Feature Article Think Small! issue 5, Volume 2:
CdTe leads the pack in thin-film solar business
by Olivier Nowak
Investment has been literally pouring
into the production of thin-film solar
cells, regardless of the technology under
consideration. However, the market
for each technology differs greatly.
While just one year ago it could appear
that cadmium telluride was still in the
starting blocks (after many years of
waiting for the signal to run), such cells
are clearly winning the race for now.
In order to understand the
rabid interest in thin-film solar cells
within the context of the itself booming
photovoltaic market, WTC
studied this market in Europe.
Advantage thin film
Thin-film cell manufacturing has
many intrinsic advantages beyond
independence from silicon. Thin
film technology is more adaptable to
operational conditions than waferbased
cells. Wafer-based cells attain
their excellent efficiencies (20% at
25°C) only under optimal conditions,
i.e. facing the sun under a cloudless
sky, with a temperature of 40°C.
Any deviation causes the efficiency to
plummet to 10%. Weather patterns
therefore have a strong influence on the
quantity of electricity that is produced and the return for the investor.
Thin-film cells on the other hand
are much more dependable at producing
electricity in varied conditions.
Combined with the lower cost of the
modules, this explains why many
German PV power plants have opted
for thin-film cells and why the market
potential of thin-film is not limited
to being a “stop-gap” alternative
until the price of silicon stabilizes.
Lower efficiency, but more
electricity output
Thin-film modules work in cloudy
weather (ambient, diffuse sunlight) and
low light conditions. Thin film technology
also works well at inclination
angles less than 10°, whereas crystalline
silicon (wafer based) modules require higher inclinations
(there is no minimum tilt angle, but
a recommended tilt, as high as 35°).
This is especially important for roof
installations in commercical and industrial
buildings, where the structures are
usually flat. A wafer-based PV installation
requires mountings that incline
the panels, which creates shadows.
A fringe benefit of flat installations
is the additional thermal isolation.
Meanwhile, temperatures of 40°C
reduce the efficiency of thin-film cells.
CdTe modules have a temperature
coefficient of -0.25% per ° above 25°C,
which is about half that of crystalline
silicon modules. At 40°C, the peak
power loss in thin film cells is 3.75%
lower than at 25°C, opposed to 7.5%
lower for wafer based cells, i.e. loss
in wafer cells is about double that
of thin film at high temperatures.
By replacing glass as the substrate,
thin-film modules can be more than
50% lighter than wafer-based modules.
For example, modules from
the company Uni-Solar with a 100-
μm stainless steel substrate weigh
around 4 kg/m2 compared to around
11 kg/m2 for glass-encapsulated
wafer-based modules. Installations
can thus be placed on very large roofs
without overloading the structure,
or on the roofs of older buildings.
Although public taste may evolve
with the proliferation of photovoltaic
installations, thin-film panels offer
a much more pleasant colour than
the fractured blue of polycrystalline
panels. Actually, their quasi-uniform
dark colour offers a good match for
many corporate architectural projects.
Thin-film cells also allow semitransparent
modules integrated with
other functions in so-called “smart
window panes”. Since 2006, Sharp
has sold panels that integrate thinfilm
cells and LEDs. The panels are
mainly targeted at the retrofit business.
However, this aspect of thin-film has
yet to find traction in the market.
Now that more large-scale experience
has been accumulated, the
integrators and end customers are
familiar with thin film units and the
market is rapidly segmenting into
wafer-based and thin-film cells.
Applications on flat rooftops and power plants
Photovoltaic power plants increasingly compete with other forms of land exploitation. Large installations are now emerging, whose sheer size is hard to overlook. Meanwhile, Building Integrated PV or BIPV is still embryonic and therefore will not impact solar farming. As a result and in spite of the positive image accorded to photovoltaics— and the lower impact on the surroundings compared to wind farms—there is increasing resistance to these large installations.
CdTe seals First Solar fortune
Compared to other thin-film solutions,
CdTe cells owe their success
primarily to three major advantages:
• module price is rather low—
down to 1.4 €/W in 2007
• efficiency now lies around
9.5-10% and climbing
• reliability has been demonstrated.
Currently, the American company
First Solar dominates the CdTe market. First Solar is probably the first thin-film
company to have entered in the virtuous
circle that sees integrators trusting the
company’s products and ordering them
in large quantities. The company can
therefore lower prices (which also feeds
the aforementioned trust!) and gain
new customers. First Solar proactively
addressed potential concerns relating to
the use of cadnium (which is poisonous)
by validating the environmental
and safety aspects and by offering a
recycling service for used modules.
First Solar’s rise can be qualified as
meteoritic (see chart above), with $77
million in sales for the quarter April-
June 2007, and six months sales already
bursting past the total sales in 2006.
Shares have consequently mutlipied by
six since the company’s market entry
in November last year. As of October
26th, First Solar was valued at almost
$11 billion. By comparison, Q-Cells
was valued at around €6 billion.
This dominance will soon be challenged
by the company Calyxo. As a
100% subsidiary of Q-Cells (the second
largest manufacturer of solar cells
worldwide with $540 million in sales
in 2006), Calyxo is probably the most
serious challenger, although its initial
plans call for a capacity of just 25 MW
in Thalheim, adjacent to Q-Cells’
other operations. The company’s 8
MW pilot line is already in operation.
Meanwhile the Italian jointventure
Arendi is a newcomer. The
company could benefit from factory
engineering know-how contained in
shareholder Marcegaglia group. The
Italian equipment supplier begins
operations of a 18 MW factory in late
2008. US start-up Primestar Solar can
also count on support of new minority
shareholder GE Electric to assert
itself on the re-emerging US market.
CIGS still looking for way in
The outlook is bleaker for copper indium–gallium sulfide (CIGS) at the moment. The technology has two main proponents: Würth Solar in Germany and Global Solar in the US, both with capacities of at best 20 MW. However, the promise of the highest efficiencies (18.5% for laboratory cells) have retained both the interest and flow of projects and investors. Notable are Avencis, a JV between Saint-Gobain and Shell building a 20 MW plant in Torgau, Germany, and Honda Soltec’s new 27 MW plant in Kumamoto, Japan. Meanwhile, long-time CIGS players like US companies Miasole and Daystar, and Japan’s Shell Showa have yet to demonstrate industrial production.
Si goes multi-junction
The single junction amorphous silicon thin-film cell looks doomed to obscurity due to low efficiencies (6%) and prices not significantly lower than other technologies. Multi-junction silicon cells with efficiency around 8.5% could remain competitive in the short term if the price is right. This type of module is rare but several new factories, mostly in Asia and Europe, are expected to start operating with multi-junction silicon based cells within the next two years.
Conclusion
Many new companies have joined the PV “gold rush”, buying manufacturing equipment for silicon thin-film processing and bypassing need for a silicon and wafer supply chain. It appears that CdTe is leading in the technology stakes as a result of its higher flexibility in exploiting the sun in varying conditions, bringing more value to the investor. That said, if the added capacities are indeed to be exploited, we will have to wait and see if the market can sustain so many thin-film modules.