Researchers from the Karlsruhe Institute of Technology and the University of Valencia have developed a rapid solvent-free vacuum process that could help scale up perovskite-silicon tandem solar cells, one of the most promising technologies for next-generation photovoltaics.

A new manufacturing route for tandem solar cells
Perovskite-silicon tandem solar cells can achieve significantly higher efficiencies than conventional silicon solar cells. Yet one of the biggest barriers remains large-scale manufacturing. A new study from the Karlsruhe Institute of Technology (KIT) and the University of Valencia, published in the journal Nature Energy, introduces a solution that could help bridge the gap between laboratory performance and industrial production.
The researchers developed a rapid solvent-free vacuum deposition technique capable of creating uniform perovskite layers on textured silicon surfaces, overcoming one of the most challenging aspects of mass production.
Tandem solar cells are widely regarded as one of the most promising technologies for pushing photovoltaic efficiency beyond the limits of conventional silicon devices. By combining a perovskite top layer with an underlying silicon cell, these devices capture complementary portions of the solar spectrum. The perovskite absorbs high-energy light, while silicon converts lower-energy wavelengths, resulting in significantly higher overall energy conversion.
The challenge has always been industrialization. Depositing a thin perovskite layer quickly and uniformly across large areas, particularly on textured silicon surfaces, remains a complex technical task.
A vacuum process that could transform production
The two research teams refined a technique known as Close-Space Sublimation (CSS), a vacuum-based process that deposits perovskite precursor materials directly onto silicon cells without using solvents.
The process is relatively straightforward. The precursor materials evaporate and travel only a few millimetres before settling on the silicon surface, where they react to form the perovskite layer.
The advantages are substantial. Conversion times are reduced to just 10 minutes, material consumption remains low, and the source materials can be reused.
“For industrial production, it is not only efficiency that matters, but also speed, robustness and scalability,” said Professor Ulrich Paetzold of KIT. “We were able to demonstrate that an exceptionally fast vacuum process not only produces uniform layers but also generates efficient perovskite-silicon solar cells.”
Applying perovskite onto microstructured silicon wafers is one of the key obstacles to industrial scaling. Producing defect-free, uniform films on non-planar surfaces without voids or local composition variations is particularly difficult.
Bromine helps control the band gap
To achieve the required band gap, the researchers had to solve another challenge. During the conversion process, bromine, a critical element for increasing the band gap, tended to disperse.
The solution came through careful adjustment of the organic precursor chemistry.
“By modifying the ratio between methylammonium iodide and methylammonium bromide in the organic sources, we were able to control the bromine content in the final material,” explained Dr. Alexander Diercks of KIT, who worked on the project at the University of Valencia through the Horizon Europe Nexus programme.
Effective on smooth and textured silicon surfaces
Another critical factor for industrial deployment is the morphology of the silicon surface. High-efficiency solar cells often use textured surfaces featuring micro- or nanostructures to increase the optical path of incoming light and improve absorption. However, many deposition methods only work effectively on perfectly smooth surfaces.
The researchers tested the CSS process on three different types of silicon subcells:
- Smooth surfaces
- Nanostructured surfaces
- Microstructured surfaces
The results showed uniform perovskite layers across all three surface types without requiring changes to the process parameters.
Scanning electron microscopy and X-ray measurements confirmed consistent coverage.
“A process that only works on perfectly smooth surfaces would have limited industrial value,” said Professor Henk Bolink of the University of Valencia. “The fact that close-space sublimation produces uniform layers even on textured silicon cells makes this approach highly relevant for practical implementation.”
What this means for the future of photovoltaics
The reported efficiencies, ranging from 23.5% to 24.3%, do not set new records. The best laboratory-scale perovskite-silicon tandem solar cells already exceed 33% efficiency.
However, the most significant outcome of this research is not the efficiency figure itself. The study demonstrates that a fast, solvent-free process compatible with standard industrial silicon surfaces can produce efficient tandem devices, a crucial step toward commercial manufacturing.
The remaining challenge is long-term operational stability. Achieving thousands of hours of real-world performance while maintaining high efficiency will likely be the decisive test before perovskite-silicon tandem solar cells can be deployed at scale, a development that could have important implications for rapidly growing solar markets in Africa and other regions pursuing affordable clean energy expansion.






