Nonspiro, Fluorene‐Based, Amorphous Hole Transporting Materials for Efficient and Stable Perovskite Solar Cells

Abstract Novel nonspiro, fluorene‐based, small‐molecule hole transporting materials (HTMs) V1050 and V1061 are designed and synthesized using a facile three‐step synthetic route. The synthesized compounds exhibit amorphous nature with a high glass transition temperature, a good solubility, and decent thermal stability. The planar perovskite solar cells (PSCs) employing V1050 generated an excellent power conversion efficiency of 18.3%, which is comparable to 18.9% obtained with the state‐of‐the‐art Spiro‐OMeTAD. Importantly, the devices based on V1050 and V1061 show better stability compared to devices based on Spiro‐OMeTAD when aged without any encapsulation under uncontrolled humidity conditions (relative humidity around 60%) in the dark and under continuous full sun illumination.


General methods and materials
Chemicals were purchased from Sigma-Aldrich and TCI Europe and used as received without further purification. The 1 H and 13 C NMR spectra were taken on Bruker Avance III (400 MHz) spectrometer at RT. All the data are given as chemical shifts in δ (ppm). The course of the  A mixture of 9,9-diethyl-9H-fluorene (1.11 g, 5 mmol), paraformaldehyde (0.33 g, 11 mmol), and 33% HBr solution in acetic acid (10 ml) was heated at 60-70 °C for 20 h. Upon cooling, the precipitates were filtered off and tree times washed with water and dried in vacuum, affording 1.47 g of pale white solid (72.0%). The product was recrystallized from toluene/n-hexane 1:1 gave as white crystals. Mp 148-150°C. The NMR spectra were identical to the corresponding spectra of the product referred in [1]; Anal. calcd for C19H20Br2: C, 55.91; H, 4.94; found: C, 55.68; H, 4.81. S3 2,7-Bis(3,6-dibromo-9H-carbazol-9-methyl)-9,9-diethyl-9H-fluorene (2) A mixture of compound 1 (0.82 g, 2 mmol) and 3,6-dibromo-9H-carbazole (1,30 g, 4 mmol) was dissolved in 15 ml of tetrahydrofurane and 0.68 g (12 mmol) of 85% powdered potassium hydroxide was added in small portions during 2-3 minutes. The obtained mixture was stirred at room temperature for 6 h. The part of solvent was removed in vacuum. Then obtained crystals of product 2 were filtered off and washed with water until it was neutral and three times with ethanol. The product was recrystallized from ethanol/tetrahydrofurane 2:1 gave as white crystals (1.82 g, 83.1%), Mp 293-295°C. 1

S5
A solution of compound 2 (1.34 g, 1.5 mmol), 4,4`-dimethoxydiphenylamine (2.06 g, 9 mmol) in anhydrous toluene (17 mL) was purged with argon for 30 minutes. Afterwards, palladium(II) acetate (6.7 mg, 0.03 mmol), tri-tert-butylphosphonium tetrafluoroborate (11.7 mg, 0.04 mmol) and sodium tert-butoxide (0.86 g, 9 mmol) were added and the solution was refluxed under argon atmosphere for 24 hours. After cooling to room temperature, reaction mixture was filtered through Celite, 50 mL of distilled water were added and extraction was done with ethyl acetate and distilled water. The organic layer was dried over anhydrous Na 2 SO 4 , filtered and solvent evaporated. The crude product was purified by column chromatography using 1:4 v/v acetone/nhexane as an eluent. The obtained product was precipitated from tetrahydrofurane into 15-fold excess of hexane. The precipitate was filtered off and washed with hexane to collect V1061 as a pale yellow -green solid. (1.70 g, 77.6%). 1

Hole drift mobility measurements
The samples for the hole mobility measurements were prepared by spin-coating the solution of the HTMs on the polyester films with conductive Al layer. The layer thickness was in the range of 2-4 m. The hole drift mobility was measured by xerographic time of flight technique (XTOF) [2][3][4]. Electric field was created by positive corona charging. The charge carriers were generated at the layer surface by illumination with pulses of nitrogen laser (pulse duration was 2 ns, wavelength 337 nm). The layer surface potential decrease as a result of pulse illumination S11 was up to 1-5 % of initial potential before illumination. The capacitance probe that was connected to the wide frequency band electrometer measured the speed of the surface potential decrease dU/dt. The transit time t t was determined by the kink on the curve of the dU/dt transient in double logarithmic scale. The drift mobility was calculated by the formula d 2 /U 0 t t , where d is the layer thickness, U 0 -the surface potential at the moment of illumination.

Thermal properties
Differential scanning calorimetry (DSC) was performed on a Q10 calorimeter (TA Instruments) at a scan rate of 10 K min -1 in the nitrogen atmosphere. The glass transition temperatures for the investigated compounds were determined during the second heating scan.  Figure S1. Photoemission in air spectra of the V1061 S12 Figure S2. Photoemission in air spectra of the V1050

Ionization potential measurements
The solid state ionization potential (I p ) of the layers of the synthesized compounds was measured by the electron photoemission in air method [5][6][7]. The samples for the ionization potential measurement were prepared by dissolving materials in CHCl 3 and were coated on Al plates precoated with ~0.5 m thick methylmethacrylate and methacrylic acid copolymer adhesive layer.
The thickness of the transporting material layer was 0.5-1 m. Usually photoemission experiments are carried out in vacuum and high vacuum is one of the main requirements for these measurements. If vacuum is not high enough the sample surface oxidation and gas adsorption are influencing the measurement results. In our case, however, the organic materials investigated are stable enough to oxygen and the measurements may be carried out in the air.
The samples were illuminated with monochromatic light from the quartz monochromator with deuterium lamp. The power of the incident light beam was (2-5)10 -8 W. The negative voltage of -300 V was supplied to the sample substrate. The counter-electrode with the 4.5×15 mm 2 slit for S13 illumination was placed at 8 mm distance from the sample surface. The counter-electrode was connected to the input of the BK2-16 type electrometer, working in the open input regime, for the photocurrent measurement. The 10 -15 -10 -12 A strong photocurrent was flowing in the circuit under illumination. The photocurrent I is strongly dependent on the incident light photon energy h. The I 0.5 =f(hν) dependence was plotted. Usually the dependence of the photocurrent on incident light quanta energy is well described by linear relationship between I 0.5 and hν near the threshold [6,7]. The linear part of this dependence was extrapolated to the h axis and I p value was determined as the photon energy at the interception point.

Device characterization
The Device parameters Figure S5. Efficiencies at stabilized power output of tested compounds S17 Figure S6. External quantum efficiency of the typical PSCs parameters employing V1050, V1061 and Spiro-OMeTAD S18 Figure S7. The statistical distribution of the PSCs parameters employing V1050, V1061 and Spiro-OMeTAD S19 Figure S8. JV curves from both of forward scan and reverse scan of the typical PSCs parameters employing V1050, V1061 and Spiro-OMeTAD

Contact angle measurement
Spiro-OMeTAD V1050 V1061 S21 Figure S9. Contact angle measurement of a Spiro-OMeTAD, V1050, and V1061 droplet on the perovskite film. Solutions of the same concentrations and with the same additive amounts were used, as for the PSC fabrication.

Stability of devices
To evaluate the device stability test, the devices were kept in dark at uncontrollable humidity