European Space Agency

Aging Effects Modelling for the New Generation of GaAs/Ge Cells


The objectives of the current activity are:

  1. To acquire sufficient test data to allow appropriately accurate prediction of 'relative damage coefficients' (as input to the 'equivalent damage' method) for the following two types of triple-junction solar cells, developed by Azur:
    • 3G28% class new generation solar cells, fully European,
    • GAGET2 solar cells.
  2. To acquire complementary test data for 'component' top, middle and bottom cells for the two triple-junction solar cells.
  3. To construct a test database of the above measurements against which future models can be tested and validated by the space community.
  4. To analyse and compare performance prediction using data for component cells with prediction using data for complete triple-junction cells.
  5. To define a general process for the validation of future models, as an alternative to the 'equivalent damage' method and to apply this process to new models such as the 'displacement damage dose' model.


The construction of a coherent database of solar cell main characteristics after irradiation is of major importance since it will improve the accuracy of the end-of-life prediction of satellite power budget.

The testing of the relevance of the 'equivalent damage' method for triple junctions is a key issue as this is the reference method, widely used throughout the solar cell community and in space industry.

The testing of new methods, such as the 'displacement damage dose' or Bourgoin?s ones, is a key issue as important benefits are expected from using in the future more theoretical and less time and money-consuming methods.


The design of a satellite solar sub-system requires an accurate evaluation of the solar cells performances. Since the end of life performances of the cells are mainly driven by their sensitivity to the radiations environment, predictions of cells performances under a typical environment has to be accurately assessed during the design. Moreover, the cost of a solar generator reaches 30% to 40% of the total cost of a satellite.

In this context, the realization of a large-scale tests campaign to acquire a complete set of data is of great importance. This set of data should allow deducing the relative damage coefficients used in 'equivalent damage' method for these cells for modelling more precisely their in-orbit degradation and reducing corresponding margins.

Moreover, this should also allow analysing the relevance of the predictions performed on triple-junctions cells with the 'equivalent damage' methodology compared with new emerging approaches. Indeed, the 'equivalent damage' method necessitates the acquisition of this large amount of test data that is an expensive and time-consuming process. As a result, if the development of new methods could provide a similar or improved level of accuracy using less test data, significant benefits would result.


The study will develop in two phases.

The pre-test phase is devoted to the realization of the pre-tests and to the preparation of the test and analysis plan. The objective of this phase is to validate the experimental means (electrical measurement facilities and irradiation laboratories) and the degradation behaviours under irradiation of triple-junction and component cells. It includes the irradiation of GAGET 2 and 3G28 cells with electrons and both high and low energy protons at one selected fluence and the measurements of the irradiated cells performances (current-voltage and spectral response measurements).

The second phase is the main phase of the study and is devoted to the realization of the tests and to the exploitation of the data.

The objective of the test campaign is to perform all the tests necessary to acquire the data. They are performed in five successive steps of both irradiation and electrical measurements. In addition, to check the consistency of the results, a cross calibration of the irradiated GAGET 2 is planed after the third irradiation. Besides complementary tests are planed to evaluate the impact of an irradiation under incidence, and to measure the temperature coefficients.

The objective of the data analysis is to exploit the test results. It consists of three main activities: the compilation of tests measurements under a Microsoft access database, the extraction of the 'relative damage coefficients' and the analysis of the relevance of the current and emerging methods to evaluate the solar cells performances.

Electrical measurements are performed at Fraunhofer-Institut für Solare Energie Systeme, INTA-Spasolab, and Astrium GmbH.

GESEC R&D is creating an engineering software model in a format useful for estimating the radiation response of a triple junction solar cell, and will validate it against the data acquired during the study.


The pre-test phase consists in testing GAGET2 + 3G28:

  • 300 keV protons (5x1010 /cm2) at Philips Research Laboratory (NL) -GAGET2-,
  • 1 MeV electrons (5x1014 /cm2) at Delft University (NL) -GAGET2 + 3G28-,
  • 6.5 MeV protons (2x1011 /cm2) at Isotron (UK) -3G28-.

The second phase includes five successive increasing fluences on both GAGET2 + 3G28 at following energies:

  • 30 keV protons at Philips Research Laboratory,
  • 100 keV protons at Philips Research Laboratory,
  • 300 keV protons at Philips Research Laboratory,
  • 750 keV protons at Philips Research Laboratory,
  • 6.5 MeV protons at Isotron,
  • 0.5 MeV electrons at Delft University,
  • 1 MeV electrons at Delft University,
  • 3 MeV electrons at Delft University.

Current status

The pre-test phase is completed.

No problem was detected during irradiations, dose levels being controlled both by in-house dosimetry equipment and control samples. The foreseen schedule, based on successive periods of three weeks for irradiations and three weeks for electrical measurements seems realistic.

Before and after irradiation, standard I-V curves were recorded for all triple-junction cells and their degradation under both electron and proton irradiation is in line with previous results; a good fit is also obtained by using 'displacement damage dose' method. Concerning 'component' cells, before irradiation, standard I-V curves were also recorded, but after proton or electron irradiation, most of the I-V curves were degraded. The problem seems to be identified and should be corrected for second phase solar cells.

The second phase should start at the end of August with the purchase of solar cells, followed by their initial characterizations at Fraunhofer-Institut für Solare Energie Systeme, INTA-Spasolab, and Astrium GmbH. First round of irradiations is foreseen in the autumn.

Status date

Wednesday, February 6, 2013 - 10:53