Interfacial versus Bulk Properties of Hole-Transporting Materials for Perovskite Solar Cells: Isomeric Triphenylamine-Based Enamines versus Spiro-OMeTAD

Here, we report on three new triphenylamine-based enamines synthesized by condensation of an appropriate primary amine with 2,2-diphenylacetaldehyde and characterized by experimental techniques and density functional theory (DFT) computations. Experimental results allow highlighting attractive properties including solid-state ionization potential in the range of 5.33–5.69 eV in solid-state and hole mobilities exceeding 10–3 cm2/V·s, which are higher than those in spiro-OMeTAD at the same electric fields. DFT-based analysis points to the presence of several conformers close in energy at room temperature. The newly synthesized hole-transporting materials (HTMs) were used in perovskite solar cells and exhibited performances comparable to that of spiro-OMeTAD. The device containing one newly synthesized hole-transporting enamine was characterized by a power conversion efficiency of 18.4%. Our analysis indicates that the perovskite–HTM interface dominates the properties of perovskite solar cells. PL measurements indicate smaller efficiency for perovskite-to-new HTM hole transfer as compared to spiro-OMeTAD. Nevertheless, the comparable power conversion efficiencies and simple synthesis of the new compounds make them attractive candidates for utilization in perovskite solar cells.


Experimental methods
1 H NMR and 13 C NMR spectra were recorded with Varian Unity Inova [300 MHz ( 1 H), 75 MHz ( 13 C)] and Bruker Avance III [400 MHz ( 1 H), 100 MHz ( 13 C)] spectrometers at room temperature. The data are given as chemical shifts δ (ppm) downfield from Si(CH 3 ) 4 . Infrared (IR) spectra were recorded using PerkinElmer Spectrum GX II FT-IR System. The samples of solid compounds were prepared as powders or in the form of KBr pellets. Mass spectrometry (MS) measurements were performed with the Waters SQ Detector 2.
Differential scanning calorimetry (DSC) measurements were carried out in nitrogen atmosphere with Perkin Elmer at DSC 8500 equipment at heating and cooling rates of 10 °C/min. Thermogravimetric analysis (TGA) was performed on Perkin Elmer TGA 4000 apparatus in nitrogen atmosphere at heating rate of 20 ºC/min. Melting points were measured with Electrothermal MEL-TEMP melting point apparatus.
Absorption spectra of dilute (10 -5 M) solutions in tetrahydrofuran (THF) were recorded on an UV−vis−NIR spectrometer Lambda 950 (Perkin-Elmer). Fluorescence spectra of dilute solutions in THF or toluene (10 -5 M) and of solid films of the compounds recorded with Edinburgh Instruments LS980 spectrometer.
Cyclic voltammetry (CV) measurements were carried out using a micro-Autolab III (Metrohm Autolab) potentiostat-galvanostat. A three-electrode cell equipped with glassy carbon working electrode, Ag/Ag (0.01 M in anhydrous acetonitrile) reference electrode and Pt wire counter electrode was employed. The measurements were done in anhydrous dichloromethane with tetrabutylammonium hexafluorophosphate (0.1 M) as the supporting electrolyte under nitrogen atmosphere at a scan rate of 0.1 V/s. The measurements were calibrated using an internal standard ferrocene/ferrocenium (Fc) system. Ionization potentials (I P ep ) of the films of the compounds were measured by photoelectron emission method in air as described before.
Hole drift mobilities were estimated by a time of-flight (ToF) method.
Device fabrication: Substrate preparation. Nippon FTO glass (10 Ω sq-1) was etched by a chemical method using zinc powder and HCl 4 M solution. The substrates were cleaned by sonication using firstly HellmanexTM III (Hellma GmbH) (2 vol% in deionized water), secondly deionized water, thirdly acetone, and finally ethanol as cleaning solvents. Thus, all substrates were cleaned by UV/ozone for 15 min.
Afterwards, compact TiO 2 layer was deposited by the spray pyrolysis method. Titanium diisopropoxide bis(acetylacetonate) was diluted in absolute ethanol (99.5%, Fischer Scientific), then deposited on substrates at 450° C and annealed 30 min at 450° C. Subsequently, a mesoporous TiO 2 layer was spin coated on the compact TiO 2 (4500 rpm for 18 s with a ramp rate of 2000 rpm s −1 ) using TiO 2 paste diluted in absolute ethanol. Thereafter, the substrates were annealed under dry flow at 450° C for 30 min. They    Hole transport layer. Different concentrations of 1, 2 and 3 were prepared and doped as discussed in the corresponding section. 70 mM spiro-MeOTAD solution in CB was doped by adding tBP, Li-TFSI and FK209 as p-dopants, in 3.3, 0.5 and 0.03 molar ratio, respectively, otherwise other thing is stated. The triple-cation perovskite solution was spin coated in two consecutive steps: first, at 1000 rpm for 10 s and, second at 4000 rpm for 30 s. Five seconds before the end of the spinning during the second step, 200 µL of CB antisolvent was eventually dropped on top of the film. Next, the films were annealed at 100 °C for 60-80 min. After that, the devices could reach room temperature, at which time the different HTLs were deposited at 4000 rpm for 20 s (2000 rpm s -1 ramp rate). To end, 80 nm thick gold electrode was thermally deposited under vacuum (active area of 0.25 cm 2 ).

Device characterization:
The solar cell devices were investigated under a 300 W Xenon light source (Oriel). The spectral mismatch between AM 1.5 G and the solar simulator was calibrated by a Schott K113 Tempax filter (Prazosopms Gas & Optik GmbH). The light intensity was calibrated with a silicon photodiode with an IR-cutoff filter (KG2, Schott). Current-voltage characteristics were applied by an external voltage bias while measuring the corresponding current with Keithley 2400. The voltage scan rate was 20 mV/s. The devices were covered with a black metal mask with an active area of 0.16 cm 2 . The IPCE measurement was performed by an EQE system with an LED light source (Ariadne EQE) in the DC mode without any voltage bias. Steady-state photoluminescence. PL measurements were performed using an Andor Kymera 193i spectrometer with a 600 l/mm grating blazed at 650 nm. Samples were excited with an OBIS 660 nm CW laser.             Figure S13. -J sc , V oc , FF, and PCE data distribution of devices with HTM 1, without and with PMMA interlayer, the latter deposited from different concentrations in CB solutions, in mg/mL. The photovoltaic parameters have been extracted from the backward J−V scans from 1.2 V to short-circuit current.