Sharpening silicon

No producer wants to be caught short when new sources of demand kick in. This is one of the reasons why the PV industry is ramping up capacity even with alternative sources looking unlikely any time soon to pick up the slack from Germany, still the largest single market for PV in the world by a considerable measure.
But while the thin film PV industry has the potential, in time, to achieve higher efficiencies and reduce the production costs of solar cells dramatically, the upper limits of crystalline silicon cell, in terms of power conversion efficiencies, is being pushed. However, when the average top tier PV producer is operating a one-gigawatt annual capacity output, technologies that achieve incremental improvements to silicon cell efficiencies can be attractive and worthy investments.
SunPower remains the industry leader when it comes to producing high efficiency modules by optimizing the power conversion efficiencies of its solar cells. In June, the company announced its best performing panels yet. The E20 panels exploit SunPower’s 22.4-percent efficient Maxeon cell technology, where the contacts are on the back of the cell to allow the cell front side to capture more sunlight and conduct more current than conventional solar cells.
Funds from the U.S. Department of Energy (DOE), under its Solar America Initiative, have been used by SunPower to develop the E20 panel prototype.
John West at VLSI Research, a firm specializing in analyzing equipment demand for semiconductor and PV industries, observes: “Producers have to get the economic model right. Getting manufacturing costs down only goes so far.” The 20 or so largest PV producers together account for just over half of global output. For many of them, selective emitter and other technologies that boost power conversion efficiencies make attractive investments if they can be retrofitted into existing lines or installed without a major overhaul of existing production operations.
Selective emitter technologies describe ways of optimizing the emitter profile on the cell’s sun-facing front side, exposing it to more light.
This year Suntech Power has begun to bed in manufacturing of solar modules based on its high-efficiency Pluto-brand solar cells. Pluto is based on a passivated emitter and rear locally diffused (PERL) cell technology.
The company is switching over 200 megawatts of capacity in 2011 to produce the Pluto cells with ongoing plans to increase Pluto cell manufacturing capacity.
Says Björn Emde, a spokesperson for Suntech in Europe: “This has to be done gradually as, typically, equipment needs to be taken off line and some tweaking done, so usually roll-out is done one line at a time. Last year was very busy so there wasn’t the time to switch over, which would have incurred some downtime.” The PERL technology used to produce Suntech’s Pluto cells originated at the University of New South Wales, which had developed a PERL cell with a one-sun efficiency of 25 percent in its laboratory. Over the course of its collaboration with the university Suntech has modified PERL for commercial production, removing, replacing and adapting several of the steps, while still able to achieve higher cell efficiencies enabled by the process.
The resulting Pluto cell technology, first announced in 2009, consists of a conventional screen-printed aluminium rear contact and back surface field (BSF). Manufacturing includes an eight-step production sequence, most of which is carried out using the standard equipment from a conventional screen-printed solar cell production line.
Innovalight, a start-up based in California, has secured several licensing partnerships with Asian PV producers to integrate its silicon ink selective emitter technology into their operations. In future nanomaterials could become a rich source of PV semiconductor materials because they can be processed by printing or other room-temperature tools. In the meantime, a handful of companies including Innovalight are paving the way for these promising materials in PV applications.
Innovalight’s process drops in after the wafer texturing step and prior to the diffusion furnace. Industry standard screen printing equipment – widely used for metallization – lays down a thin layer of the silicon ink on top of a silicon solar cell, directly beneath the areas where contact fingers are attached. The ink acts as a doping agent in these areas, reducing the bulk silicon’s electrical resistance and allowing the contact wires to extract more of the free electrons knocked loose by the photons.
Average conversion efficiencies of crystalline silicon solar cells are around 18.5 to 18.6 percent. Innovalight’s technology can increase efficiencies from 18.6 to 18.9 percent. Longer term, the company is pushing for 20 percent. “Introduce the technology across the entire manufacturing operation of a solar cell producer with an annual capacity of 1GW and the producer achieves a 1.2GW output, with no added capital outlay,” explains CEO Conrad Burke.
“Makers of solar cell panels see that our technology can help protect against declining gross margins,” Burke observes. Navigant Consulting’s Paula Mints, a veteran analyst of the PV industry, adds: “Questions manufacturers ask are does it raise efficiency and does this, in the end, lower the system cost?”

Maximizing silicon’s potential

Several approaches and technologies are available to manufacturers to help them achieve these objectives, says Mints. In order to gain more control over silicon supply costs and optimize the quality of silicon cell material producers, like Suntech, are consolidating their own supply chains, investing in producers of wafers for instance. From addressing silicon wafer production, to improving yields, to thinning wafers, to raising efficiencies, there are many approaches that aim to reduce manufacturing costs and squeeze the most out of silicon.
“Improving consistency and homogeneity of solar cell conversion efficiencies, the yield management, which didn’t matter so much before has become big business in the PV industry,” says West. He notes firms such as KLA-Tencor, which has adapted its analysis tools and software, initially for semiconductor computer chips, for the PV industry in the form of surface profilers and inspection modules for stages of wafer and cell manufacturing lines.
However technologies that enhance the power conversion efficiencies of solar cells are gaining credence. Roger Little, Spire CEO, says: “I think technologies for enhancing PV power conversion efficiencies are extremely relevant right now. The higher the efficiency the lower the system cost. Increasingly balance-of-system-cost is used and if you can get more power out of the same sized module area or the same power out of smaller module area than this can make PV more competitive.” Companies with the tools to push the silicon envelop are prized as the PV industry continues to consolidate. The pro posed acquisition of Varian Semiconductor by Applied Materials would enable Applied to expand its share of the PV equipment market with tools designed to produce high-efficiency photovoltaics. Varian is the world’s leading supplier of ion implantation equipment for applications including the production of selective emitter solar cells.
While Innovalight’s technology may not achieve the highest efficiency levels compared with some other selective emitter technologies, its screen-printed ink integrates into existing PV production processes with minimum disruption, without breaking the bank. But, as Mints points out: “Whatever the technology or approach that purports to help manufacturers increase their cell efficiencies or optimize their output, they have to be able to prove the technology on their own production lines, which means internal pilots, to establish repeatability of results.” Suntech’s collaboration with the University of New South Wales to adapt and develop the PERL technology to get to the stage where it is being rolled out across the company’s industrial lines has taken more than six years.

Results

At the recent IEEE Photovoltaic Specialists Conference, held June 19 to 24 in Seattle, Washington, Innovalight discussed results of the performance of silicon ink selective emitter cells incorporated into photovoltaic modules fabricated at Hanwha SolarOne, one of Innovalight’s Chinese clients. The performance gain of the silicon ink selective emitter cells were compared with homogeneous emitter cells after module fabrication. The study, claimed Innovalight, is the first systematic and large-scale comparison of full size selective emitter and homogenous emitter solar panels produced with industry standard materials and facilities.
Ten modules were made on the module production line at Hanwha SolarOne. Each consisted of 54 silicon ink selective emitter cells produced at Innovalight. They achieved efficiencies of 16.9 percent, related to the area under the glass, excluding the frame. These modules showed efficiency improvements of over 0.94 percent absolute compared to homogenous emitter modules fabricated under identical conditions.
At its operations in California, Innovalight has a small ten-megawatt PV production line where it can analyze the interactivity of its ink particles with the bulk silicon solar cell to improve and enhance its technology. The company can receive wafers, apply its technology and provide results for clients. Recently the firm secured a grant from the U.S. Department of Energy to invest in scaling up manufacturing to meet growing demand for its technology. The company is also expanding a staff presence in China, and establishing partnerships in Japan, Taiwan and South Korea.
In addition to Hanwha SolarOne and JA Solar, Innovalight has signed up several other Asian customers including Yingli Green Energy, JinkoSolar and recently Taiwan-based Motech Industries, now one of the top ten manufacturers worldwide in terms of production capacity and output.
Both Innovalight’s silicon ink and Suntech’s Pluto cell technology have the potential to push up efficiencies by a few percent more in time but how they’ll do so remains a closely guarded secret. All Burke will say is: “On the R&D side, Innovalight is working with its partners to improve the conversion efficiencies its ink can help achieve to above 20 percent.” Pluto is a three-stage process that could take Suntech well into next year to fully implement. Explains Emde: “We are currently implementing phase one. Phase two, which we could start implementing at the end of this year or early 2012, will push conversion efficiencies to over 20 percent and phase three will push efficiencies to over 22 percent.” Burke is hoping that the race by manufacturers to supply high efficiency cells and modules will ensure more demand for Innovalight’s technology. “We’d like to think our technology is implementable across the 40-50 companies producing crystalline silicon solar cells,” says Burke.

What’s around the corner?

And the race is on. While the biggest PV producers of today push to optimize manufacturing and output, their fortunes could change sharply if a new kid on the block with a technology that dramatically cuts PV production costs and ramps up cell performance proves scalable.
The Californian start-up Solar3D isn’t touting a thin film contender or a new silicon wafer production process but is applying techniques from the fibre-optics industry to produce cells that hold onto light and electrons more effectively.
As its name suggests, Solar3D “uses a three-dimensional design to trap sunlight inside micro-photovoltaic structures where photons bounce around for longer until they are converted into electrons.” CEO Jim Nelson explains the real draw of his firm’s offering: “We’ll be using existing semiconductor technology to manufacture our product. Therefore, it is the ideal situation for a semiconductor manufacturer with excess capacity.” Furthermore, Nelson says it will be less expensive to manufacture the 3D solar cell with compared current cells. So, with a better product that can sell for more, that costs less to make, with very little capital investment, Solar3D’s manufacturing partner could do very well in terms of its return-on-investment on the project.
Nelson is bullish about commercial prospects and timeframes, with a prototype in the works by the end of the year. “We believe that we will be able to be commercial next year. We recognize that the whole industry is working toward lower cost per watt. We will be able to get us there faster. We can use any material and believe that we will have efficiency in excess of any current commercial cell.” In the push to achieve higher efficiency modules, processes that enable silicon cells to convert light to electricity more efficiently and effectively are becoming a serious business.