Page 78 - RSE - Results of the Apollon Project

Results of the APolloN PRoject ANd coNceNtRAtiNg PhotovoltAic PeRsPective

Module costs could be reduced to less than 0.99 euros/Wp with modules of 32% effciency or higher. These
higher effciencies are possible through a better control on the shape of the primary mirror, improvements in mirror
positioning and orientation, improvements in the refective secondary optics (now made manually), more research
on the AR-coating of the glass window (wavelength optimization with mirror and cell) and through the expected
signifcant increase of cell effciency over the coming years. Consider that 30% effcient APOLLON modules has been
reached using 39% effciency cells, while 44.1% effciency cells have already been announced by Solar Junction and
IQE, therefore CPV module effciency >32% can be possible. Using these measures, module costs as low as 262 euros/
module, or 0.87 euros/Wp are feasible for 34% effciency modules (see Estimate 3, Table 12).
The use of aluminum is a quite sustainable material choice because it is highly recyclable. In the development of
the prototypes, it was highly challenging to mechanically fasten relatively thin optically-coated sheets in such a way
as to make highly reproducible light paths. In addition, the steel fasteners and small structural components (4% by
weight) contribute 7% of the greenhouse gas emissions of the module. In the frst cost optimization scenario, some
of the steel fasteners will be removed from the bill of materials, because the receivers and mirrors will be fastened
permanently by other means. Advanced manufacturing techniques, such as 3D printing, may eventually be able
to provide a solution for making an aluminum, structurally sound, very light-weight, optical element that may be
affxed to a structure with causing mechanical stress on the optical elements.

Economic Analysis of Mirror-based Spectrum-splitting Systems
and Mirror-based ones

3
An economic analysis has been performed by CPpower on mirror-based spectrum splitting systems (MBS ) and
simple mirror-based ones.
The key components differing between the MBS module and the simple mirror-based module are:
3
 p electrical circuits;
 p cells;
 p mirrors (dichroic or wide spectral refectors).

One fundamental point regards solar cells: the spatial spectral splitting solution allows light spectrum usage to
be optimised, using cells of different band-gaps, with different effciencies on diverse spectral zones of the sunlight.
This can give advantages in that the cost of the cells is signifcantly lower than that of multi-junction solar cells,
giving an overall comparable effciency. In the current state of the art, III-V based single or dual junction solar cells
able to work under a reasonable concentration and with band-gaps different to that of silicon are available at costs
comparable to those of III-V based high effciency triple junction solar cells, because they are fabricated with similar
materials and processes. So, in the case of equal cost between double junction (DJ) solar cells and triple junction
solar cells, a MBS module using TJ and silicon solar cells has a better euro/W ratio with respect to a MBS module
3
3
using DJ and silicon solar cells; this is also due to the fact that the cost difference between the two different dichroic
flters is not signifcant.
Under this cell cost condition, the most signifcant economic comparison, based on the performance-to-costs
3
ratio, is between a MBS confguration using TJ and silicon solar cells and a non-dichroic confguration using only TJ
solar cells. Matching the goal for the optical effciency of 80%, the achievable effciency for a module with this design
using only TJ solar cells is of about 32%. For a module using spectrum-splitting technology, Si solar cells and TJ solar
cells, the achievable effciency is approximately 36%. To offer the same energy costs, the additional power generated
by the spectrum-splitting system must balance its additional costs.
In more details, the MBS module comprises the following set of additional components with respect to a solution
3
using only TJ cells:
p dichroic mirrors;
p silicon cells;
p receivers for silicon cells;
p cabling for receivers with silicon cells;
p bypass diodes for Si cells.

In the framework of the APOLLON project, in the design developed by CPower Srl, a module with only TJ solar
cells can use silvered mirrors instead of dichroic mirrors. Quotations obtained from industrial producers of the

77