Co(III) Complexes as p-Dopants in Solid-State Dye-Sensitized Solar Cells

Article abstract of DOI:10.1021/cm400796u

Following our recent work on the use of Co(III) complexes as p-type dopants for triarylamine-based organic hole-conductors, we herein report on two new Co(III) complexes for doping applications. With the aim of ameliorating the dopant's suitability for its use in solid-state dye-sensitized solar cells, we show how the properties of the dopant can be easily adjusted by a slight modification of the molecular structure. We prove the eligibility of the two new dopants by characterizing their optical and electrochemical properties and give evidence that both of them can be used to oxidize the molecular hole-transporter spiro-MeOTAD. Finally, we fabricate high-performance solid-state dye-sensitized solar cells using a state-of-the-art metal-free organic sensitizer in order to elucidate the influence of the type of dopant on device performance.

Full text of DOI:10.1021/cm400796u

Subscriber access provided by FORDHAM UNIVERSITY
Article
Co(III) Complexes as p-Dopants in Solid-State Dye-sensitized Solar Cells
Julian Burschka, Florian Kessler, Mohammad Khaja Nazeeruddin, and Michael Grätzel
Chem. Mater., Just Accepted Manuscript • DOI: 10.1021/cm400796u • Publication Date (Web): 25 Jun 2013
Downloaded from http://pubs.acs.org on July 10, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Chemistry of Materials is published by the American Chemical Society. 1155 Sixteenth
Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 7
Chemistry of Materials
1
2
3
4
5
6
7
8
9
Co(III) Complexes as p-Dopants in Solid-State Dye-sensitized Solar
Cells
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Julian Burschka, Florian Kessler†, Mohammad K. Nazeeruddin, and Michael Grätzel*
Laboratoire de Photoniques et Interfaces, Institut des Sciences et Ingénierie Chimiques, École Polytechnique Fédérale de
Lausanne (EPFL), Station 6, 1015 Lausanne, Switzerland.
ABSTRACT: Following our recent work on the use of Co(III) complexes as pꢀtype dopants for triarylamineꢀbased organic holeꢀ
conductors, we herein report on two new Co(III) complexes for doping applications. With the aim of ameliorating the dopant’s
suitability for its use in solidꢀstate dyeꢀsensitized solar cells, we show how the properties of the dopant can be easily adjusted by a
slight modification of the molecular structure. We prove the eligibility of the two new dopants by characterizing their optical and
electrochemical properties and give evidence that both of them can be used to oxidize the molecular holeꢀtransporter spiroꢀ
MeOTAD. Finally, we fabricate highꢀperformance solidꢀstate dyeꢀsensitized solar cells using a stateꢀofꢀtheꢀart metalꢀfree organic
sensitizer in order to elucidate the influence of the type of dopant on device performance.
Keywords: p-type doping, organic semiconductor, cobalt(III) complex, spiro-MeOTAD, dye-sensitized solar cell
used in conjunction with nꢀtype silicon nanowires(5) and more
1. Introduction
Electronic doping is a powerful tool to carefully control the
recently with organicꢀinorganic hybrid perovskite sensitizers
(6,7). A typical ssDSC is composed of a sensitized, mesoꢀ
type and density of charge carriers in both organic and inorꢀ
structured wide bandgap metalꢀoxide semiconductor that is
ganic semiconducting materials.(1) Doping of organic semiconꢀ
infiltrated with a molecular or polymeric organic holeꢀ
ductors is usually based on a chargeꢀtransfer reaction between
transporter, usually by solution processing (e.g. spinꢀcoating).
the organic host and the dopant, the latter being either a donor
In this device photoꢀexcitation of the sensitizer and subsequent
(nꢀtype doping) or acceptor (pꢀtype doping) species that leaves
ultrafast electron injection into the conduction band of the
part of the host molecules in a reduced or oxidized state. The
metalꢀoxide semiconductor is followed by holeꢀtransfer from
key parameter for efficient doping is therefore the position of
the oxidized sensitizer to the organic holeꢀtransporter. Charge
the ionization potential or electron affinity of the dopant with
transport of both the electron and the hole through the two
respect to the host’s energy levels. However, depending on the
bicontinuous phases and charge migration through the external
application, other features such as ease of processing might
circuit complete the photovoltaic operation of the cell. Even
become equally important, finally rendering a dopant suitable
though molecular pꢀtype doping of spiroꢀMeOTAD in such a
for only a specific type of device architecture and/or fabricaꢀ
device configuration has been previously introduced,(8) it is
tion process. Thus, materials such as MoO3 or WO3, both exꢀ
usually not essential due to device fabrication under ambient
tensively studied pꢀtype dopants,(2) are remarkably performing
conditions and the facile oneꢀelectron oxidation of spiroꢀ
when thermally evaporated onto or together with the organic
MeOTAD by oxygen under illumination. Nevertheless, conꢀ
semiconductor but suffer from lack of solubility when solution
trolled doping via the introduction of molecular pꢀdopants is
processing is required.
expected to be favorable in terms of controllability and reproꢀ
Recently we reported on a new class of pꢀtype dopants based
ducibility of photovoltaic results and is expected to be indisꢀ
pensable if the devices are fabricated in an inert atmosphere.(9)
on cobalt(III) complexes that have been designed to be proꢀ
cessed from solution and are particularly promising due to
In our recent work we showed that the strategy of using moꢀ
their ambient stability, ease of preparation and the possibility
lecular pꢀdopants is indeed beneficial for controlling the
to tune their chemical and physical properties.(3) We employed
tris(2ꢀ(1Hꢀpyrazolꢀ1ꢀyl)pyridine)cobalt(III)
phosphate) (FK102, Figure 1c), as pꢀtype dopant and showed
that it can be efficiently used to dope triarylamineꢀbased holeꢀ
charge transport properties of spiroꢀMeOTAD in ssDSCs and
tris(hexafluoroꢀ
were able to directly relate the conductivity of the doped holeꢀ
transporter to device performance. In this report we investigate
two new Co(III) complexes, namely tris(2ꢀ(1Hꢀpyrazolꢀ1ꢀyl)ꢀ
4ꢀtertꢀbutylpyridine)cobalt(III)
tris(bis(trifluoromethylsulfonyl)imide)) and bis(2,6ꢀ di(1Hꢀ
pyrazolꢀ1ꢀyl)pyridine)cobalt(III)
methylsulfonyl)imide)), herein denoted as FK209 and FK269
(Figure 1c), respectively, and compare them to FK102. These
new complexes were designed to develop further the previousꢀ
conductors
such
as
2,2’,7,7’ꢀtetrakis(N,Nꢀdiꢀpꢀ
methoxyphenylꢀamine)ꢀ9,9’ꢀspirobifluorene (spiroꢀMeOTAD,
Figure S1a). SpiroꢀMeOTAD is a molecular holeꢀtransporter
that is commonly used in solidꢀstate dyeꢀsensitized solar cells
(ssDSCs)(4), presently one of the most promising thirdꢀ
generation photovoltaic (PV) technologies, but has also been
tris(bis(trifluoroꢀ
1
ACS Paragon Plus Environment

Chemistry of Materials
Page 2 of 7
ly proposed concept of using Co(III) complexes as dopants.
Synthesis
dine)cobalt(III)
of
Bis(2,6-di(1H-pyrazol-1-yl)pyri-
1
2
3
4
5
6
7
8
We demonstrate that the properties of the dopant can be easily
adjusted by modifying the organic ligands and/or the counterꢀ
ion. The complexes were characterized by optical as well as
electrochemical techniques in order to scrutinize their suitabilꢀ
ity for doping applications. The degree of oneꢀelectron oxidaꢀ
tion of spiroꢀMeOTAD was determined by UV/Vis absorption
spectroscopy. Finally, highꢀperformance ssDSCs employing a
stateꢀofꢀtheꢀart organic sensitizer were fabricated in order to
probe the effect of the type of dopant on photovoltaic device
performance.
tris(bis(trifluoromethylsulfonyl)imide))
(FK269). 0.26 g (1.23 mmol, 2 eq) of 2,6ꢀdi(1Hꢀpyrazolꢀ1ꢀ
yl)pyridine was suspended in 30 mL of acetonitrile. Then 0.21
g (0.86 mmol, 1.4 eq) of CoCl2*6H2O was added and the mixꢀ
ture was stirred at room temperature for 10 minutes. Upon
adding a solution of 0.47g (0.86 mmol, 1.4 eq) of ammonium
cerium nitrate in acetonitrile the color of the solution changed
and a pale orange precipitate formed. Then a concentrated
solution of 0.62 g (2.15 mmol, 3.5 eq) of LiTFSI in water was
added and the mixture was stirred at room temperature for 3
hours. The solid cobalt complex was collected on a sintered
glassꢀfrit, washed with water and then dried in vacuum to obꢀ
tain the pure product as a pale orange solid. Yield: 0.75 g (0.60
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2. Experimental Section
1
mmol, 70%). H NMR (400 MHz, DMSOꢀD6): δ 9.65–9.53
(m, 4H, ArH), 9.34–9.24 (m, 2H, ArH), 8.86–8.71 (m, 4H,
ArH), 7.90–7.85 (m, 4H, ArH), 6.97–6.87 (m, 4H, ArH).
Materials. All chemicals were purchased from Sigmaꢀ
Aldrich (Switzerland) or Acros Organics (Belgium) and used
as received unless stated otherwise. 2,2’,7,7’ꢀtetrakis(N,Nꢀdiꢀ
Synthesis
of
Bis(2,6-di(1H-pyrazol-1-yl)pyri-
pꢀmethoxyphenylꢀamine)ꢀ9,9’ꢀ
spirobifluorene
(spiroꢀ
dine)cobalt(II) bis(bis(trifluoromethylsulfonyl)imide)). The
Co(II) analog of FK269 was synthesized using the same proꢀ
cedure as described above, but omitting the oxidation step. H
NMR (200 MHz, AcetoneꢀD6) δ 76.36 (s, 4H, ArH), 75.39 (s,
4H, ArH), 60.38 (s, 4H, ArH), 56.62 (s, 4H, ArH), 14.73 (s,
2H, ArH).
MeOTAD) was obtained from Merck KGaA (Germany) and
used as received. The synthesis of Y123 sensitizer (Figure
S1b) followed our published method.(10) 4ꢀtertꢀbutylpyridine
was purified by distillation before use. Tris(1ꢀ(pyridinꢀ2ꢀyl)ꢀ
1Hꢀpyrazol)cobalt(III) tris(hexafluorophosꢀphate) (FK102) and
the corresponding Co(II) analog tris(1ꢀ(pyridinꢀ2ꢀyl)ꢀ1Hꢀ
pyrazol)cobalt(II) bis(hexafluorophosphate) were synthesized
as previously reported.(3) The ligand 2ꢀ(1Hꢀpyrazolꢀ1ꢀyl)ꢀ4ꢀ
tertꢀbutylpyridine was synthesized according to a published
procedure.(11) 2,6ꢀdi(1Hꢀpyrazolꢀ1ꢀyl)pyridine was synthesized
in a similar manner reacting 2,6ꢀdichloropyridine with two
equivalents of pyrazole.
1
UV/Vis Absorption Spectroscopy. UV/Vis absorption specꢀ
tra were recorded on a Varian Cary 5 spectrophotometer using
a 10 mm path length quartz cuvette. SpiroꢀMeOTAD was disꢀ
solved in chlorobenzene. The dopants were preꢀdissolved in
acetonitrile
MeOTAD/chlorobezene solution.
and
then
added
to
the
spiroꢀ
Synthesis of Tris(2-(1H-pyrazol-1-yl)-4-tert-butylpyri-
Electrochemical Measurements. The electrochemical charꢀ
acterization was performed on an Autolab Potentiostat
(PGSTAT30, Metrohm) in a classical threeꢀelectrode configuꢀ
ration. Platinum wires were used as counter and reference elecꢀ
trode. A platinum wire or gold disc was used as working elecꢀ
trode. The solvent was dichloromethane containing 0.1M tetꢀ
ramethylammonium hexafluorophosphate. The measured poꢀ
tential was internally referenced versus the Ferroceniꢀ
um/Ferrocene standard. DPV and CV measurements were
performed at a scan rate of 0.1 V sꢀ1. Redox potentials were
estimated from the DPV polarograms calculating the average
of values obtained from the forward and backward scan.
dine)cobalt(III)
tris(bis(trifluoromethylsulfonyl)imide))
(FK209). 3.45 g (17.1 mmol, 3 eq) of 2ꢀ(1Hꢀpyrazolꢀ1ꢀyl)ꢀ4ꢀ
tertꢀbutylpyridine was suspended in a 2:1 mixture of water
(120 mL) and methanol (60 mL) and heated to 70 °C. Then
1.36 g (5.7 mmol, 1 eq) of CoCl2ꢁ6H2O was added and the
mixture was stirred at the same temperature for 10 minutes. 10
mL of H2O2 (30%) and 10 mL of HCL (25 %) were added to
oxidize the cobalt and the mixture was further heated to 70 °C
for 2 h. Then a concentrated solution of 8.18 g (28.5 mmol, 5
eq) lithium bis(trifluoromethylsulfonyl)imide in water was
added slowly at 50 °C to precipitate the product. The mixture
was allowed to cool to room temperature and the orange solid
was collected on a sintered glassꢀfrit and washed with water.
The solid was dried in air and then in vacuo to obtain the pure
product as an orange solid. Yield: 7.76 g (5.16 mmol, 91%).
1H NMR (400 MHz, AcetoneꢀD6): δ 9.87–9.64 (m, 3H, ArH),
8.79–8.65 (m, 3H, ArH), 8.00–7.79 (m, 6H, ArH), 7.78–7.62
(m, 3H, ArH), 7.37–7.17 (m, 3H, ArH), 1.72–1.15 (m, 27H,
tBu).
Device Fabrication and Characterization. Photovoltaic
devices were fabricated and characterized following recently
published procedures.(3) The dopants were predissolved in
acetonitrile at a concentration of 20 mg mLꢀ1 and added to the
spiroꢀMeOTAD solution prior to spinꢀcoating. JVꢀcurves preꢀ
sented herein have been measured directly after device fabricaꢀ
tion without any additional lightꢀsoaking or ageing.
Synthesis of Tris(2-(1H-pyrazol-1-yl)-4-tert-butylpyri-
dine)cobalt(II) bis(bis(trifluoromethylsulfonyl)imide)). The
Co(II) analog of FK209 was synthesized using the same proꢀ
cedure as described above, but omitting the oxidation step. No
peaks are visible in the 1H NMR (400 MHz, AcetoneꢀD6).
3. Results and Discussion
Although the earlier reported compound FK102 can be used
as an efficient pꢀdopant, we found its solubility to be one of
the major limiting factors, e.g. impeding the investigation of
high
2
ACS Paragon Plus Environment

Page 3 of 7
Chemistry of Materials
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Figure 1. A) Differential pulse voltammetry (DPV) and B) Cyclic Voltammetry (CV) measurements of the three different complexes in
CH2Cl2 solution containing the dopant, 0.1 M TBAPF6 and the Fc+/Fc standard as an internal reference. C) Molecular structures of FK102,
FK209 and FK269.
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
doping ratios. In order to achieve pꢀtype doping of the bulk
HTM, a deposition of the dopant together with the holeꢀ
conductor is required. Therefore the dopant should be soluble
in the spinꢀcoating formulation of the HTM, which comprises
spiroꢀMeOTAD, lithium bis(trifluoromethylsulfonyl)imide
(LiTFSI) and 4ꢀtertꢀbutylpyridine (TBP) in chlorobenzene. We
find that the solubility of the Co(III) complexes depends
strongly on the presence and concentration of the two latter
additives. The main reason for the relatively low solubility of
FK102 in organic solvent is its PF6ꢀ counterꢀion, whereas salts
of bis(trifluoromethylꢀsulfonyl)imide (TFSI) are usually much
more soluble. In addition to changing the counterꢀion, we also
introduced a solubilizing tertꢀbutyl group on the pyridineꢀ
pyrazole (pyꢀpz) ligand. The 2ꢀ(1Hꢀpyrazolꢀ1ꢀyl)ꢀ4ꢀtertꢀ
butylpyridine ligand has recently been successfully used for
improved solubility of cationic Ir(III) complexes.(11) The TFSIꢀ
salt of the Co(III) complex bearing this ligand is denoted as
FK209 (Figure 1c). Apart from the solubility, we increased the
oxidation potential of the dopant in order to maximize the
driving force for the doping reaction as well as making it suitꢀ
able for a larger number of host materials. This was done by
changing the aromatic ligand. Compared to pyridine (py), pyꢀ
razol (pz) is a weaker σꢀdonor as well as a stronger πꢀ
acceptor.(12) Therefore we replaced the three bidentate pyꢀpz
ligands by two tridentate pzꢀpyꢀpz ligands which is expected to
stabilize the metalꢀcentered HOMO level of the complex. The
TFSIꢀ salt of the corresponding Co(III) complex is denoted as
FK269 (Figure 1c). The Co(II) analog of FK269 has previousꢀ
ly been reported and characterized by Holland et al. and Ayers
et al.(13, 14)
All three complexes appear as orange/red solids and give
slightly orange colored solutions in organic solvent. UV/Vis
absorption spectra taken from CH3CN solutions are shown in
Figure S2. All complexes show strong absorption bands in the
UV region below 350 nm arising from ligand centered πꢀπ*
transitions. The orange color, however, results from the very
weak dꢀd transition that lies in the region between 400ꢀ600
nm, having a molar extinction coefficient of less than 30 and
600 L molꢀ1 cmꢀ1 in case of the Co(II) and Co(III) species, reꢀ
spectively. Both new complexes are therefore very suitable for
doping applications due to their visible transparency that
avoids filtering of incoming light by the dopant in the photoꢀ
voltaic device.
The complexes were characterized by differential pulse voltꢀ
ammetry (DPV) as well as cyclic voltammetry (CV) in order to
evaluate their redox potentials. Results are given in Figure
1a/b. All three complexes show oxidation/reduction waves in
dichloromethane that correspond to the Co(III)/Co(II) redox
couple situated in the region of 0.2ꢀ0.8 V vs. the ferroceniꢀ
um/ferrocene (Fc+/Fc) standard. The redox processes are irreꢀ
versible in the case of FK102 and FK269 and quasiꢀreversible
in the case of FK209. From the DPV measurements, we estiꢀ
mate the redox potentials of FK102, FK209 and FK269 to be
around 0.38, 0.32 and 0.58 V vs. Fc+/Fc, respectively. The
lower redox potential of FK209 compared to FK102 can be
rationalized by the inductive electronꢀdonating effect of the
tertꢀbutyl groups on the pyꢀpz ligand that destabilizes the
HOMO of the complex. In agreement with our predictions, the
change from a total number of 3 py/3 pz rings (FK102,
FK209) to 2 py/4 pz rings (FK269) coordinated to the Co(III)
metal center increases the redox potential, even though the
large shift of roughly 200 mV was unexpected. Considering
the first redox potential of spiroꢀMeOTAD at 0.02 V vs.
Fc+/Fc,(3) all three dopants should exhibit a driving force of
more than 300 mV sufficient for the complete one electron
oxidation.
All three Co(III) complexes compared herein are diamagnetꢀ
ic lowꢀspin complexes. The diamagnetism is confirmed by 1Hꢀ
NMR spectroscopy and the appearance of the peaks from the
aromatic ligands in the region <10ppm. Upon the reduction to
Co(II), all three complexes become paramagnetic in agreement
with a strong downfield shift (FK269) or complete disappearꢀ
1
ance (FK102 and FK209) of the HꢀNMR signals. Holland et
The oxidation of spiroꢀMeOTAD by the three Co(III) comꢀ
plexes was confirmed by UV/Vis absorption spectroscopy in
chlorobenzene solution. The gradual addition of any of the
three complexes to a spiroꢀMeOTAD solution causes a deꢀ
crease of the absorption band of spiroꢀMeOTAD at 389 nm
together with a rise of two new absorption bands at 523 and
al. found that the Co(II) analog of FK269 takes up a high spin
quartet configuration, which results from the weak splitting of
the metal d orbitals induced by the pyrazoleꢀbased ligands. A
high spin configuration was also confirmed by Saha et al. for a
Co(II) complex bearing a methylꢀsubstituted pyꢀpz ligand.(15)
3
ACS Paragon Plus Environment

Chemistry of Materials
Page 4 of 7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. AꢀC) Change of the UV/Vis absorption spectrum of a spiroꢀMeOTAD solution in chlorobenzene upon the gradual addition of
either of the three dopants A) FK102, B) FK209 or C) FK269. The change in coloration from black to red corresponds to increasing Co(III)
levels, with black being pure spiroꢀMeOTAD in chlorobenzene. Percentages given in the legend are the amount of Co(III) added to the
solution in mol% relative to spiroꢀMeOTAD. DꢀF) Plots of the amount of oxidized spiroꢀMeOTAD (mol%) calculated from the absorbance
at 523 nm using a molar extinction coefficient of 3.4 x 104 l molꢀ1 cmꢀ1 versus the amount of Co(III) added to the solution (mol%). Solid
lines are linear fits, dashed lines indicate a theoretically expected conversion of 100%.
689 nm corresponding to the spiroꢀMeOTAD radical cation
(Figure 2). We note that the addition of the Co(II) analog of
any of the complexes to a spiroꢀMeOTAD solution, does not
lead to the formation of the spiroꢀMeOTAD radical cation as
confirmed by UV/Vis absorption measurements. This proves
that the oxidation of spiroꢀMeOTAD is linked to the reduction
of Co(III) to Co(II) as expected. Monitoring the appearance of
the absorption peak at 523 nm or the disappearance of the peak
at 389 nm allows the calculation of a Co(III)ꢀtoꢀspiroꢀ
MeOTAD+ conversion yield. For the neutral spiroꢀMeOTAD
molecule we find an average molar extinction coefficient of
7.2 x 104 L molꢀ1 cmꢀ1, which is in good agreement with previꢀ
ously reported values.(8) From the disappearance of the peak at
389 nm we calculate conversion yields of 59%, 62% and 71%
for the three different oxidants FK102, FK209 and FK269,
respectively. However, this calculation might underestimate
the conversion yield as it does not take into account any possiꢀ
ble contribution from absorption by the reaction products, i.e.
the spiroꢀMeOTAD radical cation and the generated Co(II)
complex at this wavelength. Therefore monitoring the appearꢀ
ance of the absorption band at 523 nm is preferred even though
values for the extinction coefficient of spiroꢀMeOTAD+ reꢀ
ly. From these values one would expect a nearly quantitative
reaction contrary to our experimental findings. We attribute
the discrepancy to uncertainties in the estimation of the redox
potentials from the poorly reversible DPV polarograms as well
as the extinction coefficient of the spiroꢀMeOTAD cation radiꢀ
cal.
In order to confirm that all three materials are effective as pꢀ
dopants for organic holeꢀconductors, solidꢀstate dyeꢀsensitized
solar cells using the organic DꢀπꢀA sensitizer Y123 (Figure
S1b) and spiroꢀMeOTAD were fabricated. Current (J) Voltage
(V)ꢀcharacteristics measured at 100 mW cmꢀ2 AM1.5G simuꢀ
lated solar irradiance for devices containing the three different
dopants are shown in Figure 3a. In all three cases 2.0% of doꢀ
pant with respect to the molar concentration of spiroꢀ
MeOTAD was added to the spinꢀcoating formulation. Photoꢀ
voltaic parameters derived from the JVꢀcharacteristics are
summarized in Table 1.
For an undoped device that has been fabricated under identical
conditions, we find values of 878 mV, 9.1 mA cmꢀ2 and 0.29
for openꢀcircuit potential (VOC), shortꢀcircuit photocurrent
density (JSC) and fill factor (FF), respectively, corresponding to
a power conversion efficiency (PCE) of 2.3%.(3) In the presꢀ
ence of the Co(III) complexes the FF increases to 0.62, 0.69
and 0.69, the VOC to 896 mV, 925 mV and 918 mV and the JSC
to 9.9 mA cmꢀ2, 9.4 mA cmꢀ2 and 9.6 mA cmꢀ2 for FK102,
FK209 and FK269, respectively, boosting the PCE to 5.5%,
6.0% and 6.0% (Table 1). In general we would not expect the
type of dopant to have an influence on device performance
when the same doping level is maintained. In this regard, the
difference in PCE and mainly FF for devices doped with
FK102 compared to FK209/FK269 is somewhat unexpected
but is likely to arise from the difference in the nature of the
counterꢀion. On one hand, a strong influence on ssDSC perꢀ
formance has for example been reported for LiTFSI and
ported in the literature range from 2.7 to 4.1 x 104 L molꢀ1 cmꢀ
1 (9, 16, 8)
.
Using an average value of 3.4 x 104 L molꢀ1 cmꢀ1, we
estimate the conversion yield to be 65%, 68% and 78% for
FK102, FK209 and FK269, respectively. The results indicate
similar conversion yields for FK102 and FK209 and a signifiꢀ
cantly higher yield for FK269, a trend that can be rationalized
by the difference in driving force for the oneꢀelectron oxidaꢀ
tion of spiroꢀMeOTAD, which is similar for FK102 and
FK209, but much higher for FK269. Overall, our findings sugꢀ
gest that the oxidation of spiroꢀMeOTAD using the proposed
Co(III) complexes is an equilibrium reaction in solution, altꢀ
hough the equilibrium constant is likely to be shifted for the
doping of the solidꢀstate holeꢀconductor. From the offset in
redox potential we calculate equilibrium constants of 3.3 x 106,
2.7 x 105, 1.4 x 1010 for FK102, FK209 and FK269, respectiveꢀ
EMIM TFSI (EMIM = 1ꢀethylꢀ3ꢀmethylimidazolium).(17, 18)
A
more recent publication however attributes this effect to the
role of LiTFSI acting as a pꢀdopant in the presence of oxygen,
4
ACS Paragon Plus Environment

Page 5 of 7
Chemistry of Materials
its effect on chargeꢀtransport being negligible in the absence of
Table 1. Photovoltaic parameters of ssDSCs using dif-
ferent p-type dopants (AM 1.5 G, 100 mW cm-2).a
oxygen.(19) On the other hand, the greatly enhanced solubility
of FK209/FK269 compared to FK102 leads to a better disperꢀ
sion of the remaining Co(II) complex in the spiroꢀMeOTAD
matrix, enhancing charge transport within the HTM and avoidꢀ
ing the crystallization of the Co(II) salt during the solution
processing.
1
2
3
4
5
6
7
8
Blank
-
878b
9.1b
0.29b
2.3b
FK102
2.0%
FK209
2.0%
FK269
2.0%
FK209
10%
Doping Level
VOC (mV)
JSC (mA cm-2)
FF
896 ± 9
9.9 ± 0.1
925 ±3
9.4 ± 0.2
918 ± 5
9.6 ± 0.1
941 ± 7
9.1 ± 0.2
0.62 ± 0.02 0.69 ± 0.01 0.69 ± 0.01 0.73 ± 0.01
5.5 ± 0.2 6.0 ± 0.1 6.0 ± 0.1 6.2 ± 0.1
PCE (%)
a The average of three devices is given for each condition.
b Data taken from reference (3).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4. Conclusion
In conclusion we proposed two new Co(III) complexes as pꢀ
dopants for organic semiconductors to complement the previꢀ
ously reported FK102. Their redox potential and solubility was
adjusted by modifying the ligands and the counterꢀion, respecꢀ
tively. The high solubility of the TFSI salts enabled doping
levels of more than 10% for the new complexes,x underlining
their potential as a potent pꢀdopants for triarylamineꢀbased
holeꢀconductors in solidꢀstate dyeꢀsensitized solar cells.
ASSOCIATED CONTENT
Supporting Information. Molecular Structures of Y123 and spiꢀ
roꢀMeOTAD as well as additional UV/Vis absorption spectra.
This material is available free of charge via the Internet at
http://pubs.acs.org
AUTHOR INFORMATION
Corresponding Author
* michael.graetzel@epfl.ch
Present Addresses
† Siemens AG, Corporate Technology, CT RTC MAT MPV, Günꢀ
therꢀScharowsky Strasse 1, Dꢀ91058 Erlangen, Germany.
Figure 3. JVꢀCurves for photovoltaic devices based on the Y123
sensitizer and pꢀdoped spiroꢀMeOTAD, measured under 100 mW
cmꢀ2 simulated AM1.5G solar irradiance (solid line) and in the
dark (dashed line). A) JVꢀCurves of devices containing different
pꢀdopants FK102, FK209 and FK269 at a doping ratio of 2.0%. B)
JVꢀCurve of a device doped at 10% using FK209.
ACKNOWLEDGMENT
The authors thank Dr. Carole Grätzel for her assistance. This work
was supported by the Swiss National Science foundation. M.G.
thanks the Max Planck Society for a Max Planck Fellowship at the
MPI for Solid State Research in Stuttgart, Germany as well as the
European Research Council (ERC) for an Advanced Research
Grant (ARG 247404) funded under the “Mesolight” project.
To further support this hypothesis we also tested devices
with much higher doping levels. Doping with 10% FK209,
resulted in values of 941 mV, 9.1 mA cmꢀ2 and 0.73 for VOC,
JSC and FF, yielding a PCE of 6.2%. The corresponding JVꢀ
curve is displayed in Figure 3b. The increase of the doping
level improves the FF and the PCE. We note, that this efficienꢀ
cy is measured directly after device fabrication without the
need of any additional ageing or lightꢀsoaking treatments. This
data clearly shows the benefit and potential of these new doꢀ
pants that we expect to find widespread application in organic
electronic devices based on triarylamineꢀbased semiconducꢀ
tors. Future studies will focus more on the optimization of the
doping level as well as a device fabrication under inert atmosꢀ
phere as the dopants have the potential to obviate the necessity
of any oxygenꢀinduced doping for an efficient device functionꢀ
ing.
REFERENCES
(1) Walzer, K.; Maennig, B.; Pfeiffer, M.; Leo, K. Chem. Rev.
2007, 107, 1233–1271.
(2) a) Hamwi, S.; Meyer, J.; Winkler, T.; Riedl, T.; Kowalsky, W.
Appl. Phys. Lett. 2009, 94, 253307. b) Lehnhardt, M.; Hamwi, S.;
Hoping, M.; Reinker, J.; Riedl, T.; Kowalsky, W. Appl. Phys. Lett.
2010, 96, 193301. c) Meyer, J.; Hamwi, S.; Schmale, S.; Winkler, T.;
Johannes, H.ꢀH.; Riedl, T.; Kowalsky, W. J. Mater. Chem. 2009, 19,
702ꢀ705. d) Hock, R.; Mayer, T.; Jaegermann, W. J. Phys. Chem C.
2012, 116, 18146ꢀ18154.
(3) Burschka, J.; Dualeh, A.; Kessler, F.; Baranoff, E.; CeveyꢀHa,
N.ꢀL.; Yi, C.; Nazeeruddin, M. K.; Grätzel, M. J. Am. Chem. Soc.
2011, 133, 18042ꢀ18045
(4) Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson, H.
Chem. Rev. 2010, 110, 6595ꢀ6663.
(5) He, L.; Jiang, C.; Wang, H.; Lai, D.; Rusli ACS Appl. Mater. In-
terfaces 2012, 4, 1704−1708.
(6) Kim, H.ꢀS.; Lee, C.ꢀR.; Im, J.ꢀH.; Lee, K.ꢀB.; Moehl, T.; Marꢀ
chioro, A.; Moon, S.ꢀJ.; HumphryꢀBaker, R.; Yum, J.ꢀH.; Moser, J.
E.; Grätzel, M.; Park, N.ꢀG. Sci. Rep. 2012, 2, 591.
5
ACS Paragon Plus Environment

Chemistry of Materials
Page 6 of 7
(7) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.;
(14) Ayers, T.; Scott, S.; Goins, J.; Caylor, N.; Hathcock, D.; Slatꢀ
1
2
3
4
5
6
7
8
Snaith, H. J. Science 2012, 338, 643ꢀ647.
tery, S.; Jameson, D. L. Inorg. Chim. Acta 2000, 207, 7ꢀ12.
(15) Saha, N.; Kar, S. K. J. Inorg. Nucl. Chem. 1977, 39, 1236–
1238.
(16) Cappel, U., Gibson, E., Hagfeldt, A.; Boschloo, G. J. Phys.
Chem. C 2009, 113, 6275–6281.
(17) Snaith, H. J.; Grätzel, M. Appl. Phys. Lett. 2006, 89, 262114.
(18) Abrusci, A.; Kumar, R. S. S.; AlꢀHashimi, M.; Heeney, M.;
Petrozza, A.; Snaith, H. J. Adv. Funct. Mater. 2011, 21, 2571–2579.
(19) Abate, A.; Leijtens, T.; Pathak, S.; Teuscher, J.; Avolio, R.;
Errico, M. E.; Kirkpatrik, J.; Ball, J. M.; Docampo, P.; McPherson, I.;
Snaith, H. J. Phys. Chem. Chem. Phys. 2013, 15, 2572ꢀ2579.
(8)Bach, U.; Lupo, D.; Comte, P.; Moser, J.ꢀE.; Weissörtel, F.; Salꢀ
beck, J.; Spreitzer, H.; Grätzel, M. Nature 1998, 395, 583ꢀ585.
(9) Cappel, U. B.; Daeneke, T.; Bach. U. Nano Lett. 2012, 12,
4925ꢀ4931.
(10) Tsao, H. N.; Yi, C.; Moehl, T.; Yum, J.ꢀH.; Zakeeruddin, S.
M.; Nazeeruddin, M. K.; Grätzel, M. ChemSusChem 2011, 4, 591–
594.
(11) Shavaleev, N. M.; Scopelliti, R.; Baranoff, E.; Grätzel, M.;
Nazeeruddin, M. K. Inorg. Chim. Acta 2012, 383, 316–319.
(12) Jameson, D. L.; Blaho, J. K.; Kruger, K. T.; Goldsby, K. A.
Inorg. Chem. 1989, 28, 4312ꢀ4314.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(13) Holland, J. M.; Kilner, C. A.; ThorntonꢀPett, M.; Halcrow, M.
A. Polyhedron 2001, 20, 2829ꢀ2840.
6
ACS Paragon Plus Environment

Page 7 of 7
Chemistry of Materials
1
2
Table of Contents Figure:
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
ACS Paragon Plus Environment