Tandem matchmaking: pairing published perovskite, organic and silicon cells to find the four-terminal stacks worth building

We calculate 663 combinations of 64 published solar cells, 51 top cells that transmit light (“semi-transparent” in the literature; 29 perovskite, 22 organic) and 13 opaque bottom cells (crystalline silicon, Sn–Pb perovskite, organic), into four-terminal (4T) tandems, and ask which stacks are worth building. The cell-level tandem efficiency is calculated under AM1.5G from the measured transmittance of each top cell and the measured external quantum efficiency of each bottom cell, without extending any spectrum. Against 30 measured 4T tandems from 11 publications, the calculation differs from the measurement by 0.71 percentage points rms (mean −0.46, at most 1.79) over all of them; by 0.50 points rms (at most 1.18) for the 25 tandems with an air gap or no declared optical coupling, and by 1.33 (at most 1.79) for the 5 with index matching, which the transmittance measured in air underestimates. The ±1.00 point tolerance we use rests on the tandems with perovskite top cells; it is not established for organic top cells. Pairings are graded against three yardsticks: the bottom cell alone, a large-area n-TOPCon c-Si cell (26.55 %) and the c-Si record (27.8 %). With the best bottom cell of each technology, 38 pairings whose spectra overlap (at most 1 % of the bottom-cell current missing) improve on their bottom cell and 12 exceed the c-Si record (6 more as lower bounds). The main result is an extension of the figure of merit of Kirchartz et al. to bottom cells other than silicon: the fraction of its efficiency that a bottom cell keeps under a given top cell is, in the median over the top cells whose transmittance covers the c-Si response, 0.85 of its c-Si value on Sn–Pb perovskite (range 0.79–0.99) and 0.68 on OPV (0.57–0.98), so the requirement on a top cell must be stated per bottom cell. In the base case, 4 all-perovskite pairings of published cells on the best Sn–Pb cell exceed the c-Si record (3 beyond the ±1.00 point tolerance), on small-area, mostly uncertified top-cell efficiencies, and to our knowledge none of them has been reported as a measured 4T device; no organic top cell in our data set improves any bottom cell, because of its low band gap, the light it neither collects nor transmits usefully, and, at wider gaps (two cells), a larger voltage deficit. The values are cell-level estimates: they can be conservative for index-matched stacks, because the transmittance was measured in air, and they are optimistic with respect to cell area, champion efficiencies and system losses. Data and code are open.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-12
DOI
https://doi.org/10.5281/zenodo.24160966
Primary Topic
Silicon and Solar Cell Technologies
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Tandem matchmaking: pairing published perovskite, organic and silicon cells to find the four-terminal stacks worth building

Alexandros Zachariadis
Zenodo (CERN European Organization for Nuclear Research)
Silicon and Solar Cell Technologies
preprint

Tandem matchmaking: pairing published perovskite, organic and silicon cells to find the four-terminal stacks worth building

Alexandros Zachariadis
preprint en

Abstract

We calculate 663 combinations of 64 published solar cells, 51 top cells that transmit light (“semi-transparent” in the literature; 29 perovskite, 22 organic) and 13 opaque bottom cells (crystalline silicon, Sn–Pb perovskite, organic), into four-terminal (4T) tandems, and ask which stacks are worth building. The cell-level tandem efficiency is calculated under AM1.5G from the measured transmittance of each top cell and the measured external quantum efficiency of each bottom cell, without extending any spectrum. Against 30 measured 4T tandems from 11 publications, the calculation differs from the measurement by 0.71 percentage points rms (mean −0.46, at most 1.79) over all of them; by 0.50 points rms (at most 1.18) for the 25 tandems with an air gap or no declared optical coupling, and by 1.33 (at most 1.79) for the 5 with index matching, which the transmittance measured in air underestimates. The ±1.00 point tolerance we use rests on the tandems with perovskite top cells; it is not established for organic top cells. Pairings are graded against three yardsticks: the bottom cell alone, a large-area n-TOPCon c-Si cell (26.55 %) and the c-Si record (27.8 %). With the best bottom cell of each technology, 38 pairings whose spectra overlap (at most 1 % of the bottom-cell current missing) improve on their bottom cell and 12 exceed the c-Si record (6 more as lower bounds). The main result is an extension of the figure of merit of Kirchartz et al. to bottom cells other than silicon: the fraction of its efficiency that a bottom cell keeps under a given top cell is, in the median over the top cells whose transmittance covers the c-Si response, 0.85 of its c-Si value on Sn–Pb perovskite (range 0.79–0.99) and 0.68 on OPV (0.57–0.98), so the requirement on a top cell must be stated per bottom cell. In the base case, 4 all-perovskite pairings of published cells on the best Sn–Pb cell exceed the c-Si record (3 beyond the ±1.00 point tolerance), on small-area, mostly uncertified top-cell efficiencies, and to our knowledge none of them has been reported as a measured 4T device; no organic top cell in our data set improves any bottom cell, because of its low band gap, the light it neither collects nor transmits usefully, and, at wider gaps (two cells), a larger voltage deficit. The values are cell-level estimates: they can be conservative for index-matched stacks, because the transmittance was measured in air, and they are optimistic with respect to cell area, champion efficiencies and system losses. Data and code are open.

Zenodo (CERN European Organization for Nuclear Research)
Affordable and clean energy
Silicon and Solar Cell Technologies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon