Perovskite technology continues to attract attention as developers work to scale it from laboratory research to commercial manufacturing. At SNEC 2026, TaiyangNews spoke with Dr. Bin Fan, Founder and Chairman, GCL Optoelectronic about the company’s commercialization roadmap. He also discussed its 4-terminal tandem approach, advances in single-junction perovskite technology, and challenges of bringing perovskite modules to market. Following is an edited transcript of the interview.
TaiyangNews: We understand GCL’s perovskite division has important news regarding the commercialization of its technology. Can you share the latest update on your production plans?
Bin Fan: This will be our 1st year of mass production. We installed our 500 MW production line in 2025 which is now ramping up. We expect to begin shipping modules to customers by the end of Q3 2026. Our target for this year is to deliver around 50 MW to 70 MW. While that volume is small compared with silicon manufacturing, it is a significant first step for perovskite.
TaiyangNews: Could you briefly explain the advantages of GCL’s 4-terminal (4T) tandem technology over the 2-terminal (2T) approach?
Bin Fan: The biggest advantage of 4T is that it allows us to use a more stable perovskite system. In a 2T tandem, the current between the two sub-cells must be matched very precisely, limiting the perovskite bandgap to around 1.7 eV, which is relatively unstable. With a 4T architecture, we have a much wider selection range and typically use a bandgap of around 1.55 eV, which has already been proven to be stable in single-junction devices.
Using this stable perovskite system, we have achieved 30% efficiency for this module size and 27% efficiency for our 2 m² module. These modules have also passed IEC 61215 certification, demonstrating their stability. Another key difference is commercialization. We have not yet seen real-world deployment of 2T products, whereas our 4T technology already has 1 MW deployed. We expect deployments to reach tens of MWs this year.
TaiyangNews: If the advantages of your approach are so compelling, why do you think more companies are not pursuing the same path?
Bin Fan: We consider our technology to be the mainstream approach because companies with real mass-production lines are largely following the same path. Companies such as BOE, CATL and several other startups in China are developing perovskite technology on glass substrates, just as we are.
By comparison, those pursuing 2-terminal tandem structures by depositing perovskite directly on silicon are still largely at the laboratory stage and do not yet have production lines. So, if you look at actual production lines, we believe our approach represents the mainstream direction.
TaiyangNews: Besides your 4-terminal tandem technology, you also have a compelling story around perovskite itself. Can you give us some background and the latest achievements with your perovskite technology, particularly the single-junction perovskite?
Bin Fan: One of our recent important findings is that single-junction perovskite generates much more electricity during cloudy or rainy days. Based on observations from our own rooftop installation in May, the single-junction perovskite generated 11% more electricity per kilowatt.
The higher electricity output is a significant advantage. Our combined 4-terminal tandem design generates slightly less additional energy than single-junction perovskite, but it still generates more electricity than single-junction silicon.
TaiyangNews: What is your production roadmap as you enter mass production?
Bin Fan: We are focusing on universal products—standard modules that can be used across different applications, including utility-scale, distributed generation and residential markets. They all use the same technology and the same module design. We consider this to be a platform technology.
Once this platform technology is sufficiently mature, we can transfer it to other substrates, such as flexible modules. We already have a flexible module of this size certified at almost 22% efficiency, which we believe is the world record and the largest of its kind.
This platform can also be extended to other applications. Flexible modules could address areas such as consumer electronics, while space is another promising application. For space, we have achieved efficiencies of more than 30%, which is already in the range of gallium arsenide, while maintaining production costs in the range of silicon. We believe this combination makes the technology well suited for space applications.
TaiyangNews: I noticed that you have a satellite model on display. Can you tell us more about your space applications and why you think perovskite technology is more suitable for them?
Bin Fan: In China and Europe, gallium arsenide was widely used for space applications. In the US, however, silicon is more commonly used because launch costs have significantly reduced, making gallium arsenide too expensive. The drawback is that silicon has relatively lower efficiency, perhaps around 10% lower than gallium arsenide.
Perovskite combines the advantages of both technologies, offering the high efficiency of gallium arsenide and the low cost of silicon.
Another advantage is radiation tolerance. Silicon wafers are relatively thick, so radiation can cause damage and degradation. Perovskite is a very thin film, about 400 to 500 nm thick, allowing radiation to pass through more easily rather than damage the material.
The remaining challenge is temperature. On the ground, temperatures typically range from -40°C to 85°C, whereas in space they can range from -100°C to 100°C. This creates challenges for encapsulation and material selection. However, based on the combination of high efficiency and low cost, we believe these challenges can be solved.
TaiyangNews: Space does not have the humidity problem, which is a major challenge for perovskite. However, UV radiation is much stronger in space, and perovskite is also UV sensitive. How are you addressing this?
Bin Fan: The lower humidity requirements in space allow us to reduce the demands on encapsulation. A moisture barrier that may not be sufficient for terrestrial applications can already be good enough for space, which is an advantage for us.
The downside is the much stronger UV radiation. The typical strategy is to use an anti-UV thin glass, or doped glass, which is already a common approach for silicon, gallium arsenide and perovskite space applications. We can use the same UV cut-off glass, so that is not a major problem.
The real challenge is thermal cycling. It occurs very frequently—about every 80 to 90 minutes—and the temperature extremes are much greater than on Earth. We therefore need to consider the thermal behavior of all material layers to avoid cracking under these frequent and severe cycles. But that is essentially an engineering problem.
TaiyangNews: Coming back to the perovskite commercialization project, what are the key challenges you are currently facing?
Bin Fan: On the technology side, there used to be 3 major difficulties. The first was making the modules large enough. We have solved this quite well while maintaining high efficiency. The second was stability. We have now proved that our good modules can pass IEC tests, which is quite important.
The third challenge is related to production yield. This will be a real challenge in the coming years. If we produce 1,000 modules per day, how do we make sure every module is good? A good module can perform for a long time, but if a module is bad, its lifetime can be very short. So we need to deploy our modules in the real world, build demonstration projects, conduct tests, and feed the results back into our production line. This will help us fine-tune and upgrade our technology. That will be the most difficult part in the coming several years.
On the commercial side, we are still at a very small scale. So our production cost is much higher than silicon, mainly because of material costs. Based on our calculation, with the 500 MW line this year, our cost may be twice that of silicon. When we have 1 GW running, the cost can be about 50% higher than silicon. When we have 3 GW running, the cost can get very close to silicon.
It all comes down to scaling. However, if we don't have enough orders, how can we scale? If we don't scale, how can we bring down the cost? So that's a chicken and egg issue. We are going to tackle this step-by-step.
TaiyangNews: Thank you.