■ In the past 12 months since PV CellTech 2017 last year, what have been the main changes in cell manufacturing?

We mainly see developments that are related to high efficiency cell manufacturing – that’s where our products get used. Looking at that sector of the industry, we see an increasing amount of high-efficiency cell lines – primarily PERC, bifacial and heterojunction - being planned or built and brought into operation.

With these advanced technologies, we are also seeing much more interest in technologies and techniques to continuously measure and analyse the process tool results – i.e. the physical properties of the processed wafers – to aid in better line ramp-up and in on-going operations.

Geographically, we are also seeing increasing interest from Indian manufacturers, as well as in other regions, but the main locale continues to be China.

Just before PV CellTech 2017 last year, we brought out a model of our DecimaTM dopant measurement system that can measure the sheet resistance of the back surface field on homojunction bifacial cells without interference from the wafer bulk resistivity. We have successfully qualified this product with the nPERT and BiSoN bifacial designs, among others, and we now have many installed and operational in China and elsewhere. We have also continued with ongoing volume installations for our products in PERC production lines as well.

At the same time, we are advancing our process visualization product 'Aurora Veritas' that can provide detailed information about the stability and performance of processes for these types of cell structures during the fabrication of the back surface field and the emitter, as well as linking this information to the cell efficiency results that are obtained at the cell tester.

■ China is now looking to become a technology leader, not simply a high-volume cell producer. What changes will need to be done at the cell line level, to allow for this?

If you look at other industries that manufacture complex products - such as semiconductor, pharmaceuticals and chemicals – the use of multi-variate measurement and control systems to ensure yields are maximized and quality is well-controlled has been a standard and essential practice for quite some time. That’s how they make profits and minimize costly field or consumer issues.
This approach to quality control is still in its infancy within the PV industry, however. It has been adopted to a degree by some well-established producers of high-efficiency cells, but has not been used to its full extent throughout the entire industry.

In particular for advanced cell production with its complex design trade-offs and narrower process windows, it is now becoming increasingly important for us to learn from these other industries, and adopt their best-in-class techniques in beneficial ways. In those industries there is continuous measurement of critical-to-quality properties during fabrication, and the information derived from these measurements is used to control production so that – in the sense of the product design specification - exactly the same product is made every single time. This is where the cost reduction and profits are, not necessarily in getting the highest average efficiency or the most powerful champion cells.

So, in China, with their push to become a technology leader, learning from and beneficially adopting these techniques is a wonderful opportunity for them to become a leader in this area, to drive their profitability and build a reputation as global quality leaders.

■ Industry 4.0 is now emerging as a key theme – can you explain simply what this is, and how it helps the industry as a whole?

The term was initially developed in Germany. The “4.0” comes from looking as far back as the industrial revolution in the 1800’s (as “1.0”), then continuing through various innovations within industrial manufacturing to the present.

The principles of Industry 4.0 are that there will be autonomous automatically controlled and communicated smart processes within the plant; involving smart sensors, intelligent devices such as robots, and smart control systems making decisions regarding the control of processes and the flow of materials through the plant. Consequently, these plants will be highly automated with a minimal amount of human intervention.

It’s crucial to also understand that Industry 4.0 is not just a set of technologies, but also a way of looking at things; planning and operating the plants. So to implement it, one must focus on both the equipment and the operational culture as a system. It involves training and building the culture and capabilities of the plant designers, engineers and managers. It requires a high degree of planning discipline, and day-to-day operational diligence – in terms of how things are done within the company as a whole.

For Chinese manufacturers, the idea of Industry 4.0 has been adopted within the made-in-China 2025 plan that was released in 2015 by the Chinese government, but this something we hope and expect to see adopted in PV globally.

One of the expected outcomes of successfully adopting the principles of Industry 4.0 is that products will be produced with little or no quality-affecting variations. When you have a specific product design and an objective of manufacturing exactly to that specification, then it should be achieved to the fullest extent possible every single time. This dramatically reduces costs all the way from the end of the production line to the field– thereby achieving a much higher degree of profitability in day-to-day operations including repair and warranty needs.

Yes it is. As I have mentioned, the main thing is not so much the particular average cell efficiency – whether for example 25% or 25.2% - but rather that you don’t end up producing cells that are ranging widely, say between 23% and 25%. As you look along the value chain to market, such unwanted variations are the profit-killer. Advanced process lines can also have narrower process windows. So you will more readily get these unwanted variations if you do not pay close attention to what is going on in the line.

So the purpose of manufacturing process control – and inline measurements that enable it – is for example, when setting a target of 24.9%, that every single cell produced is 24.9% with the same Voc, Isc, series resistance and other key characteristics. This means that you need to design in the means to achieve this, right from the planning stage, through line design and ramp-up and into day-to-day operations.

■ How can we quantify the benefits from greater yield/efficiency, compared to the capital costs of upgrading cell lines?

The quantification tends to be on a line-by-line basis. For some of the high-efficiency designs, during ramp-up and day-to-day operations, it is more about stabilizing the process. For improvements within existing facilities, it depends on how you use the equipment.

In 2007-2008, one manufacturer conducted a rigorous experiment in their production lines – it actually became part of one person’s Ph.D. work. Over a period of several months, they were able to decrease the standard deviation of a group of selected processes by 55 percent. That’s a large number. Besides reducing the EOL bin distribution, this reduction in variation safely allowed material and recipe optimizations resulting in a 2.5 percent relative increase in average efficiency. Not bad, considering the technologies and knowledge available at the time. Just think what could be accomplished now!

Furthermore, this does not even consider the benefits downstream from the cell fab. Eliminating variation enables several profit multipliers as we move from the cell fab into module production and field deployment. One of these is increased predictability and reliability in matching cells and module material to consistently maximize module power performance. It’s difficult to do that very well if you are grappling with much cell variance. Another is a reduction in inventory and warranty provisions, especially as system operators are becoming more and more diligent in monitoring module performance over time. Lastly, non-obvious process faults that can cause accelerated degradation or outright failure in the field can be caught and corrected. In one recent case, a single undetected fault – which could not be observed in I-V results alone – caused thousands of mis-processed cells at one manufacturer. Fortunately these cells – which later failed damp heat tests - were caught using advanced process monitoring and analysis tools. If they had made it to the field, the warranty liabilities could have been in the hundreds of thousands to millions of dollars.

■ And finally, what would you most like the GW-level cell makers’ CTOs to be discussing on stage during the event?

How to grow and make money at the same time. And what they need from us (the equipment companies) for that. The manufacturers should be consulting on and defining industry goals, and then challenging everyone to contribute to them. For instance, right now we are breaking through the 100 GW production level. What about when we’re breaking the Terawatt level? It’s not really that far away. Are the same production methods and technologies going to get the job done? And what about increasing profitability? So I would be very, very interested to hear manufacturers’ thoughts and plans for how to build and run businesses that operate profitably at that scale. What will the operational challenges be and how will they be met? What will the production tools and operational processes look like? How does Industry 4.0 fit in? What do they want and expect from the business and technical communities in the industry – the researchers, equipment providers, financiers and manufacturing experts? Asking and answering these sorts of questions together is what makes industries succeed as they grow.