Integrating metalloporphycenes into p-type NiO-based dye-sensitized solar cells

Friedrich-Alexander-University Erlangen-Nuremberg, Department of Chemistry and Pharmacy & Interdisciplinary Center for Molecular Materials, Egerlandstr. 3, 91058 Erlangen, Germany. E-mail: dirk.guldi@fau.de, ruben.costa@fau.de. Friedrich-AlexanderUniversity Erlangen-Nuremberg, Department of Chemistry and Pharmacy, Henkestr. 42, 91054 Erlangen, Germany. Friedrich-Alexander-University Erlangen-Nuremberg, Department of Chemistry and Pharmacy, Computer Chemistry Center, Nägelsbachstr. 25, 91052 Erlangen, Germany.

70 ml of a saturated sodium bicarbonate solution were added and the mixture was heated at reflux for 30 minutes.The solution was allowed to cool to room temperature and the layers were separated.The organic layer was washed with water twice, stripped of solvent and purified by column chromatography (silica gel, eluent: chloroform / hexanes = 3 / 1).The green 9-formylvinyl-2,7,12,17-tetra-n-propylporphycenato nickel(II) 3 was eluted first followed by the blue 3-formylvinyl-2,7,12,17-tetra-n-propylporphycenato nickel(II) 2.  9-Formylvinyl-2,7,12,17-tetra-n-propyl-porphycenato nickel(II) 3 (20 mg, 33.9 µmol), malonic acid (35 mg, 340 µmol) and ammonium acetate (26 mg, 340 µmol) were dissolved in 6 ml of a mixture of tetrahydrofuran and acetic acid (1/1).The reaction mixture was heated at 70 °C for one hour and allowed to cool down to room temperature.After the addition of water it was filtrated.The residue was taken up in tetrahydrofuran, stripped of solvent with a stream of N 2 gas and purified by column chromatography (silica gel, eluent: methylene chloride / methanol / acetic acid = 900 / 115 / 13.5).The solvent was removed with a stream of N 2 gas to yield porphycene P2 as a dark green powder.

3-Formylvinyl
Yield: 10.5 mg (15.5 µmol, 45.9%) 1 H NMR and 13 C NMR: As already reported in the literature for similar molecules the strong aggregation does not allow an interpretation of the spectra.was recorded with a Shimadzu UV-3102 PC UV-Vis-NIR Scanning Spectrophotometer in the range 300 -750 nm.
Square wave voltammetry was performed in a range between -1.7 to +1.7 V -Figure S13 and Table S5.1), the size of nanoparticles was estimated to be at around 16.4 nm. 4 (1) L as the size of nanoparticles that can also be expressed by L = p ⋅ dhkl (p is the Number of the lattice planes, dhkl is the distance between them).The factor β is the full width at half maximum (FWHM) in radian, which is measured at the glancing angle θ.Furthermore, K is the form factor, which is defined by the form of the crystallites.If the crystallite is supposed to be spherical, K equals 0.89.
In Long term photostability for all porphycenes was proven by periodically measuring the absorption of sensitized transparent NiO electrodes by means of steady-state UV/Vis absorption spectroscopy in the wavelength range where only the absorption features of the dye were present, meaning from 550 to 750 nm.In this range absorption or reflection features of the NiO electrodes did not influence the measurements.The soaking conditions were set in correlation to the results from the absorption kinetics measurements.Starting with a measurement at every 10 min for the first 30 min, the intervals time was set to 30 min until an overall time of 120 min was reached.16 h after the beginning of the series a last measurement was performed.In order to proof if the evaluation process on the p-DSSC devices has any effect on the stability of the porphycene sensitizers steady-state UV/Vis absorption spectroscopy of the dyed electrodes was additionally measured.Therefore, the electrodes were irradiated under AM 1.5 conditions, and their UV/Vis spectra were continuously recorded every 5 min for the first two minutes, then in 10 min intervals until the time period reached 70 min -Figure S11.
Electrochemical impedance spectroscopy (EIS) was performed under light and dark conditions.Measurements were performed under V oc conditions in a frequency range of 0.01 Hz to 100 kHz.The voltage amplitude was set to be 10 mV.The received impedance data in the form of Nyquist plots was fitted by using the Nova 1.9 software to a circuit model that simulates the impedance of interfaces between the different components of DSSCs -Figure S13. 5 It consists of a sequential arrangement of sheet resistance R s and two resistances R 1 and R 2 , the latter each in parallel with the corresponding constant phase elements CPE 1 and CPE 2 .R s represents all the ohmic series resistances of the FTO substrates, and electrical contacts.The R 1 and the CPE 1 both relate to the electrolyte/platinum counter electrode interface in the high frequency range.Here, the charge transfer from the platinum counter electrode to the electrolyte takes place.In addition, charge recombination processes at NiO electrode/sensitizer/electrolyte interface is linked to R 2 and CPE 2 .The earlier presents the charge transfer resistance either in the light (R i ) or in the dark (R d ), while the latter is related to the chemical capacitance. 5,6 inally, the fitting results of the CPE 2 were transformed in pseudo capacitances C 2, giving a realistic picture of the electrical properties of at the NiO electrodes.The recombination constant k eff for the back reaction under light conditions was determined by taking the frequency at the maximum ω max of the arc in the Nyquist plot that was related to R 2 (6). 7 Finally, the hole lifetime τ h in the NiO electrode was calculated from the Bode phase plots.
The frequency of the maximum f max corresponding to the response of the NiO electrode in the low frequency range was used for this purpose (7). 8(7)   The data received from EIS in the dark gives a clear overview about the processes at the electrode/electrolyte interface when the sensitizer is not being excited. 5,9,10 Rcombination at this interface in absence of the sensitizer can be evaluated by fitting the Nyquist plot in accordance to the circuit model -Figures S14 and S15, Tables S5 and S6.

Computational calculations
2][13][14] The geometry of the NiO cluster was fixed at the experimental crystal structure of bulk NiO during the simulations.
The porphycenes P1, P2, and P3 were dissolved in DMF.The concentration was kept at c = 10 -4 M. Tetrabutylammonium hexafluorophosophate (TBAFP) was used as conducting salt (c = 0.1 M).Ferrocene redox couple (Fc/Fc + ) was used as electrochemical reference.The scan rate was adjusted to 0.05 V s -1 .The setup consisted of a graphite working electrode, a platinum wire as counter and a silver wire as a quasi reference electrode.All measurements were performed after saturating the solvent with argon for at least 10 mins and keeping the argon flow overlaying the solution during the data recording.Fabrication of transparent p-type NiO electrodes was performed by preparing a precursor solution.In details, 1 g of NiCl 2 (Aldrich, 98 %) and 1 g of Synperonic F108 (Fluka) were dissolved in a flask containing a mixture of 3 ml of Millipore water and 8 ml of ethanol.Stirring led to a clear green solution that was rested for 3 days at 30 °C.Centrifugation (Thermo Scientific Multifuge X1R equipped with Fiberlite F15 rotor) was performed at 12,000 rpm for 3 h in order to remove microcrystals, which prevent the formation of suitable films.Previous to film processing, fluorine doped tin oxide glass slides (FTO, 8 Ω/square, Pilkington, XOP Glass, Spain) were successively cleaned with isopropanol, 0.1 M hydrochloric acid solution, tenside solution, water, and isopropanol by means of ultrasonication (Elma Elmasonic P, frequency at 37 kHz, power at 100 %) for 15 min at each step, followed by cleaning in a ozone lamp (JELIGHT COMPANY, INC.UVO-Cleaner Model # 342-220) for 20 min.NiO films were processed with the help of a rackel machine (Zehnter ZAA 2300) and a Scotch® tape mask on the FTOs.After baking at 400 °C for 30 min in a muffle furnace (Nabertherm L9/11 P330) the resulting electrodes were characterized by means of field-emission scanning electron microscopy (Zeiss Gemini 55 Ultra), X-ray diffraction (Bruker D8 Advance, λ (Cu Kα) = 0.154 nm), steady-state UV-Vis absorption spectroscopy (Jenoptik Specord S600), and profilometry (Bruker Dektak XT).In line with the SEM images (Figure S5), profilometry indicates an electrode thickness of 1.2 µm.The transparency of the NiO electrodes was proven by UV-Vis transmission spectroscopy, showing a high transparency of up to 90 % from 350 to 750 nm -Figure S7.The XRD spectrum reveals the face centered cubic crystallinity of the NiO -Figure S6.Using the full width at half maximum of the (200) NiO peak and the Debye-Scherrer equation (

4 m
order to get insight into the dye adsorption behavior on the NiO electrodes steadystate UV-Vis absorption kinetics were performed.Therefore, the transparent NiO electrodes were soaked into the corresponding dye solutions of P1, P2 and P3 (c = 10 -4 M).The absorption was subsequently monitored for every 10 min during the first 30 min by following the increase of the Q-band of the dyes that were adsorbed on the electrodes.Then the time steps were increased to 30 min until 150 min overall soaking time.The stagnating absorption of the sensitized NiO electrodes that became already apparent after 90 min induces that the maximum of the dye uptake process had finished and that a monolayer of dye molecules was established on the NiO surface.Bathochrome red shifts of the observed Q-bands (P1: 620 nm → 632 nm, P2: 628 nm → 638 nm, P3: 627 nm → 641 nm) indicated successful immobilisation of the porphycenes as shown for P2 -FigureS9.The sensitized electrodes were finally assembled with a platinum covered counter electrode in order to test the different porphycenes in relation to their ability to work as sensitizers in p-type dye-sensitized solar cells (DSSCs).Construction of the counter electrode was done by drilling two holes of 1 mm in diameter into a FTO slide and subsequently cleaning of the slide as described above.Finally, the conductive surface of the FTO substrate was covered with a 5 mM solution of H 2 PtCl 6 (Aldrich, ~ 38 % Pt) by drop-casting and then backed at 390 °C for 15 min.The p-type NiO photocathode and the counter electrode were placed together using a Syrlin foil (Solaronix, 25 µm) while the space between the two electrodes was filled with electrolyte.The latter consisted of a mixture of LiI and I 2 (molar ratio of 5 : 1) in acetonitrile.The concentration of LiI was 1.0 M. The prepared p-type NiO DSSCs were measured under standard conditions, that is AM 1.5 and incident light power density P 0 = 100 mW•cm -2 .Photocurrent (J-V) curves were monitored with a potentiostat (Metrohm Autolab AUT 83394, Nova 1.9) in the voltage range -0.2 to +0.05 V. The€ L = K ⋅λ ( β ⋅cosθ )measurements were performed for at least 5 devices of each type in order to test the reproducibility of the results.By using characteristic data points in the J-V curves such as the open-circuit voltage (V oc ), the short-current density (J sc ), the current-density and the voltage at the maximum power point (J m and V m , respectively), the fill factor (2) and the efficiency (3-to-current-efficiency (IPCE) was recorded in the range between 350 to 780 nm by illuminating the p-type NiO DSSCs with a xenon arc lamp over a Cornerstone 260 1/Monochromator equipped with a Merlin digital radiometric lock-in-system.

Figure S5
Figure S5 Reduction and oxidation potentials of P1, P2, and P3 in reference to the valence band (VB) of NiO and to the redox potential of the I 3 -/I -couple.All potentials are given versus normal hydrogen electrode (NHE).

Figure S6
Figure S6 SEM images of transparent NiO electrodes -top view on the left and cross sectional view on the right.

Figure S7 X
Figure S7 X-ray diffraction (XRD) pattern of NiO on FTO.The characteristic peaks of face centered cubic NiO are clearly identified (diamonds) despite the presence of the FTO signals (circles).

Figure S8
Figure S8 Steady-state absorption and transmission spectra of the transparent NiO electrode.

Figure S9
Figure S9 Upper part -Absorption kinetics of P1 (blue), P2 (green) and P3 (grey) recorded at the maxima of steady-state absorption spectra of the dyed transparent NiO electrodes for increasing soaking times (P1 at 630 nm, P2 at 638 nm, and P3 at 641 nm).The NiO electrodes were soaked in DMF solution, c = 1.0⋅10 -4 M. Lower part -steady-state absorption spectra recorded after 90 min for P1, P2 and P3 on transparent NiO electrodes.

Figure S11
Figure S11 Absorption spectra of P1 (a), P2 (b) and P3 (c) attached onto transparent NiO electrodes over the time under ambient light (left) and 1 sun/ AM 1.5 illumination (right) conditions.

Figure S12
Figure S12 Upper-part -Relative efficiency versus time plot of P2 sensitized NiO p-type DSSC.Lower Part -Correlated current density versus illumination intensity plot.The fitting is represented by the blue line.

Figure S13
Figure S13Circuit model that was used to fit the electrochemical impedance spectroscopy (EIS) data.Here, R 2 and CPE 2 (constant phase element) represent the resistance and the chemical capacitance at the electrode/sensitizer/electrolyte interface.5

Figure S14
Figure S14 Nyquist plots of EIS data determined for p-type NiO DSSCs sensitized with P1 (a), P2 (b), and P3 (c) measured under light (open circles) and dark (filled circles) conditions.Fittings to the model in Figure S13 are presented by lines.

Figure S15
Figure S15 Bode phase plots of EIS data determined for p-type NiO DSSCs with P1 (a), P2 (b), and P3 (c) measured under light (open circles) and dark (filled circles) conditions.