Synthesis and properties of [Pt(4-CO2CH3-py) 2(mnt)]: Comparison of pyridyl and bipyridyl-based dyes for solar cells

Article abstract of DOI:10.1039/b811943k

The dye complexes [Pt(4-CO₂R-py)₂(mnt)] (R = H (3a), CH₃ (3b)) and the precursor complexes [Pt(4-CO₂R-py) ₂Cl₂] (2a, 2b) (py = pyridyl) were synthesised, characterised by electrochemical, spectroscopic, spectroelectrochemical (UV-vis-nIR and in situ EPR) and hybrid DFT computational methods and attached to a TiO₂ substrate to determine charge recombination kinetics. The results were compared to the bipyridyl analogues [Pt{X,X′-(CO ₂R)-2,2′-bipyridyl}(mnt)], (X = 3 or 4). The electronic characteristics of the bis-pyridyl complex were found to be different to the bipyridyl complexes making the former harder to reduce, shifting the lowest-energy absorption band to higher energy and showing separate degenerate LUMO orbitals on the two pyridine rings. The latter point determines that the di-reduced pyridyl complex remains EPR active, unlike the bipyridyl analogue. Complex 3a attached to nanocrystalline TiO₂ shows a long charge recombination lifetime in comparison with the analogous complex with the ubiquitous 4,4′-(CO₂H)₂-bipyridyl ligand, suggesting that pyridyl complexes may possess some advantage over bipyridyl complexes in dye-sensitised solar cells.

Full text of DOI:10.1039/b811943k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Synthesis and properties of [Pt(4-CO2CH3-py)2(mnt)]: comparison of pyridyl
and bipyridyl-based dyes for solar cells†
Lucy P. Moorcraft,a Ana Morandeira,b James R. Durrant,b James R. Jennings,c Laurence M. Peter,c
Simon Parsons,a Andrew Turner,a Lesley J. Yellowleesa and Neil Robertson*a
Received 14th July 2008, Accepted 8th September 2008
First published as an Advance Article on the web 3rd November 2008
DOI: 10.1039/b811943k
The dye complexes [Pt(4-CO2R-py)2(mnt)] (R = H (3a), CH3 (3b)) and the precursor complexes
[Pt(4-CO2R-py)2Cl2] (2a, 2b) (py = pyridyl) were synthesised, characterised by electrochemical,
spectroscopic, spectroelectrochemical (UV-vis-nIR and in situ EPR) and hybrid DFT computational
methods and attached to a TiO2 substrate to determine charge recombination kinetics. The results were
compared to the bipyridyl analogues [Pt{X,X¢-(CO2R)-2,2¢-bipyridyl}(mnt)], (X = 3 or 4). The
electronic characteristics of the bis-pyridyl complex were found to be different to the bipyridyl
complexes making the former harder to reduce, shifting the lowest-energy absorption band to higher
energy and showing separate degenerate LUMO orbitals on the two pyridine rings. The latter point
determines that the di-reduced pyridyl complex remains EPR active, unlike the bipyridyl analogue.
Complex 3a attached to nanocrystalline TiO2 shows a long charge recombination lifetime in
comparison with the analogous complex with the ubiquitous 4,4¢-(CO2H)2-bipyridyl ligand,
suggesting that pyridyl complexes may possess some advantage over bipyridyl complexes in
dye-sensitised solar cells.
in light harvesting for photochemical hydrogen production2 and
Introduction
non-linear optical studies.3 Related complexes have also been
Square planar d8 [Pt(II)(diimine)(dithiolate)] complexes have at-
prepared where the dithiolate is replaced by two monodentate
thiolate ligands.4
tracted much attention due to their attractive electronic and pho-
tochemical properties.1a,b They are known to show luminescence
in fluid solution at ambient temperatures and absorb in the visible
region with extinction coefficients e of 5000–10000 M-1 cm-1.
This solvatochromatic visible absorption has been described as a
“mixed-metal–ligand-to-ligand” charge transfer (MMLL¢CT) due
to the significant metal character in the dithiolate-based HOMO
and has been assigned as a (Pt(d)/S(p)/dithiolate → p*/diimine)
transition, which typically appears around 450–500 nm.1a
Research on a series of [Pt(II)(diimine)(dithiolate)] complexes by
Cummings and Eisenberg has shown that systematic modifications
to one or both of the ligands can predictably alter the excited state
energies by up to 1 eV.1a The assignment of the HOMO as mostly
dithiolate based and the LUMO as diimine based allowed the
separate tuning of either of these orbitals to achieve the desired
photophysical characteristics. This has been exploited for example
The high tunability and good photoluminescent properties of
[Pt(II)(diimine)(dithiolate)] complexes also mean that they are
good candidates for sensitising dyes in DSSCs. In 1991, O’Regan
and Gra¨tzel reported the design of the dye-sensitised solar cell
(DSSC), which has since shown power conversion efficiencies of
up to 11%.5 These cells consist of dye molecules anchored to the
surface of nanocrystalline TiO2, which absorb visible light and
inject an electron into the conduction band of the TiO2 from their
excited state. The dye is re-reduced by a solution redox couple
and the separated electrons and holes are collected at different
electrodes to produce the cell potential.
The tunability of [Pt(II)(diimine)(dithiolate)] complexes makes
them excellent model dyes for studies of the light-harvesting and
electron transfer processes since the HOMO and the LUMO
characteristics can each be separately altered. The tuning of the
HOMO by varying the dithiolate ligand has been investigated6,7
and tuning the LUMO has been achieved by us through the
study of a family of [Pt(II)(diimine)(dithiolate)] complexes with the
general formula [Pt{X,X¢-(CO2R)-2,2¢-bipyridyl}(mnt)] (where
X = 3, 4, or 5, R = H or Et and mnt = maleonitriledithiolate).8 For
these complexes, alteration of the position of attachment of the
carboxylic acid group is of particular importance as this controls
the electronic coupling with the TiO2 and the interfacial electron
transfer processes that drive the solar cell. The 3,3¢ analogue was
found to give the best cell performance of the series, despite the fact
that studies of metal–bipyridyl dyes have almost entirely focused
on 4,4¢-(CO2H)-bpy ligands (bpy = bipyridyl). This was due to
more favourable (i.e. slower) charge-recombination kinetics than
aSchool of Chemistry, Joseph Black Building, King’s Buildings, West Mains
Road, Edinburgh, UK EH9 3JJ. E-mail: neil.robertson@ed.ac.uk.; Fax: +44
1316 504743; Tel: +44 1316 504755
bCentre for Electronic Materials and Devices, Department of Chemistry,
Imperial College, London, UK SW7 2AZ
cDepartment of Chemistry, University of Bath, Bath, UK BA2 7AY
† Electronic supplementary information (ESI) available: Fig. S1, CV of
1b; Fig. S2, CV of 2b; Fig. S3 and Table S1, OTTLE of 1b; Fig. S4 and
Table S2, OTTLE of 2b; Fig. S5, EPR spectrum and simulation of 1b1-;
Table S3, EPR simulation parameters for 1b1-; Fig. S6, EPR spectrum
and simulation of 2b2-; Table S4, calculated atom contributions to frontier
orbitals of 3b and Table S5, TDDFT calculated selected transitions for
3b. CCDC reference number 694836. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/b811943k
6940 | Dalton Trans., 2008, 6940–6947
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
for the 4,4¢ and 5,5¢ analogues. A twist, which disrupts the planarity
of the bipyridine in the 3,3¢-dye was interpreted as the origin of
this improved performance as the reduced delocalisation inhibits
the charge recombination.9
easy binding to TiO2, whereas the ester analogues show greater
solubility in a wider range of solvents and produce clearer results
in the electrochemical characterisation techniques used.
Although no crystal structure of [Pt(4-CO2CH3-py)2mnt] was
obtained, the crystal structure of the dichloride complex 2b was
determined (Fig. 1) and, due to its cis-structure, is expected
to show similar features to 3b regarding the pyridyl units. The
molecule resides on a two-fold axis and the Pt–N bond length
This observation has now led us to investigate substituted
pyridines (hereafter abbreviated as py) as ligands in such square-
planar Pt complexes instead of the bpy moiety since this will
lead to further decoupling of the two pyridyl rings. Despite the
intense interest in the related [Pt(II)(diimine)(dithiolate)] com-
plexes, only one prior study of a [Pt(II)(py)2(dithiolate)] complex,
[Pt(4-CH3-py)2(mnt)], has previously been reported however in
this case the electronic properties have not been thoroughly
studied.10 Pyridyl ligands have occasionally been used in ruthe-
nium DSSC sensitisers11–14 but have not been widely investigated.
They could offer flexible synthetic options and tunability of the
absorption spectrum for a dye as well as possibly higher molar
extinction coefficients than analogous bpy complexes.15 Here
we report the synthesis, electronic characterisation and charge
recombination kinetics for the substituted-pyridine complex
[Pt(II)(4-CO2CH3-py)2(mnt)] along with the precursor complexes,
representing the first detailed electronic characterisation of a
[Pt(II)(py)2(dithiolate)] complex.
˚
˚
(2.018(2) A) and Pt–Cl bond length (2.2976(6) A) are within the
normal range for Pt(II) square-planar species.16 The planarity of
the Pt centre is verified by angles of ~90◦ between neighbouring
groups and ~180◦ between opposite groups. The torsion angle
N1–Pt–N1–C2 is 80.55◦ such that both pyridyl ligands are almost
orthogonal with the platinum dichloride plane and also with each
other. This is shown by the angle between the planes formed
between the C5N-rings of the two pyridyl ligands, which is 86.69◦.
The complex has cis geometry as would be expected from the
trans effect directed substitution of [PtCl4]2- and this geometry
might aid the subsequent addition of the bidentate dithiolate
ligand. Twenty-one [Pt(substituted-pyridyl)2Cl2] complexes have
been structurally characterised16 to date, however only five of
these showed cis geometry. Only one of these structures contains
pyridines substituted with esters in the 4-positions and this showed
trans geometry.17 To compare, the Pt–N bonds in the crystal
structure of [Pt{3,3¢(CO2CH2CH3)2–2,2¢-bipyridyl}(mnt)] 5b are
slightly longer than for the pyridine complex at 2.074(10) and
Results and discussion
Synthesis and structure
◦
8
˚
2.058(8) A and the N–Pt–N angle is much narrower at 79.4(3) .
[Pt(4-CO2CH3-py)2mnt] 3b was synthesised using the route
shown in Scheme 1. The reaction of two molar equivalents
of 4-CO2CH3-py with potassium tetrachloroplatinate gave the
dichloro-substituted platinum precursor 2b and further reaction
of this compound with the disodium salt of mnt gave the desired
product. This was based on a similar procedure to that for the
synthesis of the bipyridine analogues (4b, 5b), except that the
pyridyl complexes were more susceptible to degradation during
isolation, presumably due to greater lability of the monodentate
ligands and care had to be taken to avoid heating during work-up.
This illustrates the compromise introduced by the inflexibility of
the bpy ligand in comparison with two pyridyl ligands.
Fig. 1 Crystal structure of [PtCl2(4-CO2CH3-py)2] 2b wit◦h H atoms
˚
removed for clarity. Selected bond lengths (A) and angles ( ): Pt–Cl(1)
2.2967(6); Pt–N(1) 2.018(2); N(1)–Pt–N(1) 88.91(11); Cl(1)–Pt–Cl(1)
92.32(3); N(1)–Pt–Cl(1) 89.39(6), 178.20(6).
Electrochemistry
The electrochemistry of [Pt(4-CO2Me-py)2(mnt)] 3b was studied
by cyclic voltammetry (in a solution of 0.1M TBABF4–DMF)
(Fig. 2, Table 1) and shows two reversible (as evidenced by straight
Table 1 Electrochemical data (E1/2) recorded in 0.1 M TBABF4–DMF
vs. Ag/AgCl, *chemically-irreversible process
Eox
Ered(1)
Ered(2)
Ered(1) - Ered(2)
Scheme 1 Synthesis of [Pt(4-CO2R-py)2(mnt)] 3, where R = H 3a or CH3
3b. (i) K2PtCl4(aq), reflux (ii) Na2mnt, RT. Related bipyridyl complexes that
we have previously reported,8,9 4b and 5b, are shown for comparison.
1b
—
—
+1.24*
+1.39*
+1.35*
-1.69
-1.15
-1.10
-0.65
-0.59
—
—
2b
-1.32
-1.25
-1.20
-12.0
0.17
0.15
0.55
0.61
3b
4b8,9
5b8,9
Both acid and ester analogues of the dichloride and the mnt
complex were prepared. The acid species are more appropriate for
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 6940–6947 | 6941
©

View Article Online
Fig. 3 HOMO (left) and LUMO (right) isosurfaces for [Pt(4-CO2-
CH3-py)2(mnt)]. Individual atom contributions can be found in Table S4.†
Fig. 2 Cyclic voltammetry of [Pt(4-CO2CH3-py)2mnt] 3b in 0.1 M
TBABF4–DMF at 298 K, Scan rate 0.1 V s-1.
in the calculation was observed regarding the distribution of
the LUMO over the pyridyl ligands. Two essentially degenerate
LUMO orbitals are determined (the energies of these two orbitals
are very close and within the limits of the calculation may be
regarded as degenerate), however, the extent to which these spread
over the two pyridyl ligands depends upon the angle between these.
Restricting the geometry of the pyridine ligands in 3b to a torsion
angle of 90◦ in relation to one another (similar to the torsion angle
seen in the crystal structure of 2b) caused each LUMO to be based
almost entirely on only one each of the pyridine ligands rather than
being spread over both (See Fig. 3). This is then in agreement with
the experimental observation by EPR that the unpaired electron
in the SOMO is based only on one pyridine ligand (see below).
TD-DFT calculations were carried out on 3b in the presence
of DMF. This method has previously been demonstrated to give
satisfactory results for [Pt(diimine)(dithiolate)] systems and is
therefore expected to be applicable also to [Pt(py)2(dithiolate)]
complexes.9,18,19 A summary of relevant calculated transitions for
3b can be found in Table S5† and the calculated transitions relative
to the observed absorption spectrum are shown in Fig. 4. It
should be noted that the inclusion of solvent in these calculations
was undertaken as it has been observed previously that TD-DFT
studies carried out in vacuum typically vastly underestimated the
energy of the lowest energy transition in related complexes.9
line plots of imax vs. (scan rate)1/2) reduction peaks at -1.10 V and
-1.25 V and a chemically-irreversible oxidation at 1.24 V. The
electrochemistry of the related platinum dichloride complex 2b
(Fig. S2†), also shows two reversible reductions at very similar
potential. These processes are therefore assigned as sequential
reductions based on the two substituted-pyridyl ligands. The
absence of any oxidation for 2b enables the oxidation at 1.24 V
for 3b to be assigned as based on the mnt ligand.
For a dye to be viable within the energetics of a TiO2 DSSC, the
first reduction must be sufficiently negative to inject an excited-
state electron into the titania conduction band and the process at
-1.10 V shows that this is the case for 3b, as well as the excited-
state electron being positioned on the pyridyl, appropriate for
rapid injection into the band via the carboxy groups that bind to
TiO2. The first oxidation of 3b is sufficiently positive to be rapidly
reduced by the I-/I3 redox electrolyte within the cell.
-
The analogous bipyridine complexes 4b and 5b also show two
electrochemically-reversible reductions, based on the bipyridine
ligand, and an irreversible oxidation based on the mnt ligand.8,9
The first reduction of the substituted pyridyl complex 3b is at a
more negative potential than the first reduction of the bipyridyl
complexes as there is less delocalisation within the pyridyl ligands
to stabilise the addition of an extra electron. It is also very striking
that the separation of the first and second reductions for the
bpy-complexes is much greater than for 2b or 3b, indicating the
substantial pairing energy required to add two electrons to one
orbital of the bpy in comparison with adding one electron to each
pyridyl-ligand of 2b or 3b. This is discussed further in subsequent
sections. The separation between the first oxidation and the first
reduction for 3b is 2.34 V compared to 2.04 V for 4b. This reflects
the poorer electron accepting ability of the pyridyl ligands and
means that the absorption wavelength of the pyridine complex is
expected to shift to shorter wavelength.
Hybrid DFT calculations and UV-vis spectroscopy
Hybrid DFT calculations on [Pt(4-CO2CH3-py)2(mnt)] 3b (Fig. 3,
Table S4†) show that the HOMO is based mostly on the dithiolate
ligand. Similar to 4b and 5b, this orbital has some mixed-
metal contribution to the orbital with 14.4% platinum character.
The LUMO is based on the substituted pyridine part of the
molecule, with only 3.2% platinum character, agreeing with the
cyclic voltammetry, UV-vis and EPR results (see below) that also
indicate the LUMO is pyridine-ligand based. Some complication
Fig. 4 Calculated TD-DFT results of 3b shown as bar chart (left axis)
relative to the observed absorption spectrum (right axis) in DMF.
Only four transitions with significant oscillator strength (>0.1)
were found and these correlate reasonably well with the spectrum
6942 | Dalton Trans., 2008, 6940–6947
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
of 3b, which has absorptions at 36 300 cm-1, 31 600 cm-1 and
25 700 cm-1. The lowest energy band for 3b does not occur in
either 1b or 2b, consistent with assignment as ligand-to-ligand
charge transfer (LLCT), analogous to the lowest energy transition
in the bpy species 4b and 5b. The calculated composition was
found to be 76% HOMO - LUMO + 2, which interestingly does
not involve the lowest energy degenerate LUMO orbitals on the
pyridyl ligands but instead the next-highest energy orbital, which is
also pyridyl based. The calculated energy is slightly overestimated,
by 611 cm-1.
The two higher energy peaks show a marked similarity to those
displayed by the dichloride analogue 2b (Figure S4†), which has
transitions at 35 600 cm-1 and at 30 200 cm-1. Although a p→p*
absorption at 36 500 cm-1 was observed for ligand 1b, it was
of lower intensity than the 36 300 cm-1 band of 3b, suggesting
the latter has some other character. This was confirmed by the
calculations on 3b showing this band to be charge transfer in
character from HOMO - 3 (49% Pt, 46% mnt-based) to LUMO
(80% py-based). The lower-energy process (31 600, 30 200 cm-1 for
3b and 2b, respectively) also has no equivalent transition in 1b and
is also shown to be charge-transfer in nature for 3b from HOMO -
1 (32% Pt, 52% S) to LUMO + 2.
The relative size of the oscillator strengths match the relative
intensities of the observed spectra well, taking into account
the presence of two significant transitions under the band at
36 300 cm-1.
The LLCT in the neutral complex is observed at the onset of
the visible region at 25 700 cm-1 (390 nm). This is a rather high
energy for use in a solar cell as dyes with absorption maximum of
around 550 nm are more typical. This reflects the combination of
the pyridyl ligands with a dithiolate ligand (mnt2-) that leads to a
very positive oxidation potential for the dyes and further tuning
of the dithiolate ligand will readily allow absorption at a more
optimised region of the visible spectrum.
Fig. 5 UV-vis-NIR of 3b in 0, -1, -2 and +1 oxidation states. Spectra
were run in 0.1M TBABF4–DMF at 233 K. Egen = 0 V (0), -1.3 V (-1),
-1.6 V (-2) and 1.5 V (+1). The spectrum was observed to stabilise at
-1.3 V before undergoing further change at -1.6 V indicating sequential
reduction to the mono and then the dianion.
very little interaction between them. This contrasts greatly with
4b2- where the lowest energy HOMO–LUMO transition of the
neutral species 4b is completely lost due to filling of the LUMO
orbital.
Upon oxidation to [Pt(4-CO2CH3-py)2(mnt)]+, there is an
increase in intensity at 36 300 cm-1, a decrease in intensity at
31 600 cm-1 and 25 700 cm-1 and a further peak at 18 500 cm-1
grows in. Despite the irreversibility of the oxidation in the
CV, at the reduced temperature of the spectroelectrochemistry
experiment this process was chemically reversible.
The in situ EPR study of [Pt(4-CO2CH3-py)2(mnt)]2- (Fig. 6)
shows coupling constants to the Pt(II) metal centre and the
N nuclei of 41 and 6.7 G, respectively. Coupling to other
ligand nuclei are unresolved due to the large linewidth required
to successfully simulate the spectrum. In comparison with the
analogous dichloride complex 2b2- (Fig. S6†), the coupling value
to the nitrogen atom is unchanged but the electron no longer
couples as strongly to the platinum centre, with coupling reduced
from 55.5 to 41 G (Table 2). This result confirms that the LUMO
is based on the pyridine moiety. As less coupling is seen to the
platinum centre than in the dichloride precursor, this suggests
that the orbital is more delocalised once the mnt ligand has
been attached. A reduction in coupling to the platinum from the
dichloride precursor (59.5 G) to the mnt complex (56.0 G) was
also seen with the bpy complex analogues,9 also suggesting greater
electronic delocalisation once the mnt ligand is attached. Coupling
to N was larger in 2b and 3b than in the uncomplexed species
1b (Table S3†) suggesting some polarisation of the spin density
toward the nitrogen atoms when complexed to the Pt centre.
Two key differences arise in the EPR spectra of the reduced
forms of 3b in comparison with 4b. Firstly, for 4b2- the signal
The neutral bipyridine analogue 4b shows similar peaks to
3b around the 30 000 cm-1 region, which are assigned to p→p*
transitions. There is also a LLCT transition at 18 900 cm-1, at a
lower energy than in the pyridine analogue, which fits well with the
larger oxidation–reduction energy gap seen in the electrochemistry
of 3b compared with 4b.
Spectroelectrochemistry and EPR
UV-vis-NIR spectroelectrochemistry (Fig. 5) was performed to
gain further information on the frontier orbitals of [Pt(4-CO2CH3-
py)2mnt]. This involved electrochemical reduction of this com-
pound to its mono- and di-reduced states and oxidation to its
mono-oxidised states. Comparison with related studies for 1b
and 2b (Fig. S3, S4†; Table S1, S2†) gave some insight into the
absorption processes.
Reduction to 3b1- and 3b2- retains the peak at 36 300 cm-1, while
the 31 600 cm-1 peak and lower energy LLCT band increase in
intensity and shift slightly to lower energy. These spectra show
marked similarities to that of the reduced dichloride precursor
2b1-/2- supporting the conclusion that the reduction electrons enter
the 4-CO2CH3-py based orbitals. Reducing from the mono- to
di-reduced compound, the spectrum simply shows an increase
in intensity of the bands rather than formation of new ones,
consistent with sequential reduction of the pyridyl ligands with
Table 2 EPR simulation parameters for 2b2-(spectrum shown in Fig. S6†)
and 3b2- in 0.1 M TBABF4 in DMF at 233 K. All hyperfine coupling
constants given in G. D = linewidth. g = Lande´ g-factor
Parameter
2b2-
3b2-
a
a
g
iso(Pt)/G
iso(N)/G
55.5
6.7
2.0003
5.75
41
6.725
2.0043
5.5
D
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 6940–6947 | 6943
©

View Article Online
determined, using transient-absorption spectroscopy, the charge-
recombination lifetimes for 3a and 5a using the same experimental
conditions to ensure the results would be comparable. This was
achieved by sensitisation of TiO2 using 3b and 5b hydrolysed in situ
and appended to nanocrystalline TiO2. Following excitation and
injection of excited state electrons into the TiO2 conduction band,
the charge recombination was followed by monitoring the decay at
990 nm where electrons in the TiO2 conduction bands are known to
absorb. The observed dynamics could be well reproduced by single
exponential functions with charge recombination time constants
of 30 and 40 ms for 3a and 5a, respectively (Fig. 7).
Fig. 7 Transient absorption data monitoring the charge recombination
kinetics for TiO2 film sensitised with [Pt(4-CO2CH3-py)2(mnt)] (3a, dotted
black, left axis) and [Pt{3,3¢-(CO2CH2CH3)2–2,2¢-bpy}(mnt)] (5a, dotted
grey, right axis). The solid lines are single exponential fits of the data. The
excitation wavelengths are 450 and 530 nm for 3a and 5a, respectively.
At those wavelengths, the amount of photons absorbed by each dye was
comparable. The probe wavelength is 990 nm in both cases.
These lifetimes are of the same order of magnitude and very
similar in absolute value. This demonstrates that substituting the
twisted, 3,3¢-substituted bipyridine ligand with two completely-
separated pyridine ligands has retained the slow recombination
kinetics seen in the 3,3¢-substituted ligand. It should be noted that
the very-commonly used ligand in dye-sensitised solar cells is the
4,4¢-(CO2H)-bpy ligand present is 4a, which has already shown
faster charge recombination than 5a and hence also faster than
3a. Thus, we have established that the pyridyl complex displays
superior charge-recombination kinetics to the analogous complex
with this common ligand. In addition, we have also provided some
additional clarity for the superior performance of 5a over 4a where
the twist in the bpy ligand and associated loss of delocalisation
appears to play a key role.
Fig. 6 EPR spectrum and simulation of 3b2-. Spectrum was run in 0.1M
TBABF4–DMF at 233K. Egen = -1.4 V.
collapses when the second electron is added as it enters the same
orbital as the first reduction electron and gives diamagnetic proper-
ties. Holding the pyridine analogue 3b2- at very negative potentials
however does not result in a decrease in signal suggesting that
electrons enter two different orbitals (one on each pyridine ligand)
and can not pair up. Secondly, for 4b1- the electron couples to two
nitrogen atoms whereas coupling to only one nitrogen atom is seen
for 3b1-/2-, again demonstrating the confinement of the radical to
one pyridyl ligand for the latter.
Kinetics of electron transfer on TiO2
Conclusions
One of the goals for the synthesis of this class of complex was
as a model system to assess the potential of pyridyl dyes in dye-
sensitised solar cells. Our previous observations on 4a and 5a
demonstrated8,9 that the twist in the bipyridyl for 5a leads to
slower charge recombination and better solar cell performance.
We wished therefore to determine whether complete separation
of the two rings into separate pyridyl ligands would reproduce
the favourable charge-recombination kinetics of 5a. We therefore
This study has demonstrated the first electronic characterisation
of a complex of type [Pt(py)2(dithiolate)], analogues of the much-
studied [Pt(bpy)(dithiolate)] family of complexes. Electrochemical,
spectroelectrochemical, EPR and computational results all consis-
tently showed the lack of communication between the two pyridyl
rings and that the di-reduced species involves one electron on each
pyridyl. This contrasts greatly with the bpy-analogues and is the
6944 | Dalton Trans., 2008, 6940–6947
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
+
principal electronic difference between the two series of complexes.
The poorer electron accepting ability of the pyridyl leads to blue-
shifting of the lowest energy transition of the complex, which was
mainly of HOMO - LUMO + 2 character, and a shift of the first
reduction potential to more negative value.
at 473 {M - py} ; Anal. calcd for PtS2N4O4C18H14·2H2O: C 33.48,
H 2.17, N 8.68. Found: C 33.69, H 1.96, N 8.84. 1H NMR
(250 MHz, CDCl3) d/ppm: 8.7 (4H, d), 7.8 (4H, d), 3.9 (6H, s).
-1
=
∫
IR (KBr, n/cm ): 1730 (s) (C O stretch), 2203 (s) (C N stretch).
In the context of solar cell dyes, we found that replacing
a bipyridine ligand substituted in the 3,3¢-positions in a solar
cell dye with two pyridines is likely to lead to comparable
charge-separated lifetime for injected electrons in TiO2. Since we
have already demonstrated superior charge-separated lifetime for
3,3¢-(CO2H)2-bpy containing dyes on TiO2 compared with the
ubiquitous 4,4¢-(CO2H)2-bpy containing dyes, the current work
also demonstrates that pyridyl-containing dyes may share these
advantages. Subsequent work will involve the application of these
concepts to Ru-complex dyes that typically give higher efficiencies
in DSSC. It is interesting to note that Ru complexes with pyridyl
ligands have also been shown, in some cases, to give higher absorp-
tion coefficients than those with analogous bpy ligands.15 Taken
together with the possibility of favourable charge-recombination
kinetics, pyridyl ligands may deserve greater consideration in dyes
for DSSC than they have received until now.
Synthesis of [Pt(4-CO2H-py)2(mnt)] (3a)
[PtCl2(4-CO2H-py)2] (119 mg, 0.20 mmol) dissolved in dry DCM
(50 mL) and Na2(mnt) (43 mg, 0.23 mmol) in dry MeOH (20 mL)
were mixed together and stirred at room temperature for 2 h
under N2(g). The solvent was reduced to 20% in vacuo (with no
heating), isopropyl alcohol (50 mL) was added and the mixture
was stored in the freezer until a brown precipitate formed. This
was collected using filtration and dried under vacuum. Anal. calcd
for PtS2N4O4C18H10: C 33.05, H 1.72, N 9.64. Found: C 32.86, H
1
1.83, N 9.43. H NMR (250 MHz, DMSO-d6) d/ppm: 7.5 (4H,
d), 8.5 (4H, d).
X-Ray crystallography
Yellow crystals of [PtCl2(4-CO2CH3-py)2] were grown by slow
evaporation of a DCM–heptane solution. Crystal information,
data collection and refinement parameters are summarized in
Table 3. Single-crystal X-ray structure determination was carried
out using a Smart APEX CCD diffractometer equipped with
an Oxford Cryosystems low-temperature device with MoKa
radiation for data collection. An absorption correction was applied
using the multi-scan procedure SADABS.21 The structure was
solved by Patterson methods (DIRDIF)22 and refined using
SHELXL.23 The H atoms attached to C10 were located in a
difference synthesis; the CH distances and HCH angles were fixed
at ideal values, but the methyl-torsion was refined.
Experimental
Methyl isonicotinate (4-CO2CH3-py), purchased from Sigma-
Aldrich/Fluka and K2PtCl4 provided by Johnson Matthey were
used as received. Na2(mnt) was prepared as previously reported.20
Synthesis of [PtCl2(4-CO2CH3-py)2] (2b)
K2[PtCl4] (198.4 mg, 0.478 mmol) and 4-CO2CH3-py (130.3 mg,
0.951 mmol) were refluxed in deionised water (50 mL) for one hour.
The yellow precipitate was filtered, washed with diethyl ether and
then dried in a vacuum. The solid was recrystallised from a hot
saturated solution of DMF. Yield: 82% (212 mg, 0.392 mmol);
Other experimental information
All UV-vis spectra were recorded on a Perkin-Elmer Lambda
9 spectrophotometer controlled by a Datalink PC, running
UV/Winlab software.
+
+
MS (FABMS), m/z: 540 {M} also peaks at 505 {M - Cl} , 468
+
{M - 2Cl} ; Anal. calcd for PtCl2N2O4C14H14: C 31.11, H 2.59, N
5.19. Found C 31.02, H 2.61, N 5.02. 1H NMR (250 MHz, CDCl3)
d/ppm: 4.0 (6H, s), 7.8 (4H, d), 8.9 (4H, d). IR (KBr, n/cm-1):
Table 3 Crystallographic data for [PtCl2(4-CO2CH3-py)2]
=
1729 (s) (C O stretch).
Chemical formula
C14H14Cl2N2O4 Pt
Synthesis of [PtCl2(4-CO2H-py)2] (2a)
FW/g mol-1
T/K
540.26
150
Monoclinic
C2/c
13.0562(3)
8.2030(2)
15.3120(3)
97.4710(10)
1625.99(6)
6769
K2[PtCl4] (203.4 mg, 0.49 mmol) and 4-CO2H-py (114.9 mg,
0.934 mmol) were refluxed in deionised water (50 mL) for one hour.
The water was then removed in vacuo to leave a yellow solid and
recrystallised from a hot saturated solution of DMF. Yield: 63%
Lattice type
Space group
˚
a/A
˚
b/A
+
˚
(158 mg, 0.308 mmol); MS (FABMS), m/z: 512 {M} also peaks
c/A
ß/◦
V/A
+
+
at 476 {M - Cl} , 443 {M - 2Cl} . 1H NMR (250 MHz, CDCl3)
3
˚
d/ppm: 8.1 (4H, d), 9.0 (4H, d).
No. reflections for cell
No. independent reflections
2175
0.0313
30
Synthesis of [Pt(4-CO2CH3-py)2(mnt)] (3b)
Rint
◦
2qmax.
/
Z
4
[PtCl2(4-CO2CH3-py)2] (120 mg, 0.222 mmol) dissolved in dry
DCM (50 mL) and Na2(mnt) (41.3 mg, 0.222 mmol) in dry MeOH
(20 mL) were mixed together and stirred at room temperature for
2 h under N2(g). The solvent was reduced to 20% in vacuo (with no
heating) and the remaining solution was left overnight in the fridge.
The resulting precipitate was collected using filtration. Yield: 41%
Dc/Mg m-3
2.207
8.978
2175
106
0.0182
0.0431
1.122
m/mm-1
Reflections collected
Parameters
R1 [F > 4s(F)]
wR2
GOF
+
(55.5 mg, 0.091 mmol); MS (FABMS), m/z: 610 {M} also peaks
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 6940–6947 | 6945
©

View Article Online
Electrochemical studies were carried out using a DELL GX110
PC with a General Purpose Electrochemical System (GPES),
version 4.8, software connected to an autolab system containing a
PGSTAT 30 or Type III potentiostat. The techniques used a three-
electrode configuration, with a 0.5 mm diameter Pt disc working
electrode, a Pt rod counter electrode and an Ag/AgCl (saturated
KCl) reference electrode against which the ferrocenium/ferrocene
couple was measured to be +0.55 V. The supporting electrolyte
was 0.1 M tetrabutylammonium tetrafluoroborate (TBABF4).
OTTLE (Optically Transparent Thin Layer Electrode) measure-
ments were taken using a quartz cell of 0.5 mm, a Pt/Rh gauze
working electrode, an Ag/AgCl reference electrode and a Pt wire
counter electrode.24 UV-vis spectra were recorded on a Perkin-
Elmer Lambda 9 spectrophotometer, controlled by a Datalink PC,
running UV/Winlab software. Measurements on samples were
carried out at 233 K in DMF. TBABF4 (0.1 M) was used as the
supporting electrolyte in all cases.
Electron paramagnetic resonance spectra were taken using a
flat cell, a Pt/Rh gauze working electrode, an Ag/AgCl reference
electrode and a Pt wire counter electrode and generated using
a BAS CV-27 voltammograph. Spectra were recorded on an X-
band Bruker ER200D-SCR spectrometer connected to a Datalink
486DX PC with EPR Acquistion System, version 2.42 software.
The temperature was controlled by a Bruker ER4111 VT variable
temperature unit. All g values were corrected to 2,2¢-diphenyl-1-
picrylhydrazyl with gliteratue = 2.0036 0.0002.25 Measurements on
the samples were carried out at 233 K in DMF. TBABF4 (0.1 M)
was used as the supporting electrolyte in all cases.
filter to improve signal-to-noise ratio (Costronics Electronics).
The amplified signal was collected with a digital oscilloscope
(Tektronix, TDS 220), which was synchronized with a trigger
signal of the laser pulse from a photodiode (Thorlabs Inc.,
DET210). To reduce stray light, scattered light and emission from
the sample, two monochromators and appropriate optical cut-
off filters were placed before and after the sample. Owing to the
amplification and noise reduction system, the detectable change
of absorbance was as small as 10-5–10-6.
Acknowledgements
We thank Donald Robertson for providing Na2mnt and Johnson
Matthey for the supply of K2[PtCl4] for use in this study. Financial
support from EPSRC under the Supergen Excitonic Solar Cells
project is gratefully acknowledged. This work has made use of
the resources provided by the EaStCHEM Research Computing
Facility. (http://www.eastchem.ac.uk/rcf). This facility is partially
supported by the eDIKT initiative (http://www.edikt.org).
References
1 (a) S. D. Cummings and R. Eisenberg, J. Am. Chem. Soc., 1996, 118,
1949; (b) J. S. Pap, F. L. Benedito, E. Bothe, E. Bill, S. D. George, T.
Weyhermueller and K. Wieghardt, Inorg. Chem., 2007, 46, 4187.
2 J. Zhang, P. Du, J. Schneider, P. Jarosz and R. Eisenberg, J. Am. Chem.
Soc., 2007, 129, 7726.
3 C. Makedonas and C. A. Mitsopoulou, Eur. J. Inorg. Chem., 2006, 590.
4 J. A. Weinstein, M. T. Tierney, E. S. Davies, K. Base, A. A. Robeiro
and M. W. Grinstaff, Inorg. Chem., 2006, 45, 4544.
5 B. O’Regan and M. Gra¨tzel, Nature, 1991, 353, 737; M. Gra¨tzel, Nature,
2001, 414, 338; N. Robertson, Angew. Chem., Int. Ed., 2006, 45, 2338.
6 A. Islam, H. Sugihara, K. Hara, L. P. Singh, R. Katoh, M. Yanagida,
Y. Takahaski, S. Murata and H. Arakawa, New J. Chem., 2000, 24, 343.
7 A. Islam, H. Sugihara, K. Hara, L. P. Singh, R. Katoh, M. Yanagida, Y.
Takahaski, S. Murata and H. Arakawa, Inorg. Chem., 2001, 40, 5371.
8 E. A. M. Geary, L. J. Yellowlees, L. A. Jack, I. D. H. Oswald, S. Parsons,
N. Hirata, J. R. Durrant and N. Robertson, Inorg. Chem., 2005, 44, 242.
9 E. A. M. Geary, K. L. McCall, A. Turner, P. R. Murray, E. J. L.
McInnes, L. J. Yellowlees and N. Robertson, Dalton Trans., 2008, 3701.
10 C. Yang, J. Qin, Q. Yang, J. Si, S. Wang, Y. Wang, P. Ye and C. Ye,
Synth. Mater., 2001, 121, 1491.
DFT calculations were run using Gaussian 03, Revision D.01,26
The images were generated using Arguslab.27 The 6–31G* basis
set was used for the C, N, H, O, Cl and S atoms and the Hay-Wadt
VDZ (n + 1) ECP basis sets for the Pt(II) centre.28
TiO2 film-covered slides prepared using a previously reported
method29 were treated in the furnace at 300 K for 30 min. After
allowing to cool to between 50–100 ◦C the slides were introduced
into an OH bath (pH 11) and left for an hour. The slides were
removed from the basic solution and put to dry in the oven
at 110 ◦C for 30 min before being introduced into the 0.23 M
sensitizing solutions of the dye in MeCN. The prior base treatment
of the slides led to the in situ hydrolysis and subsequent attachment
of the sensitiser dye. The films were removed from the sensitizing
solutions after 48 h and time-resolved kinetics were tested using
these sensitised slides covered with a redox inactive electrolyte
(0.25 M LiClO4 in propylene carbonate).
11 F. Liu and G. J. Meyer, Inorg. Chem., 2003, 42, 7351.
12 F. Liu and G. J. Meyer, Inorg. Chem., 2005, 44, 9305.
13 F. Gajardo, A. M. Leiva, B. Loeb, A. Delgadillo, J. R. Stromberg and
G. J. Meyer, Inorg. Chim. Acta, 2008, 361, 613.
14 P. Qu and G. J. Meyer, Langmuir, 2001, 17, 6720.
15 M. Haga, E. S. Dodsworth and A. B. P. Lever, Inorg. Chem., 1986, 25,
447.
16 Cambridge Structural Database, Chem. Des. Automat. News, 1993, 8,
Transient absorption decays were measured using the “flash
photolysis” technique. The sample was excited with a dye laser
(Photon Technology International Inc., GL-301) pumped by
a nitrogen laser (Photon Technology International Inc., GL-
3300). The excitation wavelength was 450 nm for 3a dye or
530 nm for 5a dye. The excitation wavelengths were chosen so
the amount of photons absorbed by each dye was comparable.
The pulse width was 800 ps, the fluence was about 50 J cm-2
at both excitation wavelengths and the repetition frequency was
1 Hz. A 100 W tungsten–halogen lamp (Bentham, IL1) with a
stabilized power supply (Bentham, 605) was used as a probe light
source. The probe light passing through the sample was detected
with a silicon photodiode (Hamamatsu Photonics, S1722–01).
The signal from the photodiode was pre-amplified and sent
to the main amplification system with an electronic band-pass
31.
17 M. Carnalli, F. Caruso and L. Zambonelli, Cryst. Struct. Commun.,
1980, 9, 721.
18 C. J. Adams, N. Fey, M. Parfitt, S. J. A. Pope and J. A. Weinstein, Dalton
Trans., 2007, 4446.
19 C. Makedonas and C. A. Mitsopoulou, Eur. J. Inorg. Chem., 2006,
2460; C. Makedonas, C. A. Mitsopoulou, F. J. Lahoz and A. I. Balana,
Inorg. Chem., 2003, 42, 8853; C. Makedonas and C. A. Mitsopoulou,
Inorg. Chim. Acta, 2007, 360, 3997.
20 G. Markl and R. Vybiral, Tetrahedron Lett., 1989, 30, 2903.
21 G. M. Sheldrick, SADABS, Program for area detector adsorption
correction, Institute for Inorganic Chemistry, University of Go¨ttingen,
Germany, 1996.
22 P. T. Beurskens, G. Beurskens, R. de Gelder, S. Garcia-Granda, R. O.
Gould, R. Israel and J. M. M. Smits, Crystallography Laboratory,
University of Nijmegen, The Netherlands, 1999.
23 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112.
6946 | Dalton Trans., 2008, 6940–6947
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
24 S. A. McGregor, E. J. L. McInnes, R. J. Sorbie and L. J. Yel-
lowlees, Molecular Electrochemistry of Inorganic, Bioinorganic and
Organometallic Compounds, Kluwer, Dordrecht, The Netherlands,
1992.
C. Pomelli, J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. G. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision D.01), Gaussian, Inc., Wallingford, CT, 2004.
25 S. A. Al’tshuler and B. M. Kozyrev, Electronic Paramagnetic Resonance,
Academic Press, New York, 1964, p 503.
26 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C.
Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.
Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
27 M. A. Thompson, ArgusLab 4.0.1, Planaria Software LLC, Seattle,
WA, http://www.arguslab.com.
28 P. J. Hay and W. R. Wadt, J. Phys. Chem., 1985, 82, 299.
29 A. N. M. Green, E. Palomares, S. A. Haque, J. M. Kroon and J. R.
Durrant, J. Phys. Chem. B, 2005, 109, 12525.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 6940–6947 | 6947
©