The Annual Energy Outlook 2009, issued by the Energy Information Agency (EIA), forecasts an eleven-fold increase in domestic electricity generation from grid-based photovoltaics (PV) from 2.04 billion kilowatt-hours (kWh) this year to 22.51 billion kWh by 2030. The EIA expects a slight increase in the average electricity cost from 9 cents per kW to about 10.5 per kW in 2030.

Despite a forecast of strong growth ahead, solar PV power will become one of the cost-competitive sources of energy in the United States when the gap between the solar PV power cost of about 20 to 50 cents per kWh, and the average grid-based electricity cost is closed. 

According the U.S. Department of Energy (DOE), renewable energies, including wind and concentrating solar power (CSP), in utility-scale grid-connected applications are already producing electricity at a cost below 5 and 12 cents per kW, respectively. 

The Race is On - Silicon-based conventional PV modules are the core of the solar PV power market with over 90 percent of worldwide market share. Although the average cost of silicon-based modules is about 30 percent higher than that of thin film modules, the silicon solar PV systems show less degradation of performance over time and require a smaller footprint area than that of the thin film modules. If the installation cost, which runs between $4 and $5 per watt or about 50 percent of the total cost, is considered, the gap between the total cost of the silicon-based and thin film solar systems reduces to less than 15 percent. 

In an article "A Solar Grand Plan" published in Scientific American by Ken Zweibel, president of Golden, Colorado-based PrimeStar Solar, and co-authors, it is suggested that the cost of commercial solar module systems, including installation, needs to come down to about $1.20 per watt so that large solar concentrator power plants can provide solar electricity at competitive and affordable prices. Mr. Zweibel, et al believe that 35 percent of the U.S. total energy supply could come from solar power by 2050. PrimeStar Solar, which its majority equity is owned by Atlanta, Georgia-based GE Energy, a diversified global energy infrastructure division of General Electric [NYSE:GE], is a small manufacturer of high performance cadmium telluride (CdTe) thin film PV modules.

It is much easier said than done since the average retail price of solar modules, excluding the cost of installation at the point of use, as surveyed by a San Francisco, California-based research and consulting firm Solarbuzz, has held steady at around $4.80 per watt since May 2006. According to Phoenix, Arizona-based First Solar [NASDAQ:FSLR], the long-term contract prices for 2009 of their commercial modules, or "factory-gate" prices, is about a €1.54 ($2.08 equivalent) per watt. It is very conceivable that the total cost of a commercial module system still needs to come down more than 50 percent in order to meet the $1.20 per watt benchmark.

Startup companies such as  San Jose, California-based Nanosolar, which deploys high yield printing technology in its solar module manufacturing process, claims to be the first company capable of profitably selling solar panels for as little as $0.99 per watt. MIT spin-off 1366 Techologies, is touting a new cell architecture that uses innovative, low-cost fabrication methods to increase the efficiency of multi-crystalline silicon solar cells at a cost of just $1 per watt.

With a looming over-supply of solar modules and delayed orders due to a tight credit market, investment strategies in the solar sector require near-term risk assessment. Module manufacturers with sustainable long-term growth strategies, strong cash positions, low debt/equity ratio, low cost-per-watt cutting edge manufacturing processes and diversified solar PV technology portfolios should outperform the overall clean-energy market if crude oil prices and the pressure to reduce green house gas emissions are trending upward.

Silicon-Based Technology (SBT) - The key advantages of silicon-based technology are the abundance of starting materials, an in-depth understanding of silicon properties and the robust manufacturing process, developed earlier by the microelectronics industry. One of the drawbacks is the long energy payback time (EPBT), defined as the length of deployment required for a photovoltaic system to generate an amount of energy equal to the total energy that went into its production.  

Another drawback is the high cost of the starting silicon material, which accounts for about 40 percent of the final module cost. To address this issue, advanced processes for the manufacturing and handling of ultra thin solar silicon wafers to reduce the production cost of silicon solar cells are being pursued.

Commercial solar cells are manufactured from p-type boron doped silicon wafers sliced from either high purity single crystal ingots grown using the Czochralski (Cz) method or from low cost multicrystalline silicon bricks or ingots produced by the casting method. P-type silicon is used because the diffusion length of the minority carrier in p-type silicon is longer than that in n-type silicon. A Cz-single crystal silicon ingot is "pulled" from 1410 °C molten silicon at a constant rate, about 1.0-1.5 mm/min., whereas cast silicon is produced by directly casting molten silicon into a mold and allowed to solidify into a multicrystalline ingot.

Rapid cooling during the solidification of cast silicon introduces crystallographic defects such as grain boundaries, dislocations and metal precipitates, which degrade solar cell efficiency. Although a Cz-silicon ingot is free from slips and dislocations, grown-in defects such as oxide microprecipitates and vacancy clusters, which may have an impact on solar cell efficiency, are still present.

L. J. Geerligs and D. Macdonald suggested that B-O recombination active complexes formed between intentionally doped boron atoms and oxygen atoms unintentionally incorporated into silicon during the crystal growth process, as well as transition metals such as Fe, cause degradation of the minority charge carrier lifetime and hence the solar cell efficiency in p-type silicon. The lifetime degradation associated with these defect centers is less pronounced in n-type silicon. Thus, n-type silicon is also being considered as a promising candidate for future generations of high-efficiency commercial solar cells.

Despite the high cost of Cz-crystal growth equipment and about three times more electrical power consumption than that of the casting method, the average retail prices of single crystal silicon solar modules are very competitive to those manufactured from multicrystalline silicon solar cells. The conversion efficiencies of commercial single crystal and multicrystalline silicon solar cells, e.g. Q-Cells Q6LM and Q6LTT, are about 16 and 14 percent, respectively. In a laboratory environment such as at UNSW, single crystal silicon-based PV cells can achieve an efficiency of more than 24 percent. There is a strong correlation between cell efficiency and back cell temperatures in which an increase in back cell ambient temperature leads to a decrease in cell efficiency.

The first step of silicon-based PV cell manufacturing is to form a front p-n junction by phosphorus diffusion, e.g. using phosphorus oxychloride (POCl3) as the diffusion source. Various time-temperature schedules can be used to tailor the concentration profile of the phosphorus. After the homogeneous emitter junction is formed, an anti-reflection coating (ARC) with a thin layer of dielectric nitride film (SiN) is applied, e.g. using a plasma enhanced chemical vapor deposition (PECVD) from silane (SiH4) and ammonia (NH3) based gases at around 400 °C. 

Subsequently, screen-printed contacts are applied to the front and back of the cell. Screen-printing using silver and aluminum paste is relatively simple, highly efficient and low cost. A non-contact Aerosol Jet printing process, developed for high yield printing on thin silicon PV wafers, can produce narrow and high integrity collector lines, which in turn improve the conversion efficiency by 2 to 4 percent, compared to conventional screen-printed silicon solar cells.

Thin Film Technology (TFT) - Commercial thin film solar cells are manufactured by depositing thin layers of semiconductor materials including amorphous/ microcrystalline silicon,CdTe, copper indium gallium (CIS) and copper indium gallium diselenide (CIGS) onto flexible substrate such as stainless steel or polyamide using the continuous roll-to-roll manufacturing method or on a large size glass substrate, up to about 1,000 x 1,400 mm. Thin film cell manufacturing uses less semiconductor materials and hence is highly economical, compared to the conventional solar cell batch process. Therefore, the thin film technology provides a low cost solution for commercial scale high throughput solar PV cell manufacturing. According to First Solar, CdTe thin film cells can be manufactured at an average cost of about $1.01 per watt.   

The conversion efficiencies of thin film solar cells are much lower than that of silicon-based PV cells and degrade over time, depending upon the types of thin film materials. A research study conducted by the USDA found that the performance degradation of the CdTe thin film modules occurred at a rate that was much more rapid than that of amorphous-silicon modules over timescales of years.

Amorphous/Micromorph Silicon Thin Film Technology - Amorphous silicon is a non-crystalline form of silicon having the same short range order as the silicon crystal but lacks long range order. The lattice of amorphous silicon contains defects such as dangling bonds, which negatively impact the diffusion length of the minority carriers and hence the cell efficiencies. The dangling bonds can be neutralized by hydrogen passivation in which hydrogen atoms are intentionally introduced into amorphous silicon during the thin film deposition.

Page 1 of  2  next