In the last few years, we have pinned a lot of hope on the evolution of PV to make possible building integration and distributed power generation. Each is very attractive, the latter because it helps make the system potentially safe from single points of failure.
But there has always been a tension between that approach, and centralised systems feeding ever larger grids. Wind power has gravitated towards 'bigger is better' because it is - bigger turbines are more efficient, and big wind farms are best located where the wind resource is the best. With photovoltaics, big concentrator technologies have not had the same obvious compelling advantages. But now it is emerging that they do have.
A team of IBM researchers in Switzerland is working on an implementation of a High Concentration Photovoltaic Thermal system, that will start with significantly greater electrical conversion efficiencies than BIPV technology. But with well integrated thermal technology, it actually should be able to collect 80% of incoming sunlight and convert it to useful energy, while employing robust and low tech materials for its construction.
The actual solid state chips that do the electrical conversion remain high tech, as do the fundamentals of the extraordinarily effective cooling system, which is required to accommodate the heat generated by a concentration of 2000 to 5000 times the intensity of the incoming solar radiation. But the project team have thought through the rest of the system to replace expensive steel and glass with low cost concrete for the support structures, and simple pressurized metalized foils for the optics.
Emerging trends to district heating and even district cooling prompt the other synergies in the proposed technology. The very high temperature fluid cooling is coupled with desalination or district heating potential, and with district cooling using absorption chillers. Given thermal storage is already a well studied market, the promise of a centralised, but scalable solar/thermal generation is becoming very attractive.
IBM expects that the system would be approximately one third the cost third of current comparable technologies, but hasn't promised when the technology will emerge from its research phase to full prototyping.
