Elucidating the role of the hole-extracting electrode on the stability and efficiency of inverted CsSnI 3 / C 60 perovskite photovoltaics

The correlation between the stability of thin films of black (B)-γ CsSnI3 perovskite in ambient air and the choice of supporting substrate is examined for the substrates: (i) soda-lime glass; (ii) indium tin oxide (ITO) glass; (iii) copper iodide (solution processed)/ITO glass; (iv) poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS)/ITO glass; (v) and an optically thin (8 nm) gold film electrode. The performance of (ii)–(v) as the hole-extracting electrode in inverted photovoltaic (PV) devices with a simple bilayer architecture is compared for a test condition of 1 sun continuous solar illumination in air. CsSnI3 film stability is shown to depend strongly on the density of pinholes and grain boundaries, although not on the preferred CsSnI3 crystallite orientation. Solution processed CuI is shown to be unsuitable as a hole-transport layer (HTL) for inverted CsSnI3 PV devices because it is almost completely displaced by the CsSnI3 precursor solution during the spin coating process, and its large ionisation potential is poorly matched to the valence band edge of CsSnI3. Devices using an ITO (or Au) hole-extracting electrode with no HTL are found to be more stable than those using the archetypal HTL; PEDOT:PSS. Spectroscopic analysis of the CsSnI3 layer recovered from PV devices after 24 hours testing in ambient air (with no device encapsulation) shows that ≤11% of the CsSnI3 film thickness is oxidised to Cs2SnI6 due to air ingress, which shows that the deterioration in device efficiency under continuous illumination does not primarily result from a reduction in the light absorption capability of the perovskite film due to CsSnI3 oxidation. Additionally it is shown that SnCl2 added during CsSnI3 film preparation reduces the extent of p-type self-doping of the perovskite film and serves as an n-type dopant for the adjacent evaporated C60 electron transport layer.


Supporting Information
Figure S1: Evolution of electronic absorption spectrum of CsSnI 3 + 10 mol% SnCl 2 films deposited on glass (a), ITO:CuI (b) and Au (c).In all cases the CsSnI 3 solution was 8 wt% by total mass of solids which resulted in a film thickness of  50 nm.In each case the background has been subtracted.) simulation package, showing the difference in absorbance for a 50 nm CsSnI 3 film on glass and ITO coated glass assuming a ITO thickness of  140 nm and CsSnI 3 film with an idealised uniform slab-like structure.These simulations show that optical interference due to the ITO layer gives rise to a local minimum in the absorption spectrum close to a wavelength of 400 nm, the exact position of which depends sensitively on the optical properties and thickness of the ITO layer.glass (black).In all cases ITO glass was used as the background.The apparent negative absorbance in some parts of the CuI spectrum is attributed to the CuI film functioning as an anti-reflective layer over particular wavelength ranges.(lower) of the surface of a perovskite film recovered from a PV device with the structure: ITO glass / CsSnI 3 + 10 mol% SnCl 2 (8 wt%)/ C 60 / BCP / Al.The device was unencapsulated and was tested for 24 hours in ambient air under 1 sun continuous simulated solar illumination.After testing it was transferred to a nitrogen filled glovebox (< 1 ppm H 2 O and O 2 ) and the top Al electrode was peeled off using carbon tape.It was then washed by immersing in anhydrous chlorobenzene three times to remove the SnCl 2 / C 60 / BCP layers before transferring to the vacuum chamber of the XPS spectrometer without exposure to ambient air.It is assumed that air ingress into the device results in the formation of a layer of Cs 2 SnI 6 at the surface of the CsSnI 3 film that interfaces with the C 60 electron transport layer in the device.The Cs 2 SnI 6 thickness was determined using the thickogram method 4 based on the ratio of Sn 3d peak areas assigned to Sn 2+ and Sn 4+ oxidation states, which correspond to CsSnI 3 and Cs 2 SnI 6 respectively [5] , and assuming a inelastic mean free path in Cs 2 SnI 6 of 2.58 nm (estimated assuming a Cs 2 SnI 6 density of 4.521 g/ cm 3 using the method of S. Tanuma et al. [6] ).       -1 to 1 V 0.1 V/s 0.045 V/s 0.025 V/s 1 to -1 V 0.1 V/s -0.2 to 1 V 0.1 V/s 1 to -0.2 V 0.1 V/s (ESI) for Journal of Materials Chemistry A. This journal is © The Royal Society of Chemistry 2017

Figure S2 :
Figure S2: Optical simulation using the Essential Macleod (Thin Film Centre Inc.) simulation package, showing the difference in absorbance for a 50 nm CsSnI 3 film on glass and ITO coated glass assuming a ITO thickness of  140 nm and CsSnI 3 film with an idealised uniform slab-like structure.These simulations show that optical interference due to the ITO layer gives rise to a local minimum in the absorption spectrum close to a wavelength of 400 nm, the exact position of which depends sensitively on the optical properties and thickness of the ITO layer.

Figure S3 :Figure S4 :
Figure S3: Measured XRD pattern of an ITO glass slide (green) and simulated XRD pattern for ITO (black).

Figure S5 :
Figure S5: Electronic absorption spectra of a CsSnI 3 :SnCl 2 (10 mol%) film supported on ITO glass (blue); a  40 nm CuI film on ITO glass (red); a CsSnI 3 :SnCl 2 film on CuI coated ITO glass (black).In all cases ITO glass was used as the background.The apparent negative absorbance in some parts of the CuI spectrum is attributed to the CuI film functioning as an anti-reflective layer over particular wavelength ranges.

Figure S6 :Figure S7 :
Figure S6: AFM image of surface topography (a) and cross-section (b) of a film of CsSnI 3 :SnCl 2 deposited on glass.;AFM image of surface topography (c) and cross-section (d) of a film of CuI deposited on glass.;AFM image of surface topography (e) and crosssection (f) of a film of CsSnI 3 :SnCl 2 deposited onto a film of CuI on glass.

Figure S8 :Figure S9 :Figure S10 :
Figure S8: AFM images of CuI deposited on glass (a) before and (b) after spin-casting DMF on top of the film: (c) UV/ vis/ NIR spectra of films of CuI with (red) and without (black) DMF spin-casted on top.

Figure S11 :
Figure S11: UPS spectrum of a film of Cs 2 SnI 6 (supported on Au) formed by air oxidation of CsSnI 3 .The secondary electron cut off region; left, and region near to the Fermi level, including the valance band edge; right.The sample was Ar+ ion etched for 0 to 50 s, in 10 sec steps to remove surface contaminants and adsorbed water.For t = 20, 30 and 40 seconds sputtering the data sets converge to the same value (within error) giving a valence band edge energy for Cs 2 SnI 6 as 5.83 eV ± 0.05 eV below the vacuum level and work function of 5.1 eV.Literature values for the band gap of Cs 2 SnI 6 range from 1.3 to 1.6 eV, 1-3 giving a value for the CB edge between 4.53 eV and 4.23 eV below the vacuum level.

Figure S16 :Figure S17 :
Figure S16: High resolution XPS spectra of the Cl 2p and Sn 3d regions at the point marked in the photograph (lower).The photograph shows the surface of device with the structure ITO| CsSnI 3 :SnCl 2 | C 60 | BCP| Al, in which the BCP| Al and part of the C 60 layer has been removed using carbon tape to expose the interfacial region between the perovskite film and C 60 .

Figure S18 :
Figure S18: Representative log-linear dark current-voltage characteristic of the ITO| CsSnI 3 :SnCl 2 | C 60 | BCP| Al photovoltaic devices shown in Figure 5.These data show that the current in reverse bias is dramatically reduced after the device has been subjected to

Figure S19 :
Figure S19: UPS spectrum (secondary electron cut-off (a) and valence band edge (b)) of CsSnI 3 on Au without exposure to the ambient environment.

Figure S20 :Figure S21 :
Figure S20: UPS spectrum (secondary electron cut-off (a) and valence band edge (b)) of CuI on ITO without exposure to the ambient environment.

Table S1 :
Average atomic composition of five EDX scans of a film of CsSnI 3 :SnCl 2 onto a ~40 nm thick CuI film supported on an ITO glass substrate.