Highly Efficient Semi-Transparent Perovskite Solar ... - ACS Publications

1.58 eV and 1.72 eV respectively which achieved PCEs well above 17% and ... a PCE of 26.6%, corresponding to a revealing a relative improvement above...
2 downloads 0 Views 988KB Size
Subscriber access provided by Nottingham Trent University

Organic Electronic Devices

Highly Efficient Semi-Transparent Perovskite Solar Cells for Four Terminal Perovskite-Silicon Tandems Herlina Arianita Dewi, Hao Wang, Jia Li, Maung Thway, Ranjani Sridharan, Rolf Stangl, Fen Lin, Armin G. Aberle, Nripan Mathews, Annalisa Bruno, and Subodh G. Mhaisalkar ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b13145 • Publication Date (Web): 23 Aug 2019 Downloaded from pubs.acs.org on August 23, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Highly Efficient Semi-Transparent Perovskite Solar Cells for Four Terminal Perovskite-Silicon Tandems Herlina Arianita Dewi a#, Hao Wang a#, Jia Li a, Maung Thway b,c, Ranjani Sridharan b, Rolf Stangl b, Fen Lin b, Armin G. Aberle b,c, Nripan Mathews a,d, Annalisa Bruno a*, Subodh Mhaisalkar a,d

a. Energy Research Institute @ NTU (ERI@N), Nanyang Technological University, Singapore 637553

b. Solar Energy Research Institute of Singapore (SERIS), National University of Singapore, Singapore 117574

c. Department of Electrical and Computer Engineering, National University of Singapore, Singapore 117583

d. School of Materials Science & Engineering, Nanyang Technological University, Singapore 639798

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 46

KEYWORDS

perovskite, semi-transparent perovskite solar cell, silicon solar cell, tandem solar cell, efficiency

ABSTRACT

Tandem solar cells (SCs) based on perovskite and silicon represent an exciting possibility for a breakthrough in photovoltaics, enhancing solar cell power conversion efficiency (PCE) beyond the single junction limit while keeping the production cost low. A critical aspect to push the tandem PCE close to their theoretical limit is the development of highperforming semi-transparent perovskite top-cells which also allow suitable near-infrared transmission. Here, we have developed highly efficient semi-transparent perovskite solar cells (PSCs) based on both mesoporous and planar architectures, employing Cs0.05(MA0.17FA0.83)0.95Pb(I0.83Br0.17)3 and FA0.87Cs0.13PbI2Br perovskites with bandgap of

ACS Paragon Plus Environment

2

Page 3 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

1.58 eV and 1.72 eV respectively which achieved PCEs well above 17% and 14% by detailed control of the deposition methods, thickness and optical transparency of the interlayers and the semi-transparent electrode. By combining our champion 1.58 eV PSCs (PCE of 17.7%) with an industrial-relevant low cost n-type Si SCs, a 4 terminals (4T) tandem efficiency of 25.5% has been achieved. Moreover for the first time, 4T tandem SCs performances have been measured in the low light intensity regime achieving a PCE of 26.6%, corresponding to a revealing a relative improvement above 9% compared to standard 1 sun illumination condition. These results are very promising for their implementations under field-operating conditions.

INTRODUCTION Silicon solar cells (Si-SCs) dominate the photovoltaic market with their high power conversion efficiencies (PCE), long term stability, and continuous improvements in the production processes and costs 1. Heterojunction Si-SCs have recently achieved a record PCE of 26.6% 2-3, close to their theoretical Auger recombination limit of 29.4% 4.

ACS Paragon Plus Environment

3

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 46

Novel photovoltaic devices are needed to overcome the 30% efficiency ‘barrier’ without significant increase in the cost of production. Tandem solar cells, which combine multiple cells absorbing complimentary spectral regions, can exceed 30% efficiencies by minimizing thermalization and absorption losses. Although state-of-the-art multi-junction solar cells based on III-V semiconductor materials have demonstrated record efficiencies of 39.2% under 1 sun illumination, their expensive production costs limit the possibility of their widespread development 5-6. Instead, tandem SCs based on perovskite and Si-SCs represent a real possibility for breakthrough photovoltaic performance, enhancing the efficiency of the existing Si-SCs. Indeed, low cost perovskite solar cells (PSCs) can deliver high PCE 7-8 and good bandgap tenability

9-11,

maximizing bandgap matching with silicon solar cells. Over the last few

years, perovskite/Si tandem SCs 12-14 have made impressive progresses reaching record published PCEs of 25.5%

15

(announced of 28%

4-terminals (4T) tandem configurations

17.

16)

and 27.1% for 2-terminals (2T) and

The primary difference between these two

configurations is that top and bottom cells are series connected in 2T tandem while they have independent electrical connections in 4T tandem. The main advantage of a 4T

ACS Paragon Plus Environment

4

Page 5 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

architecture is to allow a broader top cell bandgap selection and independent optimization of both top and bottom SCs 12, 14, 18. The development of high performance semi-transparent PSCs is critical to push the tandem efficiency well above that of a single junction Si-SCs and close to their theoretical efficiency of 43%

12, 19.

Highly transparent interfacial layers, i.e. electron transport layer

(ETL) and hole transport layer (HTL) with suitable band alignment, able to guarantee effective charge collection, are key elements to achieve high-performing PSCs. To date, high-efficiency semi-transparent PSCs are often realized in n-i-p architecture (glass/Fluorine doped Tin Oxide (FTO)/ETL/perovskite/HTL/electrode) based on metaloxide ETLs such as mesoporous TiO2 (m-TiO2) and planar SnO2 (p-SnO2)

20-24.

Both

these ETLs have significant advantages. The highly porous thick m-TiO2 scaffold allows reproducible nucleation of the perovskite layer, mitigates active layer defects, enhance charge collection by decreasing the carrier transport distance, and prevent electrical shunts

25-27.

p-SnO2 has a deep conduction band that facilitate fast electron extraction

and promotes UV-stability

28-30.

Moreover, low temperature processed p-SnO2

architecture improves overall SC transparency and is compatible with flexible substrate

ACS Paragon Plus Environment

5

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

and 2T tandem configurations

31-32.

Page 6 of 46

HTLs such as 2,2′,7,7′-Tetrakis-(N,N-di-4-

methoxyphenylamino)- 9,9′-spirobifluorene (Spiro-OMeTAD) and poly (triarylamine) (PTAA) are extensively used as hole transporting materials with efficient hole transport and electron-blocking properties

8, 33-35,

resulting in the highest efficiencies. Another

challenge to achieve high-PCE is the transmission at wavelengths longer than the perovskite bandgap that determines current generation in the bottom Si-SCs. Top PSCs in 4T tandem have to utilize two layers of transparent electrodes, which are the main cause for parasitic absorption losses. Perovskites with ~1.55 eV bandgap only allow NIR light to reach the bottom Si-SCs, whereas PSCs with higher bandgap (>1.55 eV) result in a wider transparency window but yield lower efficiencies. Thus, a balance between overall transparency and efficiency is a critical consideration. Recent results clearly indicate that for 4T tandem perovskite-silicon cells, top cells make a major contribution towards higher efficiencies (Table S114, 17, 21, 24, 36-48). In this work, we have developed highly efficient, semi-transparent PSCs using two different bandgaps and different architectures to investigate the best performing top cell for

a

4T

tandem

configuration.

Cs0.05(MA0.17FA0.83)0.95Pb(I0.83Br0.17)3

and

ACS Paragon Plus Environment

6

Page 7 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

FA0.87Cs0.13PbI2Br perovskites with bandgap of 1.58 eV and 1.72 eV respectively, were implemented in both m-TiO2 and p-SnO2 architectures. The semi-transparent PSCs demonstrate efficiencies above 17% and above 14% for 1.58 eV and 1.72 eV band-gap devices in both m-TiO2 and p-SnO2. Combining 1.58 eV and 1.72 eV semi-transparent PSCs with bottom Si-SCs, the 4T tandem SCs achieved PCEs as high as 25.5% and 22.7% respectively. Silicon and perovskite single junctions and 4T tandem SCs have been also investigated in intensity regime between 1 and 0.1 sun in order to emulate realistic operating conditions.

4T tandem PCEs showed a relative 9.7% PCE

improvement at the lowest illumination respect to the standard (1 sun) conditions.

RESULTS AND DISCUSSION Perovskite Thin Film Characterization Varying the perovskite bandgap in PSCs is a straightforward approach to examine the effect of top cell transparency on the 4T tandem PCE. Indeed, aiming at unveil the tradeoff between transparency spectral range and efficiency in pursuit of best performing 4T tandem SCs, perovskites with different bangdap have been implemented. Perovskites

ACS Paragon Plus Environment

7

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

with

mixed

cations

and

halide

compositions

have

Page 8 of 46

been

synthetized

-

Cs0.05(MA0.17FA0.83)0.95Pb(I0.83Br0.17)349 and FA0.87Cs0.13PbI2Br - to attain bandgaps of 1.58 eV and 1.72 eV respectively (Figure 1a). These compositions have been chosen for suitable bandgap match with Si-SC, stability, and fabrication reproducibility

14, 19, 49.

The

valence band (VB) levels for these perovskites have been measured using photoelectron spectroscopy in air (PESA), whereas the conduction band (CB) levels were derived by combing them with their optical bandgaps Figure S1a. The corresponding energy levels are shown in Figure S1b. The CB values of 1.58 eV perovskite and 1.72 eV perovskite are estimated to be at -4.22 eV and -4.15 eV, which show a favourable energy level alignment with both m-TiO2 and p-SnO2 layers, suggesting a good and efficient charge injection to the electron transport layers. Both 1.58 eV and 1.72 eV perovskites thin films display continuous and compact morphologies as indicated in the Field Emission Scanning Electron Microscopy (FESEM) images (Figure 1b). Specifically, the 1.58 eV perovskite film shows a homogenous distribution of large grains while in the 1.72 eV a larger spread of grain sizes can be observed. The crystal structures of both 1.58 eV and 1.72 eV perovskites have been

ACS Paragon Plus Environment

8

Page 9 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

further characterized by X-Ray diffraction (XRD) (Figure S2). Both perovskite compositions show a PbI2 excess at 12.4° which is beneficial to both passivate the perovskite surface and to increase the charge mobility and consequently the device performance 50-52. Carrier lifetime and injection properties of 1.58 eV and 1.72 eV perovskite thin films with and without the electron quenching layer have been studied by steady state and time resolved photo-luminescence (PL) measurements. The steady state PL spectra peak at 760 nm and 710 nm for the 1.58 eV and 1.72 eV perovskites respectively and they both composition show a significant quenching (above 90%) when the thin film is deposited on top of the m-TiO2 and p-SnO2 layers confirming the efficient charge injection (Figure 1c and Table S2) from the perovskite to the ETL. Time resolved PL (TRPL) measurements of the (Figure 1d). Pristine 1.58 eV and 1.72 eV perovskites reveal long fluorescence lifetimes of 488 ns and 292 ns respectively, which confirms their good optoelectronic properties and low defect densities. Moreover, both ETLs effectively quench the fluorescence emission, reducing the 1.58 eV perovskite excitons lifetimes to 28 ns and 41 ns for m-TiO2 and p-SnO2 respectively. Using a similar structure, 1.72 eV perovskite

ACS Paragon Plus Environment

9

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 46

lifetime has been reduced to 23 ns and 60 ns on m-TiO2 and p-SnO2 respectively. The substantial PL quenching is an evidence of efficient charge separation necessary for good performance in SCs. Slightly higher quenching for m-TiO2 is hypothesized to be due to the larger surface area contact with perovskite films as compared to p-SnO2.

ACS Paragon Plus Environment

10

Page 11 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 1.

1.58 eV and 1.72 eV perovskites film optical and morphological

characterizations. (a) Tauc plot curves of Cs0.05(MA0.17FA0.83)0.95Pb(I0.83Br0.17)3 (deep red circles) and (FA0.87Cs0.13PbI2Br) (purple circles) perovskite compositions showing the 1.58 eV and 1.72 eV bandgap respectively. (b) 1.58 eV and 1.72 eV perovskite thin films on FTO/glass substrates FESEM Top-view images reveal continuous and compact perovskite surface with largest grains around 400 nm for both compositions, (c) steadystate PL and (d) time resolved PL decays of 1.58 eV and 1.72 eV perovskite thin films on glass and on top of m-TiO2 and p-SnO2 layers, obtained with excitation wavelength of 405 nm, 1.70 eV) PSCs 17, 20-21, 23, 42. By shifting to higher bandgap of 1.72 eV, the PSCs suffer from unavoidable Jsc reduction of ~3 mA/cm2, which is in consistent with their Shockley-Queisser characteristic for different bandgaps

53-54,

whereas, loss in

both Voc and FF, may indicate presence of more carrier recombination and non-uniformity in the absorber grain sizes distribution.

Table 1. Photovoltaic parameters of semi-transparent 1.58 eV and 1.72 eV PSCs with and as p-SnO2 and m-TiO2 electron transport layer.

ACS Paragon Plus Environment

17

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Perovskit

Voc

ETL

e

(V)

Jsc (mA/cm2 )

FF

Page 18 of 46

PCE champ

(%)

(%)

PCE ave (%)

77. p-SnO2

1.06

21.52

5

17.7

17.2 ± 0.5

17.1

16.8 ± 0.3

14.1

13.7 ± 0.4

14.7

14.3 ± 0.4

74. 1.58eV

m-TiO2

1.09

21.08

6 74.

p-SnO2

1.06

17.89

1 74.

1.72eV

m-TiO2

1.10

18.07

1

Both 1.58 eV and 1.72 eV PSCs fabrication processes have good reproducibility as illustrated by the low PCEs spread of ~ 0.4% for both p-SnO2 and m-TiO2 architectures (Figure 2e). In addition, the PSCs also exhibit very good shelf-life stability, with a PCE drop smaller than 10% over 80 days (1.58 eV) and 50 days (1.72 eV). The PSCs have been stored without any encapsulation in controlled environment and been measured in >70% RH ambient air (Figure 2f). These data further confirms the good stability of the perovskite layer and the effectiveness of the sputtered ITO electrode to protect underlying layers from moisture

44.

Both mesoporous and planar architectures have shown

ACS Paragon Plus Environment

18

Page 19 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

comparable high performances when the active layer film morphology is properly optimized, but the higher transparency of the planar SnO2 based PSCs make them more promising for tandem integration.

4T Perovskite-Silicon Tandem Both 1.58 eV and 1.72 eV semi-transparent PSCs with p-SnO2 ETL have been implemented in a 4T tandem configuration with small-size (1cm2) n-type monoPolyTM bottom Si-SCs (insets of Figure 3a and Figure 3b)

55.

The 1cm2 localized grown Si-SCs

has been specifically designed to eliminate the effect of edge recombination losses as compared to conventional laser cutting process from full size silicon wafer. The schematic of silicon bottom cell is shown in Figure S8. Fabrication and characterization details of the bottom Si-SCs are reported in the experimental section and will be further described in a future work. As shown in Table 2, the stand-alone Si-SC has a PCE of 21.1% under the standard AM1.5G spectrum. As the Si-SCs under tandem configurations only receive the light filtered by the top PSCs, they effectively perform under a much lower injection level

ACS Paragon Plus Environment

19

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 46

compared to standard AM1.5G conditions, resulting in much lower Voc and Jsc values. The effective add-on efficiency from the Si-SC under the 1.58 eV PSCs is 7.8% (Figure 3a). A 4T tandem efficiency of 25.5% was achieved by combining the 1.58 eV PSC with the Si-SC. Instead, when in the 4T tandem configuration the incident light is filtered by the 1.72 eV PSCs, a larger fraction light can reach the bottom Si SCs and concomitantly its PCE contribution rises to 8.3%. This higher value is mostly driven by the lower Jsc reduction (25 mA/cm2 instead 23.8 mA/cm2) (Figure 3b). Although Si-SCs contribute a higher effective add-on efficiency, the 1.72 eV PSCs PCE is still much lower than the 1.58 eV PSCs and the total 4T tandem efficiency achieved in this configuration was only f 22.4%. Interestingly, the Jsc values for the 1.72 eV PSCs and the Si-CSs filtered by 1.72 eV PSCs are comparable, indicating the potential of this perovskite composition for future 2T monolithic integration where current matching between the top and bottom cells is required. The EQE of both 1.58 eV and 1.72 eV semi-transparent PSCs show a very sharp edge at their particular bandgap indicating the good absorption of the thick perovskite layers (Figure 3c and Figure 3d). The Si-SCs EQEs reach around 75% under

ACS Paragon Plus Environment

20

Page 21 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

the perovskite filter in the infrared range closely matching the transparency value of the top PSCs. Although 4T tandem efficiency is higher than each of the stand-alone sub-cells, the EQEs clearly show presence of parasitic absorption of the top PSCs which limits the efficiency of the filtered Si-SC. The 4T tandem SCs performances of m-TiO2 1.58 eV and 1.72 eV PSCs show a similar trend to the p-SnO2 PSC. Indeed, 4T tandem PCEs reached of 24.6% and 22.7% for 1.58 eV and 1.72 eV PSCs respectively (Figure S9a and Figure S9b). The photovoltaic parameters are summarized in Table S4. These less transparent top PSCs, as compared to the p-SnO2 counterpart, resulted in lower Jsc in the filtered SiSCs as clearly shown in the Si-SCs EQE (Figure S9c and Figure S9d). For both mesoporous and planar architectures, the absolute PCE gains from standalone Si-SCs to a 4T perovskite-on-Si tandem are ~5% and ~2% with 1.58eV and 1.72eV top PSCs, respectively, highlighting the advantage of deploying a multi-junction configuration to better utilize the solar irradiance spectrum.

ACS Paragon Plus Environment

21

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 46

Figure 3. 4T tandem based on 1.58 eV and 1.72 eV p-SnO2 PSCs. (a) J-V curve of 1.58 eV PSCs (deep red circles), stand-alone Si-SC (blue circles), ‘1.58 eV filtered’ Si-SC (green circles); (b) J-V curve of 1.72 eV PSCs (purple circles), stand-alone Si-SC (blue circles), ‘1.72 eV filtered’ Si-SC (green circles). Insets in (a) and (b) show the schematic of mechanical stack 4T tandems using their respective top PSCs bandgaps. EQE curves of 4T tandems: (c) EQEs for 1.58 eV PSC (deep red), ‘1.58 eV filtered’ Si-SCs (green), stand-alone Si-SCs (blue); (d) 1.72 eV PSC (purple), ‘1.72 eV filtered’ Si-SCs (green), stand-alone Si-SCs (blue).

ACS Paragon Plus Environment

22

Page 23 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Table 2. IV parameters of p-SnO2 1.58 eV and 1.72 eV semi-transparent PSCs, Si-SCs with and without PSCs filtering, and 4T perovskite/silicon tandems

Voc

Jsc

FF

PCE

(V)

(mA/cm2)

(%)

(%)

Si unfiltered

0.66

40.2

79.0

21.1

Si filtered (1.58 eV)

0.63

15.2

81.1

7.8

Perovskite (1.58 eV)

1.06

21.5

77.5

17.7

SCs

4T tandem (1.58 eV)

25.5

Si filtered (1.72 eV)

0.63

16.4

81.4

8.3

Perovskite (1.72 eV)

1.06

17.9

74.1

14.1

4T tandem (1.72eV)

22.4

In this work, we deployed a rear-side contact passivated silicon bottom cell (monoPolyTM technology), which is considered most industrial relevant and costeffective for n-type high-efficiency silicon SCs. This is particularly important for future

ACS Paragon Plus Environment

23

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 46

commercialization implementations due to its cost effectiveness and easy process ability 55,

but here we are still currently compromising on PCE as compared to the more costly

silicon heterojunction (SHJ) and interdigitated-back-contact (IBC) n-type silicon bottom cells (compare Table S1). Contrarily, most of the previously published tandem works employed more advanced high-efficiency bottom Si SCs architectures (i.e. SHJ and IBC), which have higher PCE, due to superior values of both Voc and Jsc 17, 43-44, 48. We evaluate the feasibility to maximize our 4T tandem performance by substituting the monoPolyTM Si SC (PCE: 21.1%) with a previously reported small area high efficiency IBC Si-SC (PCE: 24.4%

56)

in conjunction with our 17.7% p-SnO2 1.58 eV ST-PSC.

Assuming the same relative drop in efficiency for the 24.4% IBC Si SCs as the one observed for the monoPolyTM SCs, a PCE of 9.1% would be achieved, when filtered with our 1.58 eV ST-PSC. This value is significantly higher to the 7.8% PCE measured for monoPolyTM Si SC. The 4T tandem configuration deploying a 1.58 eV ST-PSCs as top cell and an IBC Si SC as bottom cells could then lead to a PCE of 26.8%. Moreover light management strategies could also be implemented to further improve our 4T tandem cell efficiency. These are likely to arise from enhancement in top

ACS Paragon Plus Environment

24

Page 25 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

perovskite cell transparency by minimizing overall reflectance losses through proper coatings and reduction of parasitic absorption 46, 56-57.

4T Perovskite-Silicon Tandem under Low Light Intensity We also explore the feasibility of 4T tandem SCs for realistic operation, when both spectral distribution and intensity of the sunlight radiance reaching the SCs can vary during the daytime and the season. Therefore, understanding the PCE trend under different light conditions is critical for field-operating conditions. Here, we investigate both perovskite and silicon single junction SCs and 4T tandem SC photovoltaic parameters under different light intensities ranging from 0.1 to 1 Sun. The 4T tandem PCEs are reported in Figure 4a and b, while the detailed Jsc, Voc, and FF are shown in Figure S10a, Figure S10b, and Figure S10c. Even if theoretical studies have shown the simulated trend of both 2T and 4T tandem SCs under different illuminations

58-59,

to the best our

knowledge these are the first experimental data showing the trend of a perovskite-silicon tandem solar cells under low intensity illumination.

ACS Paragon Plus Environment

25

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 46

Silicon SCs, either under direct illumination (stand-alone) or filtered by the perovskite, exhibit very marginal PCE reduction as the illumination intensity decreases, with the highest relative drop around 5% at 0.1 sun (Figure 4a). The detailed photovoltaic characteristics for stand-alone and filtered Si SC, Figure S10, are consistent previous works

60.

with

On the other hand, the PCE of 1.58 eV semi-transparent perovskite

SCs remains almost constant in the light intensity region between 1-0.5 sun, while it slightly increases at even lower intensities (0.5-0.1 sun) reaching the maximum value of 19.5% at 0.1 sun (18% relative improvement) (Figure 4a). The perovskite SC Jsc and FF mainly drive the PCE increase. The Jsc linear trend with light intensity of single junction perovskite SC has been also previously reported, indicating good charge separation at the perovskite and transport layers interface regardless of different light intensity

61-63

(Figure S10a). The increase of FF with the intensity decrease suggests a reduced charge recombination under lower light intensity (Figure S10b). At the same time Voc only slightly decreases from 1 to 0.1 sun in agreement with the hypothesis of a reduced number of electron generated (Figure S10c) 64-65.

ACS Paragon Plus Environment

26

Page 27 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

As a combined effect of the behaviour of both sub-cells, the 4T tandem SCs exhibits a stable PCE also under lower intensities reaching 26.6% of PCE and a 9.7% relative improvement of the PCE when operated at 0.1 sun (Figure 4b), suggesting that 4T tandem configuration is versatile to be implemented in different illumination conditions. Our experimental 4T results under different light intensity are also in line with theoretical simulations which have shown that 4T tandem SCs under practical condition can achieve higher PCE, which does not critically depends on solar radiance intensity, as compared to 2T counterpart

66-67.

Although our studies also justify good performance of single

junction perovskites and silicon SCs under low light intensity regime, the presented 4T tandem is able to outperform the single junction due to better spectrum light utilized and less thermalisation losses.

ACS Paragon Plus Environment

27

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 46

Figure 4. PCE of sub-cells and 4T tandem based on 1.58 eV p-SnO2 PSCs under different light intensity. (a) PCE of sub-cells. 1.58 eV PSCs (deep red), ‘1.58 eV filtered’ Si-SCs (green), stand-alone Si-SCs (blue), (b) PCE of 4T tandem (dark yellow)

CONCLUSION In conclusion, here we have investigated the effect of perovskite bandgap and SCs structure on a 4T perovskite-on-Silicon tandem SCs. The optimized semi-transparent 1.58 eV PSCs achieved PCE higher than 17% for both m-TiO2 (17.1 %) and p-SnO2 (17.7%) architectures, leading to 4T tandem PCEs of 25.5%. Whereas, 4T tandem based on the 1.72 eV PSCs reached 22.7% PCE, mostly due to the lower efficiency of the top cells. Our results highlight the importance of perovskite bandgap selection and PSCs devices architecture to understand best performing conditions for 4T tandem SCs. We have also explored for the first time the 4T tandem SCs performances under low intensity illumination proving a relative 9.7% increase of the performances at low intensities as

ACS Paragon Plus Environment

28

Page 29 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

compared to standard 1 sun illumination. These results confirm the potential of the 4T tandem SCs in field-operating conditions, as predicted from previous theoretical simulations.

EXPERIMENTAL SECTION Perovskite deposition The 1.58 eV perovskite (Cs0.05(MA0.17FA0.83)0.95Pb(I0.83Br0.17)3) precursor solution, was prepared following ref. 49 and was spin-coated at 1000 rpm for 10 s followed by 6000 rpm for 17 s. The 1.72 eV perovskite (FA0.83Cs0.17PbI2Br) was spin-coated at 2000 rpm for 10 s followed by 6000 rpm for 30 s. Both perovskites utilized chlorobenzene as secondary solvent and undergo 100 °C for 1 h post-annealing inside argon-filled glovebox.

Perovskite film characterization Photo Electron Spectroscopy in Air (PESA) measurements was recorded with a Riken Keiki AC-2 PESA spectrometer with a power setting of 800 nW and a power number of 0.5. Absorbance (A) and Transmission (T) spectra were recorded using UV-Vis-NIR

ACS Paragon Plus Environment

29

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 46

Spectrophotometer (UV3600, Shimadzu) equipped with an integrating sphere (300-1200 nm wavelength). Morphological characterization for both top view perovskite film and cross section of the complete devices were recorded using a Field Emission Scanning Electron Microscope (FESEM, JEOL JSM-7600F, 5 kV and 10 mA). X-Ray Diffraction (XRD) patterns were recorded on (Bruker D8 Advance). Steady state photoluminescence spectra had been measured using a top-table Fluoromax-5 system. TRPL decays were collected using a micro-PL setup and a Picoquant PicoHarp 300 time-correlated single photon counting (TCSPC) system and a laser diode (Picoquant P-C-405B, λ = 405 nm,