How to make efficient solar cells from impure materials

Most solar cells are made from crystalline silicon. These solar cells are the most efficient available on the market. Crystalline silicon is expensive because it has to be very pure to make efficient solar cells from it. Unfortunately, nobody knows exactly hów pure and that is why manufacturers play it safe. In other words: they make and use more pure and hence, more expensive silicon than probably necessary. The results described in the thesis on the promotion research of Gianluca Coletti of the Energy research Centre of the Netherlands (ECN) might change this situation.

Optimization of costs
For the first time the effects of a whole range of (metal) impurities in the silicon starting material (‘feedstock’) on the efficiency of modern solar cells are described in detail. This way it becomes possible to purify the silicon in more targeted ways and to optimize the costs. Since ‘good is good enough’ unnecessary steps in the purification processes can thus be avoided.

Strong reduction of negative effects of impurities on performance
The essence of the new findings described in this thesis is that the negative effects on solar cell performance of impurities in silicon feedstock can be strongly reduced during and by some of the process steps that follow feedstock fabrication, in particular growing large crystal blocks or rods (which are subsequently cut into very thin wafers) and processing wafers into solar cells. Per type of impurity it can now be indicated how much of it can be tolerated in the feedstock if one is to make solar cells of a certain efficiency from it. This is a crucial step towards full specifications of ‘solar grade’ silicon.

Understanding relation between impurity and performance
The availability of silicon of (assumed) sufficient chemical purity has been an important limiting factor for rapid growth and price reduction during part of the past decade, and may become one again in the future. There is a strong need for low-cost and low-investment-cost technologies for the production of silicon for PV applications. However, such low-cost production technologies will most likely compromise the purity of the resultant silicon. The aim of the studies presented in this thesis is to understand the complex relation between the impurity content of the silicon starting material (the “feedstock”) and the resulting solar cell device performance, taking into account the effects of processing and device architecture.

CrystalClear project
The research work of the thesis was part of the European Integrated project CrystalClear: an extensive European Integrated Project on wafer-based silicon photovoltaics coordinated by ECN. This project was primarily dedicated to cost reduction of solar modules and contributed substantially to the development of PV sector in Europe. The CrystalClear consortium had 16 partners: 9 companies and 7 research institutes and university groups.

More information
The full text of the thesis ‘Impurities in silicon and their impact on the solar cell performance’ can be downloaded here.

Please see or visit for more information:

• PV17 Standards Technical Education Program (STEP) meeting
  8 September 2011, Hamburg http://www.semi.org/en/node/38626;
• Oral presentation “Sensitivity of Crystalline Silicon Solar Cells to Metal Impurities”: 6 September 2011, EPVSEC26, Hamburg – 2BO.6.5;
• PhD defence “Impurities in silicon and their impact on solar cell performance”: 19 September 2011, 10.30 am, Academiegebouw Utrecht University.

Contact
Gianluca Coletti
E-mail: coletti@remove-this-part-ecn.nl
Phone: + 31 224 564382

Please contact during EPVSEC26, 5 – 9 September:
+31 6 10852314 or visit the ECN booth B6/B26b

---