Control of Porphyrin Dye Aggregation Using Bis ( 4-pyridyl ) Alkanes in Dye Sensitized Solar Cells

The aggregation of sensitizer molecules on the surface of photoanode is a serious issue that can affect the photovoltaic performance of dye-sensitized solar cells. Prevention of dye agglomeration, therefore, is critical. Traditional methods of aggregation control are either synthetically challenging or technologically difficult and expensive. In this article, the use of bis(4-pyridyl)alkanes to control porphyrin dye aggregation is presented. Three bis(4-pyridyl)alkanes – bis(4-pyridyl)butane L4, bis(4-pyridyl)octane L8 and bis(4-pyridyl)decane L10 were synthesized. These bis(4-pyridyl)alkane ligands are axially attached to the metallic center of synthesized porphyrin dye P. The complexes was obtained by mixing the solutions of dye P and each ligand (L) in 2:1 ratio 1 h before the soaking step. As a result three cells were prepared: P-L4, P-L8 and P-L10. The performance of these cells were compared with a reference cell which was prepared from porphyrin dye P only. IPCE analysis demonstrated the highest dye load in P-L4 cell which was ascribed to lowered dye aggregation. Photovoltaic analysis showed improved short circuit current density due to suppressed dye aggregation caused by the complexation of the porphyrin dye P with the linker L4. As a result the overall cell efficiency increased to 42% demonstrating the successful utilization of the (4-pyridyl)alkane linker complexes with porphyrin dye. Article info Received: 4 September 2018 Received in revised form: 13 October 2018 Accepted: 10 December 2018


Introduction
Energy is a crucial element of human life.The increasing population and growing economy requires more energy.The use of fossil fuels as energy source has drawbacks due to environmental effects and supply limitations.Therefore, it is critical to search for sustainable sources of energy.Sunlight is one of green and renewable sources of energy which can fully satisfy the energy demand of whole planet [1].Dye sensitized solar cells (DSSCs) represent third-generation solar cells and are regarded as low cost, high efficiency and easily prepared cells [2].The current highest record efficiency of DSSCs is 13% which was obtained using porphyrin dye SM315 [3].Porphyrins are unique dyes due to their strong absorption around 450 nm and 600-650 nm.They have great elec-tron donating-accepting properties which is important for the sensitizers.Also, porphyrins have several positions that various substituents can be attached to.Unlike DSSCs made of Ru-dyes, the efficiency of porphyrin sensitized solar cells has been constantly increasing since 1990s [4].One of the drawbacks of porphyrin dyes is their tendency to aggregate on semiconductor surface.Dye aggregation on semiconductor surface is a serious issue that needs to be prevented for the efficient performance of dye sensitized solar cells (DSSCs).The aggregation of sensitizers decreases the charge injection kinetics.Photoexcited aggregated molecules do not participate in electron injection due to charge annihilation.Moreover, dye aggregation may cause increased charge recombination with dye cation [5].Porphyrins, a well-known class of DSSC sensitizers, also tend to aggregate on photoanode surface due to the planarity of its macrocycle [6].Methods of aggregation control include the incorporation of bulky groups on dye structure which can be synthetically challenging and/or the use of co-adsorbents that requires preparation of cocktail with optimal ratio.In this research, the authors present the use of bis(4-pyridyl)alkanes as effective spacers between porphyrin dyes that prevent dye aggregation by axially coordinating them with the dye metal atom.This allows the use of dyes with no bulky groups which simplifies the synthetic protocol and makes possible to control the distance between sensitizer molecules.

Experimental
Commercially available chemicals were obtained from Sigma Aldrich and Thermo Fisher Scientific.Some solvents, like pyrrole, were purified using rotary evaporator before use in synthesis.Most of the synthesis reactions are sensitive to moisture and oxygen, and, thus, reactions were performed in the inert atmosphere of nitrogen.Chemical structures of the synthesized products were confirmed via NMR spectroscopy method on JEOL JNM-ECA Series FT NMR spectrometer.

Synthesis of porphyrin dye
Porphyrin dye P was synthesized according to the scheme shown in Fig. 1.
2,2'-dipyrromethane (1).Formaldehyde (1.5 g, 50 mmol) was dissolved in purified pyr-role (347 ml, 5 mol) in round-bottom flask stirring for 10 min with purging nitrogen.Then, temperature of solution was increased to 55 °C and InCl 3 (1.1 g, 5 mmol) was added to stirring solution.After 2.5 h, NaOH (6 g, 0.15 mol) was added to the reaction mixture.Reaction was stopped after 1 h and cooled down to room temperature.Solution was filtrated, and solvent was evaporated using rotary evaporator.Thin-layer chromatography (TLC) showed presence of 3 products.Using column chromatography, three products of the reactions were separated and via NMR 2,2'-dippyromethane (white solid) was identified.
10,20-dibromo-5,15-diphenylporphyrin (3).5,15-diphenylporphyrin (750 mg, 1.6 mmol) was dissolved in anhydrous DCM and stirred at room temperature for 10 min.Then, temperature of the reaction was decreased to 0 °C in ice bath for further addition of NBS (N-bromosuccinimide) (313 mg, 1.7 mmol).Reaction was quenched by adding acetone after 6 h.Solvent was evaporated and product was separated from by-products in column chromatography.
Zinc-10,20-dibromo-5,15-diphenylporphyrinate (4).10,20-dibromo-5,15-diphenylporphyrin (270 mg, 3.7 mmol) and zinc acetate (1.3 g, 37 mmol) were dissolved in the mixture of DCM and methanol and stirred at room temperature for 3 h.Reaction was quenched by adding excess water.Aqueous and organic phases were separated in separatory funnel and organic phase were washed several times with DI water, dried over MgSO 4 and rotovapped to produce desired product.

, 1 -d i p h e n y l -1 0 -( 4 -o x y h e x y l p h e n y lethynyl)-20-(4-carboxyphenylethynyl)porphyrinatato zinc (P).
Tetrabutylammonium fluoride (4.9 ml, 10.2 mmol) was added to compound 5 (0.9 g, 1.02 mmol) dissolved in THF and stirred for 2 h.Then, mixture of DCM and water was added to reacting mixture to produce aqueous and organic layers.Organic layer was separated and solvent was evaporated to produce intermediate product.4-iodobenzoic acid (1.19 g, 4.8 mmol), TEA (5 ml), Pd 2 (dba) 3 (0.19 g, 20 umol), triphenylarsine (2.5 g, 8 mmol) and 4-oxyhexyl iodobenzene (1.48 g, 4.8 mmol) were added to the intermediate compound (2.1 g) dissolved in THF.Reacting mixture stirred for 5 h in nitrogen atmosphere at 85 °C, and then cooled to room temperature and solvent was evaporated.Products of reaction were separated using column chromatography and structure of desired product was confirmed by NMR.

Synthesis of the linkers
The linkers were synthesized according to the scheme shown in Fig. 2.

Bis(4-pyridyl)
alkane.4-methylpyridine (1.57ml) was dissolved in THF and temperature of the solution was decreased to -83.6 °C in ethyl acetate -liquid nitrogen bath.n-butyllithium (13.1 g) was added to the solution dropwise and reaction stirred for 2 h.Then, 10 ml of THF was added and temperature of mixture was again decreased to -83.6 °C and dibromoalkane ligand (1.1 ml) was added dropwise (ligands: L4 -1,4-dibromobutane, L8 -1,8-dibromooctane and L10 -1,10-dibomodecane).Reacting mixture was left stirring for 24 h.Reaction was quenched by adding 10 ml water.Organic layer was separated and solvent was evaporated.TLC shows three spots, indicating 3 products of the reaction, which are separated using column chromatography and NMR confirms the structure of desired product.

DSSC fabrication
Transparent TiO 2 paste (Dyenamo DN-EP03) was applied on the pre-treated with 40 mM TiCl 4 fluorine-doped tin oxide (FTO) glass by doctor blading and allowed to air dry for 2 h.Then the film was sintered stepwise in the oven (125 °C -5 min, 325 °C -10 min, 425 °C -15 min, 500 °C -30 min).After cooling it was again treated with 40 mM TiCl 4 and paste prepared from P25 (Aldrich 718467) was applied on top as scattering layer which was followed by stepwise sintering in the oven.The cooled films were post-treated with 40 mM TiCl 4 and finally sintered at 500 °C for 30 min.The cooled electrodes were soaked with 0.2 mM dye (P)/dye-linker (P:L (2:1) where dye concentration is 0.2 mM) solutions in dichloromethane-methanol (10:1) solutions for 4 h in refrigerator before assembling into DSSC using counter electrode and electrolyte.The counter electrode was made of platinum (Dyenamo DN-EP01) and the electrolyte was iodine/iodide based solution (Dyenamo DN-OD05).
Wavelength, nm Fig. 3. UV/Vis absorption spectra of the porphyrin dye P.
ThermoScientific Evolution 60S UV-Visible Spectrophotometer.The optimized structures were done in Gaussian 16 [7] using M06-2X hybrid functional [8] and Pople's split-valence double-zeta basis set with added d polarization function on non-hydrogen atoms.

Results and discussion
The obtained porphyrin dye P was analyzed by UV/Vis spectroscopy to observe its light absorption characteristics in dichloromethane-methanol (10:1) solution (Fig. 3).The dye P has two maximums: strong absorption at 451 nm and moderate light harvest at 660 nm.
The optimized structures were obtained from computational analysis performed using Gaussian 16 program as shown in Fig. 4. The analysis allowed to calculate the distance between two complexed porphyrin planes.Porphyrins are separated by 15.7 A o in P-L4 complex, the shortest distance due to the linker size.For P-L8 the distance is 20.4A o and it further increases to 22.8 A o for P-L10 complex as the linker gets bigger.
Incident-photon-to-current-conversion-efficiency (IPCE) of the cells is shown in Fig. 5.This parameter shows how effective the dyes are in converting the absorbed photon into current.Two regions can be distinguished: around 450 nm and 680 nm which correspond to UV/Vis absorption data of the dye solutions (see Fig. 3.).It is clearly observable that cell P-L4 has the highest IPCE value indicating lowered aggregation and also higher dye load.This could be due to better complexation of the linker L4 with the porphyrin dye P compared to others.On the other hand the linker L4 is the shortest in length and therefore it occupies less space compared to other dye-linker complexes which gives higher chance for the P-L4 complex to attach to TiO 2 surface.This was confirmed by optimized structures of porphyrin-ligand complexes (Fig. 4.).Photovoltaic performance of the dyes is presented in Fig. 6 and the corresponding data is shown in Table .I-V characteristics demonstrated the same trend similar to IPCE (see Fig. 5).Due to maximum IPCE value, the complex P-L4 produced the highest current, 3.23 mA/cm 2 , among all samples.The improvement in Jsc value reached over 40% compared to the reference cell P with Jsc equal to 2.27 mA/cm 2 , indicating suppressed dye aggregation in P-L4.P-L10 complex showed similar performance to reference cell P with small increase in Jsc following the trend in IPCE.The cell with dye complex P-L8 produced the lowest current of all.This was explained by the low IPCE value of the system among all samples.Unlike current density, the photovoltages and fill factors of the cells were found to be the same.The photovoltaic parameters are shown in Table .The highest current of P-L4 led to the highest efficiency among all samples -1.42%, indicating 42% improvement and confirming the beneficial effect of the L4 linker in suppressing dye aggregation.

Conclusions
Thus, this work demonstrated the use of bis(4-pyridyl)alkane linkers axially attached to porphyrin dye P in suppressing the dye aggregation.Three linkers were synthesized together with porphyrin dye.Porphyrin-linker complexes were prepared in 2:1 ratios with final dye concentration 0.2 mM.The photovoltaic measurements showed the improvement in current density in P-L4 complex leading to 42% improvement of cell efficiency.The increase in Jsc was ascribed to lowered dye aggregation which led to improved IPCE value of the cell compared to reference cell.

Table
Photovoltaic parameters of the cells