Stacking tandem cells – pv magazine International

Excerpt from pv magazine 05/2022

The latest International Technology Roadmap for Photovoltaics (ITRPV) report was released in April, and tandem and IBC [interdigitated back contact] the cells were expected to represent 5% of the world’s production capacity by 2032. As a supplier of high-efficiency photovoltaic equipment, what do you think of this figure?

We consider that to be a reasonable number. In China alone, today there is 300 GW of production capacity and we believe that will increase, not only in China but also throughout the world. So while 5% might seem like a small number, it’s a reasonably high number for a relatively new technology – especially considering it’s in the R&D or pilot stage, depending on the customer. And as we know from recent developments of heterojunction (HJT) or even PERC in the past, it took some time before production was truly gigawatt scale.

You mentioned some customers on the production of tandem cells, can you give me an overview of the level of maturity of these manufacturers and their technologies?

It really is a wide range. There are customers who are in the R&D phase, and at the watt scale. Their main goal is to go from one square centimeter solar cells to a single wafer cell, and that is their R&D approach. But others are already beyond. These companies are already producing at pilot scale and are working hard to bring this technology to gigawatt scale.

As an equipment manufacturer, we see ourselves as a partner to these scaling manufacturers – from lab to midscale to gigawatt scale. To give you some numbers, when you talk about a lab tool, it’s about 10 wafers per day. When we go to pilot tools, it’s 1,000 wafers per day. And then mass production is multi-megawatt and then gigawatt. Although there are no tandem cell gigawatt producers on the market yet.

Are the partners you work with mainly European?

Mainly in Europe, but also in the United States, and in Asia too.

For these future industrialists, how would you describe the challenges they face in terms of technology? We know that stability is a major challenge with perovskite tandems – is that still the main challenge?

It’s one of the biggest, but scaling itself is already a challenge – scaling the technology and all the process steps and getting them into production. Efficiency and stable efficiency is the other. And the availability and cost of raw materials must also be considered if a manufacturer plans to venture into gigawatt production.

And by hardware development, what do you mean?

The perovskite itself. Perovskite is a group of materials, and it has huge advantages in that you can tailor the optical properties, which means the band gap can be aligned based on the structure below, on the underlying cell . Usually, perovskites are applied in tandem devices, with a silicon-based or thin-film cell. And then the optical properties of the perovskite can be tailored to deliver high yields. And the other big advantage of perovskite is the cost-effective synthesis and production of the material itself.

So, while perovskite materials can be cost-effective, obtaining the composition of the material itself is a challenge for developers.

Material selection is a challenge at the moment, not only for long-term stability as you mention, but also for ecological sustainability – which is increasingly important. And of course, at the end of the day, the cost. But the potential for low-cost materials with the perovskite group is huge.

By “ecological”, are you referring to the lead content of certain PV perovskites under development?

Lead is an element, that’s right. A lead-free perovskite would be great, and to be able to achieve high efficiency and lead-free long-term stability. This is something our R&D partners are working on.

As an equipment supplier, we must ensure that we find scalable deposition technologies, in order to reduce capex and opex. It doesn’t make sense to just have cheap hardware that can’t be put down productively. This is not only true for the perovskite itself, but for the entire layer stack.

Regarding layer stacking and its deposition, why do you think relatively complex layer stacking in tandem can be efficiently produced and scaled?

If you look at a basic structure in a tandem device, there is always the bottom cell, which can be HJT, TOPCon [tunnel oxide passivated contact] PERC, or even IBC. And then there are thin film technologies. Above the bottom cell is a recombination layer, then the so-called electron transport layer (ETL) or hole transport layer (HTL) – depending on whether it is an n-type solar cell or a p-type. Above the ETL or HTL is the perovskite layer, then again the ETL or HTL, the TCO [transparent conductive oxide]and metallization.

We consider the layers above the absorber, such as the ETL or HTL and TCO layer, to be critical as they must not damage the layers below – and this is something we are working on. We call the coating process “damageless deposition”. Another challenge for tandem devices, in our view, is the absorber itself. There it depends if you apply a technology like evaporation or maybe a wet chemistry process, or maybe a hybrid approach. There are also question marks for R&D. With the ETL and HTL layer, what needs to be found are scalable processes – it doesn’t make sense to have high-efficiency cells with very low throughputs. We consider sputtering, which is a proven technology for high-volume manufacturing, and is a strong candidate for applying ETL and HTL layers.

And then for the absorption layer, for the perovskite itself, sputtering is probably not an option. It can be a co-evaporation process where you apply different materials for efficient absorption of blue light.

If you were to apply spraying or evaporation processes in this way, how would you describe the process flow?

You need the recombination layer, the ETL layer, the perovskite layer, the HTL layer, the top TCO so between five and eight layers to apply in a tandem structure – on top of the cell technology lower. Looking at both CAPEX and OPEX, we believe it doesn’t make sense to apply eight different pieces of equipment and then handling. A process flow should be defined where two or three or even four layers can be combined in a system. At the end, there are then two or three filing systems, in addition to the standard PERC or HJT or TOPCon processes.

Which processes can be combined?

What we think of as a recombination layer based on ITO, then ETL or HTL, from our perspective, it’s quite reasonable to do in a sputtering system. Perovskite can be deposited by evaporation or by wet chemistry, either by a vacuum system or by a wet chemistry process. And then the ETL and TCO on top of that and there we consider it quite reasonable to apply it in a spray system, where today we have rigs that deliver 10,000 throughput wafers per hour, based on M6, or even platforms exceeding 1 GW per productivity system. From these five basic layers, three systems can result.

What can you tell me about the latest Von Ardenne platforms?

As I said, we have the R&D tool which has a lot of flexibility in terms of technologies and also in terms of materials that you can apply. There you have a platform where you apply evaporation processes, sputtering processes, and also ALD [atomic layer deposition] and the interface with wet chemistry processes. Heating and annealing stations are also included, in between. This can allow R&D engineers to play with different technologies and materials to find a pleasing process flow. However, even at this stage, we are applying evolving technologies. With ALD, we see big GW-scale challenges, but a lot of R&D check-in cells are done, on a very small scale, on ALD processes – or at least some of them. For this reason, we have ALD in the R&D phase to simply transfer record yields from square centimeter to wafer size, to find benchmark processes.

Our customers are developing processes where ALD can be replaced by sputtering. If a good process flow with sputtering and evaporation only is found, then we are sure that we can scale this technology, as we have already scaled evaporation and sputtering to gigawatt scale. For the pilot tool, our customers already know the process flow and sequence, but maybe process variations such as deposition rates, temperature, productivity, etc. must be optimized. But pilot systems are usually online systems where it is possible to drop a sequence of processes. While with the R&D system it is possible to change the flow and sequence of the process, in the pilot system the direction to follow is already clearer. And final variations can help achieve efficiency and stability. But the basic materials are already defined.

Once defined criteria are passed on the pilot tool, we can then take this directly and apply it to the gigawatt-scale tool, such as the XEANova, which is already well known.

SunPower and First Solar recently announced their collaboration on tandem structures. It is not a crystalline-perovskite tandem, but probably crystalline silicon from SunPower and a thin layer of cadmium telluride (CdTe). What do you think ?

It looks like they want to combine classic thin-film technologies with silicon. And I understand that they will put this into production next year. We see this as a tandem, where sputtering and evaporation play a role.

Could it be an IBC-CdTe tandem?

I don’t want to comment too much, but just speaking of physics, silicon is efficient in the infrared and visible range, and CdTe is efficient with the highest bandgap, in the blue range. So within theoretical limits, it really is the perfect fit and a very interesting approach, in addition to perovskites.

This content is copyrighted and may not be reused. If you wish to cooperate with us and wish to reuse some of our content, please contact: [email protected]