application in perovskite solar cells

Report 8 Downloads 144 Views
Supporting Information

Parasitic absorption reduction in metal oxide-based transparent electrodes: application in perovskite solar cells

Jérémie Werner,*1 Jonas Geissbühler, 1,2 Ali Dabirian, 1 Sylvain Nicolay, 2 Monica Morales-Masis, 1 Stefaan De Wolf, 1 Bjoern Niesen, 1,2 and Christophe Ballif 1,2

1

Ecole Polytechnique Fe´de´rale de Lausanne (EPFL), Institute of Microengineering (IMT), Photovoltaics and Thin-Film Electronics Laboratory, Rue de la Maladie`re 71b, 2002 Neuchâtel, Switzerland 2

CSEM, PV-Center, Jaquet-Droz 1, 2002 Neuchâtel, Switzerland * Corresponding author, email [email protected]

S-1

15

WOx

Absorptance (%)

Absorptance (%)

15

IZO WOx/IZO

10

IZO/WOx

5

0

V2Ox IZO V2Ox/IZO

10

IZO/V2Ox

5

0 400

600

800

1000

1200

400

600

800

1000

1200

Wavelength (nm) b) Wavelength (nm) Figure S1: Comparison of the coloration due to TCO overlayer deposition in tungsten oxide (a) and vanadium oxide (b), using sputtered indium zinc oxide (IZO) as TCO. The absorption increase is only present when the TCO is deposited on top of the TMO and not when the TMO is deposited on top of the TCO. a)

Figure S2: Comparison of TMO/TCO stack absorptance spectra for several sputtered TCOs (IZO, ZnO, IO:H, ITO), (a) on WOx and (b) on MoOx. All TCOs result in the coloration of the underlying TMO, indicating that its origin is indeed in the sputtering process and not due to chemical interactions.

Figure S3: Absorptance spectra showing the effect of UV-Ozone treatments on a MoOx layer.

S-2

Figure S4: XPS full survey for a) MoOx and b) WOx layers.

S-3

Figure S5: Complete XPS analysis of MoOx and WOx films treated with Ar and CO2 plasmas. a&e) the metal core levels; b&f) the carbon peaks; c&g) the Fermi levels; d&h) the oxygen peaks.

S-4

Figure S6: Fitted XPS curves for WOx and MoOx metallic core levels (Mo3d and W4f), for the several plasma treatments

S-5

Figure S7: Raman spectroscopy of Molybdenum oxide: a) Raman spectrum of MoO3 powder showing its alpha phase; b) Raman raw spectra of MoOx 10-nm-thick layers on AF32 glass substrates, showing almost identical features as the reference uncoated glass; c) Comparison of Raman spectra of MoOx as-deposited, MoOx treated with CO2 and Ar plasma and Ar-plasma-treated MoOx, with the glass substrate background signal subtracted.

Figure S8: J-V (a) and maximum power point tracking (b) curves for semitransparent perovskite solar cells with MoO x or WOx buffer layers (with or without CO2 plasma pre-treatment). Dashed J-V curves are in forward (Jsc to Voc) direction and solid J-V curves are in reverse (Voc to Jsc) direction. S-6

Figure S9 External quantum efficiency of two perovskite/silicon heterojunction monolithic tandem solar cells, using respectively MoOx or WOx as buffer layer in their front electrode. The device architecture and processing details can be found in Ref. 9 of the main article.

Minority carrier lifetime (ms)

25

i a-Si i a-Si/MoOx

20

i a-Si/MoOx/CO2 15

10

5

0 1014

1015

1016 -3

Excess carrier density (cm )

Figure S10 Minority carrier lifetime curves of n-type float-zone silicon wafers, passivated with thin intrinsic (i) amorphous silicon (a-Si) layers on both sides, with an n-type a-Si layer at the rear side, with or without a MoOx layer on the front side. Data is given as a function of the injection level before and after CO2 plasma treatment for the sample with MoOx layer.

S-7