As concentrated photovoltaic businesses soldier on, research into novel III-V devices is rife. Compound Semiconductor looks at what the future holds for the industry.

As IBM's novel CPV system follows a steady path to commercialisation, will devices based on new combinations of III-V materials follow? Credit: IBM-Research

Just when you thought the CPV industry was petering out, industry developments indicate otherwise. Although commercial front-runners, such as JDSU and Amonix, are scaling down operations, organisations lower down the business chain are seeing a surge in funds.

For example, the Climate Investment Fund recently extended its $7.6 billion award for the development of concentrated solar plants in MENA nations, to include CPV development.

At the same time, IBM Research and partners have bagged $2.4 million from the Swiss Commission for Technology to build an affordable version of a novel CPV set-up - the High Concentration PhotoVoltaic Thermal (HCPVT) system - that incorporates a thermal system to capture lost heat with water. IBM reckons recovering waste heat in this way will boost overall system efficiency to an admirable 80%.

But it's not just large-scale projects edging towards commercialisation that are scooping cash. Recent months have seen a flurry of research grants awarded to materials scientists and physicists keen to hone the III-V semiconductors fundamental to the efficient and affordable operation of these photovoltaics.

For example, in April last year, the UK-based Engineering and Physical Sciences Research Council (EPSRC) dished out some £500,000 to a team led by Liverpool University researchers to develop nitride-based cells for use in CPV systems. Meanwhile, researchers from the Universities of Manchester and Salford have just won some £880,000 to conduct theoretical work on InAs, GaAs and CsSe quantum dots for solar cells.

And only last month University College London and University of Bristol researchers won £950,000, again from EPSRC, to fabricate III-V quantum dots solar cells on silicon substrates for CPV systems. UCL researchers will pioneer MBE growth and device fabrication while Bristol colleagues will perform modelling to optimise performance.

Huiyun Liu, a key researcher from the UCL branch of this III-V QD project has seen a steady rise in the number of solar-funded projects. “[Commercial] companies have been struggling, but my research group is not,” he says. “Our track record is in lasers, but over the last year the funding I have received has been for solar research. We have seen a dramatic shift from laser to solar research, and definitely more interest from companies.”

As Liu highlights, his latest EPSRC-funded project is also supported by the UK government owned Defence Science and Technology Laboratory (DSTL), Wales-based IQE and UK-based Sharp Laboratories of Europe.

“We've also been talking to the French oil company, Total. They are interested in this area as they want to develop high efficiency, low cost solar cells,” he says. “Right now III-V solar cells are too expensive, but they have this high efficiency... the [companies] that have come to us all say the same thing; it's not making money now but we have to try.”

Next generation cells

When designing III-V CPV solar cells, most researchers have adopted a multi-junction structure, connecting a number of semiconductor junctions with optimised bandgaps in series to boost efficiency. While research into nitride-based solar cells is well underway, many existing multi-junction cell structures have been based on GaAs layers. For example US-based Spire Semiconductor claimed record peak efficiencies of 42.3% with itsCPV InGaAs/GaAs/InGaP cells, (bandgaps of 1.89/1.42/0.94eV respectively).

But III-V quantum dots promise to bring a better solar cell. Compared to the active layers used in conventional devices, these semiconductor crystals can be tuned to absorb light over a much wider range of wavelengths, boosting conversion efficiencies.

What's more, by working with quantum dots rather than planar layers, researchers avoid the strains that build up at the interfaces between different materials, opening the door to the growth of novel material combinations that simply wouldn't be possible in conventional devices.

With this in mind, Liu and colleagues intend to develop a “completely different” quantum dot solar cell system grown on silicon, both with and without a thin germanium buffer layer. This bold move could swipe away the high GaAs substrate prices of current CPV cells and capitalise on cheap CMOS manufacturing costs. The end result could be a very cheap yet highly efficient solar cell with, as Liu says, at least a 30% conversion efficiency.

“IQE provides silicon wafers with the germanium buffer layer, and at the same time we are also working with Sharp to develop GaAs solar cells on germanium-on-silicon,” the researcher explains. “But we are also working on the direct growth of III-V multi-junction solar cells on silicon with another industrial partner. If we can use the direct growth approach on silicon, this is better... We are academics, and we want to try any possibility."

Right now, the team islooking at a range of materials systems based on III-V quantum dots that absorb near the peak of the spectrum, around 1eV. As Liu explains: “We have previously worked with lasers and have now found that using the InAs/GaAs quantum dot system is not ideal for absorption in solar cells, so we are looking at alternative systems with a different bandgap alignment and long carrier lifetime.”

The key contenders are metamorphic InAs/GaAsSb and InP/GaAsP quantum-dot systems, and the researchers will also explore such materials combinations in the context of an intermediate band structure solar cell. Here an intermediate energy band is introduced into the energy gap of a single semiconductor junction. Such a structure promises conversion efficiencies up to 63%, but only if the photo-generated carriers in the intermediate level can be channelled solely to the host material.

Clearly such a device will require much more work yet, but as Liu highlights: “EPSRC is taking a longer term view by funding research such as this. Maybe industry is struggling, but for the academic, it's definitely a good area to be working right now.“


Professor Huiyun Liu and colleagues have just won £950,000, from EPSRC, to fabricate III-V quantum dot solar cells on silicon substrates for CPV systems.