PV solar: opening up the options

Printer-friendly versionSend to friend

This article examines developments in photovoltaic cell and module technology, with reference to the US government's support programme for the industry, and highlights the likely future direction of an industry that is well-positioned for continued advance and growth.

Solar photovoltaics (PV) have many attractions as an energy option. Based on the photoelectric effect, which involves light photons displacing orbital electrons in some types of semiconductor materials like silicon, it’s a silent-operating and easy-to-fit technology, with no moving parts or plumbing requirements. But since they use specially fabricated materials, PV cells are expensive. However, prices are falling as volume production increases and technology improves. Indeed PV has one of the best unit cost/installed capacity "learning curve" slopes in the renewable energy field: see Figure 1 below.

Figure 1: learning curve for PV solar - module price (USD/watt) vs sales volume (MW)

Progress down this curve seems likely to continue given that there are many new cell technologies on the way that will increase energy conversion efficiency and reduce unit cost - since with higher efficiencies you need less cell material. For example, the US National Renewable Energy Lab (NREL) says it has achieved a new efficiency record for copper indium gallium selenide (CIGS) solar cells: 19.9%, beating the previous 19.5%, and nearing the 20.3% record for conventional multi-crystalline silicon cells.

CIGS technology leading the way

It is, of course, some way from the lab tests to commercial scale production of new cells, but it can be worth the wait since it’s not just efficiency that is important in reducing costs. New cell material can lead to new, cheaper production methods.

One of the leaders is the US company Nanosolar which has claimed very high PV cell production rates with its ink-dye-based cell printing system. The company says that it can print flexible thin-film cell material at 100 feet per minute, and is looking to manufacture 1000MW of cell each year (see the video clip here).

There may, however, be trade-offs between cell efficiency and ease of manufacture. Some questions have been raised about some of the cost claims made by Nanosolar: according to some reports it has claimed a very low cell cost of USD0.36/watt and a wholesale solar panel cost of USD0.99/watt (see, for example, the final section of the article here).

Despite having lower efficiency, so far, than conventional silicon cells, some novel, cheap-to-make thin-film cells may find niche applications. For example, the Cardiff-based G24i solar cell company claims that "Dye-sensitised thin-film solar cells are unique in that they are extremely lightweight, durable, and produce electricity in low-light and indoor conditions. As a result, our solar cells are perfect for powering mobile electronic devices such as mobile telephones, cameras, and portable LED lighting systems. On a larger scale, flexible thin film cells integrate into clothing, tents, electronic advertising displays." The company's thin-film cell manufacturing process completes all manufacturing steps in an automated, continuous process and, it is claimed, can transform a lightweight roll of metal foil into a half-mile 100lb roll of G24i's dye sensitised thin film in less than three hours.

Companies like Nanosolar and G24i may be ahead of the pack, but they are not alone. Given the expanding market for PV, others are getting into the game. According to a 2008 review by the US National Renewables Energy Labs, 40 companies are actively involved in the technology development of thin-film CIGS products and manufacturing processes . For example, the US based global electronics company Konarka Technologies has developed its own inkjet printing system, and Japan's Shell Showa is to build a 1GW p.a. plant for producing CIGS cells.

The development of new cell materials has been fortuitous in that there have recently been shortages of the high-grade silicon used for conventional cells. Initially the PV industry made use of silicon produced by the semiconductor industry, but as PV volumes have grown, this has not been enough and the PV industry is now setting up dedicated silicon production plants, as well as CIGS production facilities.

Non-CIGS and third-generation cells

Although CIGS seems to be the dominant new option, it is not the only thin-film option. There is still a large market for amorphous silicon cells, the first type of thin-film cell to be developed. Although their efficiencies are usually relatively low, amorphous-silicon cell efficiencies of 12-13% have been produced using tandem and multi-junction devices. Cadmium telluride cell technology is also widely seen as promising, with, for example, First Solar claiming efficiencies of 10.5%, manufacturing cost of USD1.14 per watt and installed prices of USD3-4/watt. However, CIGS seem to have the edge on efficiency and cost.

According to the NREL review, worldwide production capacity for all thin-film PV is estimated at more than 5000 MW in 2010, with First Solar’s global target at 1000MW by 2009 and Sharp’s target of 1000MW by 2010.

There are also many new ideas being developed for so-called third-generation solar cells, including new types of organic/plastic cells, cells exploiting quantum tunnelling effects, and thermo-electric cells – all aiming for higher efficiencies. One of the most advanced so far, developed by the US National Renewable Energy Laboratory, is an inverted metamorphic triple-junction gallium indium phosphide /gallium indium arsenide cell, which it is claimed has an energy conversion efficiency of 40.8%. In advanced designs like this, the incoming light is split and focused optically, with each junction working on different parts of the incoming light spectrum, thus increasing the overall efficiency of energy conversion. The University of Delaware has claimed that the lateral optical concentrating system in its multi-junction cell has allowed them reached efficiencies of 42.8% (see here).

There will no doubt be many problems to resolve before commercial third-generation devices emerge, cost being one, and also the fact that some of these systems need cooling to operate efficiently. The operational life of some of the newer system is also unknown - efficiency may fall off with time.

The US Federal Government is supporting a range of new cell projects: see the box below for an overview of its development programme. Similar work is going on elsewhere, notably in Germany, the UK and Japan.

<!-- Col 1 -->

As can be seen from the US programme, cell technology and production techniques are not the only issues. Cell costs and performance are only not the only factors influencing total PV module costs. Other issues include the all-important connectors and packaging - the quality of the latter can determine how long the unit will continue to work efficiently. In addition, given the cost of the cells, optical solar concentration is increasingly being adopted, so that more solar energy can be collected and focused on a smaller area of cell. This can be done at both the small and large scale, usually, for larger systems, with the addition of tracking mechanisms to follow the sun across the sky.

Deployment of large-scale grid-supply Concentrating Photovoltaics (CPV) systems reached 18MW in 2006, and many more units are planned, including a 154MW unit in Victoria, Australia (see CEI February 2008).

At the smaller scale, as we have previously reported, a novel idea has emerged from a team at Massachusetts Institute of Technology (MIT) led by, among others, Professor Marc Baldo, which has developed a solar PV concentrator system using large glass panes treated with a transparent but reflective material, which redirects the light to PV cells round the edge. That way you can have a big collector area, but fewer cells. A layer of this transparent glass system could also be put on top of a standard PV cell array, selecting out some wavelengths. Or it could just be used as what would look like a smoked glass window letting 10% of the sunlight through into the room - a "solar window". The MIT team believes it could be in production within three years.

Some developers have produced hybrid solar thermal/ PV systems with, for example, a semi- transparent PV sheet on top of a heat-absorbing solar collector. This "PV/T" approach has the big advantage of keeping the PV system cool, which can be crucial given that PV cell efficiency falls with increased temperature. In addition it doubles up on roof or wall space. One such system, a PV array integrated with a Conserval SolarWall air-heating unit, was installed on a roof in the Beijing Olympic Village.

At the very high tech end of the spectrum, there have also been proposals for putting very large PV arrays in deep-space geo-stationary orbit, mounted on thin Mylar film stretched out over a lightweight frame like a sail, and transmitting the power generated to earth receiving stations via microwave beams. That would certainly avoid the day and night-time problem, but the costs of this approach (especially launching units into orbit) are likely to be very large. Given that solar energy falls for some periods on all parts of the planet surface, it would seem more realistic to try to make use of it here.

Future prospects

As we have seen, some new options may be just around the corner and market pressures could speed them on. There is now around 9000MW of solar PV in use around the world and the market is booming - it grew by over 40% in 2007, and 7GWp more could be added in 2010 according to the European Photovoltaic Industry Association. They add that, with volume production reducing unit costs, and new technology emerging, PV should be competitive with the electricity prices of utilities in southern Europe by 2015.

So far most PV has been use for providing power to buildings but with CPV, developing grid-linked bulk power could expand. Other key technological developments may also improve the prospects for PV. For example, the obvious problem with solar energy is that it is only available directly in the day-time. That might be suited to providing power for day-time-occupancy buildings, but at present most PV systems are operated on the basis on some type of net trading scheme: excess power during the day time, and in the summer, is exported to the grid and set against power imported at other times. In isolated off-grid situations batteries can also be used. However, if new more economically viable forms of energy storage become available, then the prospects for PV could improve dramatically, both at the small scale and large scale.

For example, if the efficiency of generation of hydrogen gas by electrolysis can be improved then it may be economical to store power from PV (and of course from other variable renewables) as hydrogen, for subsequent use in a fuel cell, to supply power when needed. MIT recently claimed a breakthrough in this area, with high-efficiency room-temperature electrolysis using special catalysts, which it says could revolutionise prospects for PV.

PV cells are relative robust, but the lifetime efficiency and long term performance characteristics of some of the more advanced cells are as yet unknown – thin-film cells are potentially prone to degradation. Some cells involve toxic materials so that production operations and occupational heath and safety have to be carefully monitored. However, once fabricated, environmental impacts are low. Once installed, PV cells of course have no emissions, although care must be taken with their eventual disposal if they contain toxic materials.

PV cell manufacture has in the past been relatively energy intensive, so that energy payback times, and thus carbon payback times, have been relatively high compared with some other renewables, but this has been reduced by new cell technology. NREL has calculated that in terms of full life-cycle costing, assuming a 30-year lifetime, energy payback times for multi-crystalline modules are up to four years; for thin-film modules, energy payback is three years using recent technology; and for anticipated thin-film technology the energy payback is just one year.

Overall, it seems likely that PV will become increasingly popular as a way to provide electricity for buildings: a roofing and cladding material that actually earns its keep. In addition, PV may begin to play a significant role in bulk electricity supply.