Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
The technological potential of PV
The basics of PV systems
The fundamental element of any PV system is the PV cell. A PV cell is a particular semiconductor device that is able to convert sunlight directly into electricity (direct current). PV cells are inherently low voltage, high-current density devices. Several series-connected cells are needed to from a module , the basic commercial components of PV systems. Serval modules are then connected to from a string. The in series and/or in parallel connection of different strings makes it possible to obtain practically any desired operating voltage for the final PV system.
| How PV cells work The possibility of producing electrical energy directly from sunlight is based on some properties of semiconductor materials, and particularly on the interaction occurring in certain solid materials between photons (light packets or "quanta") and the electrons of the solid-state atomic matrix. Semiconductors are characterized by the existence of a so-called "energy band gap." This is the finite difference between the energy level of electrons in a stable position in the crystal structure - the valence band - and the next allowed electron energy band level, known as the conduction band, in which an electron can move freely through the material. The magnitude of the gap is different for each semiconductor material. In an equilibrium situation at room temperature with no external applied fields, the semiconductor valence band is completely filled by the valence (external orbital) electrons, whereas the conduction band is completely free of electrons. When the semiconductor is exposed to sunlight, photons with an energy content higher than the band gap can excite electrons from their stable energy level (the valence band) to higher energy levels (the conduction band), leaving a so-called hole in the valence band. In this state, if an external electric field is applied, the material is able to carry electricity. However, in order to produce electricity, a further step is needed. In the absence of an electric field, the excited electrons in the conduction band will recombine with the holes. That is, they will "relax" into the vacant, lower energy levels in the valence band, and no current will be observed. In effect, the electrons excited by the photons need to be oriented by an electric field to produce useful current. The basic concept of a PV device is to produce this electric field internally in the solid. This is achieved by combining semiconductor materials with different characteristics to form a junction. Two configurations, namely homojunctions and heterojunctions, are possible. In homojunctions, two differently doped layers of the same semiconductor are combined, whereas heterojunctions are made by two (or more) semiconductors with different energy gaps. In the junction zone a built-in electric field is established, which is able spatially to separate the photo-excited electrons from the holes and start them drifting in opposite directions. If the cell material quality is good enough, the carriers will reach the external electric contacts of the cell and a voltage between the latter will be observed (photovoltaic effect). If an external load is applied, direct current will pass through the electric circuit. The PV cell is then generating useful electrical power. PV cells are inherently low-voltage, high-current-density devices. A typical commercial crystalline silicon cell (10 cm x 10 cm area) can produce a current of up to 3 amperes at a voltage of only 0.5 volts. Several series-connected cells are therefore needed to form a module, the basic commercial component of PV systems. A commercial silicon module (0.4 m2 area) produces between 40 Wp and 50 Wp at 17V voltage. Several modules are then connected to form a string. The in series and/or in parallel connection of different strings makes it possible to obtain practically any desired operating voltage for the final PV system (from low voltage for household applications to 20kV for large power plants). are needed to form a module, the basic commercial component of PV systems. Several modules are then connected to form a string. The in series and/or in parallel connection of different strings makes it possible to obtain practically any desired operating voltage for the final PV system (from low voltage for household applications to 20kV for large power plants) |
Depending on the type of electrical connection, there are two main categories of PV systems: the autonomous (stand-alone) and the grid connected systems.
In the stand-alone systems, the PV field is connected to a means of energy storage (usually electrical batteries). The energy produced can thus also be used when the sun does not shine.
In the grid-connected systems, usually no accumulation systems are employed and the electricity is fed directly into the grid. Power conditioning systems are needed to operate the system at maximum power and to avoid stability problems in the electricity network. Finally, an inverter is needed to convert the direct current produced by the PV modules into useful alternating current. However, it should be noted that many appliances (e.g. all electronic devices and several high-efficiency lamps) could run on direct current. Several companies are exploring the possibility of introducing a dual supply for their products.
The final part of a PV system is its supporting structure. There are two categories of system: those mounted on purpose-built structures (i.e. power plants in open fields and PV systems on flat roofs), and building-integrated systems, which use part of the building structure for support.
In current terminology there are two basically different parts to a PV system: the PV module; and all other structures and means by which the electricity produced by the module can be delivered to the grid or to the final users. This so-called "Balance-of-System`' includes all supporting structures, power conditioning systems, wiring, and eventual energy storage systems.
PV technologies
There are many different possible technologies for manufacturing PV cells and modules. A classification can be made with regard to system types, manufacturing processes' and semiconductor materials. One main distinction can be made between crystalline semiconductor cells and thin-film devices.
Thanks to the related experience of the electronics industry, crystalline silicon cells dominate the PV market at present. In 1993 they had 84 per cent of market share (Vigotti 1994a). Crystalline silicon cells use scraps from the electronics industry as a feedstock. Today, this mature technology is the only one that can simultaneously offer high stability, long lifetimes, high module efficiencies (15.3 per cent for high-purity monocrystalline Si modules and 11.1 per cent for polycrystalline Si modules of slightly lower purity), and advanced production status (see also table 7.1).
However, this particular technology will have to give way to other PV technologies in the near-mid future, for at least two reasons. First, demand by the PV industry will soon exceed the amount of scrap material offered by the electronics industry. Second, and more important, the present technology derives directly from the electronics industry and is not optimized for PV cell production. The present manufacturing processes of crystalline silicon cells are very inefficient in terms of the consumption of both primary energy and raw materials. This is also reflected in the present high cost of PV systems.
In the near future, so-called "solar-grade" crystalline silicon cells will most likely be used. Solar-grade silicon is much less pure than electronics-grade silicon, but pure enough for PV cells. Solar-grade silicon manufacturing processes rely on completely different purification processes of metallurgical silicon. These processes are much simpler than the ones currently used for silicon cells derived from electronics scraps. They are also expected to be much more efficient as far as primary energy consumption and raw materials use are concerned. Up to 1995, there had been no large-scale production of solar-grade silicon. However, solar-grade silicon production has top priority on the agenda of several PV industries, particularly in Europe and in Japan.
In the mid-long term, the large-scale use of thin-film PV devices is expected. Thin films are based on yet another completely different approach and manufacturing process from those employed for crystalline PV cells. Whereas crystalline silicon requires a thickness of about 200 microns to absorb 90 per cent of incident light, thin films need only a few microns of active material to collect the same amount of radiation. Consequently, far less semiconductor material is needed. This greatly decreases costs and reduces primary material resource use and primary energy consumption during manufacturing. Secondly, thin-film production techniques are particularly well suited for large scale production. Thin-film deposition is done by directly spraying or sputtering the active material onto a glass or metal substrate. This continuous process is far more efficient than the batch processes of crystalline cell production. Moreover, it allows a manufacturer to produce cells as large as complete crystalline modules. This increases the effective active area and eliminates the problems and costs of connecting a number of cells together to form a module. Moreover, by stacking several thin-film layers, multifunction cells can be produced. In such a cell, each layer absorbs a different part of the light spectrum. In principle, the theoretical efficiency limit of multifunction cells is much higher than that of conventional cells. Finally, several thin-film modules are semi-transparent. Therefore, they can be used in buildings as PV "windows" or PV glazing surfaces.
Table 7.1 Conversion efficiency of venous PV technologies at the different stages of their development (%)
Cell type |
Largest standard commercial modulea |
Best prototype module |
Area (cm2) |
Best laboratory cell |
Area (cm2) |
Theoretical limit |
Production status |
| CRYSTALLINE SEMICONDUCTOR | |||||||
| Silicon | |||||||
| Monocrystalline Si | 15.3 | 20.8 | 743 | 24.0 | 4.00 | 30-33 | Large-scale |
| 19.5 | 3,080 | 21.6 | 47.00 | ||||
| Polycrystalline Si (p-Si) | 11.1 | 17.0 | 17.2 | 100.00 | 22 | Large-scale | |
| EFG-band p-Si | 14.7 | 50.00 | Small-scale | ||||
| Dendritic Web | 17.0 | 4.00 | Pilot prod. | ||||
| GaAs | |||||||
| Monocrystalline on GaAs | 29.0 | 0.05 | Small-scale | ||||
| Monocrystalline on other substrate | 17.6 | 1.00 | Pilot | ||||
| THIN FILMS | |||||||
| Amorphous silicon (a-Si) | 6.8 | 10.2 | 933 | 12.7 | 1.00 | 27-28 | Large-scale |
| 12.0 | 100.00 | ||||||
| p-Si on ceramic (100 microns thick) | 11.2 | 225 | 14.9 | 1.00 | 20 | ||
| Polycrystalline silicon | 15.7 | 1.00 | |||||
| Polycrystalline GaAs | 8.8 | 8.00 | |||||
| CdTe | 7.25 | 10.0 | 15.8 | 1.00 | 28 | Pilot | |
| 7.7 | 3,528 | ||||||
| CIS | 11.1 | 15.9 | 1.00 | 23.5 | Pilot | ||
| 9.7 | 3,880 | 13.9 | 7.00 | ||||
| Mechanically stacked a-Si and CIS | 15.6 | 4.00 | 42 | ||||
| GaAsb | 22.0 | 29.3 | 0.50 | ||||
| GaAs on GaSbb | 34.0 | ||||||
| 37.0 | 0.05 | ||||||
Sources: Green and Emery (1994), Kelly (1993), IAEA (1992),
Proceedings of the 1st PV World Conference (1995).
a. Typical commercial PV module areas range from 0.5 to 0.75 m2.
b. Concentrator systems: in these systems direct radiation
is concentrated on a small cell by means of a Fresnel lens.
In table 7.1, the conversion efficiencies1 of the various PV technologies at the different stages of their development are summarized. The important thing to note is that efficiencies much higher than those of commercial modules have been achieved in the laboratory. Thus there is still considerable potential for further improvement between the theoretical and commercial limits. Uncertainties regarding production costs, the investments needed, and the rapidity of technological improvement make it impossible to select a single best or most likely PV system for the future.
However, it is much more important to realize that all PV technologies are in rapid evolution. Figure 7.1 shows the past efficiency evolution over time for different types of cells. Moreover, the fact that several technology development paths seem capable of providing comparable PV cell efficiencies and costs in the future means that there is likely to be competition between different design approaches. In turn, this is likely to accelerate improvements in PV technology and the industrialization process and to reduce manufacturing costs.
PV applications
PV systems are highly modular and therefore offer a wide range of applications. As already mentioned, the series and/or parallel connection of different modules and/or strings allow one to obtain practically any desired operating voltage for the final PV system. Figure 7.2 summarizes the applications of PV systems as a function of installed peak power. As shown, the range of possible applications extends from very small devices such as solar calculators or watches to the grid-connected multi-megawatt power plants.
Fig. 7.1 Efficiency evolution over time per type of PV laboratory cell
Today, PV systems for communications and solar home systems have the largest market share (21 per cent and 15 per cent respectively) (EPIA 1995). Up to 2010, the two largest markets expected are (a) solar home systems in developing countries, and (b) grid connected, mainly building-mounted systems in industrialized countries (see also the section on "A PV market diffusion strategy").
PV is very well suited to provide electricity to rural and remote areas in developing countries. Whereas taking the electricity grid to the people living in those areas would imply enormous investments and might take decades, small PV stand-alone solar home systems with a small battery storage are cost effective and can meet some basic needs such as lighting and TV. This would tremendously increase the quality of life of the local population. It would also encourage people to stay in their communities rather than migrate to a megalopolis in the hopes of increasing their standard of living. Although the cost of a solar home system today (around US$500) is relatively expensive for rural conditions, it could be subsidized by low-interest financing programmes. This is currently being done by the World Bank in some Asian countries.
Fig. 7.2 PV applications
The application of building-integrated PV systems is particularly interesting, because it shows several advantages compared with "conventional" PV power plants. First, the occupation of surfaces already used for other purposes substantially reduces the main environmental obstacle to the adoption and diffusion of PV, namely land requirements. As a consequence, it greatly increases the potential applicability of PV in areas of high population density. Second, integration into already existing or planned supporting structures and the substitution of building envelope materials reduce total system costs. Because total energy consumption during the manufacturing and installation of the systems is reduced, the energy payback time2 of the PV system is also reduced and its (indirect and low) environmental impacts are further lowered. Finally, this application actually expands the technological potential of PV systems, because in buildings they can play more roles than solely producing electricity. For instance, building-integrated PV panels can save energy when used as sun shading systems. Moreover, in buildings, there is the possibility of recovering a significant fraction of the thermal energy dissipated by the PV panels. This thermal energy can be used directly for room heating in winter and for pre-heating of water in all seasons. Potentially very interesting is the coupling of PV building-integrated systems with other solar-passive, bioclimatic architecture and energy-saving measures. This has significant environmental implications. Recent Life Cycle Analysis studies show that already building-integrated PV systems can "avoid" over twice the CO2 emissions compared with conventional PV power plants (Frankl 1994)3. These environmental benefits will increase with future PV technologies as conversion efficiencies increase and energy consumption during manufacturing decreases.
This application is poised to take off right now (1995), because practically all PV manufacturers worldwide are getting involved in developing new products to be integrated with buildings. Some non PV cell producers are specializing in this sector as well, by purchasing cells and selling specific products for the buildings market. Products range from PV roof-tiles, to facades, construction materials, and colour modules. Applications range from office, residential, and industrial buildings up to carpark roofs and highway sound barriers.
This phenomenon could have enormous implications, because it raises the number of interested and involved actors by orders of magnitude, on both the supply and the demand side. The result will be to promote competition and investment in the PV sector. PV will be a subject of interest not only for the limited number of PV industries and electricity utilities around the world, but potentially for millions of architects and engineers, as well as for their clients.