Research in Photovolatics

The Issue

Flat panel photovoltaics systems (PV for friends) has been unable to exceed a on field real efficiency of 15..17%.
This is caused by fundamental physical motivations related to the multiple colors (wavelengths) in the solar spectrum and the limitation of a single junction converter (so called Shockley Read Hall limit, SRH).

Therefore, covering our energy needs with PV, requires large areas translating into high “Balance of System costs” (cabling, support structures) and real estate usage.
At the same time, the limited value of solar converter per unit of surface, determined by the amount of energy produced during its lifetime, prevents from employing “smarter” and more sophisticated solutions at the solar cell level to work around the SRH limit.

Spectral Splitting by Multi Junction cells.

A solution is to use multiple solar cells each capable of converting a specific “color” (spectral region) of the solar spectrum.
This can be done by growing sequentially multiple exotic materials on a single substrate (Lattice Matched Multi Junction Solar Cells). Process complexity and sheer material costs, however, allow using this approach only for Concentrator Photovoltaics systems (CPV) and at the highest concentrations (500 to 1000 times the natural solar intensity).
Commercial success appears to be elusive for this approach but, as my granny used to say, “there are several ways to skin a cat” (why should she says this, is still one of those troublesome questions I do not really want to investigate too deeply!)

Concentration

Using “cheap” optical elements (either lenses or mirror) to concentrate solar radiation on a high efficiency cells allows to invest more money on the cell and to develop more efficient solutions. Aiming the concentrator at the sun a large power density can be achieved.
Still conventional optics has been historically designed to transfer good images (telescopes, binoculars, glasses, microscopes….) with maximum brightness. Solar concentration, on the other side, aims solely for the optimal energy transfer and uniform target illumination.
Imaging optics is, therefore, not the optimal candidate for solar concentration as it “answers” to a slightly different problem.
Non-imaging optics requires, therefore, a different toolbox and the definition of novel set of optical element affording to a much greater freedom in design. However, such freedom comes with huge difficulties in the definition of multi-parametric surfaces. Here I happened to provide my earliest original contribution.

Tracking

Regardless of the way the concentrator is designed, the concept of concentration is associated to a limited field of view (when we use a binocular or a microscope we look at a very small region to increase the detail level). Optical concentrators require some solution to track the (apparent) sun movement (Etandue conservation). This is normally done by mechanically aligning the concentrator with the sun as in a gigantic Sunflower.
As a consequence of this, concentrator systems are usually ugly, bulky devices better suited for the deserts of Arizona than for the elegant rooftops of Malibu’! Jokes asides, the presence of mechanical tracking prevents solar concentration to enter the residential (and rooftop) solar market altogether.
But, once again, there may be alternatives!

My contributions to the field

The first task I tackled was to find a way to practically design and realize optimal reflective concentrators. As huge free-form reflective surfaces could not be practically realized, I came to understand that a class of concentrating reflector surfaces can be developed based on a set of flat reflective elements.
The un-concentrated light from each flat element superimposes on the intended target area allowing for almost uniform receiver illumination under solar illumination. While this is a very special solution of the so called inverse problem of optics (determine the shape of the reflector given source and target illumination), it is nevertheless of critical importance for the case of solar concentration. The solution can be found in closed form for any given receiver shape simply by reverse raytracing of the edge rays from the chosen receiver. This is due to the fact that, in space, the position of a plane is characherized only by 6 independent real numbers.
My earliest concentrator designs were based on this principle, even before I fully understood its power, simply because it was much easier to “cut and paste” rigid flat mirrors on a suitably shaped glass fiber structure. The support structure surface finishing has no importance for the optical device performance since the mirrors are attached to it. This allows decoupling the difficulties of large scale mold realization from those of optical quality surface finishing.
Still, in the earlier design, I was using a Silicon solar cells receiver because Multi Junction cells were prohibitively expensive.
Looking for a path to higher efficiency converters, I came to realize that, by physically splitting the different chromatic components (wavelengths groups) of solar radiation onto spatially separated receivers, I could use “specialized” solar cells operating separately with high efficiency. While the idea, as I later found, was not new, it had been forgotten for many years.
I designed, therefore, an optical flat faceted concentrator having two reflective shells with a slight angular offset. The front “reflector” was transparent to part of the solar spectrum and reflective to everything else while the back reflector was a standard one. Each concentrator, therefore, allowed creating an illuminated area containing different portions of the solar spectrum.