Yang Chen | Hangzhou Microquanta Semiconductor: Why does a 27% efficient perovskite cell fail when scaled up, and how do we fix it?
03:25 - 04:43
Other snippets from this talk
Summary of the clip:
Why does a 27% efficient perovskite cell fail when scaled up, and how do we fix it?
Transitioning perovskite solar cells from small lab-scale devices with over 27% efficiency to large-scale commercial modules presents severe manufacturing challenges. Achieving thin-film uniformity over square-meter substrates requires highly precise fluid dynamics during the deposition stage, as any slight variation destroys the device's electrical characteristics.
At Hangzhou Microquanta, this challenge was systematically addressed by abandoning traditional silicon processing lines in favor of a custom automated line built around slot-die coating. The sequence involves a slot-die head depositing a uniform wet-film, followed immediately by a laminar air drying step to initiate phase transition, and finally a vacuum annealing step at 20 minutes to lock in crystallization and long-term stability.
To resolve the spatial non-uniformity across large-area glass, the slot-die coating head's internal chamber geometry was completely redesigned. By using finite element analysis (FEA) simulation, the engineering team optimized the flow dynamics inside the slot-die lip, eliminating micro-scale thickness variations and ensuring uniform coating across the entire substrate width.
In this short video, you can learn:
* Why standard silicon manufacturing equipment cannot be easily adapted for high-quality perovskite thin-film deposition.
* The exact process sequence of slot-die coating, linear air drying, and vacuum annealing required for commercial perovskite crystallization.
* How finite element analysis (FEA) is utilized to redesign coating heads to eliminate film thickness variations at the square-meter scale.
📋 **Clip Abstract** This clip details the custom processing line developed to transition perovskite solar cells from highly efficient lab devices to large-area commercial modules. It highlights the use of finite element analysis to redesign slot-die coating heads for uniform large-scale film deposition.
#SlotDieCoating, #PerovskiteCrystallization, #FEASimulation, #ThinFilmDeposition, #PrintedElectronics, #PerovskitePhotovoltaics
This is a highlight of the presentation:
22-23 October 2025
Organised By:
More Highlights from the same talk.
04:07 - 05:01
Which efficiency split lets a semi-transparent perovskite top cell and a silicon bottom cell reach 26.63%?
Which efficiency split lets a semi-transparent perovskite top cell and a silicon bottom cell reach 26.63%?
The panel is a stack: a semi-transparent perovskite cell on top and silicon underneath. The efficiencies are reported separately — perovskite contributes 16.86%, silicon contributes 9.77% — and combined they give 26.63%. That split, with the perovskite carrying the larger share of the total, is the headline number for the stacked architecture rather than a single-junction result.
The motivation is spectral: sunlight covers a very broad spectrum, and no single material can utilize all of it. That is why the industry arrived at the tandem idea decades ago — split the spectrum between two absorbers instead of asking one junction to carry the whole range.
Perovskite became the promising partner for silicon in that pairing. Perovskite takes over the higher potentials, while silicon takes over the longer-wavelength energies. The two absorbers are therefore assigned different parts of the incident spectrum rather than competing for the same photons.
In this short video, you can learn:
* Perovskite contributes 16.86% and silicon 9.77%, summing to 26.63% in the stacked panel.
* A tandem is used because the solar spectrum is too broad for one single material to utilize fully.
* Perovskite takes over the higher potentials while silicon handles the longer-wavelength energies.
📋 **Clip Abstract** The stacked panel pairs a semi-transparent perovskite top cell with a silicon bottom cell, perovskite contributing 16.86% and silicon 9.77% for 26.63% combined. The rationale is spectral coverage: no single material spans the full solar spectrum, so perovskite absorbs the higher-potential portion and silicon the longer wavelengths.
About the speaker:
* Speaker: Yang Chen
* Company: Microquanta
* Event: Perovskite Connect 2026 | Berlin
* Location: Estrel, Berlin
Save the date: Perovskite Connect 2027 | 20 and 21 OCT 2027 | Berlin
Co-located with the Future of Electronics RESHAPED conference and exhibition
Learn more: https://www.techblick.com/perovskiteconnect
Intestered in exhibiting: learn more here: https://www.techblick.com/perovskiteconnect
#SemiTransparentPerovskite, #PerovskiteSiliconTandem, #FourTerminalTandem, #SpectrumSplitting, #PerovskitePV, #SolarCells
08:32 - 10:04
How does four-terminal tandem compare on annual energy yield and LCOE against two-terminal?
How does four-terminal tandem compare on annual energy yield and LCOE against two-terminal?
A model compared three cities using their spectrum data across the whole year, including albedo factors. In that annual analysis, the four-terminal architecture delivered a slightly better energy yield than the two-terminal design throughout the year. The comparison uses city-specific spectra and albedo over the full annual cycle.
On LCOE, the chart shows that with a larger degradation rate, the four-terminal approach makes somewhat better sense for energy cost. If degradation reaches 0.4% and even lower, the two-terminal approach behaves much better. The LCOE result is therefore tied to the degradation rate rather than only the annual energy yield.
The stated view is that two-terminal has a better future and is a perfect solution, but for now, for practical purposes, four-terminal is the product that can be provided to the market. The discussion extends to space solar, where work has been done since as early as 2020.
In this short video, you can learn:
* Across three cities and a full year including albedo, four-terminal tandem gave slightly higher energy yield than two-terminal.
* At larger degradation rates, four-terminal made better LCOE sense, while at 0.4% and lower two-terminal behaved much better.
* The practical market product is four-terminal now, while two-terminal is seen as the better long-term solution; space solar work dates to 2020.
📋 **Clip Abstract** A full-year, three-city spectrum and albedo model found four-terminal tandem slightly ahead of two-terminal on energy yield. LCOE depends on degradation: four-terminal makes more sense at larger degradation rates, but at 0.4% and below two-terminal behaves much better, so four-terminal is the near-term market product while two-terminal is seen as the better future solution.
About the speaker:
* Speaker: Yang Chen
* Company: Microquanta
* Event: Perovskite Connect 2026 | Berlin
* Location: Estrel, Berlin
Save the date: Perovskite Connect 2027 | 20 and 21 OCT 2027 | Berlin
Co-located with the Future of Electronics RESHAPED conference and exhibition
Learn more: https://www.techblick.com/perovskiteconnect
Intestered in exhibiting: learn more here: https://www.techblick.com/perovskiteconnect
#FourTerminalTandem, #AnnualEnergyYield, #LCOEDegradationRate, #SpectralAlbedoModel, #PerovskiteTandem, #SpaceSolarPhotovoltaics
05:01 - 06:19
Why choose a four-terminal tandem when the two-terminal design is more elegant and compact?
Why choose a four-terminal tandem when the two-terminal design is more elegant and compact?
Perovskite can take over for the higher potentials while silicon takes over for the longer-wavelength energies, which is the division of labour behind the tandem stack. The architecture choice then turns on wiring: the four-terminal is electrically connected in parallel, the two-terminal is electrically connected in series. That single difference, not elegance, drives the practical case.
The speaker concedes the two-terminal is a very elegant and compact design and says he very much agrees with Adam from QPPV on the practical side. But at this stage the four-terminal is fairly easy to manufacture, a decisive advantage given the circumstance that perovskite is advancing and scaling every day.
The commercial logic follows directly. A four-terminal node does not ask the entire industry or supply chain to adopt the new technology, which makes it easier to introduce that technology to the market. The speaker presents this as what the four-terminal technology node delivers at this stage, as perovskite scaling continues.
In this short video, you can learn:
* Four-terminal tandems are electrically connected in parallel, while two-terminal tandems are electrically connected in series.
* Perovskite takes over for the higher potentials and silicon takes over for the longer-wavelength energies.
* Two-terminal is called elegant and compact, but four-terminal is fairly easy to manufacture now and avoids asking the whole supply chain to adopt the new technology.
📋 **Clip Abstract** The clip weighs two-terminal against four-terminal perovskite-silicon tandem architectures, noting that the four-terminal version is wired in parallel and the two-terminal in series. Although the two-terminal design is described as elegant and compact, four-terminal is presented as easy to manufacture today and simpler to introduce to a market without wholesale supply-chain change.
About the speaker:
* Speaker: Yang Chen
* Company: Microquanta
* Event: Perovskite Connect 2026 | Berlin
* Location: Estrel, Berlin
Save the date: Perovskite Connect 2027 | 20 and 21 OCT 2027 | Berlin
Co-located with the Future of Electronics RESHAPED conference and exhibition
Learn more: https://www.techblick.com/perovskiteconnect
Intestered in exhibiting: learn more here: https://www.techblick.com/perovskiteconnect
#FourTerminalTandem, #PerovskiteSiliconTandem, #ParallelInterconnection, #TandemManufacturability, #PerovskitePhotovoltaics, #TandemSolarCells
Other talks by Yang Chen
- Outlook for Four-Terminal Perovskite Tandem Technology · Perovskite Connect 2026
- Advancing Perovskite BIPV and Real-World Deployments · Perovskites Innovation Day 2025
- · The Future of Electronics RESHAPED 2025
More highlights from Yang Chen
- ▶ Which efficiency split lets a semi-transparent perovskite top cell and a silicon bottom cell reach 26.63%? · Perovskite Connect 2026 | Berlin
- ▶ Why choose a four-terminal tandem when the two-terminal design is more elegant and compact? · Perovskite Connect 2026 | Berlin
- ▶ How does four-terminal tandem compare on annual energy yield and LCOE against two-terminal? · Perovskite Connect 2026 | Berlin
- ▶ Does a 9% silicon bottom cell undermine the business case for adding a perovskite boost layer? · Perovskite Connect 2026 | Berlin
- ▶ Why is vacuum-flash drying secretly destroying your perovskite module's long-term UV stability? · Berlin 2025
- ▶ How can a 14% efficient perovskite panel outproduce a 21% silicon panel by 30% in real-world field conditions? · Berlin 2025
