While silicon PV is currently king of the PV landscape, CIGS is a PV technology that has the potential to make significant inroads in the next several years. While progress has been slow due to the complexity of co-depositing four elements to generate the material and the need for encapsulation of the moisture-sensitive product, the high efficiency of CIGS (near that of single crystal PV), coupled with the requirement for only small amounts of the raw materials, provides one of the few paths to beating current Si-based PV in terms of cost.
The past ten years have been difficult for CIGS. During the PV bubble timeframe (2005-2008), there was much activity in the CIGS area with dozens of start-ups in the field. However, although CIGS held much promise, the combination of the financial crash of 2008, which dried up funding, the immaturity of the CIGS process, and the aggressive price reductions in silicon PV as China ramped that technology to volume resulted in most CIGS companies going bankrupt or being acquired at a deep discount.
Despite these challenges, today silicon PV is very mature and nearing its limits of cost reduction, while CIGS is still early in its ramp to high-volume production, with many opportunities for economies of scale improvements that should lower the overall price to less than that of silicon PV.
CIGS is also out of the lab and beginning to enter large-scale production, as evidenced by the construction of high-volume factories in Taiwan by TSMC, in China by Hanergy, and by Solar Frontier in Japan. Building-integrated PV (BIPV) products such as Dow’s solar shingle (available in 17 states to date) are also now on the market. Hanergy is now even selling CIGS modules at IKEA in the U.K.
There is, as a result, renewed optimism for CIGS; however, it must be tempered with some measure of skepticism. After all, everything seemed poised for rapid CIGS growth several years ago, and some of the factors that brought it down then are still here today (cost competition from silicon, need for capital for factory construction, acceptance in the market).
Our analysis shows that while there is some risk, the entry of large corporations such as TSMC and Hanergy will both provide the financing and high-volume manufacturing excellence CIGS needs to be successful and allow it to make serious inroads in the PV landscape.
Industry Consolidation and Shake Out. Are We Done Yet?
Years of consolidation look to be nearing the end for CIGS. In 2010 there were more than 25 companies that had plans for a GW or more of CIGS capacity, but only one (Solar Frontier) achieved that goal. Most of the others are now either defunct or have been acquired.
In fact, in the years after the financial crisis of 2008 and the subsequent brutal shakeout of research-based companies, there were serious questions as to whether any new capacity would be added beyond the 3 GW in production at the time.
The difference today is that the CIGS landscape has shifted from small start-ups to large corporations. At this point, TSMC, Hanergy, and Solar Frontier are the only companies in the field with ready access to the capital necessary for the next round of factory construction.
Several recent announcements by these companies show their commitment to the technology. Solar Frontier announced in December 2013 plans for a new 150-MW plant. Hanergy, in January 2014, announced long-term plans to install 3 GW of capacity, beginning with two 300-MW plants. TSMC has over 40 MW in production, which is currently being fully utilized. An additional 120 MW will be entering production by the end of 2014. While the company’s expansion plans are strong, they are currently less aggressive than the original 2011 plan to have 1 GW in place by 2015.
There is currently little or no new capacity being built outside of these three companies, however. Solexant recently came out of stealth mode as a CIGS company named Siva Power and is trying to raise money for a 300-MW line. Its goal is to realize a 40-cent-per-watt price point.
With respect to production technology, printed CIGS, which was highly touted as the low-cost method for CIGS deposition, no longer has a viable backer since Nanosolar exited the market in 2013. It now looks like batch and roll-to-roll sputtering are the most manufacturable methods for CIGS deposition. Electrodeposition is also a possible method for low-cost, lower volume production.
Enter Fresh Capital: TSMC and Hanergy
This report looks at how the growing interest in CIGS "this time around" differs substantially from last time, and at the proven fundamentals that will determine whether CIGS can achieve the volumes and money it aims for. One thing that seems to be in favor of success this time is the entry of large, well-funded companies with extensive high-volume manufacturing experience. Both TSMC and Hanergy have the funding and the experience to move into the CIGS area and be successful. The shift from a VC start-up model to a large corporate model should bode well for CIGS success.
Hanergy is continuing its buying spree of smaller CIGS companies. In mid-2013, the company acquired Global Solar Energy and its line of flexible BIPV and portable charging products. In the past year, Hanergy also purchased MiaSole and Solibro.
The firm is committed to building out up to 10 gigawatts of capacity and is running both MiaSole’s roll-to-roll sputtering process and Solibro’s batch co-evaporation process. It will be interesting to see if the company settles on one or the other or maintains both.
TSMC, meanwhile, runs technology licensed from Stion.
Higher Efficiency and More Flexible Modules: the Path to Market Share for CIGS?
A significant milestone for CIGS was achieved in October 2013 when ZSW (Center for Solar Energy at Baden-Wurttemberg) eclipsed the efficiency of polysilicon-based cells (20.4 percent) with a 20.8 percent efficient CIGS cell. A large part of the appeal of CIGS—now and in the past—is that it is well-suited to building on flexible substrates. Adding high efficiency approaching that of silicon just increases its value proposition.
The various manufacturing aspects that make the use of flexible substrates possible are pointed out in the main body of this report, but we point out here that the use of flexible substrates means two things for CIGS' commercial prospects. First, it enables the production of flexible devices with twice the efficiency of existing amorphous silicon flexible products. Second, flexible CIGS also promises money savings through roll-to-roll manufacturing and the use of flexible substrates that are cheaper than glass.
For flexible CIGS, however, there are questions that must be addressed. Will flexible PV devices really be as big a market segment as the CIGS manufacturers and developers say they will be? Will roll-to-roll manufacturing produce real savings in manufacturing costs? And will the CIGS devices and processes be better than flexible options based on other PV technologies, mainly thin-film silicon? We discuss these issues in detail in this report.
CIGS and BIPV: A Match Made for Rooftops?
CIGS became very closely associated with BIPV even in its earliest days, and this relationship has not been diminished by the economic crisis and recession. Flexibility has always been a part of the value proposition of CIGS, at least for some applications, and in some cases the applications are enabled by that flexibility itself.
Flexible BIPV products fall squarely in this category and represent the highest volume and money-making prospects for flexible CIGS. This flexible BIPV sector is currently filled only by thin-film silicon-based products, which, while slightly lower in cost, have much lower efficiencies than competing CIGS products. We pay special attention to products like the Dow Powerhouse shingle, which certainly seems to have the potential of being a breakthrough product. Since its introduction in late 2011, the company has expanded sales of the product to 17 states at last count.
But the BIPV market is not made up of only flexible products, so where are the CIGS-based rigid BIPV products? While there is no technical reason that rigid CIGS could not be used for BIPV, those who have tried to date have failed, mainly due to increased encapsulation costs compared to silicon and CdTe modules with similar efficiencies. With the exit of Wurth Solar from the ridged BIPV market in late 2013, it looks like this segment of the PV market will continue to be dominated by silicon-based products.
There is also the question of how far BIPV can take CIGS as housing and construction markets continue to languish. Certainly there are now pockets of growth in the construction industry, but with many of the CIGS-based BIPV products under development targeting consumers, it is uncertain whether CIGS-based BIPV will be a real growth segment compared to, say, conventional panels.
Flexible CIGS' Achilles Heel: Lifetimes and Encapsulation
Despite all of the excitement surrounding flexible CIGS-based BIPV products, the most substantial barrier to these products' viability—beyond the question of whether they can be manufactured cost-effectively in high volume—which is their sensitivity to moisture, cannot be ignored.
Conventional panels and other rigid products don't have this problem, because they can be effectively encapsulated in impermeable glass. Flexible CIGS development, on the other hand, has required the simultaneous development of robust, reasonably-priced, flexible encapsulation.
We examine the status of flexible encapsulation development for CIGS and look at how it is influencing the commercialization of flexible CIGS products, both in terms of timing and the types of products. We also explore the question of whether the ability to make flexible products will get ahead of the ability to exclude moisture from them and give them the lifetimes they need for market viability.
Will the money spent on better flexible encapsulation translate into more money for manufacturers, or is it best to leave lifetimes a bit shorter if the costs are too high? Reliability has greatly improved for CIGS flexible cells, but more work needs to be done to reduce costs to make it competitive with ridged modules.