respective band edges, and as a result, the
reaction rate should be dependent on the number of electron−
hole pairs generated and independent of wavelength. Our
experimental result clearly indicates that the energy of the
excited electron−hole pairs is also important.
Figure 3. Dependence of CH3OH and H2O (from BBO sites) TPD
yields on laser irradiation time, at 355 nm (left panel) and 266 nm
(right panel). The initially prepared surface was TiO2(110) with 0.5
ML of adsorbed CH3OH. The background H2O TPD signal (t = 0)
from CH3OH dissociation on the BBO vacancy sites has been
subtracted from the total H2O TPD signals. Solid squares and circles
are the experimental data (calibrated), while the solid lines are
exponential functions that appear to follow the data very well.
One possible mechanism by which photon energy could
influence the dissociation rate would be through dissociative
electron attachment (DEA).24 A 266 nm photon has an energy
of 4.67 eV, and the work function of a CH3OH-covered
TiO2(110) surface is only about 4 eV.34 Free electrons could
thus be produced during irradiation, and these free electrons
could induce DEA of CH3OH.51 However, the density of
electron states near the Fermi level (EF) of reduced TiO2(110)
surfaces is quite small,52,53 and the density of free electrons
would be very low. Furthermore, DEA of gas-phase CH3OH is
generally observed for electron attachment resonances that lie
>6 eV.51 Thus, DEA by a free electron is expected to make little
or no contribution to the photoinduced dissociation of CH3OH
on TiO2(110).
background TPD H2O signal at t = 0 from the spontaneous
dissociation of CH3OH on BBOv has been subtracted from the
data. These two plots, corresponding to 255 and 266 nm
irradiation, show the correlation between the depletion of the
CH3OH TPD signal and the increase of the H2O TPD signal,
with increasing irradiation time.
DISCUSSION
■
The results in Figure 3 show clearly that the depletion of the
CH3OH TPD peak and the increase of the H2O TPD peak are
almost exactly anticorrelated at both wavelengths, suggesting
that the H2O peak comes entirely from H atoms on the BBO
sites. From the plots of TPD H2O product yield as a function of
irradiation time (Figure 4), we found the rise times at 90% of
the asymptotic value to be 280 s for 355 nm and 2.8 s for 266
nm. A more precise determination of the difference in reaction
rates at the two wavelengths may be obtained from quantum
Electrons excited into the conduction band could also be
transferred to unoccupied levels of CH3OH or CH3O (e.g., the
lowest unoccupied molecular orbital, LUMO), that lie at or
below the energy of the photoinduced (“hot”) electrons, and
19042
dx.doi.org/10.1021/ja4114598 | J. Am. Chem. Soc. 2013, 135, 19039−19045

Journal of the American Chemical Society
Article
Figure 5. H2O product yield from 0.5 ML CH3OH adsorbed TiO2(110) as a function of photon flux with 355 and 266 nm irradiation. The
irradiation times chosen for 355 and 266 nm are 60 and 2 s, respectively. The surface temperature (∼110 K) was nearly unchanged during
irradiation. These data were collected with a new sample (i.e., a different sample than was used in the other experiments). Although the sample was
prepared in the same way at the previous sample, the product yields from the new sample were lower.
then anion excited states of the adsorbate which are formed by
electron transfer at the TiO2(110) interface could result in
dissociation of the adsorbate. This dissociation mechanism
might seem unlikely, because the LUMO of the CH3OH
molecule is calculated to be +0.69 eV,54 making it higher than
the vacuum level. However, when molecules adsorb on a
surface, the LUMO generally shifts ∼1−2 eV lower in energy.55
Assuming a width of 1 eV (based on the observed full width at
half-maximum of the low-energy resonance for electron
attachment to gas-phase CH3OH),51 it would be possible for
a conduction-band electron to transfer to the LUMO of
CH3OH.
The transfer of a hole to the CH3OH might be another
mechanism for dissociation. Based on ultraviolet photoelectron
spectroscopy data,21,25,56 the highest occupied molecular orbital
(HOMO) of CH3OH adsorbed on TiO2(110) relative to the
valence band maximum (VBM) is reported to be ∼2 eV.
Previous studies show that the photocatalytic dissociation of
CH3OH on TiO2(110) can occur even with 400 nm
irradiation.40,44 There will be a large energy mismatch between
photogenerated holes in the valence band by 400 nm (3.1 eV)
irradiation and the electronic states of CH3OH. Thus, hole
transfer to CH3OH is unlikely.
Figure 5). Within experimental error, the H2O yield at 355 nm
is linear with photon flux from 0 to 2.6 × 1017 photons cm−2
s−1. The H2O yield at 266 nm is approximately linear up to
about 7.1 × 1016 photons cm−2 s−1, and then it appears to begin
to saturate as the power is raised to 1.2 × 1017 photons cm−2
s−1. With both irradiation wavelengths, there is no evidence for
the exponential increase in yield that would be expected for a
multiphoton, or ladder-climbing, process. If a dissociation step
proceeds on the ground electronic state potential energy
surface, then it may be a single-photon process where the
photon energy is rapidly converted to high phonon excitation
that can, in turn, rapidly couple to the reaction coordinate in
the adsorbate (CH3OH or CH3O).
The apparent saturation in the H2O yield with high photon
fluxes of 266 nm light might be the result of the photoejection
of HBBO atoms, which is possible at this short wavelength (see
Figure S2). The loss of HBBO atoms would reduce the
availability of hydroxyls at BBO sites for the production of
H2O. The fact that the saturation is very mild at the laser
powers used suggests that the ejection of HBBO plays a minor
role and should have no bearing on the conclusion about the
strong photon-energy dependence of the dissociation yield of
CH3OH on TiO2(110).
Another possible explanation for photoinduced dissociation
of CH3OH on TiO2(110) is that the dissociation occurs on the
ground electronic state surface. In this case, the initially formed
electrons would thermalize to the conduction band minimum
(CBM) within hundreds of femtoseconds15 and then
recombine with the holes, in a process that is analogous to
internal conversion in the gas phase.57 As a result, the photon
energy would be fully converted to local phonon energy in a
short time. Through vibrational redistribution between the
highly excited phonon modes and the vibrational modes of
CH3OH or CH3O in a localized area, the adsorbate could gain
sufficient energy to dissociate before the phonons dissipated
their energy into the bulk. Sufficient energy for dissociation on
the ground state might become available through a single-
photon event, as implied above, or through a multiphoton
process involving the excitation of many phonons. The
efficiency of this “ladder-climbing” process would be expected
to be nonlinear with photon flux. We thus investigated the yield
of H2O as a function of photon flux at 266 and 355 nm (see
CONCLUSION
■
In summary, our wavelength-dependent TPD investigation
provides evidence that the relative quantum yield of HBBO from
the photoinduced decomposition of CH3OH on TiO2(110) at
266 nm is more than 100 times higher than that at 355 nm,
indicating that the reaction is strongly dependent on the
irradiation wavelength. This result contradicts a widely accepted
model of photocatalysis on TiO2(110), which assumes that
charge carriers in TiO2 rapidly thermalize to their respective
band edges via strong coupling with phonon modes and
reaction is thus only dependent on the number of electron−
hole pairs created during photoexcitation. Plausible explan-
ations for the wavelength-dependent rate are reaction on the
ground-state potential energy surface or transfer of a
conduction band electron to the LUMO of CH3OH or
CH2O. The importance of the electron−hole pair energy
demonstrated in this work calls for the development of a more
sophisticated surface photocatalysis model that incorporates the
19043
dx.doi.org/10.1021/ja4114598 | J. Am. Chem. Soc. 2013, 135, 19039−19045

Journal of the American Chemical Society
Article
(18) Nilius, N.; Ernst, N.; Freund, H.-J. Chem. Phys. Lett. 2001, 349,
351−357.
(19) Yamada, Y.; Kanemitsu, Y. Phys. Rev. B 2010, 82, 113103.
(20) Yamada, Y.; Kanemitsu, Y. Phys. Status Solidi C 2011, 8, 104−
107.
effect of photon energy. Such development is expected to
enhance our understanding of fundamental processes in
photocatalysis and guide the development of more efficient
photocatalysts in the future.
(21) Sporleder, D.; Wilson, D. P.; White, M. G. J. Phys. Chem. C
2009, 113, 13180−13191.
(22) Diwald, O.; Thompson, T. L.; Goralski, Ed. G.; Walck, S. D.;
Yates, J. T., Jr. J. Phys. Chem. B 2004, 108, 52−57.
(23) Brusa, M. A.; Grela, M. A. J. Phys. Chem. B 2005, 109, 1914−
1918.
ASSOCIATED CONTENT
* Supporting Information
Additional information as noted in the text. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
(24) Kim, S. H.; Stair, P. C.; Weitz, E. J. Chem. Phys. 1998, 108,
AUTHOR INFORMATION
Corresponding Authors
guoqing@dicp.ac.cn
tminton@montana.edu
xmyang@dicp.ac.cn
■
5080−5088.
(25) Grela, M. A.; Colussi, A. J. J. Phys. Chem. B 1999, 103, 2614−
2619.
(26) Stafford, U.; Gray, K. A.; Kamat, P. V. J. Catal. 1997, 167, 25−
32.
(27) Kawai, T.; Sakata, T. J. Chem. Soc. Chem. Commun. 1980, 694−
695.
(28) Henderson, M. A.; Otero-Tapia, S.; Castro, M. E. Surf. Sci. 1998,
412/413, 252−272.
(29) Farfan-Arribas, E.; Madix, R. J. Surf. Sci. 2003, 544, 241−260.
(30) Zhang, Z. R.; Bondarchuk, O.; White, J. M.; Kay, B. D.;
Present Address
⊥Also with School of Physics and Optoelectric Engineering,
Dalian University of Technology, Dalian, Liaoning 116023,
China
Author Contributions
∥These authors made similar contributions to this work.
́
Dohnalek, Z. J. Am. Chem. Soc. 2006, 128, 4198−4199.
(31) Sanchez de Armas, R.; Oviedo, J.; San Miguel, M. A.; Sanz, J. F.
J. Phys. Chem. C 2007, 111, 10023−10028.
Notes
The authors declare no competing financial interest.
(32) Oviedo, J.; Sanchez de Armas, R.; San Miguel, M. A.; Sanz, J. F.
J. Phys. Chem. C 2008, 112, 17737−17740.
ACKNOWLEDGMENTS
■
(33) Farfan-Arribas, E.; Madix, R. J. J. Phys. Chem. B 2002, 106,
10680−10692.
This work was supported by the Chinese Academy of Sciences,
National Science Foundation of China, and the Chinese
Ministry of Science and Technology. We also wish to thank
Prof. Hongjun Fan for many helpful discussions and theoretical
calculations during the course of this work.
(34) Onda, K.; Li, B.; Zhao, J.; Petek, H. Surf. Sci. 2005, 593, 32−37.
(35) Wang, L.-Q.; Ferris, K. F.; Winokur, J. P.; Shultz, A. N.; Baer, D.
R.; Engelhard, M. H. J. Vac. Sci. Technol., A 1998, 16, 3034−3040.
(36) Sanchez, V. M.; Cojulun, J. A.; Scherlis, D. A. J. Phys. Chem. C
2010, 114, 11522−11526.
(37) Martinez, U.; Vilhelmsen, L. B.; Kristoffersen, H. H.; Stausholm-
Moller, J.; Hammer, B. Phys. Rev. B 2011, 84, 205434.
(38) Zhao, J.; Yang, J. L.; Petek, H. Phys. Rev. B 2009, 80, 235416.
(39) Wong, G. S.; Kragten, D. D.; Vohs, J. M. J. Phys. Chem. B 2001,
105, 1366−1373.
(40) Zhou, C.; Ren, Z.; Tan, S.; Ma, Z.; Mao, X.; Dai, D.; Fan, H.;
Yang, X.; LaRue, J.; Cooper, R.; Wodtke, A. M.; Wang, Z.; Li, Z.;
Wang, B.; Yang, J.; Hou, J. Chem. Sci. 2010, 1, 575−580.
(41) Zhou, C. Y.; Ma, Z. B.; Ren, Z. F.; Mao, X. C.; Dai, D. X.; Yang,
X. M. Chem. Sci. 2011, 2, 1980−1983.
(42) Shen, M.; Henderson, M. A. J. Phys. Chem. Lett. 2011, 2, 2707−
2710.
(43) Ahmed, A. Y.; Kandiel, T. A.; Oekermann, T.; Bahnemann, D. J.
REFERENCES
■
(1) Fujishima, A.; Honda, K. Nature 1972, 238, 37−38.
(2) Linsebigler, A. L.; Lu, G.; Yates, J. T. Chem. Rev. 1995, 95, 735−
758.
(3) Fujishima, A.; Zhang, X.; Tryk, D. A. Surf. Sci. Rep. 2008, 63,
515−582.
(4) Gaga, U. I.; Abdullah, A. H. J. Photochem. Photobiol., C 2008, 9,
1−12.
(5) Hoffmann, M. R.; Martin, S. T.; Choi, W. Y.; Bahnemann, D. W.
Chem. Rev. 1995, 95, 69−96.
(6) Thompson, T. L.; Yates, J. T., Jr. Chem. Rev. 2006, 106, 4428−
4453.
(7) Henderson, M. A.; Otero-Tapia, S.; Castro, M. E. Faraday Discuss.
1999, 114, 313−319.
Phys. Chem. Lett. 2011, 2, 2461−2465.
(44) Guo, Q.; Xu, C.; Ren, Z.; Yang, W.; Ma, Z.; Dai, D.; Fan, H.;
Minton, T. K.; Yang, X. J. Am. Chem. Soc. 2012, 134, 13366−13373.
(45) Shen, M.; Henderson, M. A. J. Phys. Chem. C 2012, 116, 18788−
18795.
(8) Gong, X.-Q.; Selloni, A.; Dulub, O.; Jacobson, P.; Diebold, U. J.
Am. Chem. Soc. 2008, 130, 370−381.
(9) Pang, C. L.; Lindsay, R.; Thornton, G. Chem. Soc. Rev. 2008, 37,
2328−2353.
(46) Shen, M.; Acharya, D. P.; Dohnal
Phys. Chem. C 2012, 116, 25465−25469.
(47) Ren, Z.; Guo, Q.; Xu, C.; Yang, W.; Xiao, C.; Dai, D.; Yang, X.
Chin. J. Chem. Phys. 2012, 5, 507−512.
́
ek, Z.; Henderson, M. A. J.
(10) Fujishima, A.; Zhang, X.; Tryk, D. Surf. Sci. Rep. 2008, 63, 515−
582.
(11) Diebold, U. Surf. Sci. Rep. 2003, 48, 53−229.
(12) Enache, D. I.; Edwards, J. K.; Landon, P.; Solsona-Espriu, S.;
Carley, A. F.; Herzing, A. A.; Watanabe, M.; Kiely, C. J.; Knight, D. W.;
Hutchings, G. J. Science 2006, 311, 362−365.
(13) Khan, S. M.; Al-Shahry, M.; Ingler, W. B., Jr. Science 2002, 297,
2243−2245.
(14) Green, I. X.; Wang, W.; Neurock, M.; Yates, J. T., Jr. Science
2012, 333, 736−739.
(15) Henderson, M. A. Surf. Sci. Rep. 2011, 66, 185−297.
(16) Gundlach, L.; Felber, S.; Storck, W.; Galoppini, E.; Wei, Q.;
Willig, F. Res. Chem. Intermed. 2005, 31, 39−46.
(17) Gundlach, L.; Ernstorfer, R.; Willig, F. Phys. Rev. B 2006, 74,
035324.
(48) Zehr, R. T.; Henderson, M. A. Surf. Sci. 2008, 602, 1507−1516.
(49) Guo, Q.; Xu, C.; Yang, W.; Ren, Z.; Ma, Z.; Dai, D.; Minton, T.
K.; Yang, X. J. Phys. Chem. C 2013, 117, 5293−5300.
(50) Lu, G.; Linsebigler, A.; Yates, J. T., Jr. J. Phys. Chem. 1994, 98,
11733−11738.
(51) Ibanescu, B. C.; Allan, M. Phys. Chem. Chem. Phys. 2008, 10,
̃
5232−5237.
(52) Henderson, M. A.; Epling, W. S.; Perkins, C. L.; Peden, C. H. F.;
Diebold, U. J. Phys. Chem. B 1999, 103, 5328−5337.
(53) Kurtz, R. L.; Stockbauer, R.; Madey, T. E.; Roman, E.;
Desegovia, J. L. Surf. Sci. 1989, 218, 178−200.
19044
dx.doi.org/10.1021/ja4114598 | J. Am. Chem. Soc. 2013, 135, 19039−19045

Journal of the American Chemical Society
Article
(54) The LUMO of the CH3OH molecule is calculated to be +0.69
eV by a Hartree−Fock approach using Gaussian 09. (Geometry
optimization: ccsd/cc-pVTZ basis set; single point energy: ccsd(t)/
aug-cc-pVTZ basis set.)
(55) Feulner, P.; Menzel, D. Laser Spectroscopy and Photo-
Chemistry on Metal Surfaces, Part II. In Advanced Series in Physical
Chemistry; Hai-Lung Dai, W. H., Ed.; World Scientific: Singapore,
1995; Vol. 5; p 627.
(56) Wang, L.-X.; Ferris, K. F.; Winokur, J. P.; Shultz, A. N.; Baer, D.
R.; Engelhard, M. H. J. Vac. Sci. Technol., A 1998, 16, 3034.
(57) Felker, P. M.; Zewail, A. H. J. Chem. Phys. 1985, 82, 2961−2974.
19045
dx.doi.org/10.1021/ja4114598 | J. Am. Chem. Soc. 2013, 135, 19039−19045