CIGS-based Solar Cells for the Next MillenniumHans-Werner Schock1and Rommel Noufi2*1Universitat Stuttgart, Institut fur Physikalische Elektronik, Pfa?enwaldring 47, D-70569 Stuttgart, Germany2National Renewable Energy Laboratory, 1617 Cole Blvd., Golden, CO 80401, USAThin-film photovoltaic (PV) solar cells based on Cu(In,Ga)Se2(CIGS) have twokey distinctive features: highest performance of any true thin-film solar cell (18.8% e?cient) and leading performance on the module level. There is no evidence of limits to further improvement of the e?ciency. Device stability is not curtailed by intrinsic material properties. The obstacles to large-scale production and commercialization of Cu(In,Ga)Se2-based modules are the complexity of the material and the manu- facturing processes. Published in 2000 by John Wiley however, we hope that here we will capturesignificant topics to stimulate discussions and raise questions.FUNDAMENTAL ASPECTS OF THE MATERIALThe many elements in CIGS solar cells can form a great variety of compounds during cell processing; therefore, the CIGS system is very complicated. On the other hand, it is very tolerant to defects and impurities because the chemistry, as well as the structure, can adjust itself in many possible ways.1,2 Fortunately, this adjustment generally occurs in a way that is favorable to photovoltaic properties. An amazingly high level of performance of the solarcells in the laboratoryand in pilot production of modules has been achieved on an intuitive basis and byempirical optimization of the deposition process and devicePublished in 2000 by John Wiley contract/grant number: 0328059D-F Contract/grant sponsor: DDE; contract/grant number: DE-AC36-98-G010337 Foxit Reader - (C) Foxit Software Company,2005-2006 structure. The questions now are how far we will get with this development, how we can get closer to the theoretical limit, and whether there is an intrinsic mechanism that could limit the performance of solar cells fabricated with this material. An important example of the empirical development of the material isthe deliberate incorporation of sodium into CIGS absorber films. Known as a detrimental impurity insemiconductor technology, sodium has a positive influence on the properties of CIGS films. The challengeis to sort out how to influence material properties to specifically tailor properties such as carrier concentration, di?usion length, and lifetime and to apply optimization procedures that are known from single-crystal-based solar cells. We have to ask whether we have the appropriate tools for the design and manufacture of very high-e?ciency devices. The knowledge of basic material properties has increased considerably in recent years. By combining chemistry/defect chemistry issues, solid-state physics, semi-conductor physics, and material science, a new view of this solar cell material has emerged. The specificissues to be addressed are the mechanisms of film growth, the control of defects, role of grain boundaries, properties of the surface, and junction formation.Film growthThe most striking feature of CIGS is the tolerance of the parameters in manufacturing processes. It isimportant to note that films for very high-e?ciency devices can be realized along di?erent routes.However, some basic rules must be followed to obtain device-quality films. Recipes for the growth of films generally include a Cu-rich growth step that leads to the formation of large grains.3However, the simple coevaporation of the elements or the adjusting of the composition by di?usion Cu into InxSeyfilms in thepresence of Se vapor yield films that could lead to devices with e?ciencies exceeding 15%. The nucleationphase of the films is important and should have controlled conditions so as not to create large, isolatedgrains and to obtain smooth, dense films.One prerequisite for high-quality films is the presence of Na during film growth. The benefit of Na incorporation has not yet been fully explained in terms of simple models. In view of the amount of Na (0.1 at.%) necessary for optimum film fabrication, arguments that consider the e?ect on the film growthslightly outweigh those dealing with the incorporation of Na into the completed film. The explanations ofthe beneficial e?ect of Na on the properties of CIGS films and solar cell performance are numerous, andthe Na incorporation most likely has a variety of consequences. During film growth, the incorporation ofFigure 1. The best one-of-a-kind laboratory cell e?ciencies for thin film (standard conditions)Published in 2000 by John Wiley on the other hand, grain boundaries do not have asignificant influence on the electronic properties of the CIGS layers. Both of these findings are somewhat contradictory, because the high barrier at the surface, which is the prerequisite for an e?cient buried p/n junction in the absorber layer, would considerably enhance grainboundary recombination.Surface/interface chemistry and