Photovoltaics: Energy for the New Millennium

Thomas Surek



Photovoltaics (PV) is a semiconductor-based technology that directly converts sunlight to electricity. The stimulus for terrestrial PV started more than 25 years ago in response to the oil crises of the 1970s resulted in major government programs in the United States, Japan, Europe, and elsewhere. Ongoing concerns with the global environment, as well as the worldwide efforts to seek alternate, local sources of energy, continue to drive the investment in PV research and deployment. Today, the manufacture, sale, and use of PV has become a billion-dollar industry worldwide, with more than 200 megawatts (MW) of PV modules shipped in 1999.

Over the past 25 years, research and development has led to the discovery of new PV materials, devices, and fabrication approaches; continuing improvements in the efficiency and reliability of solar cells and modules; and lower PV module and system costs. This article reviews the rapid progress that has occurred in PV technology from the laboratory to the marketplace, including reviews of the leading technology options, status and issues, and key industry players. Major ongoing efforts involve a better understanding of the crystal and film growth processes and the resulting material properties. New processes for fabricating PV materials and devices, and innovative PV approaches with low-cost potential are elements of an ongoing research program aimed at future improvements in PV cost and performance.

Progress, Status, and Research Directions of Photovoltaic Technologies

Photovoltaic technologies can be divided into two main areas: flat-plates and concentrators. In the flat-plate technologies, semiconductor material is used to cover as much area as possible on a flat surface, while balancing tradeoffs between material cost and conversion efficiency of light into electrical power.

Flat-plate technologies include thick cells of crystalline silicon (from both ingot and sheet-growth techniques) and thin-films. Thin-films are typically less than 100m m of material (e.g. amorphous silicon, copper indium diselenide, cadmium telluride, and polycrystalline silicon) deposited using vapor deposition, electrodeposition, or wet chemical process. Present thin-film approaches generally do not allow conversion efficiencies as high as those demonstrated by crystalline silicon modules. In spite of this, development of thin-film approaches is an active area of research, since thin-film cells require 1/10th to 1/100th the amount of expensive semiconductor material compared to that required by crystalline silicon cells with equal collection areas.

Table 1 presents cell and module conversion efficiencies (percentage of sunlight converted to electricity under standard conditions) for both ingot- and non-ingot-based crystalline silicon technologies. Although there are specific areas for improvement associated with each of the crystalline silicon sub-technologies, general research areas that apply to crystalline silicon include: a) manufacturing yield and throughput, b) impurity/defect gettering and passivation, c) low-cost, high-efficiency processes, d) environmentally benign processing and waste stream reduction, e) manufacturing automation and module packaging for 30-year life, f) thinner wafers or sheets and associated handling, g) wire-saw slurry recycling (ingots only), and h) new processes to produce "solar-grade" silicon.

Table 1. Crystalline Silicon PV Conversion Efficiencies (%)










Cast Polysilicon



EFG Ribbon



Dendritic Web



String Ribbon



Thick Silicon Substrate



Table 2 presents conversion efficiencies for thin-film cells and modules. Similar to crystalline silicon, increased manufacturing throughput and yield and improved conversion efficiency are primary concerns for all thin films. Special attention is directed towards reducing the gap between laboratory cell efficiencies and production module efficiencies. Specific amorphous silicon research is directed to the following areas: a) novel growth techniques that allow higher growth rates and better materials and b) improved fundamental understanding with the goal of improved material stability and long-term field performance. Current cadmium telluride R&D includes work addressing the issues of: a) improved film deposition, b) better contacting techniques for extracting electrical power from the cells, and c) low-cost module packaging for long-term reliability. Current R&D areas for copper indium diselenide are: a) scalability of production processes, b) new deposition techniques and materials that lend themselves to lower temperature and non-vacuum approaches, and c) improved understanding of the device physics at the active semiconductor junction. R&D for thin-film silicon includes developing techniques for high-rate deposition of large-grain-size films on foreign substrates.

Table 2. Thin-Film PV Conversion Efficiencies (%)




Amorphous Silicon



Cadmium Telluride (CdTe)



CuInSe2 (CIS)



Polycrystalline Silicon



Concentrator technologies are generally of two types: low concentration (typically 10 to 40 suns) which uses line or one-dimensional focus, and high concentration (typically 100 to 1000 suns) which uses point or two-dimensional focus. In addition to balancing material cost and conversion efficiency, portions of the more expensive semiconductor material in flat-plate systems are replaced with a system of lenses or reflectors that can be made from less expensive materials. This replacement may, however, be at the expense of overall system efficiency, and thus one should consider each system as a whole in evaluating its benefits.

Table 3 presents solar conversion efficiencies for various materials that lend themselves well to the somewhat higher module operating temperatures often found in concentrator systems. Note that the efficiencies are reported at particular concentration ratios since the efficiencies are a function of measurement conditions, including light intensity. Module efficiencies are in the range of 15% to 17 % for silicon-based systems, with prototypes of more than 20%. Modules using GaAs cells have efficiencies of more than 24%. A prototype module with a three-junction GaInP2/GaAs/Ge cell measured at 28% efficiency (10 suns).


Table 3. Cell Efficiencies for Concentrator PV Systems (%)


Concentration (suns)



Up to 400


Gallium Arsenide (GaAs)

Up to 1000









40 to 560


General issues for concentrator systems include the structural characteristics of the system that lend themselves to larger applications and make the highly visible and currently more-prevalent small-application market less useful to concentrators in terms of establishing market position. An additional concern is that concentrator systems use essentially only direct radiation, and therefore their areas of best application require high-intensity sunlight, such as the southwest United States. Areas of R&D that are important for concentrators include, as in flat-plate PV, manufacturing yield and throughput and higher conversion efficiency to reduce ultimate energy cost. Higher efficiencies are expected from multijunction structures, including 3- and 4-junction devices. Novel concentrating techniques may also ultimately be incorporated into successful concentrator systems.

Markets and Applications

Photovoltaic systems may be used for almost any situation requiring electrical power, either tied to or independent of the utility grid. Systems may include energy storage and power conditioning to convert from DC to AC power. The size of the application may vary from milliwatts (as in calculators) to kilowatts (as in grid-tied, roof-mounted systems) to megawatts and larger (as envisioned with central-station generating systems). Consequently, the list of applications is long and includes developing country applications such as lighting, water pumping, health clinics, and village power; U.S. rural applications such as electric fences, water pumping for livestock, and irrigation; remote applications such as telecommunications and signaling; and grid-connected power generation on commercial and residential buildings.


While major market opportunities continue to exist in developing countries, where sizable populations are without any electricity, today's manufacturing expansions are fueled by market initiatives for grid-connected PV in residential and commercial buildings. The combinations of increased production capacities, with the attendant cost reductions as a result of economies of scale, are expected to lead to sustainable markets. A key to achieving the ultimate potential of PV is to continue to increase the sunlight-to-electricity conversion efficiencies and translate the laboratory successes to cost-competitive products. Building a robust technology base is essential to overcoming the perceived high-risk transition into grid-connected PV. Such a base will make PV a globally significant contribution to our energy supply and environment.

Thomas Surek

National Renewable Energy Laboratory

1617 Cole Boulevard, Golden, Colorado 80401, U.S.A.


(1) "Photovoltaics: Energy for the New Millenium, The National Photovoltaics Program Plan, 2000-2004," DOE/GO-10099-940 (January 2000).

(2) "Terrestrial Photovoltaic Technologies -- Recent Progress in manufacturing R&D", C. E. Witt, T. Surek, et al., Proceedings of the 34th Heat Transfer Conference (August 2000).