In December 2007, more than 1,500 participants from industry, academia, and military and government institutions converged on Fukuoka, Japan, to attend the Seventeenth Annual Photovoltaic Science and Engineering Conference (PVSEC). From the opening of the conference, there was palpable excitement about the environmental and political benefits of photovoltaic technology. And judging from the large number of companies entering the PV space, there are financial rewards as well. Indeed, the industry's promise of boundless opportunity evokes the semiconductor industry of the early 1980s.
PV technology offers a secure domestic power source that's free from the leash of the oil industry, low in maintenance, and high in modularity and scalability. It is the only technology currently capable of meeting the world's long-term energy needs without emitting greenhouse gases. In fact, a solar array 150 x 150 km could, in principle, meet all of North America's energy needs. But PV technology is still more expensive than grid power, lacks a suitable load-balancing solution and requires considerable space for electricity generation.
There are four general categories of solar cells, of which the first three are in full-scale production.
• The materials in the III-V group, also known as multijunction concentrators, are direct-bandgap compounds and have the highest conversion efficiency (and cost). They are used primarily for satellite and military applications and generally require light-concentrating optics and sophisticated tracking systems to deliver 40+ percent efficiencies.
• Bulk-silicon solar cells, using single-crystalline silicon (cSi) or multicrystalline silicon (mcSi), achieve conversion efficiencies above 22 percent in some cases.
• Thin-film solar cells have efficiencies in the range of 10 percent for amorphous Si-based cells to almost 18 percent for the II-VI materials, which include cadmium telluride (CdTe) and copper indium gallium diselenide (CIGS). Thin-film methods are believed to pose the greatest opportunity for cost reductions in the long term because of their potential for introducing new or modified materials into existing processes.
• The final category is a catchall for emerging technologies, such as dye-sensitized thin-film cells, which already demonstrate high conversion efficiencies, and organic-based cells, which to date have relatively low conversion efficiencies.
The simplest definition for photovoltaic conversion efficiency is a single value that shows what fraction of the photons striking the PV array is converted to usable current for the attached load. In practice, efficiency is more complicated; just absorbing the light and generating free carriers isn't sufficient. To generate energy for use, the electron and hole carriers must reach the cell electrodes. If the electron-hole pairs formed by the incident photon recombine too quickly on their way to the electrode, they cannot contribute to the photocurrent.
Various issues can reduce carrier mobility (or increase recombination). Crystalline-silicon bulk solar cells have the fewest issues and have very high efficiencies for single-junction cells. This is because moving carriers from the junction at which they are generated to the cell electrodes from which they are supplied to a load is relatively simple: Electron-hole pairs split and move through the n-type or p-type material as appropriate.
Multicrystalline-silicon PV cells operate in a similar fashion; but because of the presence of grain boundaries, increased electron-hole pair recombination occurs, reducing mobility and conversion efficiency to a level approaching CIGS-based thin-film cells (approximately 15 to 18 percent).
Although thin-film cells use less Si (or none at all) and may provide the best path in the future to grid parity through cost reductions, they exhibit lower conversion efficiencies than do bulk-Si cells. Although CIGS cells approach mcSi in conversion efficiency, the CIGS-based PV-cell processes have shown higher degrees of variability than have bulk-Si-based processes. Although the reason is not yet clear, it is thought that nanoscale-level segregation control of the p-type alpha and n-type, indium-rich beta phases of CuIn3Se5 may be critical to forming a consistent bandgap of the cells.
Bulk-silicon PV cells constitute about 94 percent of the current global production capacity, of which Japan and Germany possess the lion's share. China is rapidly growing into a global PV production powerhouse and will likely surpass Japan (the current production leader) within the next few years. Although the main focus of most companies is on increasing cell- and module-production capacity, many new entrants hope to capitalize on the disparity between supply and demand. Not all of these new entrants will be successful in the execution of their plans, however, in large part because of the high risk associated with both inexperience and the introduction of new technology.
With the introduction of feed-in tariffs by many countries (a subsidy by which the government commits to purchase PV-generated energy at a predetermined price), global production capacity has been growing by more than 50 percent every year for the past five years and is expected to continue growing at a high rate.
Revenues of $10 billion in 2006 increased to more than $13 billion in 2007. Almost $16 billion was invested in capacity expansion and R&D over that same period. For many companies, the importance of the supply chain has been a hard-learned lesson; the consistency and reliability of cells and modules depend on the consistency and composition of the starting material. From silicon feedstock to wafers and ingots, the worldwide shortage of silicon has governed the silicon-hungry PV market and has prompted efforts to find solutions, sometimes with unexpected results.
For China's Suntech Power, expanding the supply chain to beat demand initially led to Si-procurement and -quality issues, prompting the company to address impurity variability with proprietary substrate-treatment processes. Those processes have improved efficiency and increased solar-cell stability.
Germany and Japan are still ramping production capacity at a high rate, though growth of the PV industry in China will soon enable it to lead the world as a cell and module supplier. Although the United States is not a significant global supplier of PV cells and modules at this time, it has declared its intent to reach grid parity by approximately 2012, significantly earlier than the rest of the world. The United States has shown strong interest in exploring disruptive technology to achieve grid parity, and, with a few notable exceptions, is not heavily focused on bulk-Si technology (Sunpower manufactures cSi cells with the highest conversion efficiency in the world, and will generate approximately $1.2 billion in revenue in 2008).
Since the introduction of the feed-in tariff system, PV installations have proliferated in Germany. Domestic cell and module production has also been doubling every year. Q-Cell, a major competitor based in the region, provides bulk-silicon modules to the domestic and global markets. Recognizing the importance of nonbulk-Si technologies, however, it has invested in a variety of thin- film PV technologies via merger and acquisition activities.
Japan has stated a goal of becoming energy-independent and self-sustaining, as well as of becoming the world leader in PV power generation. It produces 50 percent of the world's solar cells, a figure that ranks it first in global production. Although the bulk of this region's production is in bulk-Si-based solar cells (major suppliers are Sharp and Sanyo), Japan is also producing thin-film products (capacity increases are planned for the amorphous thin-film cells manufactured by Kaneka and the CIGS PV cells manufactured by Showa Shell).
Sleeping giant: IP
With most of the focus on increasing capacity to generate revenue, there is little concern at this time about intellectual-property issues, though there is activity in generating IP from the research under way. This activity includes improvements in starting materials to increase mobility and in interface engineering to enhance light scattering and reflection (which increases absorption). It also includes improving adhesion of the various layers of composite thin films to increase mobility and conversion efficiency.
As feed-in tariffs and other forms of subsidies decrease each year, innovative techniques will emerge at a faster rate to provide differentiation within and across technologies. When that occurs, such techniques will play a significant role in a company's success.
This activity and focus can be measured, to some extent, by a careful study of the patent activity, the broadest view being the "patent velocity," or rate of IP being filed. In the PV area, the high number of thin-film patents filed over the past decade makes it clear that thin-film PV is a key focus.
As in the semiconductor industry, protecting and enforcing these competitive advantages will become a necessity as latecomers with deep pockets challenge the early players.
We can be sure that almost every large semiconductor manufacturer has a PV program of some sort and is assessing strategies to leverage its manufacturing experience and depreciating assets. Expect consolidation of the industry as companies with solid road maps to grid parity--and with the IP assets to back them--flex their technological muscle.
In the meantime, the industry will enjoy the rewards during the honeymoon period of the next few years.
John Boyd (firstname.lastname@example.org) is a technology analyst at Semiconductor Insights (Kanata, Ontario), a Techinsights company. He holds more than 60 U.S. patents and has more than 40 pending.