The aim of this work package is to develop the metrology required to measure the photovoltaic properties of III-V materials used in triple-junction (3J) solar cells at the single layer level. Current industry standards are lattice-matched triple-junction cells GaInP/GaInAs/Ge with conversion efficiencies of around 41 %. Industrial and academic roadmaps have identified four directions to improve the efficiency of the 3J solar cells:

• Optimise the band-gap combination of the three individual junctions to meet the ideal efficiency of the conversion of the energy spectrum. This can be achieved either by a band-gap engineering approach of existing material through a microstructure change or by using new alloys such as GaInNAs (see WP3);
• Optimise the growth process to reduce defects in the layers. This can be achieved by using improved lattice-matching of the sub-cell, metamorphic buffers or through high quality, mechanical bonded stacks;
• Increase the overall photocurrent production as, due to the nature of its series connection, the output current of the multi-junction solar cell is limited by the lowest of the currents produced by any individual junction. This can be done by a fine adjustment of the thickness of the layers, which affect the absorption coefficient, the diffusion length and the degree of transparency;
• Optimise the tunnel-junction interconnects in terms of band-offsets and low resistance.

The high number of layers in III-V multi-junction solar cell structures makes a purely experimental optimisation very expensive and protracted. Optimisation is currently performed mainly by predictive numerical modelling, which is extremely challenging for these sophisticated structures due to the complex electrical and optical interactions between the different layers and the high number of material parameters and physical phenomena that need to be considered. For most new materials used in these complex structures, there is also a significant lack of material properties data, which limits the modelling capabilities so that discrepancies from 20 % to 30 % are currently observed between measured and predicted efficiencies.

In order to validate new experimental cell structures and to better understand the parameter spaces of the devices, accurate and spatially resolved metrology for III-V heterostructures is required.

The heat absorbed in the cell and dissipated through the substrate by Joule effect or other mechanisms and this can lead to thermal runaway with dramatic consequence on solar cell operation. In addition thermal transport in very thin film is still controversial as the transport mechanisms are not fully understood and the models are not reliable for very thin films.

WP1 will develop accurate measurements of the physical parameters that are of particular importance to understand the electrical transport phenomena in these heterostructures: band-gap, work function, dopant distribution, photocurrent, carrier density, diffusion length, doping dependent minority carrier lifetime, absorption coefficients and series-resistances. Composition, thickness and structural properties of bulk and nanoscale III-V material will also be addressed to highlight the effect of defects concentration, microstructure and interfaces on recombination mechanisms of charge carriers.

WP1 will also improve understanding of tunnel-junction interconnects through the design, fabrication and characterisation of tunnel junctions made with new materials lattice-matched or nearly lattice matched to GaAs. A numerical model will be developed to describe electronic transport in these structures.

This WP will provide accurate material properties data of III-V materials used for MJSC in order to reduce the discrepancies between models and experiment from the current 20 %-30 % to at least 10 %. This will help designers to optimise the choice of materials for each sub-cell in terms of band-gaps, thicknesses and optoelectronic properties in order to push the multijunction solar cells towards their theoretical maximum efficiency.

LNE Near field scanning microwave microscope

---

NPL Ultra-High Vacuum Atomic Force Microscope

---

METAS Microwave measurement head

---

NPL Variable temperature, environmentally controlled probe station

---

CEA-LETI cathodoluminescence system.

The aim of this work package is to establish traceable and reliable laboratory calibration methods for III-V multi-junction solar cells around Europe. Creating standards for the determination of the devices' short circuit current as well as power conversion efficiency is crucial since stakeholders need such certifications to enable new MJSC products to enter the market entrance of. So far, only MJSC with a maximum of three active layers (3J) can be calibrated in NMIs/DIs across Europe and the devices are measured in solar simulators and traced to reference cells which are calibrated during balloon flights directly to the sun's spectrum at Air Mass zero (AM 0). Under such realistic and consequently non-controllable conditions a lower calibration reproducibility and a higher uncertainty is likely compared to primary calibrated reference samples investigated e.g. by differential spectral responsivity (DSR). In addition to reducing the uncertainty, there is a need to extend the calibration methods to enable measurements on MJSC with more than three junctions to meet industry's needs in this area.

To achieve these goals, the development of novel calibration setups, such as DSR setups with additional monochromatic bias towers as well as optimised sun simulators is necessary. State of the art solar cell measurement setups at NMIs/DIs in Europe allow either the calibration of component solar cells with low uncertainties (0.5 %) using DSR at PTB or the direct calibration of 3J devices with a multi-source solar simulator at INTA. INTA will use its experience on 3J device calibration to determine the device characteristics of solar cells, which will then be used later during the JRP by the JRP-Partners to test their setups. PTB will improve their capabilities to allow DSR calibrations on 3J devices avoiding the disadvantages of balloon calibrations while TUBITAK will extend its calibration capabilities on MJSC by DSR measurements. MIKES will extend its goniometric reflectance measurement system to measure MJSC reflectance up to 2 µm. This will allow a unique network for MJSC calibrations to be created across Europe. Furthermore, for the calibration of MJSC with more than three junctions, a LED-based solar simulator will be developed at PTB.

In Task 2.1 AZUR SPACE will characterise and deliver component cells, which will subsequently be studied together with cells from industrial collaborators by INTA and PTB. In Task 2.2, broadband tuneable light sources for DSR and sun simulators will be developed by TUBITAK and PTB. Task 2.3 will address the implementation and testing of the equipment developed in Task 2.2 as well as evaluation of the setups of all JRP Partners. Two good practice guides will be developed in Task 2.4, both on procedures for MJSC calibration including information about the capabilities developed in the JRP, but one targeted at the JRP Partners and the other at stakeholders. To transfer the JRP's results to the European industry reports will be provided to stakeholders and standardisation institutions to help enable an internationally accepted standard for MJSC calibrations.

---

MIKES goniometric reflectance measurement system

---

LNE LED-based sun simulator

The aim of this WP is to develop traceable measurement and characterisation capabilities needed at the R&D laboratory level in order to highlight and understand the transport mechanisms, the influence of quantum confinement, dopant location and activation, and interface effects in several innovative structures, which are being proposed by the PV community for the development of next generation of solar cells. Indeed, several innovative approaches are being studied by several PV research teams to further enhance the effective efficiency of III-V MJSC leading to a considerable breakthrough for their commercial use in terrestrial applications.

Three main research directions will be addressed in WP3:

• III-V MJSC with more than three junctions (Task 3.1): as MJSC efficiency increases with the increasing number of junctions (a theoretical efficiency of 86.8 % is achievable with infinite number of band-gaps), 4, 5 or 6 multi-junction solar cells are the first approach to further increase the efficiency compared to that of standard 3J cells. As the number of junctions increases the solar spectrum is partitioned into narrower wavelength ranges allowing a better conversion of solar energy. However, as the sensitivity to spectral variations increases with the number of sub-cells, it is important to investigate the possible gain in energy production before a decision is made on the number of sub-cells and the material combination to use. Two approaches will be investigated in this WP: i) the use of a dilute nitride cell, GaInNAs, with a ~ 1 eV band-gap ; ii) the use of a new direct wafer bonding technique to produce multi-junction stacks of high structural quality. Furthermore, the advantage of using III-V materials will be to hybridise a solar cell with thermoelectric structure to improve efficiency. However, to use these materials for harvesting additional energy, other properties such as thermoelectric properties have to be exploited and hence need to be well characterised.
• III-V MJSC on Si (Task 3.2): the main objective of this approach is to reduce the cost of III-V material based solar cells making them highly cost effective by replacing classical Ge or GaAs wafers with Si as the bottom sub-cell for instance in GaInP(1.9 eV)/GaInAs(1.4 eV)/Si(1.1 eV). The main issues concern the effect of III-V/Si interface on the quality of the layers and their electrical transport properties.
• Quantum dots (QDs) based solar cell (Task 3.3): the efficiency of MJSC devices could be improved by incorporating nanoscale semiconductor quantum dots. Quantum dots are size-dependent electronic structures and therefore their optoelectronic properties are tuneable. Hence, the use of nanostructures, such as quantum wells, quantum wires, superlattices, nanorods, or nanotubes, offers the potential for high photovoltaic efficiency by tailoring the properties of existing materials, and for reducing the cost using self-assembling of nanostructures. Measurement techniques and methodologies developed in WP1 will be used to characterise structural and electronic transport properties of these devices in order to establish the relationship between the size of the quantum dots and their band-gap.

---

LAAS Molecular Beam Epitaxy system(MBE)

---

NPL Thermoelectric efficiency facility

The aim of this work package is to ensure a broad impact of this JRP within the European community of stakeholders. To reach this goal, a Stakeholder Committee will be formed to progressively provide feedback on the relevance of the different metrology techniques and methodologies developed during the lifetime of the JRP.

In order to ensure dissemination to industry, academia and standards bodies, the JRP will organise industrial workshops, attend technical and scientific conferences and publish results in international peer-reviewed scientific journals and appropriate trade press.

A particular focus of the JRP will be to deliver good best practice and pre-standards to the community of stakeholders and standardisation bodies. The JRP-Consortium already has strong links with advisory and standardisation committees. These include: standards Working Groups for Photovoltaic (IEC/TC82 and ISO/TC180), and for nanoelectronics (IEC/TC113 and ISO/229), The European Commission's PV Joint Research Centre (JRC), CATRENE (Cluster for Application and Technology Research in Europe on Nanoelectrics) support group; Co-Nanomet (European nanometrology group).

---