Alkyl chain-functionalized hole-transporting domains in zinc(II) dye-sensitized solar cells

Article abstract of DOI:10.1016/j.dyepig.2015.01.008

FTO/TiO₂ electrodes have been functionalized with {Zn(tpy_anchor)(tpy_ancillary)}²⁺ dyes (tpy = 2,2′:6′,2″-terpyridine) using a stepwise method to sequentially introduce (i) the anchoring ligand tpy_anchor (either a dicarboxylic acid or a diphosphonic acid), (ii) Zn²⁺ ions, and (iii) chromophoric ancillary (4-([2,2′:6′,2″-terpyridin]-4′-yl)-N,N-bis(4-alkoxyphenyl)aniline ligands. A comparison of unmasked and fully masked DSSCs containing representative dyes shows a significant drop in photon-to-current efficiency upon masking. Solid-state absorption spectra of the dye-functionalized electrodes confirm that the intensity of absorption decreases with the steric demands of the ancillary ligand. DSSC measurements show that the {Zn(tpy_anchor)(tpy_ancillary)}²⁺ dyes give poor photon-to-current efficiencies, values of the short circuit current density (J_SC) and the external quantum efficiency (EQE) spectra are consistent with very poor electron injection. Introducing longer alkoxy chains in place of methoxy substituents in the hole-transporting domains in tpy_ancillary is beneficial, resulting in increased J_SC and V_OC, although values remain low despite the 'push-pull' design of the sensitizers.

Full text of DOI:10.1016/j.dyepig.2015.01.008

Dyes and Pigments 116 (2015) 124e130
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Alkyl chain-functionalized hole-transporting domains in zinc(II)
dye-sensitized solar cells
Nik Hostettler, Sebastian O. Fürer, Biljana Bozic-Weber, Edwin C. Constable*,
Catherine E. Housecroft*
Department of Chemistry, University of Basel, Spitalstrasse 51, 4056, Basel, Switzerland
a r t i c l e i n f o
a b s t r a c t
FTO/TiO2 electrodes have been functionalized with {Zn(tpyanchor)(tpyancillary)}2þ dyes (tpy ¼ 2,20:60,200-
terpyridine) using a stepwise method to sequentially introduce (i) the anchoring ligand tpyanchor (either a
dicarboxylic acid or a diphosphonic acid), (ii) Zn2þ ions, and (iii) chromophoric ancillary (4-([2,20:60,200-
terpyridin]-40-yl)-N,N-bis(4-alkoxyphenyl)aniline ligands. A comparison of unmasked and fully masked
DSSCs containing representative dyes shows a significant drop in photon-to-current efficiency upon
masking. Solid-state absorption spectra of the dye-functionalized electrodes confirm that the intensity of
absorption decreases with the steric demands of the ancillary ligand. DSSC measurements show that the
{Zn(tpyanchor)(tpyancillary)}2þ dyes give poor photon-to-current efficiencies, values of the short circuit
current density (JSC) and the external quantum efficiency (EQE) spectra are consistent with very poor
electron injection. Introducing longer alkoxy chains in place of methoxy substituents in the hole-
transporting domains in tpyancillary is beneficial, resulting in increased JSC and VOC, although values
remain low despite the 'pushepull' design of the sensitizers.
Article history:
Received 3 December 2014
Received in revised form
6 January 2015
Accepted 7 January 2015
Available online 22 January 2015
Keywords:
Zinc
Sensitizer
DSSC
Surface-functionalization
'Surfaces-as-ligands'
2,20:60,200-terpyridine
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
[11]. Compared to the former method where an equivalent of Lan-
is wasted, the latter is atom-efficient (Scheme 1) and was
cillary
Organiccompounds [1] andd-block metal complexes [2]are both
used as coloured compounds in dye-sensitized solar cells (DSSCs).
Among metal-containing dyes, ruthenium(II) complexes are the
most widely explored [3], but there is increasing interest in sensi-
tizers based on Earth abundant metals [4]. Bis(diimine)copper(I)
dyes were first evaluated in DSSCs by Sauvage and coworkers [5]
and, generations of dyes later, a record photoconversion efficiency
of 4.66% has been reported with a heteroleptic copper(I) complex
assembled using the HETPHEN approach [6]. Heteroleptic
[Cu(L)(L0)]þ species usually equilibrate in solution to give a mixture
of homo- and heteroleptic complexes. The HETPHEN strategy [7]
minimizes ligand dissociation by using bulky substituents [6,8]. In
contrast, wehavedeveloped a 'surfaces-as-ligands' [9] methodology
in which the TiO2 surface is first functionalized with an anchoring
ligand, Lanchor, to give an active domain that can either undergo
ligand exchange with a homoleptic [Cu(Lancillary)2]þ complex [10] or
be treated with [Cu(MeCN)4][PF6] followed by an ancillary ligand
developedusing 2,20:60,200-terpyridine (tpy)ligands forthe assembly
of dyes containing {Zn(tpy)2}2þ domains (Scheme 1) [12].
With the exception of zinc(II) porphyrin dyes [13,14], little effort
has been made to optimize zinc(II) sensitizers for DSSCs. An
advantage of developing {Zn(tpyanchor)(tpyancillary)}2þ dyes is that it
allows a library of light-harvesting ligand domains to be combined
through facile coordination to zinc(II) (Scheme 1). In zinc(II) com-
plexes, the visible region photoresponse is determined by the li-
gands, making these dyes distinct from those containing the d10
Cuþ ion which exhibit metal-to-ligand charge transfer (MLCT)
bands. Our initial trials of {Zn(tpyanchor)(tpyancillary)}2þ dyes used 1
or 2 (Scheme 2) as the ancillary ligand. Ligands 1 and 2 are chro-
mophoric by virtue of an intra-ligand charge transfer (ILCT) band
with the tertiary amine acting as an electron donor [15], and pho-
toconversion efficiencies of up to 0.71% (relative to 7.29% for N719)
were achieved for unmasked DSSCs containing TiO2-supported
[Zn(1)(3)]2þ dyes. We have now extended these studies using
masked DSSCs (which prevents overestimation of cell performance
[16]) and have introduced alkyl chains onto the periphery of the
hole-transporting domains in an attempt to reduce charge recom-
bination events detrimental to DSSC performance [17e21].
* Corresponding authors. Fax: þ41 61 267 1018.
E-mail address: catherine.housecroft@unibas.ch (C.E. Housecroft).
http://dx.doi.org/10.1016/j.dyepig.2015.01.008
0143-7208/© 2015 Elsevier Ltd. All rights reserved.

N. Hostettler et al. / Dyes and Pigments 116 (2015) 124e130
125
Scheme 1. 'Surfaces-as-ligands' approach to the assembly of copper(I) and zinc(II)-based dyes. For the copper(I) dyes, the ligands are diimines (e.g. derivatives of 2,20-bipyridine)
and for the zinc(II) dyes, the ligands are 40-functionalized 2,20:60,200-terpyridines.
2. Experimental
butoxydiphenylamine, 4-isobutoxyaniline, 1-bromo-4-n-octox-
ybenzene and 4-n-octoxyaniline were prepared by the method
reported for the analogous hexoxy compounds [23] and spectro-
scopic data compared to those reported [24e27]. Bis(dibenzylide-
neacetone)palladium(0), [Pd(dba)2] was purchased from Strem
Chemicals. The dye N719 was purchased from Solaronix.
2.1. General
1H and 13C NMR spectra were recorded on a Bruker Avance III-
500 or 400 MHz NMR spectrometer; chemical shifts were refer-
enced to residual solvent peaks with respect to v(TMS) ¼ 0 ppm.
Solution absorption spectra were recorded using a Cary 5000 or
Agilent 8453 spectrophotometer, and FT-IR spectra using a Perkin
Elmer Spectrum Two with UATR for solid samples. MALDI-TOF,
electrospray ionization (ESI) mass spectra were recorded on
Bruker esquire 3000plu and Bruker Daltonics Inc. microflex in-
struments, respectively; LC-ESI-MS mass spectra used a combina-
tion of Shimadzu (LC) and Bruker AmaZon X instruments.
2.2. 1-Bromo-4-isobutoxybenzene
1-Bromo-4-isobutoxybenzene was prepared by the method re-
ported for the analogous hexoxy compound [23]. 4-Bromophenol
(5.00 g, 28.9 mmol) was dissolved in acetone (60 mL). After the
addition of K2CO3 (7.99 g, 57.8 mmol), the white suspension was
stirred for 5 min, then 1-bromo-2-methylpropane (4.76 mL,
43.4 mmol) was added and the mixture heated at reflux for 3 d.
After the mixture was cooled to room temperature, it was filtered
Compound 3 [12] and 40-(4-bromophenyl)-2,20:60,200-terpyr-
idine [22] were prepared as previously reported; 4,40-di-n-
Scheme 2. Structures of ancillary ligands 1, 2, 5e7, and anchoring ligands 3 and 4.

126
N. Hostettler et al. / Dyes and Pigments 116 (2015) 124e130
and the filtrate concentrated in vacuo to give a colourless oil. This
was dissolved in CH2Cl2, washed with aqueous NaOH (3 ꢀ 50 mL,
2.0 M) and water (2 ꢀ 50 mL), then dried over MgSO4. After
filtration, the filtrate was concentrated under reduced pressure to
yield 1-bromo-4-isobutoxybenzene as a light yellow oil (1.78 g,
155.9 (CD4/B2), 155.85 (CD4/B2), 150.0 (CC4), 149.9 (CC1), 149.2 (CA6),
140.4 (CD1), 137.0 (CA4), 129.6 (CB4), 128.0 (CC2), 127.1 (CD2), 123.8
(CA5), 121.5 (CA3), 120.0 (CC3), 118.1 (CB3), 115.5 (CD3), 68.1 (COCH ),
2
OCH2CH2CH2
31.6 (COCH CH ), 19.4 (C
), 14.0 (CMe). IR (solid, cmꢂ1) 2951
2
2
(w), 2927 (w), 2869 (w), 1598 (m), 1582 (s), 1505 (vs), 1465 (s), 1391
(m),1335 (m),1241 (vs), 1202 (s), 971 (m), 823 (vs), 790 (vs), 732 (s),
660 (m), 546 (s), 515 (m). ESI-MS: m/z 621.4 [M þ H]þ (calc. 621.3).
HR ESI-MS: m/z 621.3226 [M þ H]þ (calc. 621.3224).
7.75 mmol, 26.8%) 1H NMR (500 MHz, CDCl3)
d/ppm 7.36 (d,
J ¼ 9.1 Hz, 2H, HAr2), 6.77 (d, J ¼ 9.0 Hz, 2H, HAr3), 3.68 (d, J ¼ 6.6 Hz,
2H, HOCH ), 2.07 (m, 1H, HCH), 1.01 (d, J ¼ 6.7 Hz, 6H, HMe).
2
2.3. 4,40-Di-isobutoxydiphenylamine
2.6. 4-([2,20:60,200-Terpyridin]-40-yl)-N,N-bis(4-isobutoxyphenyl)
aniline (6)
A round-bottomed flask was charged with K2CO3 (1.21 g,
8.73 mmol), CuI (83.1 mg, 0.436 mmol) and
L-proline (100 mg,
The method was as for 5 starting with 40-(4-bromophenyl)-
2,20:60,200-terpyridine (113 mg, 0.290 mmol) and bis(4-
isobutoxyphenyl)amine (100 mg, 0.319 mmol) and NaOtBu
(36.2 mg, 0.377 mmol). The catalyst was [Pd(dba)2] (3.34 mg,
0.873 mmol) under an N2 atmosphere and then DMSO (20 mL) was
added. The mixture was stirred until most of the solid had dis-
solved. Then 1-bromo-4-isobutoxybenzene (1.50 g, 6.55 mmol) and
4-isobutoxyaniline (721 mg, 4.36 mmol) were added and the
mixture heated at 90 ꢁC for 3 d. After cooling, water was added and
extracted with EtOAc (3 ꢀ 25 mL). The combined organic phases
were washed with brine and dried over MgSO4. The solvent was
removed under reduced pressure and the brown oil obtained was
purified by column chromatography (SiO2, hexane: EtOAc 10:1
changing to 10:2 then 10:3 then 0:1). 4,40-Di-isobutoxydiphenyl-
amine was obtained as a brown oil and was used without further
5.80 m mL, 5.80 mmol). The product was purified by
mol) and PtBu3 (6
column chromatography (SiO2, toluene:EtOAc 1:0 changingto 20:1 to
10:1to10:2to2:1to1:1to0:1)andcompound6 wasisolated as apale
yellow solid (81.0 mg, 0.130 mmol, 44.8%). M.p. 65.9 ꢁC. 1H NMR
(500 MHz, THF-d8) d
/ppm 8.79 (s, 2H, HB3), 8.70 (m, 2H, HA3), 8.67
(ddd, J ¼ 4.8,1.9, 0.9 Hz, 2H, HA6), 7.88 (m, 2H, HA4), 7.73 (d, J ¼ 8.7 Hz,
2H, HC2), 7.35 (m, 2H, HA5), 7.09 (d, J ¼ 9.0 Hz, 4H, HD2), 7.03 (d,
J ¼ 8.7 Hz, 2H, HC3), 6.88 (d, J ¼ 9.0 Hz, 4H, HD3), 3.73 (d, J ¼ 6.4 Hz, 4H,
purification (103 mg, 329 mmol, 7.6%). 1H NMR (500 MHz, CDCl3)
d/
HOCH ), 2.11 (m, 2H, HCH), 1.04 (d, J ¼ 6.7 Hz, 12H, HMe). 13C NMR
2
ppm 6.92 (d, J ¼ 8.0 Hz, 4H, HD2), 6.81 (d, J ¼ 8.9 Hz, 4H, HD3), 3.68
(126 MHz, THF-d8) d
/ppm 157.0 (CA2), 157.0 (CD4), 156.6 (CB2), 150.8
(d, J ¼ 6.7 Hz, 4H, HCH ), 2.05 (m, 2H, HCH), 1.02 (d, J ¼ 6.7 Hz, 12H,
(CC1),150.3 (CC4),149.8 (CA6),141.1 (CD1),137.2 (CA4),130.4 (CB4),128.3
(CC2),127.6 (CD2),124.3 (CA5),121.4 (CA3), 120.6 (CC3),118.2 (CB3),115.9
2
HMe). MALDI-TOF MS (m/z): 313.0 [M]þ (calc. 313.2).
(CD3), 75.0 (COCH ), 29.2 (CCH), 19.4 (CMe). IR (solid, cmꢂ1) 2957 (w),
2
2.4. 4,40-Di-n-octoxydiphenylamine
2916 (w), 2867 (w), 1583 (m), 1503 (vs), 1467 (m), 1234 (s), 1033 (m),
825(m), 791(s), 737(w), 660(w), 522(w). ESI-MS:m/z621.6[Mþ H]þ
(calc. 621.3). HR ESI-MS: m/z 621.3223 [M þ H]þ (calc. 621.3224).
The method was as for 4,40-di-isobutoxydiphenylamine starting
with 4-(octyloxy)aniline (1.50 g, 6.78 mmol), K2CO3 (1.87 g,
13.6 mmol), CuI (129 mg, 0.678 mmol),
L-proline (156 mg,1.36 mmol)
2.7. 4-([2,20:60,200-Terpyridin]-40-yl)-N,N-bis(4-n-octoxyphenyl)
and 1-bromo-4-(octyloxy)benzene (2.90 g, 10.2 mmol). The reaction
time was 4 d. The product was purified by column chromatography
(SiO2, hexane: EtOAc 1:0 changing to 20:1 to 10:1 to 10:2 to 0:1) and
was recrystallized twice from hexane. 4,40-Di-n-octoxydiphenyl-
amine was isolated as a white solid (320 mg, 0.752 mmol, 11.1%). 1H
aniline (7)
The method was as for 5 starting with 40-(4-bromophenyl)-
2,20:60,200-terpyridine (166 mg, 0.427 mmol) and 4,40-di-n-octox-
ydiphenylamine (200 mg, 0.470 mmol) and NaOtBu (53.4 mg,
NMR (400 MHz, CDCl3)
d
/ppm 6.92 (d, J ¼ 8.9 Hz, 4H, HD2), 6.81 (d,
0.555 mmol). The catalyst was [Pd(dba)2] (4.91 mg, 8.54
mmol) and
J ¼ 8.9 Hz, 4H, HD3), 3.91 (t, J ¼ 6.6 Hz, 4H, HOCH ), 1.75 (m, 4H,
PtBu3 (8
m
L, 8.54 mol). The product was purified by column
m
2
HOCH CH ),1.45 (m, 4H, H
),1.39e1.23 (m, 16H, HCH2 ), 0.88 (t,
chromatography (SiO2, toluene:EtOAc 1:0 changing to 10:1 to 10:4
OCH2CH2CH2
2
2
J ¼ 6.9 Hz, 6H, HMe). LC-ESI: m/z 426.4 [M þ H]þ (calc. 426.3).
to 2:1 to 1:1 to 0:1) and 7 was isolated as a pale yellow oily-solid
(250 mg, 0.341 mmol, 79.9%). 1H NMR (500 MHz, CDCl3)
d/ppm
2.5. 4-([2,20:60,200-Terpyridin]-40-yl)-N,N-bis(4-n-butoxyphenyl)
aniline (5)
8.72 (ddd, J ¼ 4.8, 1.8, 0.9 Hz, 2H, HA6), 8.68 (s, 2H, HB3), 8.65 (m, 2H,
HA3), 7.86 (m, 2H, HA4), 7.74 (d, J ¼ 8.8 Hz, 2H, HC2), 7.34 (ddd, J ¼ 7.5,
4.8, 1.2 Hz, 2H, HA5), 7.10 (d, J ¼ 8.9 Hz, 4H, HD2), 7.01 (d, J ¼ 8.8 Hz,
Solid 40-(4-bromophenyl)-2,20:60,200-terpyridine (500 mg,
1.29 mmol), bis(4-n-butoxyphenyl)amine (404 mg, 1.29 mmol) and
NaOtBu (161 mg, 1.67 mmol) were placed in a flask under an N2
atmosphere. Toluene (20 mL) was added and the mixture was
stirred vigorously for 5 min. After the addition of [Pd(dba)2]
2H, HC3), 6.85 (d, J ¼ 8.9 Hz, 4H, HD3), 3.95 (t, J ¼ 6.6 Hz, 4H, HOCH ),
2
1.79 (m, 4H, HOCH CH ), 1.47 (m, 4H, H
), 1.40e1.26 (m, 16H,
OCH2CH2CH2
2
2
HCH ), 0.90 (m, 6H, HMe). 13C NMR (126 MHz, CDCl3)
d
/ppm 156.7
2
(CA2), 155.9 (CD4 þ B2), 150.0 (CC4/C1), 149.9 (CC4/C1), 149.2 (CA6), 140.4
(CD1), 137.0 (CA4), 129.6 (CB4), 128.0 (CC2), 127.1 (CD2), 123.8 (CA5),
mol) and PtBu3 (26
mL, 25.8
mmol), the mixture was
121.5 (CA3), 120.0 (CC3), 118.1 (CB2), 115.5 (CD3), 68.4 (COCH ), 32.0
2
(14.8 mg, 25.8
m
CH2
OCH2CH2
heated at 100 ꢁC for 20 h. The hot mixture was filtered into a hot
Erlenmeyer flask to give a reddish filtrate. The solvent was removed
under reduced pressure and the solid residue was recrystallized
from EtOH and then subjected to column chromatography (basic
Al2O3, hexane:EtOAc 10:1). Compound 5 was isolated as a pale
yellow solid (152 mg, 0.245 mmol, 19.0%). M. p. 122.9 ꢁC. 1H NMR
(CCH ), 29.6 (C ), 29.5 (C
), 29.4 (CCH2 ), 26.3 (COCH2CH2CH2 ),
2
22.8 (CCH ), 14.3 (CMe). IR (solid, cmꢂ1) 2924 (m), 2854 (w), 1599
(w), 1583 (m), 1503 (vs), 1467 (m), 1235 (s), 826 (s), 792 (s), 738 (m),
660 (m), 521 (m). ESI-MS: m/z 733.7 [M þ H]þ (733.4), 755.7
[M þ Na]þ (755.4). HR ESI-MS: m/z 733.4482 [M þ H]þ (calc.
733.4476), 755.4300 [M þ Na]þ (calc. 755.4295).
2
(500 MHz, CDCl3)
d
/ppm 8.72 (ddd, J ¼ 4.8, 1.9, 1.0 Hz, 2H, HA6), 8.68
(s, 2H, HB3), 8.65 (m, 2H, HA3), 7.86 (m, 2H, HA4), 7.74 (d, J ¼ 8.8 Hz,
2H, HC2), 7.34 (m, 2H, HA5), 7.10 (d, J ¼ 8.9 Hz, 4H, HD2), 7.01 (d,
J ¼ 8.8 Hz, 2H, HC3), 6.85 (d, J ¼ 9.0 Hz, 4H, HD3), 3.96 (t, J ¼ 6.5 Hz,
2.8. DSSC fabrication
€
DSSCs were prepared based on the procedure of Gratzel and
coworkers [28,29]. Solaronix Test Cell Titania Electrodes with a
scattering layer were used for the photoanodes. The electrodes
4H, HOCH ), 1.78 (m, 4H, H
), 1.51 (m, 4H, HOCH2CH2CH2 ), 0.99 (t,
OCH2CH2
2
J ¼ 7.4 Hz, 6H, HMe).13C NMR (126 MHz, CDCl3)
d
/ppm 156.6 (CA2),

N. Hostettler et al. / Dyes and Pigments 116 (2015) 124e130
127
were sintered at 450 ꢁC for 30 min, cooled to ca. 80 ꢁC and then
soaked in a DMSO solution of 3 or 4 (1 mM) for 24 h. The colourless
slides were then removed from the solution, washed with DMSO
and EtOH, and dried. Then, the electrodes were soaked in an EtOH
solution of ZnCl2 (0.5 mM) for 24 h, and after washing with EtOH
and drying, were immersed in a CH2Cl2 solution of ligands 5, 6 or 7
(0.5 mM) for 64 h. The now orange electrodes were washed with
CH2Cl2 and dried. Solaronix Test Cell Platinum Electrodes were used
for the counter electrodes, and cleaned of volatile impurities by
heating at 450 ꢁC for 30 min. The DSSC was assembled by
combining the electrodes using thermoplast hot-melt sealing foil
(Solaronix Test Cell Gaskets) and pressing them together. The
electrolyte was introduced into the DSSC by vacuum backfilling and
comprised LiI (0.1 M), I2 (0.05 M), 1-methylbenzimidazole (0.5 M)
and 1-butyl-3-methylimidazolinium iodide (0.6 M) in 3-
methoxypropionitrile. The hole in the counter electrode was
sealed using hotmelt sealing foil (Solaronix Test Cell Sealings) and a
coverglass (Solaronix Test Cell Caps).
CuI/L-proline catalysed [23] Ullmann coupling, and subsequent
Hartwig-Buchwald amination with 40-(4-bromophenyl)-2,20:60,200-
terpyridine yielded compounds 5e7 in moderate to high yields. In
the electrospray mass spectra of the compounds, peaks assigned
to [M þ H]þ were observed at m/z 621.4 for 5, 621.6 for 6 and 733.7
for 7. Satisfactory elemental analyses could not be obtained, but
there was good agreement between the observed and calculated
m/z values in the high resolution mass spectra of 5e7 (see
experimental section). Fig. 1 shows a representative 1H NMR
spectrum (of compound 5); minor organic impurities observed in
the 1H NMR spectra persisted in all compounds after chromato-
graphic workup. The 1H and 13C NMR spectra were assigned by
COSY, NOESY, DEPT, HMQC and HMBC methods. NOESY cross-
peaks between signals for HB3/HC2 and between HC3/HD2 allowed
the pairs HC2/HC3 and HD2/HD3 to be distinguished (see Scheme 3
for atom labelling). The spectra were consistent with the struc-
tures shown in Scheme 3.
Electrodes for solid state absorption measurements were pre-
pared in the same way as the photoanodes (above) using Solaronix
transparent Test Cell Titania Electrodes without a scattering layer.
3.2. Solid state absorption spectra of TiO2-adsorbed zinc(II)-
containing dyes
Electrodes for solid state absorption measurements were pre-
pared by stepwise dipping of commercial FTO/TiO2 electrodes
without a scattering layer into solutions of (i) anchoring ligands 3
or 4, (ii) zinc(II) chloride, and (iii) ancillary ligands 5, 6 or 7. Details
of solvents and soaking times are given in the experimental sec-
tion. After the electrodes had been washed and dried, trans-
mission absorption spectra were recorded. The spectra were
corrected with respect to a spectrum of a blank electrode. Ab-
sorption maxima are given in Table 1; a small blue-shift of the ILCT
band is observed on changing the anchoring domain from car-
boxylic acid to phosphonic acid. Fig. 2 compares the spectra of one
pair of dyes with a common ancillary ligand (6) but different
anchoring ligands (carboxylic acid 3 versus phosphonic acid 4),
and also compares the series of dyes with a common anchoring
ligand (4). An increase in absorbance is seen comparing
[Zn(3)(6)]2þ to [Zn(4)(6)]2þ, a trend also observed on going from
2.9. DSSC and external quantum efficiency (EQE) measurements
The DSSC measurements were carried out using fully masked
cells; the mask comprised a black-coloured copper sheet with one
aperture of average area 0.06012 cm2 (standard deviation of 1%)
placed over the active area of the cell. The area of the aperture in
the mask was smaller than the surface area of TiO2 (0.36 cm2). Black
tape was applied to complete the masking of the cell. Measure-
ments were made by irradiating the DSSC from behind using a
SolarSim 150 instrument (100 mW cmꢂ2 ¼ 1 sun), and the simu-
lated light power was calibrated by using a silicon reference cell.
The external quantum efficiency (EQE) measurements were
performed on a Spe-Quest quantum efficiency setup from Rera
Systems (Netherlands) equipped with a 100 W halogen lamp (QTH)
and a lambda 300 grating monochromator from Lot Oriel. The
monochromatic light was modulated to 3 Hz using a chopper wheel
from ThorLabs. The cell response was amplified with a large dy-
namic range IV converter from CVI Melles Griot and then measured
with an SR830 DSP Lock-In amplifier from Stanford Research.
[Zn(3)(5)]2þ to [Zn(4)(5)]2þ, and from [Zn(3)(7)]2þ to [Zn(4)(7)]2þ
.
Although not definitive, the data are consistent with enhanced
surface coverage by the phosphonic acid anchoring ligand
compared to the carboxylic acid. Fig. 2 reveals that with a constant
anchor 4, the absorbance of the functionalized titania
decreases as the ancillary ligand is changed from 5 to 6 to 7, and
3. Results and discussion
the
same
trend
is
observed
for
the
series
3.1. Ligand syntheses and characterizations
[Zn(3)(5)]2þ > [Zn(3)(6)]2þ > [Zn(3)(7)]2þ. This follows the order
of the steric demands of the alkoxy-substituents, implying that
fewer surface coordination sites can be occupied as the steric
demand of the ancillary ligand increases.
The synthetic route to compounds 5e7 is summarized in
Scheme 3. The secondary amine precursors were prepared by a
Scheme 3. Synthesis of 5e7. Conditions: (i) K2CO3, CuI, L-proline in DMSO, 90 ꢁC, 4 days; (ii) 40-(4-bromophenyl)-2,20:60,200-terpyridine, NaOtBu, Pd(bda)2/PtBu3, toluene, 20 h. Atom
labels refer to NMR spectroscopic assignments.

128
N. Hostettler et al. / Dyes and Pigments 116 (2015) 124e130
Fig. 1. 500 MHz 1H NMR spectrum of compound 5; chemical shifts in v/ppm. * ¼ residual CHCl3; ꢀ ¼ H2O; o ¼ traces of ethyl acetate.
Table 1
direct comparison with dyes containing the alkoxy tails. Duplicate
DSSCs were made for each sensitizer.
Solid state absorption maxima for electrodes func-
tionalized with zinc(II) dyes.
External quantum efficiency (EQE) spectra for the DSSCs were
recorded and Fig. 3a shows spectra for cells containing [Zn(4)(2)]2þ
,
Dye
lmax/nm
[Zn(4)(5)]2þ, [Zn(4)(6)]2þ and [Zn(4)(7)]2þ with the phosphonic
acid anchor. Fig. 3b shows the EQE spectrum of a reference DSSC
containing the ruthenium(II) dye N719. Despite the orange colour
of the zinc(II)-containing electrodes and their solid state absorption
[Zn(3)(5)]2þ
[Zn(3)(6)]2þ
[Zn(3)(7)]2þ
[Zn(4)(5)]2þ
[Zn(4)(6)]2þ
[Zn(4)(7)]2þ
446
447
448
443
443
444
3.3. EQE spectra
DSSCs were assembled by the stepwise process described above,
using commercial electrodes with a scattering layer (see Experi-
mental section). In addition to ancillary ligands 5e7, DSSCs were
also fabricated using ancillary ligand 2 which contains methoxy
substituents (Scheme 1). After immersion in the dye bath con-
taining 2, 5, 6 or 7, the electrodes changed from colourless to or-
ange. Previously, we have reported efficiency data for TiO2-
supported dyes [Zn(3)(2)]2þ and [Zn(4)(2)]2þ in unmasked DSSCs
[12]. More recently, we have made the transition to fully masked
[16] DSSCs, and we have reinvestigated the performances of
masked cells incorporating [Zn(3)(2)]2þ and [Zn(4)(2)]2þ to allow a
Fig. 3. (a) EQE spectra of DSSCs containing the dyes [Zn(4)(2)]2þ [Zn(4)(5)]2þ
, ,
Fig. 2. Solid state absorption spectra of electrodes (without scattering layer) with
[Zn(4)(6)]2þ and [Zn(4)(7)]2þ. (b) EQE spectrum of reference DSSC containing standard
dye N719.
adsorbed dyes [Zn(4)(5)]2þ, [Zn(4)(6)]2þ, [Zn(4)(7)]2þ and [Zn(3)(6)]2þ
.

N. Hostettler et al. / Dyes and Pigments 116 (2015) 124e130
129
Table 3
spectra, both of which confirm light-harvesting in the visible re-
gion, the EQE spectra in Fig. 3a reveal extremely poor electron in-
jection. The EQEmax are <1.4% at lmax ¼ 430e450 nm for all the
DSSCs, and the corresponding values for DSSCs containing
[Zn(3)(2)]2þ, [Zn(3)(5)]2þ, [Zn(3)(6)]2þ and [Zn(3)(7)]2þ (carboxylic
acid anchors) are even lower (<1.0%). Nonetheless, we note that the
introduction of the alkoxy chains has a beneficial effect, further
substantiated by the DSSC parameters discussed below. For com-
parison, the EQE spectrum of the N719 reference cell (Fig. 3b) is
characterized by EQEmax of 67.5% at lmax ¼ 540 nm extending to a
high energy shoulder (lmax ¼ 410 nm) with EQEmax ¼ 54.3%.
DSSC performance data for duplicate masked cells using the zinc(II) dyes with
dicarboxylic acid anchoring ligand (3) compared to DSSC containing N719
measured under the same conditions. Measurements were made on the day of
fabricating the cells.
a
Dye
JSC/mA cmꢂ2
VOC/mV
ff/%
h/%
[Zn(3)(2)]2þ
[Zn(3)(2)]2þ
[Zn(3)(5)]2þ
[Zn(3)(5)]2þ
[Zn(3)(6)]2þ
[Zn(3)(6)]2þ
[Zn(3)(7)]2þ
[Zn(3)(7)]2þ
N719
0.14
0.15
0.38
0.45
0.30
0.32
0.26
0.27
14.59
345
376
414
425
386
396
392
392
609
55
64
60
65
66
64
64
64
73
0.03
0.04
0.09
0.13
0.08
0.08
0.07
0.07
6.44
3.4. DSSC performances
Performance characteristics of DSSCs containing the
{Zn(tpy)2}2þ dyes measured on the day of assembling the cells are
shown in Tables 2 and 3. Duplicate cells were prepared for each dye
and the data are compared to those measured for an N719 cell; the
DSSCs correspond to those used for the EQE spectra discussed
above. Data in Tables 2 and 3 refer to dyes containing the phos-
phonic and carboxylic acid anchors 4 and 3, respectively. It is
important to note the reproducibility of data for pairs of duplicate
cells which lends credence to the data, despite the poor efficiencies.
Control measurements were made using a blank electrode (i.e. no
adsorbed anchoring ligand or dye), yielding values of the short-
circuit current density (JSC) z 0.3 mA cmꢂ2, open-circuit voltage
(VOC) z 460 mV and an efficiency (h) z 0.10%. Thus, detailed dis-
cussion of the performance parameters is not justified and we focus
only on trends.
A critical observation is the effect of masking the DSSC con-
taining the dyes [Zn(3)(2)]2þ and [Zn(4)(2)]2þ. With unmasked
DSSCs efficiencies of 0.46 and 0.63% are observed, respectively [12],
but a fall to <0.05% occurs upon masking (Tables 2 and 3). This is
consistent with previous comparisons of masked and unmasked
DSSCs containing copper(I)-based dyes [30].
dye-functionalized electrodes for which the intensities are
[Zn(tpyanchor)(5)]2þ > [Zn(tpyanchor)(6)]2þ > [Zn(tpyanchor)(7)]2þ
both for tpyanchor ¼ 3 and 4 (Fig. 2).
4. Conclusions
An atom-efficient stepwise assembly procedure has been
developed for functionalization of FTO/TiO2 electrodes with
{Zn(tpyanchor)(tpyancillary)}2þ dyes incorporating chromophoric
ancillary ligands 2, 5, 6 and 7. The intensities of the solid-state
absorption maxima of the dye-functionalized TiO2 decrease in the
order [Zn(3)(5)]2þ
>
[Zn(3)(6)]2þ
>
[Zn(3)(7)]2þ
,
and
[Zn(4)(5)]2þ > [Zn(4)(6)]2þ > [Zn(4)(7)]2þ. Assuming that equal
soaking times in solutions of an anchoring ligand results in similar
surface coverages, this is consistent with fewer surface sites being
occupied (bottom of Scheme 1) as the steric demand of the ancillary
ligand increases. Despite the absorption properties of the dyes,
DSSC measurements reveal poor photon-to-current conversion
efficiencies. A major contributing factor is the poor electron injec-
tion, seen in low values of JSC and EQEmax. Introducing alkoxy chains
(in 5, 6 and 7) into the hole-transporting domains in tpyancillary is
beneficial, and leads to increased JSC and VOC, although absolute
values remain low.
A comparison of Tables 2 and 3 suggests that the phosphonic
acid anchoring ligand
4
is favoured over the dicarboxylic
€
acid anchor 3, an observation that has been noted by Gratzel [31]
and is also true for copper(I) dyes [10,30,32]. However, for
both anchoring domains, values of JSC show that electron
injection is extremely poor. This is consistent with the
EQE spectra, and is the dominant factor leading to the
meagre global efficiencies of {Zn(tpyanchor) (tpyancillary)}2þ dyes.
Replacing methoxy substituents by butoxy groups on the pe-
riphery of the ancillary ligand results in an enhancement in both
JSC and VOC, and a consequent gain in the efficiency. For both 3
and 4, the trend in DSSC performances follows the order of
ancillary substituents n-butyl > isobutyl > octyl > methyl. This
is consistent with the solid state absorption maxima of
Acknowledgements
The project is supported by the European Research Council
(Advanced Grant 267816 LiLo), the Swiss National Science Foun-
dation (200020_144500) and the University of Basel. We thank M.
Waser, FHNW, Basel for the preparation of ligand 4, and Dr Collin
Morris and Dr Heinz Nadig for recording LC-ESI and HR-ESI mass
spectra. The graphical abstract was illustrated by Danielle Hayoz.
References
€
[1] Mishra A, Fischer MKR, Bauerle P. Metal-free organic dyes for dye-sensitzed
Table 2
solar cells: from structure-property relationships to design rules. Angew
Chem Int Ed 2009;48:2474e99.
[2] Kalyanasundaram K, editor. Dye-sensitized solar cells. Boca Raton: CRC Press,
Taylor & Francis; 2010.
[3] Vougioukalakis GC, Philippopoulos AI, Stergiopoulos T, Falaras P. Contribu-
tions to the development of ruthenium-based sensitizers for dye-sensitized
solar cells. Coord Chem Rev 2011;255:2602e21.
[4] Bozic-Weber B, Constable EC, Housecroft CE. Light harvesting with Earth
abundant d-block metals: towards a sustainable materials chemistry. Coord
Chem Rev 2013;257:3089e106.
DSSC performance data for duplicate masked cells using the zinc(II) dyes with
phosphonic acid anchoring ligand (4) compared to
measured under the same conditions. Measurements were made on the day of
fabricating the cells.
a DSSC containing N719
Dye
JSC/mA cmꢂ2
VOC/mV
ff/%
h/%
[Zn(4)(2)]2þ
[Zn(4)(2)]2þ
[Zn(4)(5)]2þ
[Zn(4)(5)]2þ
[Zn(4)(6)]2þ
[Zn(4)(6)]2þ
[Zn(4)(7)]2þ
[Zn(4)(7)]2þ
N719
0.21
0.20
0.51
0.72
0.45
0.45
0.41
0.36
14.59
379
377
440
467
428
421
436
411
609
64
55
63
67
66
64
65
63
73
0.05
0.04
0.14
0.23
0.13
0.12
0.12
0.09
6.44
[5] Alonso-Vante N, Nierengarten J-F, Sauvage J-P. Spectral sensitization of large-
band-gap semiconductors (thin films and ceramics) by
bis(1,10-phenanthroline)copper(I) complex. J Chem Soc Dalton Trans 1994:
1649e54.
a carboxylated
[6] Sandroni M, Favereau L, Planchat A, Akdas-Kilig H, Szuwarski N, Pellegrin Y,
et al. Heteroleptic copper(I)-polypyridine complexes as efficient sensitizers for
dye sensitized solar cells. J Mater Chem A 2014;2:9944e7.

130
N. Hostettler et al. / Dyes and Pigments 116 (2015) 124e130
[7] Schmittel M, Luning U, Meder M, Ganz A, Michel C, Herderich M. Synthesis of
[21] Suresh T, Ganesh K, Singh SP, Islam A, Han L, Chandrasekharam M. Func-
tionalized phenyl bipyridine ancillary ligand as double recombination inhib-
itor in ruthenium complex for dye solar cells. Dyes Pigments 2013;99:850e6.
[22] Wang J, Hanan G S. A facile route to sterically hindered and non-hindered 40-
aryl-2,20:60,20-terpyridines. Synlett 205; 1251e1254.
[23] Shallaiah M, Rajan YC, Lin H-C. Synthesis of novel triarylamine-based den-
drimers with N4,N6-dibutyl-1,3,5-triazine-4,6-diamine probe for electron/
energy transfers in H-bonded donor-acceptor-donor triads and as efficient
Cu2þ sensors. J Mater Chem 2012;22:8976e87.
sterically encumbered 2,9-diaryl substituted phenanthrolines. Key building
blocks for the preparation of mixed (bis-heteroleptic) phenanthroline cop-
per(I) complexes. Heterocycl Commun 1997;3:493e8.
[8] Sandroni M, Kayanuma M, Planchat A, Szuwarski N, Blart E, Pellegrin Y, et al.
First application of the HETPHEN concept to new heteroleptic bis(diimine)
copper(I) complexes as sensitizers in dye sensitized solar cells. Dalton Trans
2013;42:10818e27.
€
[9] Schonhofer E, Bozic-Weber B, Martin CJ, Constable EC, Housecroft CE,
Zampese J. A. 'Surfaces-as-ligands, surfaces-as-complexes' strategies for cop-
per(I) dye-sensitized solar cells. Dyes Pigments 2015;115:154e65.
[10] Bozic-Weber B, Constable EC, Housecroft CE, Kopecky P, Neuburger M,
Zampese JA. The intramolecular aryl embrace: from light emission to light
absorption. Dalton Trans 2011;40:12584e94.
[24] Odom SA, Lancaster K, Beverina L, Lefler KM, Thompson NJ, Coropceanu V,
et al. Bis[bis-(4-alkoxyphenyl)amino] derivatives of dithienylethene, bithio-
phene, dithienothiophene and dithienopyrrole: palladium-catalysed synthesis
and highly delocalised radical cations. Chem Eur J 2007;34:9637e46.
~
[25] Blasco E, Schmidt BVKJ, Barner-Kowollik C, Pinol M, Oriol L. Dual thermo- and
[11] Brauchli SY, Malzner FJ, Constable EC, Housecroft CE. Influence of a co-
adsorbant on the performance of bis(diimine) copper(I)-based dye-sensi-
tized solar cells. RSC Adv 2014;4:62728e36.
photo-responsive micelles based on miktoarm star polymers. Polym Chem
2013;4:4506e14.
[26] Vasconcelos UB, Schrader A, Vilela GD, Borges ACA, Merlo AA. Buchwald
protocol applied to the synthesis of N-heterotolan liquid crystals. Tetrahedron
2008;64:4619e26.
[12] Bozic-Weber B, Constable EC, Hostettler N, Housecroft CE, Schmitt R,
10
€
Schonhofer E. The d route to dye-sensitized solar cells: step-wise assembly
of zinc(II) photosensitizers on TiO2 Surfaces. Chem Commun 2012;48:5727e9.
[13] Griffith MJ, Sunahara K, Wagner P, Wagner K, Wallace GG, Officer DL, et al.
Porphyrins for dye-sensitized solar cells: new insights into efficiency-
determining electron transfer steps. Chem Commun 2012;48:4145e62.
[14] Yella A, Lee H-W, Tsao HN, Yi C, Chandiran AK, Nazeeruddin Md K, et al.
Porphyrin-sensitized solar cells with cobalt(II)/(III)-based redox electrolyte
exceed 12 percent efficiency. Science 2011;334:629e34.
[27] Kim D-Y, Lee S-A, Choi Y-J, Hwang S-H, Kuo S-W, Nah C, et al. Thermal- and
photo-induced phase-transition behaviors of a tapered dendritic liquid crystal
and photochromic azobenzene mesogens and a bicyclic chiral center. Chem
Eur J 2014;20:5689e95.
€
[28] Ito S, Murakami TN, Comte P, Liska P, Gratzel C, Nazeeruddin MK, et al.
Fabrication of thin film dye sensitized solar cells with solar to electric power
conversion efficiency over 10%. Thin Solid Films 2008;516:4613e9.
ꢀ
[15] Walters KA, Kim Y-J, Hupp JT. J Electroanal Chem 2003;554e555:449.
[16] Snaith HJ. How should you measure your excitonic solar cells? Energy Environ
Sci 2012;5:6513e20.
[29] Ito S, Chen P, Comte P, Nazeeruddin MK, Liska P, Pechy P, et al. Fabrication of
screen-printing pastes from TiO2 powders for dye-sensitised solar cells. Prog
Photovoltaics Res Appl 2007;15:603e12.
€
[17] Kroeze JE, Hirata N, Koops S, Nazeeruddin Md K, Schmidt-Mende L, Gratzel M,
[30] Bozic-Weber B, Brauchli SY, Constable EC, Fürer SO, Housecroft CE, Wright IA.
Hole-transport functionalized copper(I) dye sensitized solar cells. Phys Chem
Chem Phys 2013;15:4500e4.
et al. Alkyl chain barrier for kinetic optimization in dye-sensitized solar cells.
J Am Chem Soc 2006;128:16376e83.
[18] Koumura N, Wang Z-S, Mori S, Miyashita M, Suzuki E, Hara K. Alkyl-func-
tionalized organic dyes for efficient molecular photovoltaics. J Am Chem Soc
2006;128:14256e7.
[19] Chang Y-C, Wang C-L, Pan T-Y, Hong S-H, Lan C-M, Kuo H-H, et al. A strategy
to design highly efficient porphyrin sensitizers for dye-sensitized solar cells.
Chem Commun 2011;47:8910e2.
[20] Xue X, Zhang W, Zhang N, Ju C, Peng X, Yang Y, et al. Effect of the length of the
alkyl chains in porphyrin meso-substituents on the performance of dye-
sensitized solar cells. RSC Adv 2014;4:8894e900.
ꢀ
[31] Pechy P, Rotzinger FP, Nazeeruddin Md K, Kohle O, Zakeeruddin SM,
Humphry-Baker R, et al. Preparation of phosphonated polypyridyl ligands to
anchor transition-metal complexes on oxide surfaces: application for the
conversion of light to electricity with nanocrystalline TiO2 films. J Chem Soc
Chem Commun 1995:65e6.
[32] Bozic-Weber B, Chaurin V, Constable EC, Housecroft CE, Meuwly M,
Neuburger M, et al. Exploring copper(I)-based dye-sensitized solar cells: a
complementary experimental and TD-DFT investigation. Dalton Trans
2012;41:14157e69.