M. Ziolek presenting his poster on International Conference on
Hybrid and Organic Photovoltaics, in Assisi, Italy, 2010.
M. Ziolek presenting his poster on International Conference on
Hybrid and Organic Photovoltaics, in Valencia, Spain, 2011.
Example of the solar cell constructed during the performance of NANOSOL project
and the measurements of its voltage under room light conditions.
Significant results:
The following main scientific results of the NANOSOL project can be highlighted:
1) The caging effect of the photochromic proton transfer dye (salizylaldehyde azine)
encapsulated in zeolites NaX and NaY has been observed using steady-state and
femtosecond to nanosecond time-resolved spectroscopy. In solution, an excited-state
intramolecular proton- transfer (ESIPT) reaction takes place in less than 80 fs, leading
to a keto-type tautomer. In contrast within NaX zeolite, a zwitterionic (Z) form is
present both at S0 and S1 states, and a large hypsochromic shift of the stationary
emission spectrum is observed. The increase in fluorescence lifetime upon
encapsulation (from 54 ps to 0.2 – 2.8 ns) is mainly due to hindrance in twisting
motion of the confined Z structure imposed by the nanocage. A significant dependence
of the lifetimes on the guest concentration inside the zeolite indicates an interaction
between neighboring guest molecules leading to a quenching of the fluorescence
(Figure 1). The analysis of emission decays using stretched-exponential model
suggests that the excited-state interactions between neighboring dyes play a key role in
Emission intensity / kCounts
the deactivation of the trapped Z fluorophores.
10
A
conc. x 1
conc. x 10
conc. x 100
conc. x 1000
conc. x 100 000
5
0
0
4
5
8
Time / ns
12
16
B

4
Figure 11.
Time / ns
Left: Effect of the used initial concentration of the dye resulting in quenching of the ps - ns fluorescence decays.
3
Right: cover of the J. Phys. Chem. C issue with the article about salizylaldehyde azine in zeolites
2
2
1
3
0
0
1
2
3
4
5
2) The presence of an equilibrium between the normal and anion form of the
triphenylamine dye of the best solar cell efficiency (TPC1) in solution was observed.
In polar solvents the equilibrium between TPC1 normal and anion structures was
found to depend on solvent H-bond acceptor ability and concentration of the dye.
Remarkable, a correlation between the contribution of the normal form in the total
absorption spectrum and solar energy conversion efficiency of the photovoltaic
devices prepared under different baths was observed, being high in dichloromethane
and low in tetrahydrofurane (see Table 1). The preliminary studies reveal that such
equilibrium is common for many molecules from the triphenylamine family having
cyanoacrylic acid group as an electron acceptor.
Table 1.
Correlation of the solar cell efficiency for photovoltaic devices prepared under different bath with the
contribution of normal form in th Uv-visible absorption spectrum.
Solvent
H-bond acceptor ability
parameter ()
Contribution of
normal form, %
Solar cell
efficiency, %
DCM
0.10
100
5.33
ACN
0.40
52
4.59
EtOH
0.75
20
4.29
THF
0.55
9
3.61
3) Complete characterization of the spectroscopic properties of the TPC1 dye in the
solvents of different properties was performed. Both normal and anion structures
exhibit a large charge transfer character in the excited state manifested by a large
Stokes shift between absorption and emission maxima (up to 9 000 cm-1 in
acetonitrile), see Figure 2. The lifetime of the relaxed state of the normal structure
varies significantly with the solvent polarity (from 80 ps in acetonitrile to 1.8 ns in nhexane), and it is considerably shorter than that of the anion one (1.2 – 2.6 ns). It was
also found that the ultrafast relaxation processes of TPC1 in solution are dominated by
the solvation dynamics which is the fastest in acetonitrile (below 1 ps) and the slowest
in ethanol (about 25 ps). The typical emission transients and time-resolved emission
spectrum is shown in Figure 3.
Normalized absorbance
1.0
0.8
0.6
0.4
0.2
0.0
300
Normalized intensity
1.2
1.0
DCM,
ACN,
THF,
350
HEX
EtOH
DCM plus DBU
400 450 500
Wavelength, nm
HEX,
ACN,
THF,
EtOH,
550
DCM+DBU
DCM
0.8
0.6
0.4
0.2
0.0
450 500 550 600 650 700 750 800
Wavelength, nm
Figure 2.
Uv-visible absorption (left) and emission (right) spectra of TPC1 in the indicated solvents.
Normalized intensity
4.5
750 nm
4.0
700 nm
3.5
3.0
650 nm
2.5
600 nm
2.0
550 nm
1.5
500 nm
1.0
475 nm
0.5
457 nm in pure DCM (IRF)
0.0
0
1
2
3
4
Time, ps
Intensity
1.0
0.05 ps
0.1 ps
0.2 ps
0.5 ps
1 ps
1.6 ps
3 ps
10 ps
100 ps
0.5
TRES
CT →
LE
0.0
500
550
600 650 700 750
Wavelength, nm
Figure 3.
Left: Femtosecond emission transients of TPC1 in dichloromethane. Right: Time-resolved emission spectrum
(TRES) of TPC1 in dichloromethane. The spectra show the rapid (<100 fs) decay of the initial locally excited
state (LE) and the red-shift of the subsequent charge-transfer state (CT) during the solvation process.
4) The dynamics of electron injection and back electron transfer was measured for TPC1
interacting with titania materials in suspension. The comparison has been performed
for different titania materials: nanoparticles of different size (P13 and P25), nanotubes
(NT) and nanorods (NR). As can be seen in Figure 4A the electron injection was
found to be faster in NT comparing to other materials, which was explained by the
caging effect. The electron recombination transients were similar in all materials
(Figure 4B).
(A)
Normalized intensity
1.0
0.6
0.4
0.2
0.0
(B)
emission at 650 nm in ACN
TPC1 / P25
TPC1 / P13
TPC1 / NT
TPC1 / NR
0.8
0
2
4
6
8 10 12 14 16
Time, ps
0.0
A (normalized)
-0.2
-0.4
-0.6
P25
P13
NT
-0.8
-1.0
0
5
10
15
20
Time, ps
100
1000
Figure 4.
(A) Comparison of the femtosecond emission transients of TPC1 in ACN solvent and interacting with the
P13, P25 NR and NT of titania upon excitation at 400 nm and observation at 650 nm. (B) Comparison of the
bleach signal observed at 480 nm in the transient absorption spectra and upon excitation at 400 nm of TPC1
attached to P13, P25 and NT in ACN suspension.
5) The dynamics of electron injection and back electron transfer was measured for TPC1,
TPC3 and TPC5 dyes (Scheme 1) interacting with vertically oriented titania nanotube array
(inner diameters 36 nm and 70 nm), standard titania nanoparticle (25 nm diameter) and, as
a reference, alumina nanoparticle (13 nm diameter) films. The studied dyes differ by the
substituents in a spacer group that causes a shift of its absorption spectra. Despite a redshift of the dye absorption band resulting in an improved response to the solar spectrum,
smaller electron injection rates and smaller extinction coefficients result in reduced dye
sensitized solar cell (DSSC) conversion efficiencies. For the most efficient dye, TPC1,
electron injection from the hot locally excited state to titania takes place on a time scale of
about 100 fs, but from the relaxed charge transfer state it proceeds in a non-exponential
way with time constants from 1 ps to 50 ps. The latter process involves the trap states
below the conduction band edge, localized close to the dye radical cation, and is
accompanied by fast electron recombination to the parent dye’s ground state. This process
limits the efficiency of DSSCs made using these type of organic dyes. The residual, slower
recombination can be described by a stretched exponential decay with a characteristic time
of 0.5 s and a dispersion parameter of 0.33. Both the electron injection and back electron
transfer dynamics are similar in titania nanoparticles and nanotubes and are summarized in
Scheme 2. Variations between the two film types are only found in the time resolved
emission transients, which are explained in terms of the difference in local electric fields
affecting the position of the emission bands.
TPC1 R=H
R
N
TPC3 R=Cl
COOH
R
TPC5 R=CN
NC
Scheme 1.
Structure of TPC1, TPC3 and TPC5 dyes.
Energy
titania
conduction
band
e
(electron
thermali- ~100 fs
zation)
e- injection
~100 fs
LE state
S1
1-50 ps
CT state
trap states
e- (interface
bound
trapped
e- recombination
electrons)
1-50 ps
e- (separated
trapped
electrons)
1 ns – 100 s
400 nm
excitation
S0
Scheme 2.
Schematic representation of the photoinduced electron transfer processes, taking place in TPC1 interacting with
titania nanotubes and nanoparticles.
6) TPC1 encapsulated in titanium-doped mesoporous silica structures as alternative materials
for dye sensitized solar cells has been studied by means of stationary absorption and
emission as well as ultrafast emission spectroscopy. A TPC1 complex with titanium atoms
within the mesoporous silica in dichloromethane (DCM) solution is formed, having a red
shift of the visible absorption band by about 30 nm with respect to that of the TPC1 in
DCM (from 455 to 485 nm). For the complexes, multi- exponential emission quenching of
the relaxed singlet excited state occurs with time constants from 300 fs to 30 ps, and is
assigned to confined electron injection process into the Ti-O chromophore. The averaged
electron injection rate from the higher energy levels gets smaller values for less energetic
probing, from 2.71012 s-1 at 600 nm to 1.51012 s-1 at 700 nm. However, in the titaniumdoped samples prepared by an impregnation method we observed about 2-3 times slower
Normalized emission
injection (Figure 5).
MCM-41
Ti-MCM-41
TiO2-SBA-15
0.8
0.4
0.0
0
1
2
3
Time, ps
4
5
Figure 5.
Comparison of the fs-ps emission transients gated at 700 nm, and upon excitation at 420 nm of the indicated
samples in a DCM solution.
7) We measured femto- to nanosecond emission studies of the interaction of TPC1 with
different semiconductor particles in a dichloromethane (DCM) suspension. We used ZnO,
ZrO2 and Al2O3 nanoparticles employed in dye-sensitized solar cells as active electron
collection materials or insulating layers. Steady-state absorption and emission spectra
reflect strong complex formation between TPC1 and the used materials. The fs-ns emission
transients of the interfacial systems show a non-exponential behaviour with an averaged
half-lifetime of 4 ps, 11 ps and 150 ps for ZnO, ZrO2 and Al2O3, respectively. For the latter,
we observed the effect of dye’s concentration indicating the action of a fluorescence selfquenching mechanism. For ZnO and ZrO2 samples, the lifetime of the complexes is
determined by electron injection rate to the conduction band and trap states of these
semiconductor samples. The electron injection does not occur from the high vibrational
levels of TPC1 at the S1 state, and the sub-ps dynamics is dominated by solvation with a
time similar to that of TPC1/DCM (1.4 ps). It is in contrast with the previously observed
strong emission quenching of hot S1 state when interacting with titania. The example
emission transient and an interaction scheme are given in Figure 6.
Emission intensity
1.0
-
e
N
O
C
0.8
NC
O
h
Al2O3
0.6
ZrO2
0.4
ZnO
TiO2
0.2
0.0
0
2
4
6
8 10
200
400
600
Time / ps
Figure 6.
Up: fs-emission transients of TPC1 interacting with different semiconductors show electron injection dynamics
from 100 fs to 10 ps. Down: relative presentation of energy diagram of TPC1 S0 and S1 states and of the used
semiconductor material at their valance (VB) and conduction (CB) bands. The scheme gives the value of the
averaged fluorescence time constants resulting in the interaction of TPC1 with the semiconductors and electron
injection.
8) We have studied the interaction between an organic dye for photovoltaics (TPC1) and
titanium-doped mesoporous MCM-41 materials in semi- and complete solar cells. We
found that, from the point of view of interfacial charge separation dynamics, this material
performs better than the classical titania nanoparticles film, having similar electron
injection time from the dye to titania, slower electron recombination between titania and
the dye (Figure 7), and faster dye regeneration by electrolyte. However, despite potentially
larger absorption surface, the total dye loading in MCM-41 based structures is still lower
than for nanoparticles. Moreover, there is a severe limitation in the short circuit
photocurrent (Figure 8), probably due to low electron conductivity along silica-based
channel walls of MCM-41. Therefore, further modifications of MCM-41 structure is
necessary to improve dye loading and electron transport in order to take advantage of these
materials in constructing more efficient solar cells.
(A)
0.4
Normalized A
0.2
0.0
-0.2
-0.4
TPC1/NP
TPC1/TiMCM-41
-0.6
-0.8
-1.0
450
500
550
600
650
700
Wavelength, nm
(B)
Normalized A
0.0
-0.5
TPC1/NP,=7 s
TPC1/TiMCM-41, =19 s
-1.0
0
10
20
30
40
50
60
Time, s
Figure 7.
(A) Normalized transient absorption spectra of TPC1 interacting with nanoparticles (TPC1/NP) and
TPC1/TiMCM-41 films measured in flash photolysis (B) Normalized bleach recovery kinetics (measured at 470
nm) and 1-exponential function fit (with the given time constant) for TPC1/NP and TPC1/TiMCM-41.
Current, mA cm
-2
2.0
1.5
TPC1/NP
TPC1/TiMCM-41 x 10
1.0
0.5
0.0
0
100 200 300 400 500 600 700
Voltage, mV
Figure 8.
J-V curves for TPC1/NP and TPC1/TiMCM-41 solar cells. The photocurrent values of TiMCM-41 sample are
multiply by 10 for a better presentation.
Statement on the use of resources
All the facilities and equipment of the host group have been used according to the
workplan, except for the single molecule spectroscopy setup due to the lack of proper system
to study (dyes confined in mesoporous material of long enough fluorescence lifetime).