Characterization of Photovoltaic Absorbers for High Throughput Processing

  • PDF / 1,190,853 Bytes
  • 6 Pages / 612 x 792 pts (letter) Page_size
  • 93 Downloads / 187 Views

DOWNLOAD

REPORT


Characterization of Photovoltaic Absorbers for High Throughput Processing Jean-François Guillemoles§*, Amaury Delamarre§,*, Gilbert El-Hajje§, P. Rale§, D. Ory§, Myriam Paire§, Daniel Lincot§, Kentaroh Watanabe*, Masakazu Sugiyama* and Laurent Lombez§ §

IRDEP, Institute of R&D on Photovoltaic Energy, UMR 7174, CNRS-EDF-Chimie ParisTech 6 Quai Watier-BP 49, 78401 Chatou cedex, France * NextPV, RCAST/CNRS, Tokyo University, 4-6-1 Komaba, Meguro-ku Tokyo 153-8904, Japan. Email: [email protected];

ABSTRACT We analyze photoluminescence (PL) and electroluminescence (EL) using a hyperspectral imager that records spectrally resolved luminescence images of solar cell absorbers. The system is calibrated to yield the luminescence flux in absolute values. This system enables to quantitatively image physical parameters such as the photovoltage with an uncertainty of less than 30mV. The wide field illumination, low power excitation and fast acquisition brings new insights compare to classical setups such as confocal microscope. Several types of absorbers have been analyzed. For instance, we can investigate spatial fluctuations of the Quasi Fermi Levels splitting in CIGS polycristalline absorbers and link those fluctuations to transport properties. The method is general to the point that third generation PV cells absorbers can also be evaluated. We illustrate the great potential of our setup by imaging quasi Fermi levels splitting in Intermediate Band Solar cells. Such techniques, directly evaluating the performance of photovoltaic absorbers and devices are needed for fast, high throughput investigations of combinatorial experiments such as the projects carried out for the material genomics programme.

INTRODUCTION Search for new functional materials are presently being carried out at an unprecedented scale within programs such as the “Materials Genome Initiative”. While materials with differing properties can be synthesized rather efficiently using robots or using non-uniform deposition, one of the bottlenecks is still the possibility to evaluate rapidly the potential of these materials. In the case of Photovoltaics, while local composition and structure can be quickly searched, the analysis of properties related to photovoltaic conversion remains tedious. A technique that would enable directly, and in a massively parallel fashion, the evaluation of the photovoltaic performance is obviously an objective with high stakes. It turns out that photoluminescence (PL) and electroluminescence (EL) spectra allow the determination of important optoelectronic parameters of solar cells, such as the open circuit voltage from PL [1]

or the external quantum efficiency (EQE) from EL [2]. To get an accurate measurement of those parameters and thus a correct understanding, experimental conditions have to be properly set. We propose here to use a hyperspectral imager that can record spectrally resolved images. The absolute calibration of the photons flux and the wide field illumination allows to image several physical parameter