The market has taken a shine to an array of nano-based technologies that now challenge traditional photovoltaics on their home turf, says Tom Cheyney.Although crystalline-silicon (c-Si), wafer-based solar cells and modules dominate the photovoltaic marketplace with more than 90% of market share, sales of thin film-based products are growing, and at twice the rate of the entrenched technologies. Leading the challenge to mainstream silicon’s reign are thin-film contenders: amorphous silicon (a-Si), cadmium telluride (CdTe), and copper-indium-(di)selenide/copper-indium-gallium (di)selenide (CIS/CIGS). Organic PV is also gaining momentum. Micro- and nano-enabled thin-film photovoltaics provide an attractive and increasingly cost-competitive cleantech alternative to carbon-based energy solutions. And the newer technologies are appearing in an array of forms, from huge solar glass panel-based power stations to buildings constructed or coated with PV materials, to portable, flexible solar charged batteries.
It’s true that c-Si has the advantage of maturity, material stability, the highest conversion efficiencies, and an already-established global manufacturing infrastructure. But, thin film PV has many advantages, too. It’s lightweight, requires relatively minuscule amounts of active semiconducting materials, and can be applied on a variety of substrates. Thin-film advocates also point to production advantages, such as inherent process scalability and lower cost, claiming potentially exponential savings compared to the capital outlay required for crystalline manufacturing facilities. Several thin-film companies say they will offer full solar panels at around a dollar per watt and grid parity of about $0.10 per kilowatt-hour within the next few years. A recent report on the thin-film, organic, and printable PV sectors from NanoMarkets forecasts the overall thin-film PV markets to grow from just over US $1 billion dollars in 2007 to nearly $7.2 billion in 2015. Three segments – large projects and utilities, commercial and industrial buildings, and residential buildings – will account for more than $5.4 billion of the projected 2015 total. In addition, industry analysts at Photon Research Associates say thin-film PV will experience a 63% compound annual growth rate (CAGR) through 2010, accounting for 2000 MW of the 10,000 MW of total solar cell and module production forecasted for 2010.Thin-film’s players
The state of thin-film PV is a mixed bag, with scores of companies ranging in maturity and size, from early-stage development to volume production operations, and from venture-seeded start-ups to deep-pocketed global entities. Glass substrates with a-Si and CdTe films for large and very large-scale applications represent the biggest thin-film market niches. Market-leader Sharp plans to build a 1 GW capacity silicon thin-film cell plant by 2010. CdTe module pioneer First Solar has also secured several major long-term supply deals, collectively worth more than $1 billion, and will be adding 360 MW of capacity at three new plants in Malaysia. As another example, roll-to-roll web-processing leader United Solar Ovonic has embarked on a multi-stage production expansion of its flexible, triple-junction a-Si and new nano-crystalline-Si cells, expecting to reach nearly 120 MW of capacity by year end, and 300 MW by 2010, at its factories in Michigan, USA.
A silicon cluster tool used for thin-film PV at NREL’s new Process Development and Integration Laboratory
Further down the supply chain, Applied Materials and Oerlikon Solar have both scored impressive customer wins during the past year with their lines of thin-film PV production tool sets. But thin-film PV is not just about big glass. The fledgling CIGS sector seems poised for take-off, with market-segment leader Global Solar already selling flexible, portable solar chargers and adding roll-to-roll manufacturing capacity at its Tucson, Arizona, USA, facility as well as its new fabrication plant near Berlin, in Germany. The material has the highest lab, cell, and module conversion efficiencies, just a few percentage points behind c-Si-based technologies, and is still improving. CIGS, like c-Si-based technologies, is ‘relatively unique among thin films, in that it has no intrinsic degradation mechanism’ susceptible to light, voltage, or heat ‘under normal operating conditions,’ says B.J. Stanbery, CEO and founder of HelioVolt.
Companies such as Miasolé and NanoSolar have already shipped product and have begun their respective ramps to volume. HelioVolt, recent recipient of $77 million in additional investment funds, is moving forward on its plan to build the company’s first fabrication plant, which is to be located in the US, and get product to market in 2009. ‘We have commenced commercial shipments for revenue,’ says Dave Pearce, Miasolé’s Chairman and Chief Strategy Officer, although he declined to provide details of the company’s production ramp to high volume: ‘There are simply too many variables at this stage to provide an accurate forecast of a ramp. These range from efficiency improvements to tool uptime to module certification.’ Pearce did, however, offer specifics of his company’s wide-area, stainless-steel-foil roll-to-roll production process, which features internally designed and built deposition tools. ‘Miasolé’s approach is to continuously sputter all of the [seven] layers of the solar cell in a single pass through a high-throughput vacuum system. Our system is capable of processing more than 250,000 square metres of cell material per year, based on 80% system uptime and the current two-linear-feet-per minute line speed we are operating at.’ Pearce cites ‘the CIGS layer itself’ as ‘the most challenging step of the process,’ noting that Miasolé has a ‘plan of record and is able to consistently make product on a reproducible basis run-to-run.’
HelioVolt’s Stanbery points to one particularly troublesome element in the CIGS cocktail: ‘Selenium presents the most significant challenge in any CIGS manufacturing process. Insufficient selenium in CIGS results in defects, which reduce performance, and selenium is extremely volatile at the high temperatures used to form device-quality CIGS absorber layers,’ he says. His company’s FASST production process uses a printing plate, which ‘prevents the loss of selenium and provides other more subtle benefits as well.’ Stanbery also claims the process is ‘highly flexible and controllable…which allows us to explore many different options to optimize throughput and produce competitive manufacturing costs in a relatively low volume production facility compared to other methods.’ FASST also ‘facilitates the formation of performance-enhancing nano-structures within the device’ and ‘produces high-quality CIGS absorber layers much more rapidly than any other method.’
Using a scanning Fourier transform infrared microscope to examine novel PV materials
Although NanoSolar CEO Martin Röscheisen offered little insight into his company’s fabrication ramp status and equipment set, he says NanoSolar is ‘on track’ with its plans ‘to commence commercial production by the end of this year.’ Röscheisen cites ‘incoming materials quality control’ as ‘the issue that’s taking the most work,’ while noting that ‘we understand our own process control very well.’ The flexible substrate on which NanoSolar deposits its CIGS nano-particle inks using a proprietary print process is ‘an ultrasmooth metal foil … more than 20 times more conductive than stainless steel,’ according to Röscheisen. Published reports say the company uses an encapsulate nano-laminate made of 1000 layers, which self-assembles as a single printing step.
Miasole’s proprietary roll coater is a key part of the company’s thin-film CIGS PV process
The materials path
As CIGS-based cells and modules have moved closer to volume production, purveyors of organic photovoltaics have trod a separate path toward materials improvement and market acceptance. Plextronics’ prototype single-layer organic cell with a certified 5.4% efficiency may be impressive, but Konarka Technologies will soon be shipping sample products from its roll-to-roll pilot facility, operated with its manufacturing partner, Kurz. Using a dye-sensitized polymer material, the company ‘has brought proven coating and printing know-how from the chemical and flexible electronics industries to energy via a new class of nano-structured materials,’ explains Konarka’s CEO and President Rick Hess. He also sees a relatively smooth path to volume production: ‘The material ramp-up is straightforward, and our materials’ suppliers are major chemical corporations. The main change will be moving to larger material widths and faster web speeds. This will require process engineering, but not significant innovation. Due to the nature of the printing process, we expect that defects will be low and yields will be high,’ he says. Several encouraging developments have recently bolstered Konarka’s efforts to commercialize its OPV technology. In addition to landing $45 million in private capital financing and winning a NIST Advanced Technology Program award of $4.7 million with joint venture partner Air Products, it has signed a deal with Toppan Forms, in which the former ‘will provide its Power Plastic films and Toppan will integrate the films with their products and market the final product,’ according to Hess. First products should hit the market in 2008. The company has also produced a ‘working solar fibre,’ which could enable applications such as ‘wearable power generation for portable electronics.’
Shaking up the photovoltaic space
Other disruptive technologies could still shake up the photovoltaic manufacturing space, and NanoGram thinks its approach has a legitimate shot at achieving costs of well below $1/Wp. ‘We enter the value chain for silicon solar cells near the very beginning – at the basic raw material stage,’ notes VP and COO Ron Mosso. ‘NanoGram’s process involves the direct deposition of a thick silicon film from a silane precursor, much like thin-film solar cells, except that our films are 20 to 30 times thicker. The crystal grain sizes in our films are more than 2000 times larger than in micromorph crystalline thin film. We expect these two film attributes will combine to enable much higher efficiency than can be obtained in a thin-film cell,’ he adds. Mosso says the company’s process ‘collapses the value chain for making high efficiency silicon solar cells. Since the films are deposited in a large-area format, the cell junctions and contacts are fabricated at the module scale, eliminating many of the wafer-handling steps required to ‘string up’ an array of cells to make a module.’ NanoGram has hit its thick-film layer structure and critical materials targets on the lab scale, according to Mosso, and ‘is investing in improved prototype equipment to accelerate multi-crystalline film development. We are [also] developing our junction and contact fabrication techniques on commercial multi-crystalline wafers… and expect to have our first small module demonstration prototypes around the turn of the calendar year. Our priorities are to demonstrate our product in a module prototype and build a pilot plant to demonstrate the process and its economics [next year].’ With its future module costs modelled at 100 MWp to 300 MWp per year, Mosso likes his company’s prospects. ‘All the [solar] market projections show continued, massive growth over the next five years, with volumes tripling. New capacity will be needed, and existing capacity will be retired. The risk of NanoGram’s process to an existing manufacturer today is that the efficiency and cost targets still need to be proven. The reward opportunity is having the highest-value product on the market,’ he says.
Tom Cheyney is a freelance journalist based in the USA. This article first appeared in REW’s sister publication Small Times www.smalltimes.com
NCPV fosters collaboration
As part of the US National Renewable Energy Laboratory (NREL), the National Center for Photovoltaics (NCPV) in Golden, Colorado, features a world-class R&D organization, fostering a collaborative environment for government, academic, and industrial partners in the PV sector. John Benner, manager of the centre’s electronic materials and devices group, says, ‘In the electronic materials group, one of our major roles has been to create some of the proofs of concept that have helped launch many of these companies. The 19.5% efficiency CIGS device, the 16.5% efficient CdTe … these didn’t hurt any of the companies in terms of making the pitch for their business plans.’ Benner continues: ‘The key project for NREL right now is developing the process and development integration facility. This new research facility is intended to take some of the variables out of the research … by developing cluster tools that will allow us to make [PV] cells without breaking vacuum, without exposing the samples to air or other adverse environments, controlling the transfer lines and all of the other variables that go into making a complete device.’ He concludes: ‘When you look a little closer at the inorganic thin films, First Solar is thriving on the fact that their material [CdTe] goes down very fast. The total process time for that module is exceedingly short. CdTe is a binary compound and tends to come out with the right composition, while CIGS can take on a lot of different compositions and phases and has proven to be a little bit more difficult to put into large-scale production.’