Tool standardization a la the semiconductor industry hasn’t yet evolved for thin-film solar PV manufacturing, but in the meantime there’s a lower-cost alternative to squeeze out manufacturing costs, explains Lauren Ravenscroft from Owens Design.
By Lauren Ravenscroft, national sales manager, Owens Design
November 16, 2011 – Conventional wisdom has held for many years that for the solar industry to succeed, it must be able to achieve cost/watt parity with conventional energy sources, such as oil, natural gas and coal. Achieving this goal has been frustrated by two main issues, the relatively low cost of conventional power compared to solar and the cost of solar manufacturing, which is one of the most significant contributor to solar energy costs.
The challenge of reaching cost-per-watt parity has been mitigated in recent years by two factors: the rise of the cost of oil, and the growing concern worldwide about climate change, which has highlighted some of the hidden costs of conventional energy sources. The cost of using conventional sources has been further raised by regulations requiring the mitigation of various emissions. Despite the success of the Chinese photovoltaic (PV) industry, which has succeeded in lowering the cost/W below $1.10 thanks to government subsidies and low labor costs, the PV industry as a whole is still pursuing the cost/W parity Holy Grail.
A large contribution to solar energy costs comes from cell-level manufacturing; therefore, lowering these manufacturing costs would be one of the keys to ensuring widespread adoption of solar energy. At first glance, since thin-film PV manufacturing uses similar processes to those used in the semiconductor industry, it would seem that the semiconductor industry offers a clear path to reducing manufacturing costs: tool standardization.
When the semiconductor industry was in its infancy, the major chip manufacturers such as Intel and IBM built the equipment for manufacturing their individual chip designs. During this period chip prices were relatively high, and were limited to certain applications by cost as much as by functionality. Development of a semiconductor capital equipment industry — with standardized manufacturing processes and form factors — helped drive down the cost of semiconductors to the point where they have become ubiquitous in a variety of applications, from communication to computing to automobiles.
Today, there are currently two main types of PV cells: crystalline silicon (c-Si) and thin-film. Since most c-Si solar cells are essentially the same form factor (with only a couple of size variations), there has been some success in developing process tools that can be sold to a variety of solar cell manufacturers. Unfortunately, while solar modules built using c-Si are among the most efficient on the market, that efficiency is offset by the high cost of silicon.
As an alternative to c-Si PV products, many PV manufacturers (particularly in the US and Europe) have moved to thin-film alternatives that depend primarily on the deposition of films on substrates other than c-Si. For now, thin-film PV cells are less efficient at converting solar energy than c-Si cells, but a number of thin-film PV manufacturers are pursuing technology approaches hoping that lower material costs will offset the lower conversion efficiency, enabling the achievement of the cost/W parity needed to drive widespread adoption. Since the majority of these thin-film deposition techniques are established semiconductor manufacturing processes, it would be natural to assume that PV thin-film solar manufacturers could easily support a PV equipment industry.
Why then, has the thin-film solar industry not been able to pursue this path as an effective means of achieving cost/W parity? In fact, this approach has been tried by a variety of companies, such as Applied Materials and Veeco, with limited success. Veeco, in fact, recently announced it was exiting the solar capital equipment market.
Unfortunately, unlike the semiconductor industry in which devices are routinely built on silicon wafers and have essentially the same form factor, the thin-film solar industry is still in an experimental stage using a variety of different manufacturing technologies. This includes different manufacturing approaches such as plating, sputtering, and implantation, as well as varying materials including copper indium gallium selenide (CIGS), cadmium telluride (CdTe) and amorphous silicon. In addition, there is a wide variety of substrates in use, including glass panels, glass wafers, metallic or plastic roll-to-roll processing, as well as silicon cells of various shapes, thicknesses, and sizes.
To realize tool standardization in thin-film solar manufacturing, one of these manufacturing approaches must demonstrate a clear advantage and win out over the others. Until then, the market to support “off-the-shelf” solar tools for this non-c-Si portion of the market is not likely to emerge. Quite simply, the demand for standardized thin-film tools is not sufficient to make the equipment supply business profitable. Even if one of these technologies emerges as a clear winner, it is likely that some of the other approaches will continue to play a role in niche markets.
With manufacturing costs largely contributing to the cost of solar energy, and lacking the robustness to support an “off the shelf” equipment market, how can thin-film solar manufacturers achieve cost/W parity needed to fuel widespread adoption? In fact, a lower-cost alternative to developing and building tools in-house already exists, one that has remained highly useful in the semiconductor industry despite process and tool standardization and is being increasingly employed by solar thin-film companies: collaboration with an experienced manufacturing tool design house. In this paradigm, the design house focuses on such issues as automation, product handling, and tool manufacturing that are designed to help lower manufacturing costs — freeing the thin-film solar manufacturer to focus on improving its core processes and increase the solar efficiency of its product.
A successful collaborative approach can benefit not only the creation of new tool platforms, but also play a role in the continued evolution of existing tools. An established PV OEM or PV manufacturer can use this collaborative approach to reduce the cost of ownership of existing process tools, while adding new functionality to the existing tool platforms. In the case of emerging companies, the collaborative development approach enables them to accelerate product development and meet growing customer demand without having to grow internal engineering resources too quickly.
Unlike the c-Si solar industry, the thin-film PV industry currently employs a wide range of manufacturing processes and materials. As a result of the technological fragmentation, the thin-film PV market lacks the coherency and robustness to support a standardized PV thin-film equipment market. (Although, if one of these key thin-film technologies wins out over the others, this situation will likely change.)
In the meantime, thin-film PV manufacturers are not without recourse while trying to reduce tool costs and improve productivity in pursuit of cost/W parity. Manufacturing tool design houses that help semiconductor tool manufacturers speed the time to market and reduce the cost of ownership of their tools can provide the same benefits to thin-film PV manufacturers. Leveraging the expertise gained from involvement in multiple tool development projects per year, manufacturing tool design and build houses can speed tool development times and reduce tool development costs, while helping to improve tool productivity. This will help thin-film PV manufacturers achieve their desired goal of cost-per-watt parity with conventional energy sources.
Lauren Ravenscroft is national sales manager for Owens Design in Fremont, CA. She joined Owens from Applied Materials, where her roles included business management, key account management, and regional sales management. Prior to that, she held various marketing positions at AMD and National Semiconductor.