By Geoffrey Duggan

The Market Today

Currently there is a lot of variety when it comes to CPV module design--particularly when it comes to choosing the optical system. Whether it is between reflective mirrors or Fresnel lenses as primary optics and then reflective, refractive or no secondary optic; the decision is at the heart of the design of CPV modules and systems and helps determine the ultimate power conversion efficiency that can be delivered by the complete system. In addition to the debate over the primary and secondary optical elements, at an even more fundamental level there is the question of optimal PV cell size and concentration factor.

These variations and options merely highlight that the current CPV market offerings are incredibly diverse, with many companies approaching the technical challenge of making CPV a truly commercial technology from many different perspectives. However, this also demonstrates that there is little consensus among the various players, and as yet there is no ‘referred method‘ of developing CPV components and systems. As a result, it is difficult for any company (or group of companies) to truly take advantage of the economies of scale necessary to drive down costs and enhance the commercial appeal of CPV systems.

In the next 3 years, lowering manufacturing costs will be crucial to the CPV industry. But as well as the gains from adopting best practices and economies of scale, a part of the cost reductions will come from advances in cell manufacturing techniques to lower the amount of cell material required in each cell. Exploiting increasingly optimized bandgap combinations, either by metamorphic growth or by layer transfer techniques, will produce cells with higher fundamental efficiency limits.

Efficiencies

In the past few years, we have seen efficiency gains at module and system level accelerate. For instance, Circadian Solar recently announced that it has achieved a temperature-corrected peak module aperture efficiency of 30% during trials at its test site in Lisbon1). We expect the current trend of ~1% annual increases in research cell efficiency (from the 2010 level of ~42%2)) to continue, although advances in cells with more optimum bandgap combinations could deliver more significant increases. Production cell efficiencies, meanwhile, will most likely continue to lag behind world record research cell efficiencies by 2%-3%. Overall system efficiencies are expected to rise to around 32% by 2013.

Commercially the emphasis will increasingly be placed on Levelized Cost of Electricity (LCOE), rather than just system efficiency and system price per watt, since LCOE is the key determining factor in commercial payback and return on investment. That is not to say that improving efficiencies is not still a critical factor in the future of CPV systems, but as already indicated in this article, commercial concerns will have to increasingly be at the forefront of the CPV system manufacturers thoughts.

Trends in CPV Module Design: 4 Key Elements

More specifically, there are four key technology elements that we expect to see significant developments in--delivering both technical and commercial improvements to CPV systems.

Module Construction     

The housing of the module itself is the first key area where the increasing prominence of commercial concerns will have an impact. Although we have started to see some modules with composite materials3) emerging that potentially deliver lower costs, but also some uncertainty about their durability, aluminium is likely to remain the material of choice in the near term.

However, there undoubtedly needs to be a continued effort to find cheaper methods of manufacture such as pressing, as used in the automated manufacture of the aluminium back pan of Solfocus‘ reflective CPV system. This method of manufacture is something that a number of companies have been looking at, but there remains scope for further cost savings to be achieved in the manufacturing process and the sourcing of the basic material.

The Lens Array

Nevertheless, there remains a lot of scope for further efficiency gains in the fundamental technology--at system and, of course, module level. Given the nature of CPV, the lens array is obviously the first area that will see continued development and improvements in the next few years. Many companies have been working with Fresnel lenses, as some reflective systems can lead to less optical power reaching the cell.

However, although Fresnel lenses deliver a number of benefits, one of the key trends is the perceived move of an increasing number of module manufacturers to the adoption of Fresnel lenses based on Silicone on Glass (SOG). While PMMA may continue to be the material of choice for many CPV modules, scrutiny of PMMA’s performance under the harsh conditions seen in Sunbelt regions where there are concerns about, for example, abrasion and lens expansion induced by high ambient and module temperatures, is leading to an increasing number of companies moving to SOG as their preferred solution, and we expect that trend to steadily continue.

The Receiver--Secondary Optics

There is now almost 100% use of a secondary optical element in CPV modules, most popularly a refractive trapezoidal light pipe or a reflective cone or trapezoid4). However, these optical elements still have a number of non-trivial problems: they are either too expensive, or could involve a mechanical fixing, which, in turn, adds cost and complexity in assembly. These issues present a significant hindrance to producing large volumes of CPV modules cheaply and efficiently.

One of the key trends we expect to see developing is the use of alternative solutions, such as ‘free form’ refractive elements4) and ball lenses5). Regardless of which alternative gains mass market acceptance, it is clear that this will be a key area for CPV module development.

The Receiver--Cells

The cell itself is the final area where there are still significant improvements to be found and where we expect to see a lot of progress in the next few years. Triple junction cells are currently the preferred technology and efficiencies have increased significantly in recent years as discussed above. However, in the longer term, we expect to see further innovation. For example, Circadian has been working with researchers at Radboud University working on removing triple junction cells from their substrates, which can be reused, leading to substantial cost savings. This collaboration has resulted in the formation of a joint venture, tf2 devices6).

This Epitaxial Lift-Off (ELO) technique will help create flexible, high-efficiency and lower-cost triple junction cells. This is a key trend for the CPV industry and we expect to see a growing interest in development of thinner cells that will be cheaper than current generation cells, and, moreover, offer better thermal properties7).

Driving CPV Forward: Technical and Commercial Requirements

These four areas are the critical developments we expect to see over the next few years. Allied to these improvements in individual elements, CPV module developers must also take into account the need for as much of the module assembly to be as automated as possible--as we have made clear, CPV should not simply be seen as a technical or academic exercise with a focus purely on developing the most efficient cell, or indeed any individual element. CPV is not as easy as piecing together all of the ‘best’ parts. It is important to adopt a holistic module and system design that not only takes into account performance but also acknowledges the future need of high volume manufacturing.

Certainly the crossover between technical and commercial gain is something that must be taken very seriously. As important as driving up fundamental efficiencies has to be the readiness for a highly commercialized and commoditized industry in the future. Although it is extremely difficult to forecast, we anticipate that there could be 20 GW of CPV installations globally, for both on and off-grid applications by 2020. For that to be a reality, and to enable even more growth, commercial realities must be considered now. 

Geoffrey Duggan is Director of Research & Development for Circadian Solar (www.       circadiansolar.com), where a part of his role includes management of the low-cost, high-efficiency CPV cell R&D program with Radboud University, Nijmegen.

He graduated in Physics from the University of Leeds in 1973 and remained there to complete his Ph.D. in theoretical solid-state physics. Duggan began his working career at Philips Research Labs in the U.K. During this period, he co-invented the III-V, quantum well solar cell’s technological approach now being exploited by one of the new generation of solar cell suppliers.

REFERENCES

1) http://www.circadiansolar.com/news/news_16-12-2010.html

2) http://phx.corporate-ir.net/phoenix.zhtml?c=76421&p=irol-newsArticle&ID =1479650&highlight=

3) http://www.soliantenergy.com/products/se1000x_spec_mechanical.php

4) For an overview see, for example, J.C. Minano et al, p81, Proceedings of 6th International Conference on Concentrating Photovoltaic Systems, 2010 (ed. A.W. Bett, F. Dimroth, R.D. McConnell and G. Sala) AIP Conference Proceedings

5) Wolfgang Wagner, et al, 3rd Workshop on Concentrating Photovoltaic, Optics and Power, 2010, Bremerhaven

6) www.tf2devices.com

7) S. Burroughs et al, p163, Proceedings of 6th International Conference on Concentrating Photovoltaic Systems, 2010 (ed. A.W. Bett, F. Dimroth, R.D. McConnell and G. Sala) AIP Conference Proceedings