PV loeng, Green
Introduction
Photovoltaics involves the direct conversion of sunlight into electricity in thin layers of
material known as semiconductors with properties intermediate between those of metals and
insulators. Silicon, the material of microelectronics and the information age, is the most
common semiconductor. In the latter half of the 20th century, silicon photovoltaic solar cells
started to be used mainly to generate small amounts of electricity in remote areas where there
was no conventional source of electricity. In the 21st century, photovoltaics will grow to
maturity. Almost everyone will be aware of photovoltaics since photovoltaic solar cells will
be on the roof of their home or that of their neighbours will be they in one of the growing
megacities across the globe .
Brief history
Solar cells have their origins from some of the most important scientific developments of the
20th century, combining the Nobel Prize winning work of several of the most important
scientists of that century. The German scientist, Max Planck, began the century engrossed in
the problem of trying to explain the nature of light emitted by hot bodies, such as the sun. He
had to make assumptions about energy being restricted to discrete levels to match theory and
observations. This stimulated Albert Einstein at the year1905, to postulate that light was made
of small “particles”, later called photons, each with a tiny amount of energy that depends on
the photon's colour. Blue photons have about twice the energy of red photons. Infrared
photons, invisible to the eye have even less energy. Ultraviolet photons are also invisible but
carry even more energy than the blue ones.
Einstein's radical suggestion led to the formulation and development of quantum mechanics,
culminating in 1926 in Edwin Schrödinger's wave equation. Wilson solved this equation for
material in solid form in 1930. This allowed him to explain the difference between metals,
good conductors of electricity and insulators; also the properties of semiconductors with their
intermediate electrical properties. Electrons, the carriers of electrical charge, are free to move
around in metals, allowing electrical currents to flow readily. In insulators, electrons are
locked into the bonds holding the atoms of the insulator together. They need a several value of
energy to free them from these bonds, so they can become mobile. The same applies to
semiconductors, except a smaller value of energy is needed -even the red photons in sunlight
have enough energy to free an electron in the archetypical semiconductor, silicon.
Russel Ohl discovered the first silicon solar cell by accident in 1940. He was surprised to
measure a large electrical voltage from what he thought was a pure rod of silicon when he
shone a flashlight on it. Closer investigation showed that small concentrations of impurities
were giving portions of the silicon properties dubbed “negative” (n-type). These properties are
now known to be due to a surplus of mobile electrons with their negative charge. Other
regions had “positive” (p-type) properties, now known to be due to a deficiency of electrons,
causing an effect similar to a surplus of positive.
William Shockley worked out the theory of the devices formed from junctions between
“positive” and “negative” regions ( p-n junctions) in 1949 and soon used this theory to design
the first practical transistor. The semiconductor revolution of the 1950s followed, which also
resulted in the first efficient solar cells in 1954. This caused enormous excitement and
attracted front-page headlines at the time.
The first commercial use of the new solar cells was on spacecraft, beginning in 1958. This
was the major commercial application until the early 1970s, when oil embargoes of that
period stimulated a re-examination of the cells' potential closer to home. From small
beginnings, a terrestrial solar cell industry took root at this time and has grown rapidly,
particularly over recent years, to US$1 billion per year in sales, by the end of the 20th century.
Increasing international resolve to reduce carbon dioxide emissions as a first step to reigning
in the `Greenhouse Effect, combined with decreasing cell costs, sees the industry poised to
make increasing impact over the first two decades of the new millennium.
3. Operating principles
Fig. 1 is a schematic of a solar cell under illumination. Light entering the cell through the gaps
between the top contact metal gives up its energy by temporarily releasing electrons from the
covalent bonds holding the semiconductor together; at least this is what happens for those
photons with sufficient energy. The p-n junction within the cell ensures that the now mobile
charge carriers of the same polarity all move off in the same direction. If an electrical load,
such as the lamp shown in Fig. 1, is connected between the top and rear contacts to the cell,
electrons will complete the circuit through this load, constituting an electrical current in it.
Energy in the incoming sunlight is thereby converted into electrical energy consumed by this
load.
The cell operates as a `quantum device, exchanging photons for electrons. Ideally, each
photon of sufficient energy striking the cell causes one electron to flow through the load. In
practice, this ideal is seldom reached.
Some of the incoming photons are reflected from the cell or get absorbed by the metal
contacts (where they give up their energy as heat). Some of the electrons excited by the
photons relax back to their bound state before reaching the cell contacts and thereby the load.
The electrical power consumed by the load is the product of the electrical current supplied by
the cell and the voltage across it. Each cell can supply current at a voltage from 0 V to a
maximum in the 0.5-1.0 V range, depending on the particular semiconductor used for the cell.
Cell technology
Silicon wafers
The technology used to make most of the solar cells, fabricated so far, borrows heavily from
the microelectronics industry. The silicon source material is extracted from quartz, although
sand would also be a suitable material. The silicon is then refined to very high purity and
melted. From the melt, a large cylindrical single crystal is drawn, usually of 10-15 cm
diameter and 1 m or more in length, weighting several tens of kilograms. The crystal, or
`ingot, is then sliced into circular wafers, less than half a millimetre thick, like slicing bread
from a loaf. Sometimes this cylindrical ingot is `squared-off before slicing so the wafers have
a quasi-square shape that allows processed cells to be stacked more closely side-by-side.
Most of this technology is identical to that used in the much larger microelectronics industry,
benefiting from the corresponding economies of scale. Since good cells used in
microelectronics, additional economies are obtained by using `off-specification silicon and
`off-specification a silicon wafers from this industry.
Fig 2
As the photovoltaics industry matures, it will increasingly use technology optimised for its
own requirements. An example is the increasing use of multicrystalline silicon wafers (Fig. 2)
The starting ingot is formed simply by solidifying the molten silicon slowly in its container.
This ingot can be massive, weighing several hundreds of kilograms. It is sawn into pieces of a
more manageable size and then sliced into wafers. Techniques for growing silicon in the form
of ribbons from the melt have also been developed. These have the advantage that no slicing
is required.
4.2. From wafers to cells
Some manufacturers make their own wafers while others buy them from wafer suppliers. In
either case, the first step in processing a wafer into a cell is to etch the wafer surface with
chemicals to remove damage from the slicing step.
The surface of crystalline wafers is then etched again using a chemical that etches at different
rates in different directions through the silicon crystal. This leaves features on the surface,
with the silicon structure that remains determined by crystal directions that etch very slowly.
The square-based pyramids of Egypt. These pyramids are very effective in reducing reflection
from the cell surface. (Light reflected from the side of a pyramid will be reflected downwards,
getting a second chance to get coupled in).
The all-important p-n junction is then formed. The impurity required to give p-type properties
(usually boron) is introduced during crystal growth, so it is already in the wafer. The n-type
impurity (usually phosphorus) is now allowed to seep into the wafer surface by heating the
wafer in the presence of a phosphorus source. This gives a thin skin of phosphorus-doped
material around the entire wafer. The skin along the wafer edge is removed (that along the
rear is rendered inactive during the rear contacting step).
Next, the top and bottom contacts are applied using metal particles (usually silver) suspended
in a paste with other additives. This paste is `screen-printed onto the cell surface in the desired
pattern using a simple process similar to that used to print patterns onto T-shirts. After
printing, the paste is dried and heated at high temperature, leaving the metal particles
agglomerated together. A very thin layer of insulating material is sometimes added to the top
cell surface as an antireflection coating, similar to the coating used on high-quality camera
lenses. Such coatings are always used for `multicrystalline wafers, since the pyramidaltexturing approach is not effective for such wafers and this alternative approach is essential to
control reflection.
All the equipment required for this process is available “off the shelf” from the
microelectronics industry (the `hybrid or “thick film” industry sector). This, ombined on the
surface, with the silicon structure that remains determined by crystal directions that etch very
slowly (Fig. 3)
The square-based pyramids apparent in Fig. 3 that are formed by this process are similar in
shape, if not in size, to the great pyramids of Egypt. These pyramids are very effective in
reducing reflection from the cell surface. (Light reflected from the side of a pyramid will be
reflected downwards, getting a second chance to get coupled in).´The all-important p-n
junction is then formed. The impurity required to give p-type properties (usually boron) is
introduced during crystal growth, so it is already in the wafer. The n-type impurity (usually
phosphorus) is now allowed to seep into the wafer surface by heating the wafer in the
presence of a phosphorus source. This gives a thin skin of phosphorus-doped material around
the entire wafer. The skin along the wafer edge is removed (that along the rear is rendered
inactive during the rear contacting step). Next, the top and bottom contacts are applied using
metal particles (usually silver) suspended in a paste with other additives. This paste is
“screen-printed” onto the cell surface in the desired pattern using a simple process similar to
that used to print patterns onto T-shirts. After printing, the paste is dried and heated at high
temperature, leaving the metal particles agglomerated together. A very thin layer of
insulating material is sometimes added to the top cell surface as an antireflection coating,
similar to the coating used on high-quality camera lenses. Such coatings are always used for
multicrystalline wafers, since the pyramidal-texturing approach is not effective for such
wafers and this alternative approach is essential to control reflection.
A more recently commercialised approach is the Heterojunction with intrinsic thin layer
(HIT) cell of Fig. 7. This combines crystalline silicon technology with that of amorphous
silicon, discussed below. The HIT cell would be expected to give some of the improvements
of the buried contact sequence in the areas mentioned, although not to the same extent. The
rear processing of the cell is improved compared to the buried contact sequence and the cell
responds to light from both directions, a feature that can be used to advantage in some
applications.
For the processing of multicrystalline wafers, the use of silicon nitride as an antireflection
coating has advantages known for some time. These arise from the presence of hydrogen in
this layer, arising from its presence in one of the source gases (SiH4) used in the deposition
process. The hydrogen diffuses into the silicon and is effective in reducing detrimental
activities at the boundaries between the individual grains in multicrystalline material. The use
of nitride is expected to be more widely adopted in the future for such multicrystalline
material, in particular.
. Thin-film solar cells
Thin-film advantages
The potential for on-going cost reductions is the key reason for confidence in a significant
role for photovoltaics in the future. Rather than the wafer-based technology of the previous
section, the future belongs to thin-films), thin layers of semiconductor material are deposited
onto a supporting substrate, or superstrate, such as a large sheet of glass´
.
Typically, less than a micron thickness of semiconductor material is required, 100-1000 times
less than the thickness of a silicon wafer. Reduced material use with associated reduced costs
is a key advantage. Another is that the unit of production, instead of being a relatively small
silicon wafer, becomes much larger, for example, as large as a conveniently handled sheet of
glass might be. This reduces manufacturing costs. Silicon is one of the few semiconductors
inexpensive enough to be used to make solar cells from self-supporting wafers. However, in
thin-film form, due to the reduced material requirements, virtually any semiconductor can be
used. Since semiconductors can be formed not only by elemental atoms such as silicon, but
also from compounds and alloys involving multiple elements, there is essentially an infinite
number of semiconductors from which to choose.
At present, solar cells made from five different thin-film technologies are either available
commercially, or close to being so. Over the coming decade, one of these is expected to
establish its superiority and attract investment in major manufacturing facilities that will
sustain the downward pressure on cell prices. As each of these thin-film technologies has its
own strengths and weaknesses, the likely outcome is not clear at present.
5.2. Amorphous silicon alloy cells
5.2.1. Properties
Given its success in wafer form, silicon is an obvious choice for development as a thin-film
cell. Early attempts to make thin-film polycrystalline silicon cells did not meet with much
success. However, starting from the mid-1970s, very rapid progress was made with silicon in
`amorphous form. In amorphous silicon, the atoms are connected to neighbours in much the
same way as in the crystalline material but accumulation of small deviations from perfection
means that the perfect ordering over large distances is no longer possible. Amorphous
material has much lower electronic quality, as a consequence, and originally was not thought
suitable for solar cells. However, producing amorphous silicon by decomposing the gas,
silane (SiH4), at low temperature, changed this opinion. It was found that hydrogen from the
source was incorporated into the cell in large quantities (about 10% by volume), improving
the material quality. Hydrogenated amorphous silicon cells very quickly found use in small
consumer products such as solar calculators and digital watches, their main use so far.
The problem with outdoor use is that some of the beneficial effect of hydrogen becomes
undone under bright sunshine and the cell performance degrades. Initially, there was hope that
some simple material-related solution could be found. When this did not happen, the only
alternative was to design around it. Cells had to be developed that could work well with
material of degraded quality, rather than of the starting quality.
5.2.2. Amorphous cell design
Since the amorphous silicon quality is much poorer than crystalline silicon, a different cell
design approach is required. The most active part of a p-n junction solar cell is right at the
junction between the p- and n-type regions of the cell (Fig. 1). This is due to the presence of
an electric field at this junction.
With amorphous silicon cell design, the aim is to stretch out the extent of this junction region
as far as possible so almost all the cell is junction. This is done by having the p- and n-type
doped regions very thin, with an undoped region between them. The strength of the electric
field established in this undoped region is nearly constant and depends on this region's width.
The poorer the quality of amorphous silicon, the stronger the field needs to be for the device
to work well and hence the thinner the device needs to be. For degraded material, it turns out
that the cell needs to be thinner than the thickness required to absorb all the useable incident
sunlight. The way around this is to stack several cells on top of one another so that light not
absorbed by an upper one passes through to an underlying cell.
This works best if the material in the underlying cells is varied so that each responds
progressively better to the redder light that is transmitted to it. By alloying silicon with
germanium, a material chemically similar to silicon but much scarcer, this is readily achieved.
The best commercial amorphous silicon cells presently use three cells stacked on top of one
another, with progressively more germanium in the bottom. Each cell is very thin, only 100-
200 nm thick. This ensures reasonable stability (only about 15% degradation in output when
exposed to bright sunlight). However, the stabilised efficiency is quite poor, only 6-7% for the
best commercial modules, according to manufacturers' data sheets.
This low efficiency, even with the sophisticated cell design involved, is expected to make it
difficult for this technology to be competitive in the long term. However, the low
temperatures involved in making these cells mean that they can be deposited onto lowtemperature substrates such as plastics. This makes them especially suitable for consumer
products.
5.3. Thin-film, polycrystalline compound semiconductors
Many semiconductors made from compounds can absorb light more strongly than the
elemental semiconductors, silicon and germanium, for reasons that are well understood but
quite subtle (silicon and germanium are indirect rather than direct bandgap materials). This
means compound semiconductor cells can be thin but still operate efficiently. Most compound
semiconductors, when formed in polycrystalline form, have poor electronic properties due to
highly deleterious activity at grain boundaries between individual crystalline grains in the
material. A small number maintain good performance in polycrystalline form for reasons that
are not usually well understood. These are the candidates for thin-film polycrystalline
compound semiconductor solar cells. One such semiconductor is the compound cadmium
telluride (CdTe). Technically, it is an ideal material, giving properties suitable for making
reasonable solar cells even with relatively crude material deposition approaches (such as
electrodeposition, chemical spraying, and so on). The junction in these cells is again between
p- and n-type materials, but for the latter, a different compound semiconductor, cadmium
sulphide, gives best results. CdTe cells have been used mainly in pocket calculators to date,
but large area, moderate performance modules have also been demonstrated. The main
concern with this technology is the toxicity of the materials involved, even though very small
amounts are used in the modules. At the very least, this would mean that modules would have
to be carefully disposed of or, preferably, recycled after their useful life was finished.
However, there may be some problems in gaining market acceptance in what is likely to be
mainly a `green market over coming years.
There are also only limited known resources of tellurium. If all identified reserves were
converted into cells of the present designs overnight, they could generate 10% of the world's
present electricity use (a steadily decreasing percentage indefinitely, if recycled at end of life).
An even more promising technology at the moment, in the author's opinion, is one based on
the ternary compound, copper indium diselenide (CuInSe2). As if three elements were not
enough, this compound is often alloyed with copper gallium diselenide (CuGaSe2) and copper
indium disulphide (CuInS2), giving material with up to five elements involved. The n type
layer in these devices consists of a layer of cadmium sulphide, as in the previous cadmium
telluride cells. An alternative for this layer is being sought, to eliminate the toxic cadmium.
Small area laboratory cells have demonstrated efficiency close to 19%, despite the finegrained polycrystalline material used. Modules of this material are now commercially
available in small volumes with efficiency up to 12% demonstrated in pilot production. This
is not far behind what is achieved with standard crystalline silicon wafer modules.
Apart from the use of cadmium and even more limited known resources of indium than
tellurium, an often quoted limitation of this technology is manufacturability. This is often
interpreted as meaning it is difficult to diagnose problems in production with this material,
since the difference between good and bad material is not sufficiently well understood to
allow differentiation and control during the various manufacturing steps.
Thin-film polycrystalline silicon cells
As previously mentioned, silicon is a weak absorber of sunlight compared to some compound
semiconductors and even to hydrogenated amorphous silicon. Early attempts to develop thinfilm solar cells based on the polycrystalline silicon did not give encouraging results since the
silicon layers had to be quite thick to absorb most of the available light.
However, in early 1980s, understanding of how effectively a semiconductor can trap weakly
absorbed light into its volume greatly increased. Due to the optical properties of
semiconductors, particularly their high refractive index, cells can trap light very effectively if
the light direction is randomised, such as by striking a rough surface, once it is inside the cell.
Optically cell can appear about 50 times thicker than its actual thickness if this occurs. Such
`light trapping removes the weak absorption disadvantage of silicon.
Work on polycrystalline thin-film solar cells is proceeding in two areas. A variety of `high
temperature approaches such as suggested by Fig. 9 are being explored.
There generally involve either high-temperature deposition of silicon onto a substrate or
melting the silicon after deposition, to obtain large grain size in the final film. Although
preparation details are sketchy, the thin-film silicon product available from the US company,
Astropower, is the most developed representa-tive of this class of approach. In this case, the
silicon is deposited onto an expansion-matched ceramic substrate. The final material consists
of millimetre sized grains and is similar in appearance to multicrystalline silicon wafers.
Small area cell performance in the 16-17% range has been demonstrated, similar to that from
such cells on moderate-quality multicrystalline silicon. The performance of large arrays of
such cells has also been similar.
The other type of approach is a `low-temperature approach, generally based on amorphous
silicon technology. One approach is to deposit the silicon in amorphous form and then
crystallise it by heating for prolonged periods at intermediate temperatures. This `solid-phase
crystallisation approach has produced cells of quite reasonable performance. Another
approach has involved changing the amorphous silicon deposition conditions, to produce a
nanocrystalline phase of silicon. The potential of this approach was highlighted by early
results with the `micromorph solar cell. More recently, cell efficiency above 10% has been
confirmed with this approach. There are plans to use such cells as the lower cell in a tandem
configuration with an amorphous silicon upper cell, with commercial product targeted for
2002. Also targeted for commercialisation in the same timeframe is a polycrystalline silicon
on glass product shown in Fig. 10, based on an amorphous silicon precursor
5.5. Nanocrystalline dye cells
A completely different thin-film approach is based on the use of ruthenium-based organic
dyes [8,9]. Dye molecules are coated onto a porous network to titanium dioxide particles and
immersed in an electrolyte (Fig. 11). In a process bearing some relationship to photosynthesis,
light absorbed by the dye photoexcites a electron into the titanium dioxide which completes
the circuit through the external load and the electrolyte. Interestingly, the dye only absorbs a
band of photon energies, rather than all photons of energy above the prospects for unique
devices, such as transparent windows that convert the infrared wavelengths while letting
visible light through.
6. Longer-term developments
As outlined above, photovoltaic technology is on the verge of a major transition from `firstgeneration silicon wafer-based technologies to a `second generation thin- film product. This
transition should proceed at an accelerating pace over the coming decade. Once this transition
is complete, is there anywhere else for the technology to go?
Although there will be evolutionary improvements in both first- and second-generation cell
performance, the efficiency of standard cells is restricted to less than 33% by quite
fundamental considerations. Basically, standard cells are quantum converters, ideally
converting one sunlight photon to an electron in the load. This alone limits efficiency to about
44%. An additional loss arises since the cell can supply this current at a voltage somewhat
less than the potential corresponding to its energy bandgap.
However, the thermodynamic limit on the conversion of sunlight into electricity is a more
impressive 93%. Is it possible for solar cells to come closer to this limit? The answer seems to
be `YES, although new ideas are probably required to make this a reality. One idea that is
well established is based on splitting sunlight into different wavelength bands and sending to
cells of bandgap optimised for each band. Fortunately, this splitting can be achieved more
simply merely by stacking cells on top of one another, with the highest bandgap cells
uppermost. In the limit of an infinite number of such cells, the limit on conversion efficiency
is increased to 70%, and further to 87% if focussed sunlight is used (the diffuse component of
sunlight would have to be wasted in the latter case, giving a lower effective efficiency in
practice).
Three-cell stacks are now in production for high-performance cells for spacecraft and to
improve the stability of amorphous silicon cells.
Are there other ways of obtaining such performance that may be simpler in practice? A few
other approaches have been suggested. These should become experimentally more feasible
with on-going improvements in material science over the coming decades.
One class of device is based on `hot carriers. Although the solar cells stay cool, the charge
carriers in such devices ideally reach distributions typical of much higher temperatures by
avoiding energy loss by collisions with the atoms making up the cell,a fundamental loss
mechanism in standard cells. Efficiencies similar to the infinite tandem cell are possible in
this case. Another approach is to use not just excitations between two bands of carrier energy
as in a normal cell, but excitations in materials designed to have more than two bands of
allowed energy participating in the process. Again, efficiencies approaching the infinite
tandem cell case are feasible, in principle.
7. Policy implications
The history of the evolution of solar cell costs versus accumulated production volume is
shown in Fig. 13. The straight line fitted to the data shows an 80% learning curve, i.e. the
behaviour expected if cell costs decreased to 80% of their original for each doubling in
production volume. This straight line can be seen to be quite a good fit to the data actually
observed. Almost all the data shown apply to wafer based technology, with thin-film
technology yet to make its impact. With on-going improvements in cost expected from both
second- and third-generation technology, the type of `learning behaviour shown in Fig. 13 is
expected to be maintained well into the future, or even accelerated when new technology
makes its appearance. If this scenario is accepted, the key to reducing photovoltaic costs lies
in increasing the quantity sold. This encourages both efficiencies in production and the
investment in the development and commercialisation of new generations of technology.
However, at present costs, photovoltaics are competitive mainly in remote areas.
Markets here are small in the developed world and potentially enormous in the developing
world, but largely inaccessible due to lack of access to finance.
Fortunately, a market-pull mechanism seems to have been recently put into place that seems
likely to provide healthy growth and cost reductions over the coming decade. This is by the
subsidisation of rooftop mounted systems in urban areas of the developed world (Green,
2000).
Interest in this rooftop application started in the USA in the late 1970s. Japan took over the
running in the mid-1980s with the construction of a test bed for over 200 residential systems
on Rokko Island in 1986. After several years of evaluating technical issues related to their
grid connection, a subsidised installation program was launched by the Japanese government
in 1993. By the end of 1999, over 20,000 rooftop photovoltaic systems had been installed in
urban areas of Japan, with a massive 1.5 million being targeted for 2010.
Note to be outdone, the US Government announced a `million roof program in mid-1997,
targeting this number of systems also for 2010. At the end of 1997, the European Union also
agreed to a similar target for 2010, with half the systems to be installed inside Europe and half
outside. Since then, individual countries such as the Netherlands, Germany, Italy and
Australia have announced targets for photovoltaic roofs.
These programs generally involve a three-way subsidy between the government, home-owner,
and the local utility. Government subsidies are usually large in the early stages of the program
(50-70%) but reduce with time as system costs drop (currently 33-1/3% in the Japanese
program). This allows more systems to be subsidised for the same sum as the programs
develop.
Even with this government subsidy, it would still be cheaper for the home-owner to buy
power from the grid. Those participating are usually attracted by the opportunity to do
something positive for the environment, rather than waiting for governments or utilities to
take concerted action.
The systems operate without energy storage by selling excess power generated in the daytime
to the utility operating the local grid. At night, power flows in the opposite direction. In some
parts of the world, utilities are required by legislation to pay the same for electricity supplied
by a photovoltaic system as they charge for electricity sent the other direction. Other utilities
do this voluntarily. By adopting such ”net metering”, utilities also help improve the viability
of these systems by effectively providing free storage for the solar generated electricity.
Utilities are already well placed to provide such storage, since `spinning reserves and stand-by
units are always available to meet a sudden increase in demand or the unscheduled outage of a
large, conventional power station. It is not until photovoltaics were supplying more than about
15% of the electricity in any particular network that any significant change in operation of the
grid network would be required.
As a result of these residential programs and others likely to be stimulated by them, there is
expected to be a rapidly increasing demand for photovoltaics over the coming decade. This
should continue driving down the costs of photovoltaic systems as suggested by Fig. 13, to
the level where systems should be attractive with minimal subsidies. It is not difficult to
imagine that, at some stage in the future, all new housing might require its own photovoltaic
system to make it greenhouse neutral. Reduced costs will progressively open up other
applications such as large centralised photovoltaic power stations providing bulk electricity to
the grid in direct competitive to conventional, but less benign, large-scale generators.
If this prognosis is correct, appropriate policies are those that encourage the success of these
urban rooftop photovoltaic systems, while simultaneously encouraging the development and
commercialisation of improved photovoltaic technology. In schemes where premiums for
sustainable energy generation are available, it is important for photovoltaic development that
`portfolio rather than `least-cost strategies be adopted. In portfolio approaches, some portion
of the funds generated by the premium would be assigned to photovoltaically generated
electricity. Even though it is presently not directly competitive with electricity from other
sustainable sources such as wind or biomass, this strategy would recognise the greater longterm potential of photovoltaics to impact energy generation and use.
8. Conclusion
As we enter the new millennium, photovoltaics are poised to make a more significant impact
upon energy use. Although still dominating the marketplace, `first generation a technology
based on silicon wafers is starting to be challenged by `second generation thin-film
technology. This has the advantage of much lower material costs and of being better suited to
high-volume manufacture.
There is also scope for a third generation of technology, based on principles not yet fully
developed, offering prospects for significantly enhanced efficiency at some stage further into
the future. Accompanying the rapidly growing demand for photovoltaics and such on-going
improvements in technology, photovoltaic costs have been steadily decreasing and are
expected to do so well into the future.
The application driving market growth at the present is in urban residential rooftop systems.
This is expected to continue to be the most important commercial application over the coming
decade, although dependent on subsidies for its viability. The required level of subsidy is
expected to decreases over the coming decade, with this application fully economic by the end
of the decade. Policies that support this market pull mechanism while encouraging the
development and commercialisation of improved photovoltaic technology are believed to be
the most appropriate.