Improved Performance Induced by in-situ Ligand. Exchange Reactions of Copper Bipyridyl Redox. Couples in Dye-Sensitized Solar Cells

Transcription

1 Electronic Supplementary Material (ESI) for Chemical Communications. This journal is The Royal Society of Chemistry 2018 Improved Performance Induced by in-situ Ligand Exchange Reactions of Copper Bipyridyl Redox Couples in Dye-Sensitized Solar Cells Yujue Wang and Thomas W. Hamann* Department of Chemistry, Michigan State University, 574 S Shaw Lane, East Lansing, USA, Experimental Synthesis of copper complexes Copper(II/I) bis(6,6-dimethyl-2,2-bipyridine) triflate ([Cu(dmbpy)2] + /[Cu(dmbpy)2] 2+ ) complexes were made using previous published method. 1 Tetrakisacetonitrile copper(i) triflate/copper(ii) triflate and 6,6 -dimethyl-2,2 -bipyridine (dmbpy) were mixed in a 1 to 2.1 ratio in acetonitrile/dichloromethane mixture. Red/green powders crashed out upon the addition of diethyl ether. Further purification was achieved by recrystallization of the red/green powders from acetonitrile or dichloromethane with diethyl ether. Copper(II) tetra(4-tert-butylpyridine) triflate ([Cu(TBP)4] 2+ ) was prepared by mixing copper(ii) triflate and 4-tert-butylpyridine in a 1 to 4.1 ratio (or higher) in acetonitrile. Purple powder crashed out upon the addition of diethyl ether to the solution. Further purification was achieved by recrystallization from acetonitrile with diethyl ether. 1

2 b) Figure S1. Crystal structure of a) [Cu(dmbpy)2] + and b) [Cu(dmbpy)2] 2+, with triflate as counter ion. 2

3 c) Figure S2. Proton NMR of a) free dmbpy ligand, b) [Cu(dmbpy)2] + and c) [Cu(dmbpy)2] 2+ in AcN-d3, with DCM as an internal standard. 3

4 Electrochemistry Cyclic voltammetry (CV) measurements were performed with an Autolab PGSTAT128N under an inert atmosphere. A Pt disk or a glassy carbon disk was used as working electrode. A high surface area Pt mesh was used as counter electrode. A home-made Ag/AgNO3 reference electrode was used for all the three-electrode measurements. Ferrocene was also used as an internal standard to calibrate the potential of the reference electrode. 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) was used as supporting electrolyte. a) b) Figure S3. CV s of 4mM a) [Cu(dmbpy)2] + and b) [Cu(dmbpy)2] 2+ in acetonitrile with different scan rates. Colors from light to dark represent scan rate from to 0.5 V/s. A peak current versus square root of scan rate (ip vs. v 1/2 ) is plotted in the insert. The linear behavior and the ratio between the cathodic and anodic peak current indicates the good electrochemical reversibility of the redox couple. Glassy carbon disk was used as working electrode and the same reversibility was observed also on Pt disk electrode. 4

5 Figure S4. CV of 4 mm [Cu(dmbpy)2] 2+ with the addition of aliquots of TBP on glassy carbon working electrode. From top to bottom, the equivalents of TBP to the copper complex were 0, 3, 13 and 25. The scan rate was 0.1 V/s. The absorption spectra of all copper complexes were measured with Lambda 35 (PerkinElmer) spectrometer. Solutions were prepared in the glovebox under an inert atmosphere with dry solvent. Screw-cap quartz cuvette with 1 cm path length was used for the solution measurement. A solution of 4 mm [Cu(dmbpy)2] 2+ was made in dry acetonitrile under an inert atmosphere. Aliquots of TBP, from 0 to 40 eq. were added. UV-Vis spectra were measured for each sample using a screw-cap cuvette. 5

6 Titration of the [Cu(TBP)4] 2+ solution with aliquots of dmbpy does not show a complete transformation to [Cu(dmbpy)2] 2+, as indicated by the absorption spectra, with up to 6.5 equivalents. As shown in figure S5, a product with a higher extinction coefficient (180 M -1 cm -1 ) was observed. This green colored complex was not isolated, but it shows absorption peaks at 707 nm and 950 nm which matches the black curve in figure 2b in the main text. Therefore, we believe this complex is one of the intermediate species formed during the titration of [Cu(dmbpy)2] 2+ with TBP. This result shows that the ligand substitution reactions of the Cu(II) complexes are reversible and can be described generally by the equilibrium: Figure S5. Titration of [Cu(TBP)4] 2+ with aliquots of dmbpy in acetonitrile. The base concentration of [Cu(TBP)4] 2+ was 8 mm, with 0, 0.53, 1.1, 1.5, 2.0, 2.4, 3.0, 4.0, 5.3, 6.5 equivalents dmbpy (bottom to top) introduced to the solution. The end-point spectrum showed the same feature as the intermediate complex s spectrum, the black curve in figure 2b in the main text. 6

7 CV s of a solution containing 4 mm [Cu(TBP)4] 2+ in acetonitrile with 0.1 M LiOTf as a supporting electrolyte showed no well-defined redox peaks, shown below in figure S6 a). The UVvis spectrum of this solution is also shown in figure S6 b). Upon the addition of 8 mm dmbpy (corresponding to 2 equivalents), a distorted redox wave appeared which is comparable to wave shown in figure 1 of the main text and figure S5 above where 4 mm [Cu(dmbpy)2] 2+ was titrated with TBP. The absorption spectrum shown in figure S6 d) also changed to resemble the intermediate complex [Cu(dmbpy)(TBP)x] proposed. Bulk electrolysis was performed by applying a potential sufficiently negative to reduce Cu(II) species to Cu(I). After 1.7 C charge was passed, the anodic current increased in the CV shown in figure S6 e) consistent with formation of [Cu(dmbpy)2] +. The corresponding absorption spectrum from both the extinction coefficient and the absorption peak wavelength confirmed the near complete formation of ~4 mm [Cu(dmbpy)2] +. These results show that the ligand exchange is reversible and that the [Cu(dmbpy)2] + complex is strongly favored over a [Cu(TBP)x] + complex which we attribute to the chelate effect. 7

8 a) b) c) d) e) f) Figure S6. Plots of CV s (left) and absorption spectra (right) for solutions containing: a, b) 4 mm [Cu(TBP)4] 2+ in acetonitrile containing 0.1 M LiOTf. (blue lines); c, d) 4 mm [Cu(TBP)4] 2+ in acetonitrile containing 0.1 M LiOTf with 8 mm dmbpy (green lines); e, f) 4 mm [Cu(TBP)4] 2+ in acetonitrile with 8 mm dmbpy after bulk electrolysis at V vs. Fc +/0 which resulted in passing 1.7 C of charge cathodically (red lines). Absorption spectrum of 10-fold diluted spectrum is shown in yellow line of plot f). Working electrode was platinum disk and the scan rate was 0.1 V/s. 8

9 [Cu(dmbpy)2] 2+ solutions were made in AcN-d3 and aliquots of TBP were added: from 0 to 28 eq. Proton NMR was measured using a J-Young NMR tube under an inert atmosphere with Agilent DDR2 500 MHz NMR spectrometer at room temperature. a) b) c) Figure S7. Proton NMR of [Cu(dmbpy)2]2 + with a) 0 eq. of TBP, b) 5 eq. TBP, c) 25 eq. TBP. In all samples, DCM was added as an internal standard. 9

10 Cell assembly Photoelectrodes were prepared on 12 Ω/cm 2 FTO-coated glass (Hartford Glass) cleaned by sonicating in soap water solution, deionized water, isopropanol, and acetone, and then heated to 450 C for 30 min. Blocking layers of TiO2 were deposited on the precleaned FTO-coated glass using 500 ALD cycles of titanium isopropoxide (TIPS, Aldrich) and water as precursors with a Savannah 200 instrument (Cambridge Nanotech, Inc.). TiO2 was grown at 225 C using reactant exposure times of 0.3 s and s for TIPS and H2O respectively, and nitrogen purge times of 5 s between exposures. The thickness of the TiO2 blocking layer was determined to be 10 nm by ellipsometry performed on Si samples coated concurrently in the ALD reactor. A transparent TiO2 nanoparticle layer (electrode area 0.36 cm 2 ) was prepared by doctor blading a paste of TiO2 nanoparticles (Ti-Nanoxide HT/SP, Solaronix) on the FTO. The resulting electrodes were annealed at 325 C for 5 min, 375 C for 5 min, 450 C for 5 min, and 500 C for 15 min in air. The nanoporous TiO2 film thickness, d, was measured using a Dektak3 Surface Profiler to be ~8 μm. TiCl4 post-treatment was done by immersing the electrodes in 40 mm TiCl4 solution at 80 C for 30 min, followed by annealing at 500 C for 30 min. The anodes were immersed in dye solution (0.2 mm solution of 3-{6-{4-[bis(2',4'- dibutyloxybiphenyl-4-yl)amino-]phenyl}-4,4-dihexyl-cyclopenta-[2,1-b:3,4-b']dithiophene-2- yl}-2-cyanoacrylic acid, D35cpdt, in 1:1 acetonitrile and tert-butyl alcohol). 2 mm chenodeoxylicacid were added during dye soaking. After hours, they were rinsed with acetonitrile. A ~25 μm thick Surlyn film (Solaronix) was sandwiched between the dyed anode and the platinized FTO electrode. Pressure was applied at 150 C to seal the cells. The electrolyte was filled by capillary force through the two pre-drilled holes on the platinum counter electrode and sealed with micro glass and Surlyn film. 10

11 The electrolyte used in DSSCs were prepared by the addition of 0.2 M [Cu(dmbpy)2]OTf, 0.02 M [Cu(dmbpy)2](OTf)2, 0.1 M LiOTf and different concentrations (0.1, 0.2, 0.3, 0.4, 0.5 M) of TBP, in acetonitrile. Solution potentials are shown below. Table S1. Solution potentials of the electrolyte with different concentrations of TBP. The solution potential was measured on both Pt and glassy carbon electrodes, versus Ag/AgNO3 (0.01M in AcN) reference electrode. Concentration of TBP / M Esol / V vs. Ag/AgNO

12 Device characterization Photoelectrochemical measurements were performed with Autolab PGSTAT128N interfaced with a Xe Arc Lamp. An AM 1.5 solar filter was used to simulate sunlight at 100 mw/cm 2. A 400 nm long-pass filter was used to prevent direct excitation of the TiO2 in all light measurements. A Horiba Jobin Yvon MicroHR was used for monochromatic light for IPCE measurements. The scan rate of both J-V measurements under illumination and in dark condition was 0.01 V/s. Figure S8. The dark current density of sandwich cells with electrolytes containing different concentrations of TBP. From right to left: 0, 0.1, 0.2, 0.3, 0.4, 0.5 M. 12

13 Electrochemical Impedance Spectroscopy All EIS measurements were performed in the dark with Autolab PGSTAT128N interfaced FRA2 module and controlled by NOVA software. The impedance spectra were recorded at applied voltages from -0.4 to -1.2 V, stepped in 25 mv increments, with a 10 mv alternating potential superimposed on the direct bias. Each impedance measurement consisted of frequency sweeps from 0.05 to 10 5 Hz in equally spaced logarithmic steps. Figure S9. Equivalent circuit used to fit the EIS data measured for sandwich cells under dark condition. RS is the series resistance. RT is the charge transport resistance inside TiO2 nanoporous structure. RCT is the charge transfer resistance between TiO2 and liquid electrolyte interface. Cμ is the chemical capacitance of TiO2. RPT and CPT are the resistance and capacitance on the counter electrode/electrolyte interface. 2 Figure S10. Chemical capacitance measured by EIS of DSSCs applying [Co(bpy)3] 3+/2+ electrolyte. Red: 0.2 M [Co(bpy)3] 2+ and 0.02 M [Co(bpy)3] 3+, 0.1 M LiOTf. Black: 0.2 M [Co(bpy)3] 2+ and 0.02 M [Co(bpy)3] 3+, 0.1 M LiOTf and 0.5 M TBP. 13

14 References 1. Williams, R. M.; Cola, L. De; Hartl, F.; Lagref, J.-J.; Planeix, J.-M.; Cian, A. De; Hosseini, M. W., Coord. Chem. Rev., 2002, 230, Bisquert, J., Phys. Chem. Chem. Phys., 2000, 2,