Photocatalytic reactions at macrocrystalline fac-Mn(CO)3(η2-Ph2PCH2PPh 2)Cl-electrode-aqueous (electrolyte) interfaces

Article abstract of DOI:10.1021/ja9908974

The electrochemical oxidation of fac-Mn(CO)₃(η²-dpm)Cl (dpm = Ph₂PCH₂PPh₂) is extensively photocatalyzed when a microcrystal-electrode-aqueous (electrolyte) interface is irradiated with light having a wavelength corresponding to that of the 385 nm charge-transfer band. Investigations of the voltammetry of the solid in the presence and absence of light and monitoring the course of the reaction by EPR, electron probe, and electrochemical quartz crystal microbalance techniques indicate that the catalysis is associated with enhanced charge transport caused by doping of fac-Mn(CO)₃(η²-dpm)Cl with photooxidized mer-[Mn-(CO)₃(η²-dpm)Cl]X, where the anion, X^-, is incorporated into the solid after mass transport from the aqueous electrolyte. Photocatalysis also is found for the electrochemical oxidation of cis,mer-Mn(CO)₂(η¹-dpm)(η²-dpm)Br.

Full text of DOI:10.1021/ja9908974

8306
J. Am. Chem. Soc. 1999, 121, 8306-8312
Photocatalytic Reactions at Microcrystalline
fac-Mn(CO)3(η2-Ph2PCH2PPh2)Cl-Electrode-Aqueous (Electrolyte)
Interfaces
John C. Eklund and Alan M. Bond*
Contribution from the Department of Chemistry, Monash UniVersity, Clayton, Victoria 3168, Australia
ReceiVed March 19, 1999
Abstract: The electrochemical oxidation of fac-Mn(CO)3(η2-dpm)Cl (dpm ) Ph2PCH2PPh2) is extensively
photocatalyzed when a microcrystal-electrode-aqueous (electrolyte) interface is irradiated with light having
a wavelength corresponding to that of the 385 nm charge-transfer band. Investigations of the voltammetry of
the solid in the presence and absence of light and monitoring the course of the reaction by EPR, electron
probe, and electrochemical quartz crystal microbalance techniques indicate that the catalysis is associated
with enhanced charge transport caused by doping of fac-Mn(CO)3(η2-dpm)Cl with photooxidized mer-[Mn-
(CO)3(η2-dpm)Cl]X, where the anion, X-, is incorporated into the solid after mass transport from the aqueous
electrolyte. Photocatalysis also is found for the electrochemical oxidation of cis,mer-Mn(CO)2(η1-dpm)(η2-
dpm)Br.
Introduction
electrode surfaces. The electrode is then placed in an aqueous
electrolytic medium in which the microcrystalline material is
insoluble. Attachment of solids to an electrode in this manner
leads to the formation of an array of microscopically small sites
where redox chemistry can take place. As would be expected
when an array of microcrystals is attached to an electrode
surface, a three-phase electrode/solid/solution boundary is
obtained, which allows electrolyte counterions from the solution
phase to be transported into or out of the crystal structure upon
its oxidation/reduction, to achieve charge neutralization.17,19
In solar energy conversion devices, separation of charge in
semiconductors occurs upon excitation of an electron from a
valence to a conduction band11,14,15 and this is translated into
current flow in an external circuit. In voltammetric studies on
solids, charge separation in the crystalline materials usually is
generated via application of an external electric field rather that
a photolytically induced electron transition. However, there are
several examples of organometallic molecules which can be
photolyzed by UV/visible light in the solution phase to form a
new species in solution that may be oxidized or reduced
heterogeneously at an electrode surface. For example, in
acetonitrile, the photoelectrochemical oxidation of fac-Mn(CO)3-
In recent years, fundamental studies on the voltammetry of
solid materials has received much attention1-10 due to the large
number of important applications in fields ranging from novel
solar energy devices11 to molecular electronics.2,12,13 For
example, the voltammetry of mixed valence polyoxotungstate
compounds3-6,10 has been studied extensively and charge
transport within the solid has been defined to occur through
electron hopping and counterion diffusion mechanisms,1,2 in
contrast to solution-phase voltammetry where the charge is
primarily transported by mobile supporting electrolyte ions.14,15
In one of the most versatile forms of solid-state electro-
chemistry,16-24 redox-active microcrystals are attached to
* To whom correspondence should be sent: E-mail alan.bond@
sci.monash.edu.au. Fax: +61 3 9905 4597.
(1) Bruce, P. G., Ed. Solid State Electrochemistry; CUP: Cambridge,
1997.
(2) Kulesza, P. J.; Cox, J. A. Electroanalysis 1998, 10, 73.
(3) Kulesza, P. J.; Faulkner, L. R. J. Am. Chem. Soc. 1988, 110, 4905.
(4) Kulesza, P. J.; Faulkner, L. R. J. Electroanal. Chem. 1988, 248, 305.
(5) Kulesza, P. J.; Faulkner, L. R. J. Am. Chem. Soc. 1993, 115, 11878.
(6) Kulesza, P. J.; Faulkner, L. R. J. Am. Chem. Soc. 1991, 113, 379.
(7) Parthasatathy, A.; Martin, C. R.; Srinivasan, S. J. Electrochem. Soc.
1991, 138, 916.
(8) Gorski, W.; Cox, J. A. J. Electroanal. Chem. 1992, 323, 163.
(9) Jaworski, R. K.; Cox, J. A. Anal. Chem. 1991, 63, 2984.
(10) Karwowska, B.; Kulesza, P. J. Electroanalysis 1995, 7, 1005.
(11) Pleskov, Y. V. Solar Energy ConVersion, A Photoelectrochemical
Approach; Springer-Verlag: Berlin, 1990.
(12) Chao, S.; Wrighton, M. S. J. Am. Chem. Soc. 1987, 109, 2197.
(13) Mirkin, C. A.; Wrighton, M. S. J. Am. Chem. Soc. 1990, 112, 8596.
(14) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamen-
tals and Applications; John Wiley: New York, 1980.
(15) Brett, C. M. A.; Oliveira-Brett, A. M. Electrochemistry, Principles,
Methods and Applications; OUP: Oxford, 1993.
(16) Bond, A. M.; Cooper, J. B.; Marken, F.; Way, D. M. J. Electroanal.
Chem. 1995, 396, 407.
(17) Bond, A. M.; Colton, R.; Daniels, F.; Fernando, D. R.; Marken, F.;
Nagaosa Y.; Van Steveninck, R. F. M.; Walter, J. N. J. Am. Chem. Soc.
1993, 115, 9556.
(20) Bond, A. M.; Fletcher, S.; Marken, F.; Shaw, S. J.; Symons, P. G.
J. Chem. Soc., Faraday Trans. 1996, 92, 3925.
(21) Dostal, A.; Meyer, B.; Scholtz, F.; Schroder, U.; Bond, A. M.;
Marken, F.; Shaw, S. J. J. Phys. Chem. 1995, 99, 2096.
(22) Downard, A. J.; Bond, A. M.; Hanton, L. R.; Heath, G. A. Inorg.
Chem. 1995, 34, 6387.
(23) Bond, A. M.; Colton, R.; Marken, F.; Walter, J. N. Organometallics
1994, 13, 5122.
(24) Bond, A. M.; Colton, R.; Mahon, P. J.; Snook, G. A.; Tan, W. T.
J. Phys. Chem. B 1998, 102, 1229.
(25) Compton, R. G.; Barghout, R.; Eklund, J. C.; Fisher, A. C.; Bond,
A. M.; Colton, R. J. Phys. Chem. 1993, 97, 1661.
(26) Bond, A. M.; Colton, R.; McCormick, M. J. Inorg. Chem. 1977,
16, 155.
(27) Bond, A. M.; Grabaric, B. S.; Grabaric, Z. Inorg. Chem. 1978, 17,
1013.
(18) Bond, A. M.; Marken, F. J. Electroanal. Chem. 1994, 372, 125.
(19) Shaw, S. J.; Marken, F.; Bond, A. M. J. Electroanal. Chem. 1996,
404, 227.
(28) Colton, R.; McCormick, M. J. Aust. J. Chem. 1976, 29, 1657.
(29) Carriedo, G. A.; Riera, V.; Santamaria, J. J. Organomet. Chem. 1982,
234, 175.
10.1021/ja9908974 CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/31/1999

Microcrystalline-Electrode-Aqueous Interfaces
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8307
(η2-dpm)Cl, where dpm is Ph2PCH2PPh2, occurs via the
experiments were undertaken with an ADI instruments Maclab/4e
potentiostat system controlled by a Macintosh Powerbook microcom-
puter and potentials are quoted versus the aqueous Ag/AgCl (3 M KCl)-
reference electrode.
following photo-CE mechanism:25
C: fac-Mn(CO)3(η2-dpm)Cl
9
hν8
mer-Mn(CO)3(η2-dpm)Cl (1)
E: mer-Mn(CO)3(η2-dpm)Cl a
mer-[Mn(CO)3(η2-dpm)Cl]+ + e- (2)
For photochemical experiments, UV/visible irradiation was provided
by a broad band 300 W xenon arc lamp (Cermax LX300; ILC
technology, Sunnyvale, CA) enclosed in a R400 lamp holder (ILC
technology). The infrared component was removed by an infrared
transparent mirror after which the light was passed through a series of
lenses and focused into an optic fiber (ILC technology) directed at the
surface of the working electrode. The wavelength of the incident light
was controlled via a range of UV/visible colored-glass filters (Jena
Glaswerk, Schott & Gen., Mainz, Germany). Quantitative voltammetric
experiments were conducted using a filter which had a wavelength range
of 300-450 nm and yielded an internal transmittance of 50% or above,
with 90% transmittance between 350 and 400 nm and less than 1%
between 500 and 680 nm. The intensity of this filtered light source
was calibrated using the known25 photoinduced isomerization rate of
fac-Mn(CO)3(η2-dpm)Cl (absorption band of compound centered at 385
nm). Typically intensities of 10 mW cm-2 were obtainable in the 300-
450 nm UV/visible region. In some experiments a further blue-green
filter was utilized which had >95% transmittance between 330 and
550 nm. To obtain action spectra, a sequence of colored glass filters
were used for which 50% internal transmittance (ϑ50) occurred at 230,
335, 360, 385, 395, 400, 420, 435, 455, 475, and 550 nm, respectively.
The transmittance of these filters rose sharply on moving from lower
to higher wavelengths. A rise from 10 to 90% was attained over a 10-
40 nm range, the filters allowing >99% transmittance for all wave-
lengths 40-80 nm greater than the quoted ϑ50 values. Thus, an estimate
of the minimum wavelength available with the significantly transmitted
filtered light could be made.
X-band EPR measurements were made with a Varian E-12
spectrometer. EPR spectra were recorded at 77 and 295 K. Gains in
the range 103-104 and modulation amplitudes of 4.0 G were used.
Simultaneous voltammetric and mass balance experiments were
undertaken with an electrochemical quartz crystal microbalance (EQCM)
consisting of an Elchema (Postdam, NY) model EQCN-701 nanobal-
ance and model PS-205 potentiostat. The system was controlled by a
486 PC running VOLTSCAN software (Intellect Software, Postdam,
NY). The working electrode for the EQCM measurements was one
side of a 13 mm diameter AT-cut quartz crystal (Bright Star Crystals,
Rowville, Victoria, Australia) that had gold disks (5.0 mm diameter)
vapor deposited on each side and oscillated at a frequency of 10 (
0.05 MHz. The method of calibration of the EQCM was as described
previously.30 Solid microcrystalline material was attached to the gold
electrode surface using the same technique described above for the
graphite electrode.
If mechanisms of these kinds occur in the solid state upon
UV/visible irradiation, then the voltammetry of a microcrys-
talline organometallic material attached to a solid electrode
surface can be envisaged to give rise to two prime possible
outcomes: (1) As in the case of semiconductor electrodes,
charge separation could occur within the microcrystalline
environment upon irradiation, resulting in a current flow and
concurrent charge balance through diffusion of ions into the
crystalline material. (2) The material attached to the electrode
could undergo a chemical transformation to form a material
within the solid environment that can be oxidized or reduced
at a different potential from that of the starting material and
hence give rise to a photocatalytic reaction.
In this work, microcrystals of the fac-Mn(CO)3(η2-dpm)Cl
manganese(I) organometallic compound have been attached to
an electrode surface and the solid-state photoelectrochemistry
of this complex has been investigated. Knowledge of the
solution-phase voltammetry25-27 and the ready accessibility of
both Mn(I) and Mn(II) oxidation states suggested that both
above-mentioned routes could be available upon photolysis of
microcrystals of this compound attached to a solid electrode
surface that is in contact with an electrolytic aqueous medium.
Experimental Section
Reagents, Compounds, and Solvents. All electrolytes (NaCl,
NaClO4, LiCl, BaCl2, HCl, KCl, NaOH, NaNO3, Na2SO4, and NaF)
were of analytical or electrochemical grade purity. Triply distilled water
was used for preparation of all the aqueous electrolyte solutions.
Dichloromethane and acetonitrile were of HPLC grade (99.9%,
Mallinckrodt) and were dried for at least 12 h over molecular sieves
prior to use. The chemical oxidant, NOBF4, was used as supplied by
Aldrich (Minnesota, WI). Fac-Mn(CO)3(η2-dpm)Cl28 and cis,mer-Mn-
(CO)2(η1-dpm)(η2-dpm)Br29 were prepared using methods based on
standard literature procedures. All solutions were thoroughly purged
of oxygen by outgassing with nitrogen that had been presaturated with
the appropriate solvent.
Instrumentation and Procedures. Solid-state photoelectrochemical
voltammetric experiments were undertaken using in-house fabricated
5 mm diameter pyrolytic graphite disk electrodes housed in a Teflon
mount, an Ag/AgCl (3 M KCl) reference electrode, and a platinum
wire counter electrode. Small amounts of fac-Mn(CO)3(η2-dpm)Cl or
cis,mer-Mn(CO)2(η1-dpm)(η2-dpm)Br crystalline solid were placed on
a coarse grade filter paper and the material was ground to microcrys-
talline size (e10 µm as determined by electron microscopy; the
instrumentation used is described below) using the flat side of a spatula.
A cotton bud was then rubbed over the microcrystalline material and
adhered solid transferred to the electrode by rubbing the cotton bud
end over the electrode surface. Finally, the microcrystal coated electrode
was placed into the electrochemical cell which contained an electrolytic
aqueous medium. A water-jacketed cell, connected to a Grant Instru-
ments water bath/pump circulating system, was utilized to maintain a
constant temperature of (22 ( 1)°C for experiments with and without
irradiation. A quartz disk (Herbert Groiss, Victoria, Australia) was
attached to the bottom of the cell to allow unhindered passage of UV/
visible irradiation to the electrode surface. The quartz disk was attached
to the cell using a spring clip and a good seal was maintained using a
rubber “O-ring” between the plate and the cell. All voltammetric
Surface elemental compositions of the manganese compounds
attached to a 5 mm diameter pyrolytic graphite electrode were
determined by the electron microprobe technique. The instrumentation
used for these measurements consisted of a scanning electron micro-
scope (JEOL JSM 840A) coupled to an X-ray analyzer (Moran
scientific; 20 KV accelerator voltage). For these experiments, the
electrode to which solid had been attached was placed in the aqueous
electrolyte solution for 5 min in the absence and then presence of light.
The graphite disk electrode was then removed from its Teflon mount,
rinsed with distilled water, and fixed with double-sided tape onto a
stub and gold plated in a Balzers sputter-coating unit.
Results and Discussion
A. Voltammetry of fac-Mn(CO)3(η2-dpm)Cl at a Solid-
Electrode-Aqueous (0.1 M NaCl) Interface in the Presence
and Absence of Light. Initial solid-state voltammetric experi-
ments were conducted at a microcrystalline fac-Mn(CO)3(η2-
dpm)Cl-graphite electrode-0.1 M NaCl aqueous electrolyte
interface. In the absence of irradiation with light, repeated
cycling of the potential (scan-rate, 100 mV s-1) over the range
-0.20 to +1.20 V vs Ag/AgCl produced a barely detectable
(30) Shaw, S. J.; Marken, F.; Bond, A. M. Electroanalysis 1996, 8, 732.

8308 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Eklund and Bond
Figure 2. Phototransient response obtained for the oxidation of solid
fac-Mn(CO)3(η2-dpm)Cl attached to a graphite electrode and placed
in contact with aqueous (0.1 M NaClO4) electrolyte after irradiation
by light of different wavelengths. The electrode was held at a potential
of +1.10 V vs Ag/AgCl.
Figure 1. Solid-state cyclic voltammograms obtained at a scan rate
of 100 mV s-1 for oxidation of fac-Mn(CO)3(η2-dpm)Cl attached to a
pyrolytic graphite electrode and placed in contact with aqueous 0.1 M
NaCl electrolyte: (- - -) 10th “dark” cycle, (s) 10th “light” cycle
(wavelength, 300-450 nm; intensity, 10 mW cm-2), and (‚‚‚) 10th
“dark” cycle following 10 “light” cycles of the potential.
ox
oxidation process at Ep ) +0.90 V with a corresponding
red
ox
reduction process at Ep ) +0.62 V (Figure 1), where Ep
red
and Ep are oxidation and reduction peak potentials, respec-
tively. Repetitive cycling of the potential (50 cycles) failed to
yield any significant improvement in the voltammetric response,
relative to that shown in Figure 1. A weak, barely detectable
chemically reversible response also is observed at Epox ) +0.23
red
V and Ep ) -0.10 V after 10 cycles of the potential. This
overall behavior appears to be closely related to the square
reaction scheme26,27 observed in dichloromethane. In organic
solvent solution-phase voltammetry, fac-Mn(CO)3(η2-dpm)Cl
is oxidized to the 17-electron cationic fac+ form which then
rapidly isomerizes to the mer+ form, which then gives rise to a
chemically reversible mer+/0 couple at a less positive potential.
By analogy, in the solid-state situation, the processes with peak
potentials of +0.90 and +0.62 V are postulated to be associated
with the fac-[Mn(CO)3(η2-dpm)Cl]+/0 couple, while the pro-
cesses at less positive potentials are attributable to the mer-
[Mn(CO)3(η2-dpm)Cl]+/0 couple, with of course substantially
different mechanisms applying to the solution and solid-state
voltammetry.
Figure 3. Dependence of photocurrent on electrode potential for the
oxidation of solid fac-Mn(CO)3(η2-dpm)Cl attached to a pyrolytic
graphite electrode and placed in contact with 0.1 M NaCl aqueous
electrolyte.
current, I, initially rises rapidly and then decays rapidly with
time, t, in a Cottrellian manner (I proportional to t-1/2
)
presumably due to a diffusional relaxation process (see later).
If the magnitude of the photocurrent is measured as a function
of wavelength, using a sequence of colored glass filters, then
the maximum value is observed between 350 and 450 nm, with
the photocurrent tailing off to zero for wavelengths greater than
500 nm. The wavelength for the maximum photocurrent
correlates with the 385 mm charge-transfer absorption spectrum
in the electronic spectrum of fac-Mn(CO)3(η2-dpm)Cl. Thus,
the photocurrent can be attributed to a process associated with
absorption of light by fac-Mn(CO)3(η2-dpm)Cl.
Figure 3 displays the photocurrent associated with the
oxidation of fac-Mn(CO)3(η2-dpm)Cl (300-450 nm wavelength
light) as a function of applied potential. An increase in
photocurrent is observed only over the potential range where
oxidation of the fac-manganese(I) species occurs
In the presence of 300-450 nm irradiation, the solid-state
voltammetry for oxidation of fac-Mn(CO)3(η2-dpm)Cl becomes
far better defined (Figure 1). Now, after 10 or more cycles of
the potential and using a scan rate of 100 mV s-1, a clearly
detected fac/fac+ response is observed (Ep ) +0.98 V; Ep
ox
red
) +0.71 V), while the peak current associated with the mer/
mer+ process is only slightly enhanced (Epox ) +0.23 V, EP
red
) -0.10 V). Additional cycling of the potential beyond the
first 10 cycles yielded little further change in the voltammetry
of fac-Mn(CO)3(η2-dpm)Cl attached to the electrode surface.
The photolytic fac oxidative peak current enhancement de-
creased as the scan rate increased, with negligible photoen-
hancement being observed at scan rates greater than 1 V s-1
.
Related results to those described above are observed when
the managanese(I) carbonyl compound is attached to a platinum
electrode and placed in contact with an aqueous electrolyte (0.1
M NaCl) solution. The peak potentials are now as follows, for
B. Effect of the Nature of the Electrolyte. Table 1
summarizes the data obtained for the solid-state voltammetry
of fac-Mn(CO)3(η2-dpm)Cl in the presence of 300-450 nm
irradiation when the electrolyte present in the solution phase is
varied.
ox
ox
fac/fac+ Ep ) +0.91 V and for mer/mer+ Ep ) +0.18 V
red
and EP ) -0.06 V, but no peak for reduction of fac+ could
be detected above the background signal.
(i) Anion Dependence. For the F-/NO3-/ClO4 series of
-
When the potential of the graphite electrode is more positive
than for the fac oxidation (+1.10 V), the phototransient response
illustrated in Figure 2 is obtained. Under these conditions, the
sodium salts, two processes again are observed upon irradiation
with 300-450 nm UV/visible light. However, the potentials are

Microcrystalline-Electrode-Aqueous Interfaces
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8309
Table 1. Voltammetric Data Obtained at a Scan Rate of 100 mV s-1 When Solid fac-Mn(CO)3(η2-dpm)Cl Attached to a 5 mm Diameter
Pyrolytic Graphite Electrode in Contact with 0.1 M Aqueous Electrolyte Solution Is Irradiated with 300 to 450 nm UV/Visible Light (10 mW
a
cm-2
)
electrolyte
Epox(fac)/V
Epred(fac)/V
Epox(mer)/V
Epred(mer)/V
comments
NaCl
NaNO3
NaF
Na2SO4
NaClO4
KCl
BaCl2
LiCl
0.98
0.89
0.67
0.86/1.05
0.98
0.97
0.87
0.89
0.71
0.70
0.45
0.63
0.78
0.63
0.57
0.63
0.23
0.26
0.45
0.18/0.35
0.63
0.33
0.18
0.45
-0.10
-0.10
-0.02
-0.11
-
-0.05
-0.05
-
well-resolved peaks
well-resolved peaks
broad mer peaks
split oxidation peaks
broad peaks
well-resolved peaks
broad peaks
well-resolved peaks
a Peak potentials are reported from the tenth cycle of the potential in the presence of light.
Figure 5. EPR spectra obtained in the ex situ mode at 77 K after
fac-Mn(CO)3(η2-dpm)Cl is attached to a graphite electrode (open circuit
potential) and then placed in contact with 0.1 M NaCl aqueous
electrolyte for 5 min: (lower curve) absence of light and (upper curve)
presence of 300-450 nm UV/visible light (intensity, 10 mW cm-2).
For KCl and LiCl the oxidation potentials are similar to those
recorded in 0.1 M NaCl and the photovoltammetric response is
similar to that in Figure 1. However, with 0.1 M LiCl as the
electrolyte, the mer+ reduction peak was only observed at slow
scan rates (10 mV s-1). When BaCl2 was used as the electrolyte,
well-defined voltammetry also was only observed at slow scan
rates (10 mV s-1).
C. EPR and Electron Microprobe Studies. To identify the
origin of the photocatalysis, solid fac-Mn(CO)3(η2-dpm)Cl was
mechanically attached to the graphite surface. The electrode was
then placed in aqueous 0.1 M NaCl at open circuit potential.
After 5 min, the electrode was removed from the solution and
left to dry. The manganese material was then scraped off the
electrode surface and placed in an EPR tube. Figure 5 shows
that in the absence of light, a single line EPR response is
observed at 3415 G. This signal is associated with the graphite
material now present as a mixture with manganese compound
(as confirmed by recording the EPR spectrum of the graphite
in the absence of any attached manganese compound). In the
presence of light, of wavelength 300-450 nm, an additional
six-line EPR spectrum is obtained, which is consistent with the
formation of an oxidized manganese(II) complex (spin I ) 5/2).
The spectrum shown in Figure 5 is consistent with the presence
of high-spin Mn(II). However, because of the relatively high
intensity usually associated31 with isotropic sharp-line spectra
obtained with the high-spin state versus the low-intensity broad
anisotropic spectra commonly found with the low-spin state, it
Figure 4. First 10 cycles of solid-state cyclic voltammograms obtained
at a scan rate of 100 mV s-1 for fac-Mn(CO)3(η2-dpm)Cl attached to
a pyrolytic graphite electrode and placed in contact with aqueous (a)
0.1 M NaNO3 and (b) 0.1 M Na2SO4 electrolyte following irradiation
by 300-450 nm UV/visible light. The initial “light” cycle is represented
by a dotted (‚‚‚) line.
considerably different from those observed when 0.1 M NaCl
is present. The dramatic effect of irradiation is shown in Figure
4a when the aqueous medium contains 0.1 M NaNO3. In the
initial cycle, the voltammetric responses are very weak, even
in the presence of light. However, after 10 cycles of the
potential, both reversible oxidation processes are very well
defined. In the presence of 0.1 M Na2SO4 as the electrolyte,
the fac and mer oxidation peaks are split (Figure 4b) presumably
2-
-
because either SO4 or HSO4 may be incorporated into the
microcrystalline structure to give different potentials for the Mn-
(I) to Mn(II) processes, or two different phases of fac- and mer-
[Mn(CO)3(η2-dpm)Cl]2SO4 exist, each having slightly different
potentials.
(31) (a) Cookson, D. J.; Smith, T. D.; Boas, J. F.; Hicks, P. R.; Pilbrow,
J. R. J. Chem. Soc., Dalton Trans. 1977, 211. (b) Carriedo, G. A.; Connelly,
N. G.; Perez-Carreno, E.; Orpen, G. A.; Rieger, A. L.; Rieger, P. H.; Riera,
V.; Rosair, M. J. Chem. Soc., Dalton Trans. 1993, 3103.
(ii) Cation Dependence. A series of chloride salts were
examined to probe the effect of the cation in the electrolyte.

8310 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Eklund and Bond
Figure 6. Schematic diagram of the EQCM cell used in photoelectrochemical experiments.
is not possible on the basis of the EPR evidence to exclude the
presence of a high concentration of a low-spin manganese(II)
complex. Importantly, despite uncertainties in the assignment
of the spin state, the EPR result implies that even in the absence
of any externally applied potential, the manganese(I) complex
attached to the electrode surface in contact with an aqueous
(electrolyte) solution phase can be oxidized to a Mn(II) complex
upon photolysis. Furthermore, it can be noted that the EPR
spectrum also is similar to that observed after chemical oxidation
of fac-Mn(CO)3(η2-dpm)Cl.26,27 The EPR spectra in Figure 5
were obtained at 77 K. At ambient temperature (293 K), EPR
spectra were barely detectable above background, but have the
same basic features as those observed at 77 K.
Figure 7. EQCM data obtained when solid fac-Mn(CO)3(η2-dpm)Cl,
attached to a gold-coated quartz crystal and placed in contact with
aqueous 0.1 M NaCl electrolyte, is irradiated periodically with 300-
450 nm light.
Electron microprobe experiments on the nonphotolyzed solid
confirmed that the elemental distribution of manganese, chlorine,
and phosphorus is in the 1:1:2 proportion expected for the
formulation Mn(CO)3(η2-dpm)Cl. After application of UV/
visible light for 5 min, the relative level of chlorine in the solid
material attached to the electrode surface, as determined after
removal from an electrode placed in 0.1 M NaCl, which was
thoroughly rinsed with water, is significantly enhanced by up
to 50% whereas the manganese and phosphorus levels always
remained in the expected 1:2 ratio. The fact that sodium was
not detected confirmed that all NaCl from the electrolyte had
been removed by the rinsing procedure. This result suggests
that irradiation leads to a photooxidation process which in turn
causes chloride anions from the aqueous electrolyte to become
incorporated into the manganese(II) microcrystals to form [Mn-
(CO)3(η2-dpm)Cl]Cl on the surface. Thus, both the EPR and
microprobe data infer that the manganese(I) species attached
to the electrode surface are partially oxidized to the manganese-
(II) state upon irradiation. The electron probe data confirm that
the chloride anion is transported from the solution to solid phase
as a consequence of photooxidation.
Figure 7 shows that an increase in mass is recorded upon
irradiation of fac-Mn(CO)3(η2-dpm)Cl attached to the gold
electrode surface, while upon terminating the irradiation of the
electrode surface, the mass decreases. Upon each subsequent
irradiation burst, the mass increases further, with the background
dark mass also becoming greater on the time scale of the
switching employed in Figure 7. Results contained in Figure 7
confirm that no mass changes occur upon irradiation of a blank
gold electrode to which no solid compound has been attached.
These in situ EQCM experiments provide further evidence that
irradiation of the manganese material is accompanied by anions
being incorporated into the structure (increase in mass) which
is as expected if photooxidation to a Mn(II) cation occurs.
E. Voltammetry of fac-Mn(CO)3(η2-dpm)Cl Doped with
the mer-Mn(CO)3(η2-dpm)Cl]BF4. The above results lead to
the conclusion that with irradiation of fac-Mn(CO)3(η2-dpm)-
Cl attached to the electrode surface, the rate of transport of
electrons and ions within the solid is greatly enhanced via
incorporation of a manganese(II) salt. This doping is assumed
to increase the conductivity of nonconducting organometallic
species (resistivity typically g109 Ωcm17) and thereby enhance
transport of charge to and from the electrode surface via an
“electron-hopping” mechanism.1 This doping mechanism is
believed to be the origin of the photocatalysis observed in the
solid-state voltammetry of fac-Mn(CO)3(η2-dpm)Cl. If this
D. Electrochemical Quartz Crystal Microbalance Studies.
The EPR and electron probe techniques were applied in an ex
situ manner. To obtain in situ data, EQCM experiments were
conducted using a cell of design shown in Figure 6. Data were
obtained from fac-Mn(CO)3(η2-dpm)Cl attached to a gold
electrode in contact with 0.1 M NaCl aqueous solution. The
electrode was held at a potential of 0.00 V (vs Ag/AgCl) where
oxidation of neither fac- or mer-Mn(CO)3(η2-dpm)Cl occurs.
Phototransient experiments were conducted when 300-450 nm
irradiation (intensity 10 mWcm-2) was focused periodically on
the electrode surface.
(32) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry; John
Wiley: New York, 1988.

Microcrystalline-Electrode-Aqueous Interfaces
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8311
Figure 8. Mechanism proposed to explain the photocatalysis of the electrochemical oxidation process at the fac-Mn(CO)3(η2-dpm)Cl-electrode-
aqueous electrolyte interface.
mechanism is correct, it would be predicted that chemical doping
of the solid with an oxidized manganese(II) form of the
compound also would lead to current enhancement in the solid-
state voltammetry of fac-Mn(CO)3(η2-dpm)Cl.
(I) and Mn(II) are formed, each being doped on the surface
with small amounts of material in the other oxidation state, in
a manner akin to that achieved by photooxidation of fac-Mn-
(CO)3(η2-dpm)Cl.
F. Mechanism of the Photocatalysis of the Solid-State
Voltammetry. The combination of results obtained by EPR,
electron microprobe, EQCM, and chemical doping experiments
suggest that irradiation of fac-Mn(CO)3(η2-dpm)Cl attached to
an electrode surface in contact with 0.1 M NaCl electrolyte
results in photooxidation of the manganese(I) compound to a
manganese(II) cation with concomitant injection of chloride
anions from the electrolyte occurring to neutralize the excess
positive charge generated.
Figure 8 represents a schematic diagram of a mechanism
proposed to explain the photocatalysis. It is suggested that upon
photolysis of the microcrystalline material, the manganese(I)
species forms a charge-transfer excited state, which is then
rapidly oxidized to a fac-manganese(II) cationic product that
can then isomerize to the mer+ form. From solution-phase
studies, it is known that photolysis of fac-Mn(CO)3(η2-dpm)Cl
induces isomerization to the mer form.26,27 Thus, it is also
possible that it is this mer isomer generated by photoisomer-
ization that is in turn photooxidized directly to its mer cationic
form. Presumably, reduction of water at the interface and transfer
of chloride ion from the electrolyte to solid phases complete
the photoredox reaction. Voltammetric evidence for the forma-
tion of the mer isomeric form upon photolysis is provided by
the fact that a mer+/0 couple is observed in initial cycles in the
presence of light at a less positive potential than that for
oxidation of the fac isomer (see Figure 1, 4, and 5). Photooxi-
dation and charge neutralization results in the holes of Mn(II)
being created at the surface. The mer+ doped material is now
in a better state to transport charge when oxidation of the fac-
Mn(CO)3(η2-dpm)Cl occurs at the solid-electrode interface and
thus photocatalysis of the current occurs upon electronic
excitation of the material attached to the electrode surface.
Chemical doping of fac-Mn(CO)3(η2-dpm)Cl with mer-Mn-
(CO)3(η2-dpm)Cl may be achieved as follows. A 10 mM
dichloromethane solution of fac-Mn(CO)3(η2-dpm)Cl was pre-
pared to which small molar quantities of the one-electron oxidant
32
NOBF4 was added to generate mer-manganese(II) tetrafluo-
roborate salt via the oxidation-isomerization26,27 reaction:
NOBF4 + fac-Mn(CO)3(η2-dpm)Cl f
fac-[Mn(CO)3(η2-dpm)Cl]BF4 + NOv f
mer-[Mn(CO)3(η2-dpm)Cl]BF4 (3)
Doping levels up to the 10% level were achieved via this
method. One milliliter of the resulting dichloromethane solution
mixture of the fac-Mn(CO)3(η2-dpm)Cl and mer-[Mn(CO)3(η2-
dpm)Cl]BF4 was then placed on the surface of a freshly cleaved
pyrolytic graphite electrode surface. After evaporation of the
solvent, the required solid mixture was deposited onto the
electrode surface, and the voltammetric properties were probed.
For purposes of comparison, fac-Mn(CO)3(η2-dpm)Cl that had
not been doped with its cationic form also was attached to the
electrode after evaporation from a dichloromethane solution.
A substantial effect in the voltammetry was observed after
doping of fac-Mn(CO)3(η2-dpm)Cl with 5% of the manganese-
(II) cation. Thus, for the doped sample and in the dark, a well-
defined fac-Mn(CO)3(η2-dpm)Cl oxidation process was observed
on the first oxidative voltammetric scan (scan rate, 100 mV
s-1), and upon successive oxidative voltammetric scans, a
positive shift in the peak potential of the fac-Mn(CO)3(η2-dpm)-
Cl oxidation process also occurred as is the case with the
photocatalysis experiments described previously. In the experi-
ments described above, presumably microcrystals of both Mn-

8312 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Eklund and Bond
Phototransient experiments (Figure 2) therefore are analogous
to switching on the oxidizing power of the electrode via a
potential step to an appropriate potential and observing the
current decay due to mass transport processes associated with
anion diffusion.
If the above ideas are correct, then other compounds having
properties related to fac-Mn(CO)3(η2-dpm)Cl also should exhibit
voltammetric photocatalysis. The solid-state photovoltammetry
of cis,mer-Mn(CO)2(η1-dpm)(η2-dpm)Br adhered to an electrode
in contact with 0.1 M aqueous NaCl was therefore examined.
The organic solvent solution-phase voltammetry33 of interest
for this species consists of the one-electron oxidation process,
cis,mer-Mn(CO)2(η1-dpm)(η2-dpm)Br h
Figure 9. Solid-state voltammograms obtained in the presence and
absence of 300-450 nm UV/visible irradiation when cis,mer-Mn(CO)2-
(η1-dpm)(η2-dpm)Br is attached to a pyrolytic graphite electrode and
then placed in contact with an aqueous 0.1 M NaCl electrolyte. Initial
cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)Br]+ + e-
f
trans-[Mn(CO)2(η1-dpm)(η2-dpm)Br]+ (4)
cycle shown at a scan rate of 100 mV s-1
.
A cis-fac-[Mn(CO)2(η1-dpm)(η2-dpm)Br]+ intermediate may
also be relevant to the isomerization reaction in eq 4. It can be
seen from Figure 9 that significant photocatalysis is observed
for electrochemical oxidation of cis,mer-Mn(CO)2(η1-dpm)(η2-
dpm)Br in the presence of 300-450 nm UV/visible irradiation,
the wavelength of which corresponds to a cis,mer-Mn(CO)2-
(η1-dpm)(η2-dpm)Br charge-transfer band (λmax ) 398 nm in
acetonitrile solution).
may be catalyzed if irradiation with light of a wavelength
corresponding to an absorption band can induce a suitable
doping mechanism. This possibility has been demonstrated in
the case of photoactive manganese carbonyl complexes which
can exist in two readily accessible oxidation states and two
isomeric forms having different reversible potentials.
Acknowledgment. The authors express their appreciation
to Dr. Georgii Lazerev and Professor John Pilbrow for assistance
with the EPR studies, Richard Morrison for generous provision
of the light source, and Adrian van den Bergen for the electron
microprobe data. Financial assistance from the Australian
Research Council also is gratefully acknowledged.
Conclusions
The voltammetry of poorly conducting solids attached to
electrode surfaces that are in contact with a solvent (electrolyte)
(33) Eklund, J. C.; Bond, A. M.; Colton, R.; Humphrey, D. G.; Mahon,
P. J.; Walter, J. N. Inorg. Chem. 1999, 38, 2005.
JA9908974