FREE subscription
Subscribe for free to receive each issue of Semiconductor Today magazine and weekly news brief.


16 June 2008


IBM and Tokyo Ohka Kogyo to co-develop CIGS PVs

IBM Corp of Yorktown Heights, NY, USA and process equipment maker Tokyo Ohka Kogyo Co Ltd (TOK) have agreed to collaborate on jointly developing processes, materials and equipment for the production of more affordable and easier-to-install thin-film CIGS (copper indium gallium diselenide) solar cell modules.

Currently, about 90% of photovoltaic (PV) cells use silicon, but their greater thickness and weight limit how they can be deployed. In contrast, CIGS solar cells can be 100 times thinner than silicon-wafer solar cells, can be deposited on either inexpensive glass substrates or flexible substrates, and are suitable for the use on the tops and sides of buildings, tinted windows, and other surfaces.

The most common CIGS manufacturing process is co-evaporation, in which active chemicals are immersed in a solution before removal in a vacuum. Using a co-evaporation process, in April the US Energy Department’s National Renewable Energy Laboratory increased the efficiency record for CIGS PV cells to 19.9% (nearing the record for multicrystalline silicon cells of 20.3%). However, no firms have yet approached this in manufacturing, where cells must be quickly mass produced in a cost-effective manner. Existing thin-film CIGS products on the market (e.g. from Nanosolar Inc of San Jose, CA, Miasolé of Santa Clara, CA, HelioVolt Corp of Austin, TX and Global Solar Energy of Tucson, AZ) have efficiencies of 9-10%. In May, HelioVolt said that it had achieved 12.2% in a ‘champion cell’ using its reactive transfer printing process (which is faster than co-evaporation), while Global Solar expects to achieve 13-14% this year.

However, based on work done about 10 years ago on flexible electronics, IBM Research has now developed a new, non-vacuum, solution-based manufacturing process for CIGS solar cells that involves dissolving chemicals in a liquid and then drying them, doesn’t require as much energy to run, and is faster than co-evaporation, it is claimed. Solution processing allows printing onto a rolled backing of a flexible module or a glass plate, eliminating many of the high-energy and equipment-intensive processes typical of conventional PV manufacturing. The firm is targeting efficiencies of about 15% and higher.

Combining IBM’s technology with the coating technique and the high-purity chemicals of TOK (building on its experience of both semiconductor and LCD panel manufacturing) has the potential to bring large-scale production of thin-film solar cells to market, the firms say.

“Our goal is to develop more efficient photovoltaic structures that would reduce the cost, minimize the complexity, and improve the flexibility of producing solar electric power,” says Dr Tze-Chiang Chen, VP of Science and Technology at IBM Research. “IBM’s technology combined with TOK’s expertise in equipment design and manufacture, have the potential to broaden the use of alternative energy sources,” he adds.

“We believe that this joint development is a great opportunity to expand the applications of our technologies into the photovoltaic industry,” says TOK’s president & CEO said Yoichi Nakamura.

IBM has already built a prototype device, according to a Reuters report. Once made at large volumes on a glass substrate, the cells are expected to deliver electricity at less than the long-targeted cost of $1 per watt at peak times. “I think that if we can get to a module cost of less than $1 per watt, and be able to keep a handle on the system costs, then one should be able to get to grid parity [the level at which solar power is competitive with traditional forms of electricity generation].” Photovoltaics still need roughly a twofold improvement in efficiency, he adds.

However, the firms do not plan to manufacture solar modules themselves, but aim to develop technology that can be licensed to manufacturers within the next two to three years. “We’ve already been in discussions with photovoltaic manufacturers,” said Supratik Guha, who leads IBM Research’s PV activities. “There are problems to be resolved,” he said, “but this is the time we’re starting to talk to them.”

IBM intends to provide more technical detail of its solutions-based process in an advanced-material paper in about a month’s time, says Guha.

*IBM Research says that it is exploring four main areas of PV research: using current technologies to develop cheaper and more efficient silicon solar cells; developing new solution-processed thin-film PV devices; concentrator PVs (CPVs); and future-generation PV architectures based on nanostructures such as quantum dots and nanowires.

See related items:

Thin-film solar market to reach 9GW in 2012

HelioVolt hits 12.2% thin-film PV efficiency with 6 minute printing process

Solar VC investment reaches record $280m in Q1

Thin-film to grow from 10% to 19% of PV market by 2012

Search: IBM CIGS Solar cell modules Nanosolar HelioVolt