Champion concentrator cell efficiencies have surpassed 40% and now many are asking whether the efficiencies will surpass 50%. Theoretical efficiencies of >60% are described for many approaches, but there is often confusion about ''the'' theoretical efficiency for a specific structure. The detailed balance approach to calculating theoretical efficiency gives an upper bound that can be independent of material parameters and device design. Other models predict efficiencies that are closer to those that have been achieved. Changing reference spectra and the choice of concentration further complicate comparison of theoretical efficiencies. This paper provides a side-by-side comparison of theoretical efficiencies of multijunction solar cells calculated with the detailed balance approach and a common one-dimensional-transport model for different spectral and irradiance conditions. Also, historical experimental champion efficiencies are compared with the theoretical efficiencies. Copyright # 2008 John Wiley & Sons, Ltd.
INTRODUCTIONAs the world's interest in renewable energy has increased, the investment in solar has also grown. New approaches to achieving high efficiencies have been proposed by numerous groups. Sometimes these new approaches are described as having a theoretical efficiency >60%. Will these new cells be superior to today's cells, which have now surpassed 40%? 1 Predicting the answer to this question requires two steps: (1) when comparing the theoretical efficiencies of existing and new approaches, the same methodology must be used for both, and (2) an assessment should be made of how close the experimental efficiency may, ultimately, approach the theoretical efficiency for each case. The first step is obvious, but is often neglected, resulting in comparison of numbers that were derived in different ways. When making such comparisons, the choice of the methodology is much less important than the consistency of the methodology. The second step is aided by a review of the efficiencies that have been achieved so far.Over the years, many studies have presented theoretical multi-junction solar cell efficiencies as a function of the materials' band gaps. [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] In general, no two studies use exactly the same assumptions, resulting in a range of theoretical efficiencies. The efficiencies are especially dependent on the conditions (spectrum, concentration, and temperature). Concentrating the sunlight to the maximum value ($46 000 suns) results in theoretical efficiencies 10-20% (absolute) higher than the corresponding one-sun efficiencies. 4 Changing the spectrum has a smaller effect (typically 1-4% absolute). 17 The choice of the theoretical model, or assumptions used in the model, also affects the calculated efficiency. Marti et al. compared two of the most common models. 18 They defined a ''standard'' model using Shockley's diode equation, modeling a one-dimensional p-n junction with appropriate values for surface recombination, layer thicknesses, etc. They co...