Search for content and authors
 

Applications of group III-nitride alloys for multijunction solar cells

Wladek Walukiewicz 

Lawrence Berkeley National Laboratory (LBNL), 1 Cyclotron Road, Berkeley, CA 94720, United States

Abstract

The direct band gaps of the group III-nitride alloys can be continuously tuned in a very large energy range from 0.65 eV in InN to 6.2 eV in AlN.  In addition, changing the composition from InN to AlN allows for varying the electron affinity of the alloys from 5.8 to 2.1 eV.  These unique properties offer a possibility of using group III-nitride alloys for high efficiency tandem cells.  A successful implementation of group III-nitrides for multijunction PV devices requires p- and n-type doping in the whole alloy composition range. As grown, undoped InGaN films are always n-type.  Reproducible p-type doping with Mg acceptors in the whole composition range has been reported by several groups.  Any design of the multijunction cells based on InGaN or InAlN alloys must consider the n-type surface inversion layer that is experimentally found for p- InxGa1-xN with x > 0.4.  A number of possible cell designs will be presented, including structures with graded band gaps. In addition results on a hybrid InGaN/Si double junction (2J) cell will be discussed in detail. Our experimental work has established that for x=0.5 the conduction band of InxGa1-xN aligns with the valence band of Si. Thus, a low resistance ohmic contact is expected at a junction between n-type In0.5Ga0.5N and p-Si, eliminating the requirement of heavily doped layers needed for the tunnel junctions in the current multijunction cells. This greatly simplifies the design of a 2J hybrid cell. Efficiencies exceeding 30% are predicted for an optimized p-InGaN/n-InGaN/p-Si/n-Si cell based on realistic absorption and transport parameters for InGaN.  It should be noted that similar considerations apply to the InyAl1-yN/Si system with y=0.7. Latest results on the performance of the InGaN based tandem cells will be presented and the remaining obstacles in achieving higher power conversion efficiencies will be discussed.

This work was supported by the US DOE and Rose Street Labs Energy, Phoenix, Arizona
 

Legal notice
  • Legal notice:
 

Related papers

Presentation: Keynote lecture at E-MRS Fall Meeting 2009, Symposium A, by Wladek Walukiewicz
See On-line Journal of E-MRS Fall Meeting 2009

Submitted: 2009-05-25 10:36
Revised:   2009-06-07 00:48