Too Many Changes At the Same Time

TaiyangNews: The impact of COVID-19 on solar demand does not seem to be as bad as anticipated in a number of markets; what is your view on the market for this year and next?

Mark Ma: We think this year the total global PV installation would be a little bit lower than last year, but it’s going to be flat for major markets like China. The US is also not heavily impacted. For next year, we believe that the installations will come back stronger and may even increase over this year’s level as well as that of 2019. We already see that several countries are coming out of the COVID-19 situation and installations are coming back. Due to the higher module price in the second half of this year, some of the installations will be postponed to next year as well. That’s why we are very optimistic on installations next year and expect the demand to be pretty high.

Industry has strong confidence in the future; that is why we see a lot of capacity expansion. There could be some oversupply issues in the short term, but in the long-term solar installations are increasing at a fast pace every year. Maybe in a couple of years, we will reach the equilibrium of supply and demand. If you look at the first half of this year, the price has dropped a lot, which is actually not sustainable. That is why we see such a big increase in the second half. Next year, the price may drop a little bit but may not get to the level where it was in the first half of this year. It is going to be a more sustainable price and if demand is not that dramatic I think the price will drop gradually.

TaiyangNews: What are the key topics for the solar sector as it is now often the lowest cost power generation source—and continuing to decrease costs and prices—and adding more capacities than any other power generation source every year? 

Kaushik Roy Choudhury: We are seeing significant innovation on the technology side  in modules and some in inverters as well. A lot of new technology changes are being pushed into the market, for example, larger modules, half cut cells, which are almost mainstream now, new types of cell doping to mitigate LID. These innovations are coming through rather fast, considering the pace at which the solar industry has moved in earlier decades. The important thing with moving forward with these innovations in performance and quality is that they need to be balanced with overall long-term durability of these solutions. With each new innovation, there are bound to be some reliability and durability concerns. So the balance between new technology being pushed into the market and its perceived longer term durability will play a key role in the next phase of PV development, say in the coming 8 to 10 years.

Mark Ma: Reliability also influences bankability, which is especially important for new technologies. If we could help the industry to strengthen the confidence on reliability, that is going to greatly help the adoption of these new technologies. More so now as PPAs are executed for longer and longer periods; all the economic calculations are based on assuming that modules last and produce power for whatever their expected lifetime is—25, 30 or even 35 years.

TaiyangNews: Looking at the recent technology developments, modules based on larger wafer formats are mostly topping the list now. What is the influence of module designs based on larger wafers on reliability?

Kaushik Roy Choudhury: If you look at larger modules, they have made the industry move towards half cut cells in order to pack more power into each module, as a result of which the industry has also gone into different kinds of ribbon technologies. As the modules are becoming larger, module makers are also seriously considering and, in fact, even implementing the approach of bringing the cells closer to each other. So, the inter-cell gaps are reducing quite a bit, again to pack more power into each module. Tiling is one example of a technology that features  these changes. On the other hand, these technologies also influence module reliability. Concerns exist around new interconnection solutions such as tiling . For example what kind of localized strains is it causing in modules that are deployed in the field.

Additionally, at the module  level, with the higher power generated, there are chances of modules running at higher temperatures. Hot spots and general temperature handling capabilities of the module can potentially become a concern. In the context of bifacial modules, power generation is even higher and, therefore, the concern can grow further. Also, as modules become larger and larger, the mechanical stability of these modules will also be challenged in the current racking and tracking systems. How that’s handled is a question, in the case of, say, double glass modules, where two pieces of semi-tempered or heat strengthened 2 mm glass is used on both sides of the module. It would be an interesting challenge to see how modules handle mechanical stress; it is unproven in the field. We have some test results showing that a single piece of 2 mm glass or even sandwiched pieces of 2 mm glass are much less strong than what is required to resist the impact of mechanical loading. 

Mark Ma: I also feel that there are too many changes at the same time. Starting from the cell, we are changing to half cut, ribbons, encapsulation material, module size and then changing the structure to glass-glass. The industry is moving very fast, which may raise unexpected issues in these early stages and it may take a couple of years to see more mature products. I personally think it would be better to have some gradual change rather than a dramatic one as we are seeing now.

TaiyangNews: Can you share any relevant experiences with your field-testing results?

Kaushik Roy Choudhury: As you know, DuPont takes field reliability of the modules very seriously. We run a global field program. A case in point, which underscores Mark’s concern that the industry is probably moving too fast with many technological changes being introduced at the same time, is about backsheet technologies. There were a few new backsheet materials that were launched 7 to 9 years ago. The driver was definitely cost at that point and maybe the option of other materials that looked somewhat similar at the beginning of testing. But what was unknown at that point was their overall long term reliability in the field, which is difficult to assess. Typical IEC testing does not really test for that, because the protocols focus on meeting quality standards at the infancy stage of a panel.

The field program has shown that after a few years of their deployment in the field, some of the not so rigorously tested materials or technologies are showing defects. At that point, it is causing more concern to the asset owner than what it would have if some of the technologies had been tested properly before deployment. For example, we are finding more and more cases where one of the leading backsheet materials in the industry, polyvinyl difluoride (PVDF) is showing backsheet cracking in the field.

This material did not show any defects in the first four to five years. The starting properties of a lot of these materials are pretty good. However, in some cases, the long-term reliability is not vetted until you see it in the field. That’s why we always go back to the field and try to get as much data from the field on technologies that have been out there and try to figure out how we can integrate that into the material design and, therefore, in the module design.

TaiyangNews: What are the initiatives from DuPont to educate the module makers as well as the end customers about module reliability?

Kaushik Roy Choudhury: We look at it from all the different technical angles. Internally, as well as in collaboration with pretty much all the major test labs and research organizations in the world from Fraunhofer ISE in Europe and Sun Yat-sen University in China, IIT and NCPRE in India, to PVEL and NREL in the US and several others, we have been trying to develop field-relevant accelerated testing to be able to predict what happens in the next 25 to 30 years in the field. This cannot be done in isolation. We need the whole industry to agree on certain things.  We don’t want to introduce any kind of testing, that is not predictive of the field. While doing so, we have been engaging technology experts from the industry all along the value chain, from module makers to independent technology experts in test labs as I have mentioned. That’s on the technology side.

On the field side, as I mentioned, we have a  global program where we put the information out every year. There, we are trying to help the asset owner to understand the breadth of component degradation issues and module failures that can occur in the field. Our field program looks holistically at all aspects of module degradation, it’s a complete look at the bill of materials, mainly for crystalline silicon solar panels. By providing this information, we believe we empower end-users so that they are aware that certain type of materials perform better in the field. They can avoid making choices which result in premature power degradation and catastrophic panel failures.

TaiyangNews: What is your impression of the current industrial practices?

Mark Ma: I feel there is a fundamental either short-term or long-term mindset when developing or launching a new product. There is an interesting gap we observed in the industry when we talked with downstream customers. On the one hand, they keep stretching the lifetime of the module in their financial models to 30 or even 35 years. This is based on the assumption that a module can last for 25 years, and maybe it can also last for another 5 or even 10 years. On the other side, some module makers are trying everything to cut corners, making the modules just good enough to pass the current test standards. I am less confident that the current lot of modules can last for 25 years compared to the earlier ones due to measures taken to cut costs.

I think the whole industry is having a big gap and unfortunately this gap is getting bigger and bigger. From our perspective, we are a material science company with 218 years history, so the key difference is we have a lot of information at our disposal. The more we know, the less bold we are. This may not be the case with others, who might have not seen field failure cases of AAA backsheets and PVDF backsheets or those that primarily focus on short-term results, which is why they are promoting the idea that just passing the IEC test is sufficient. This could become an issue for the industry in the long term when long-term durability is taken into account. Our approach is to fundamentally understand and make sure the material can last 25 to 30 years.

The good news is, many 3^rd party testing institutes such as NREL, PVEL etc. are working toward developing better lab testing protocols to better simulate long-term field aging mechanism. So downstream owners could differentiate the long-term reliability performance of difference modules and reduce their risk.

 TaiyangNews: What are your key takeaways from various field testing and collaboration programs you have initiated?

Kaushik Roy Choudhury: When any new material is introduced to the market or changes to existing material are made, we think the material should not only be tested to pass the lowest standard as Mark has mentioned to get in the market, but it should be tested to field relevant conditions that would simulate the performance, or lack of it, for the entire lifetime that’s expected of that product; in this case like 25, 30 or 35 years.

There is a lot of PV deployment already in the market, spanning over different generations of technologies. We need to get information from the field and use that to design a product so that we don’t make the same mistakes from the past. This concern is spurred by the fact that we are seeing some pretty advanced technologies, including bifacial ones, failing prematurely in the field.

The other thing that is really important, as Mark just pointed out is that, instead of taking really a short-term cost per watt peak approach, we should take into account the overall value over the system’s expected lifetime. That’s where we will get a real sense of the cost of electricity or the return on investment.

TaiyangNews: What is your opinion about the current testing protocols and where are they lagging in terms of assessing the field degradation and thus the long-term reliability?

Kaushik Roy Choudhury: The main focus of IEC qualification tests when they were written was to ensure that a material or a module meets the required quality standards when it enters the market. Most IEC tests are not really longer-term reliability assessment tests. Although modules in the field are simultaneously subjected to different stresses, UV, temperature cycling, the temperature at which it resides, moisture, mechanical stresses, in IEC testing each module is typically put under and assessed against a single stress. In degradation of both inorganic and organic materials, a combination of stresses can lead to a very different point than what a single stress would lead to. That’s one of the major issues with standard tests. The magnitude and the amount of stress is not really commensurate with what backsheets, modules, other packaging materials or cells would see over a 25 to 30-year lifetime. On the other hand, in some cases the individual stress tests are somewhat overblown. For example, unusual importance is given to the damp heat test. We have seen module makers extending damp heat tests to 3,000 or 4,000 hours.

However, from the field, we have repeatedly seen for both polymeric backsheet and glass-glass modules that the kind of degradation seen in most standard testing, for example in damp heat tests, doesn’t really correlate to what happens in the field.

There are also questions around potential induced degradation (PID) on the rear-side of a module as a result of polarization in bifacial PERC cells. Current testing is done in the dark, but, it has been shown that this type of degradation will most likely not even happen in the field, because polarization-PID losses can be almost 100% recovered in the presence of light. Current changes are being proposed where a light stabilization will be introduced at the end of testing, so this potentially artificial degradation does not prevent a module from being qualified. These are examples of improvements needed in qualification testing standards so that results correlate with the field.

TaiyangNews: How are you trying to fill these gaps with product innovations from DuPont, for example Clear Tedlar for bifacial modules?

Kaushik Roy Choudhury: Clear Tedlar was produced even before the bifacial technology came on. Backsheets based on Clear Tedlar have been in the field for over 20 years. The earlier generation of Clear Tedlar was actually used in building integrated photovoltaics (BIPV). There are quite a few installations out there. We have gathered the data that shows that Clear Tedlar has been out in the field for two decades without any significant amount of degradation; no yellowing, delamination or cracking. The new generation of Clear Tedlar improves the performance and reliability even further. This product innovation helps accelerate the progress of bifacial technology.

 Our experience of close to 40 years of field operation of the Tedlar backed monofacial modules, gives us the confidence about the reliability of this new product, since fundamentally the polymers are same.

There are other significant advantages when compared to the incumbent dual glass structure such as cleaning efficiency and weight.. With larger and larger modules, the difference in weight will increase. At the operations and maintenance end, we know that the Tedlar polymeric film, is hydrophobic in nature. This prevents the accumulation of electrostatic dust particles. Generally, it attracts less dust. Even when the dust does deposit on the panel, it’s easier to clean with water due to the low surface tension  Tedlar backed backsheet compared to typical solar glass. These advantages make it an attractive choice. There are also some performance related advantages. Some field studies have reported that Tedlar backed modules have been generating slightly more power than equivalent double glass modules. This is mainly from the improved thermal dissipation properties of these modules, which are enabling the modules to run at lower temperatures.

TaiyangNews: DuPont supplies only one component of the backsheet; what special care you take from your side that Tedlar based backsheet as a whole meets the reliability requirements?

Mark Ma: Indeed, DuPont is just the supplier of one component of the backsheet, but it’s a very important component. As for ensuring reliability of the total backsheet, we work very closely with all the backsheet suppliers and module makers who adopt our product.

Right from our product launch, we insist on the most stringent testing protocols, so that all the products we launch have to go through long and appropriate reliability tests to make sure they can last 25 to 30 years. We invest a lot of resources to do that. As mentioned before, we are working with other testing institutes on how to improve and how to better evaluate the different materials. That’s the effort we are making; first to ensure our products have 25 to 30 years reliability and also help the industry to better evaluate other different materials.

TaiyangNews: It’s nice to see that transparent backsheet based module products are also now offered with 30 years of power warranty. However, first JinkoSolar and now Jolywood are the only two companies that came forward, whereas glass-glass structure enjoys this privilege exclusively. What is the reason in your view?

Kaushik Roy Choudhury: Many players often consider a single sheet of glass and then equate its properties to the double glass packaging structure in a module. In a module, the double glass structure is part of a package that tries to protect the other components. However, there are certain components inside the module that are also degrading at the same time. So just saying that a piece of glass that’s out there will survive 30 years shouldn’t really directly translate to module reliability made with two pieces of glass on both sides. For example, the impermeability of glass, as we have seen in the past, plays a big role in causing delamination. Additionally, the glass that’s being used in bifacial modules has undergone several changes; the thickness changed from 3.2 mm to 2.5 mm and now it’s 2 mm. The very nature of glass has also changed from being fully tempered to the semi-tempered or heat strengthened variety, changing the mechanical characteristics.

Mark Ma: I think the estimated durability is based on lab testing. The glass-glass structure may show pretty good performance in the damp heat lab test, but there is a significant gap between lab tests and field aging mechanisms. For instance, time; in the lab testing time is short, while in the field we have seen many cases of moisture ingression into glass-glass structures over several years, resulting in the delamination of glass-glass modules. So glass may not be as robust as one imagines. 

TaiyangNews: Thank you for the interview.