As a researcher at UNSW, I worked in a team to develop a Bragg reflector in GaAs photovoltaic technology for space applications. 

Product design | 6 Months | Client (Academic) UNSW

Humanity’s interest in the heavens has been persistent and ubiquitous. The drive to explore the unknown, discover new worlds, and push the boundaries of our scientific and technical knowledge is deeply rooted in the human condition. The desire to explore and challenge the boundaries of what we know and where we have been has provided universal benefits to society for centuries. Human space exploration is the current macro-frontier to helping us address fundamental questions about the history of our solar system and our place in the Universe. The continual development of technology to address the risks and difficulties of space exploration produces collateral benefits in many areas of our society from the redirection of those technologies for social gain. This relentless chase inspires the human spirit and accepting its challenges is necessary to see those pursuits realised.

Figure 1: impression shows how the solar wind shapes the magnetospheres of Venus (top), Earth (middle) and Mars (bottom). Unlike Venus and Mars, Earth has an internal magnetic field which makes its magnetosphere bigger. The lines coming out of the Sun symbolise the outward propagation of the solar wind. The planet’s distances are not shown to scale. 

Outside the protective cocoon of the Earth’s atmosphere and magnetic field is a universe full of radiation. Particles trapped in the Earth’s magnetic field; particles shot into space during solar flares emitted from the sun and galactic cosmic rays (high-energy protons and heavy ions from outside our solar system), make space a hostile environment. Radiation levels differ greatly from one space environment to another. In consequence, solar-powered spacecraft present many challenges and even the most efficient photovoltaic technologies have not yet been optimised for such places. Time delay and large distances make servicing such power systems virtually impossible. A highly reliable and efficient power supply is absolutely essential to ensure successful voyages into space. Amongst the highest performers in photovoltaics, GaAs solar cells are by far the most reliable, resilient to radiation and power efficient (more about this in the literature review). 

Furthering the destructive environment of space, as satellites equipped with photovoltaic devices leave the atmosphere, these particles are able to penetrate into the photovoltaic cells. GaAs solar cells are a zinc blende structure composed of two interpenetrating FCC sub-lattices which is similar to silicon in the fact that it also has face centred lattices but on one of those layers, we have arsenic sitting on one sub lattices and the gallium’s sitting on the other sub lattice. This combination, compared with other photovoltaics like silicon or perovskite, are relatively resistant to these radiation attacks although not completely. This can be tested by measuring the power and quantum efficiency of the solar cells before and after radiation damage. The efficiency drops over time for solar cells. Generally we see an 11% degradation over the lifetime of a solar panel with high energy electrons in geostationary orbits.

A number of methods improve radiation hardness (radiation resistance). One such method is through the implementation of Multijunction cells. Thinning a solar cell has been demonstrated to increase radiation tolerance. Light trapping is another method to increase the amount of power absorbed by the solar cells and is usually achieved by changing the angle at which light travels in the solar cell by having it be incident on an angled surface. Light trapping is usually achieved by changing the angle at which light travels in the solar cell by having it be incident on an angled surface. A Lambertian back reflector is a special type of rear reflector which randomises the direction of the reflected light. High reflection off the rear cell surface reduces absorption in the rear cell contacts or transmission from the rear, allowing the light to bounce back into the cell for possible absorption. Randomising the direction of light allows much of the reflected light to be totally internally reflected. Bragg reflectors can be used to control how much light leaves and how much light stays inside the solar cell. Specific wavelengths can be reflected by Bragg reflectors, which then reflect them back into the solar cell before they can escape. The effects of Bragg reflectors on the quantum efficiency of GaAs solar cells have received less attention in the literature, and there has been less work done to simulate these models using simulation programs such as Solcore.

Figure 2: Image on the left showing the reflectance inside the solar cell and the propagation of light once it has reached a specific location in the depth of the solar cell. Image on the right shows the shift in reflectance R as the wavelength changes. Specific wavelengths can be targeted or removed depending on the purpose of the solar cell

The compound solar cells represented by GaAs have advantages such as high-efficiency potential, possibility of thin-films, good temperature coefficient and radiation-resistance and potential of multi-junction application when compared with their crystalline Si solar cell relatives. The semiconductor can be combined with a number of other elements, unlike silicon which is much less versatile, offering scientists a wider pallet of experimental material.  These details become relevant when we begin to explore how space environments affect solar cells in different ways. The radiation begins knocking the atoms off their positions within the solar cell and the PV device breaks down. The GaAs blend of the material makes it more impervious to these attacks and hence more suitable for space applications. More on this in the section Weather conditions in orbit. As well as radiation, GaAs is also a better performing solar panel type when it comes to efficiency. With a more efficient solar panel, less of them are needed to power spacecraft and with smaller solar panels and hence a cheaper spacecraft to develop. The main driver for the use of GaAs in space applications however, is its reliability and less about cost, an absolute for voyages that are often one way and unaccompanied possibility forever. Very rarely are solar panels able to have access to maintenance, although not always the case which we have found for the international space station which has had the chance to have several maintenance check ins  during its life.

Building on industry work completed by a phd candidate (Vasilev, 2021), a simulated demonstration using python and solcore was able to be produced. This simulation forms the grounding for future optimisation.

Figure 3: Showing four different refractive indexes: Air, Si02, TiO2 and Gallium oxide. Clearly the higher index per wavelength belongs to Si02, followed by Gallium oxide, which is then followed by Si02 and then lastly air.

From these base level reflectance material data shown in figure 16, three different solar cells were constructed. One solar cell with a simple GaAs solar cell, the other with 3 layer semi-conductive solar cells and the other with 5 layer semi-conductive solar cells. These solar cells were tested across a spectrum of simulated light waves and plotted in the figure below.

Figure 4: The shift in slope between a single Au-thin oxide-n-GaAs solar cell, a three layer semi conductive Bragg reflector solar cell and a 5 layer semiconductive bragg reflector solar cell.

A five layer semiconductor photovoltaic solar cell was able to achieve significantly higher reflectance values and a sharper reflectance area than that using 3 semiconductor layer photovoltaic cells or simple single GaAs solar cells. Theoretically, this increase in slope could be optimised by increasing the number of layers in the multi-junction GaAs solar cell to an infinitesimally sharp area.

Figure 5: comparison of 3 layer Bragg reflector versus a 5 layer Bragg reflector. An increase in approximately 0.8 R was able to be achieved.

The results showed a sharper increase in reflectance per wavelength as the number of layers increased from 1, 3 and 5 in the simulated data. Theoretically the increase in the slope will approach an optimal value per layer added. Finding this optimal layer number for reflectance output will form the base for further work. Changing and modifying the materials used for the simulation should also be a part of further work for this research project.

Aim: To obtain baseline non-irradiated, non Bragg introduced GaAs solar cell efficiency data. 

Hypothesis: The experiment will produce data that is similar or within a measurable uncertainty, to that which is already recorded for GaAs solar cells. 

Methodology: 

Methodology is available upon request. All steps involved are part of the UNSW standard equipment workflow for the Light IV machine.

Figure 6: Basic structure of a simple IV tester. The current and voltage are measured separately to overcome contact resistance problems (National Renewable Energy Laboratory, 2019).

Figure 7: Image on the left showing the first cell tested using the light IV machine. Image on the right shows the experimenter preparing the solar cell for testing.

Of the 100 solar cells tested, the best performing 25 power output was plotted in the following figure. The differences in power produced varied.

Figure 9: 25 highest performing solar cell power output.

The results showed a wide discrepancy between the known ideality factor for GaAs solar cells. Due to the age of the GaAs solar cells (10 years), the veracity of results is outside the allowable tolerance for power and ideality factor. Further radiation could make further power or ideality factor data gathering impossible to collect. Due to time restrictions and the shutdowns from Covid-19 we were unable to complete the experimentation using EQE and probe machines and thus that information has been omitted from this report. From this data, as a team we determine that most of the solar cells do not resemble what we understand the power and ideality factor should be and hence they will not be used to conduct further radiation testing with. New solar cells should be ordered as part of further work and a retrial of laboratory testing should be completed. Simulated data should be used to continue the study in the meantime.

Laboratory time was very difficult to obtain due to the 2020-2021 Covid lock down restrictions. Because of this our team was unable to finish testing the GaAs solar cells as part of our experiment and hence the radiation testing could not be conducted either. Future attempts to finish the initial and post-radiated experiments outlined in the experimentation section of this report will be conducted at a further date. 

We need more laboratory time to finish the base line experimentation, manufacture the solar cells with the Bragg reflector included and order new base-line solar cells to be used as a control for all experiments. We still need to run the radiation testing as well, so there is a lot to do on this project.  

In conclusion, this research project was able to achieve a basic Bragg reflector design using Solcore, python and simulated its reflectance data in a multi-junction semiconductor GaAs solar cell. The reflectance showed an increase in overall reflectance for the given wavelengths. Further work in optimising the results as well as laboratory testing should be completed.

Pretty impressive right? We should talk about it!