The solar-powered family car Stella Vie
– a closer look at the solar cells

On June 21, the Solar Team Eindhoven launched its 5-person, solar-powered family car “Stella Vie”. The car, which has been developed by students at the Eindhoven University of Technology (TU/e), will take part in the cruiser class event of the World Solar Challenge in October this year. During this challenge, the Stella Vie will compete in a solar-powered car race from Darwin to Adelaide on a 3000 km long highway which crosses the center of Australia. Although a lot of attention has been devoted to the (very nice looking!) car and its construction team, little attention has been paid to the enabling technology and true driving force of this car: its solar cells. What kind of solar cells have been used? How much power can they deliver? And how does the TU/e contribute to the development of affordable solar energy? Questions that will be treated in this blog.

Figure 1. The Stella Vie, a 5-person solar-powered family car developed by students from the Eindhoven University of Technology (Solar Team Eindhoven). Photo: Bart van Overbeeke.

Solar cell technologies

Many different types of photovoltaic (PV) solar cells could be used to convert sunlight into electricity to power the car. This also becomes clear from Figure 2, which provides an overview of various solar cell concepts and their record energy conversion efficiencies.

Figure 2. Overview of the highest conversion efficiencies for various photovoltaic (PV) technologies. Courtesy of the National Renewable Energy Laboratory, Golden, CO, USA.

So which PV technology did the team choose? Did they just pick the most efficient solar cells?

While top efficiencies of 46.0% have been reached by advanced solar cell concepts (meaning that 46% of the energy of the incoming sunlight is converted into usable electricity), such very high efficiency solar cell concepts are mainly used in space applications, where cost is not a limiting factor. The world’s most widely-used solar cell technology is, rather, based on crystalline silicon (c-Si). c-Si solar cells are those typically visible on rooftops and which are used in large-scale photovoltaic power plants. Particularly the combination of low cost (<1$/Watt), and reasonably high conversion efficiencies for commercial c-Si cells in the range of 18-24% have made c-Si the dominant PV technology with a market share >90% (obviously the values for commercial solar cells are somewhat lower than the record, lab-scale efficiencies shown in Fig. 2). Such c-Si based cells are also used in the Stella Vie.

Besides their widespread availability, another motivation for the choice of c-Si solar cells is the regulations of the World Solar Challenge. The overall area of the cars which is allowed to be used for the collection of light for solar energy conversion has been regulated, and this area is scaled inversely with the record efficiency of the PV technology used. Thus, for the very advanced PV technologies which yield the highest efficiencies, the allowed area of solar cells is smaller (see Table 1). In this way, the total power which PV systems using different technologies can produce is kept approximately the same, and the race becomes more accessible, as not only the few teams who can afford the most expensive ultra-high efficiency solar cells stand a chance. Ironically, at the time the regulations were made, the power of the three included PV technologies (using the allowed areas) would have been exactly the same, at 1280 Watts. However, in the meantime the record efficiency of c-Si has already been broken ¹. In addition, the efficiency of commercially available c-Si solar cells is already relatively close to the reported record efficiency. The efficiency of Stella Vie’s cells is 24.3% (see Table 2), compared to a record efficiency of 26.6% for c-Si solar cells ¹. Thus, the actual power which is harvested by using c-Si can be higher than for the other technologies. Finally, it is worth mentioning that the specified area is the total area which can be used for solar energy collection. So, if lenses and mirrors are used to concentrate sunlight, the area they take up has to be accounted for. This is part of the reason why having concentrator solar cells is typically not viable for the solar cars.

Table 1. Allowed solar cell area in the cruiser class of the World Solar Challenge for different solar cell technologies, and their record efficiencies as taken from Fig. 2. The output power under standard test conditions (1000 Watts/m², 25 ^oC, spectrum ‘air mass 1.5g’) has been calculated.

*At the time the regulations were set-up (30 June 2016), the record conversion efficiency of c-Si was 25.6% ².

Table 2. Summary of the PV technology used in Stella Vie.

A close look at the solar roof:

Figure 3. The Solar Roof of the Stella Vie. One can distinguish the individual cells of the Sunpower brand  (model: E60). Photo: Bart van Overbeeke.

Let’s have a closer look at the solar roof of the Stella Vie. It becomes clear that the roof, just like commercial solar panels, is built up of individual cells. On the Stella roof, these cells appear to be very dark blue, nearly black. However, in most commercial PV modules, grey lines are typically observed on the cells (see e.g. Figure 4). These grey lines are metal wires which conduct electricity out of the solar cell. The metal wires at the front side connect to the negative contacts of the cell, whereas the metal at the rear connects to the positive ‘pole’ of the solar cells (similar to the situation in a battery). As metal is not transparent to sunlight, the metal grid shades about 5% of  the solar cell, reducing its energy production(note that the shaded area still counts as area for solar energy collection in the World Solar Challenge regulations). To prevent such shading losses, in the solar cells of Stella Vie the positive and negative contacts are all positioned at the rear side of the cell. Such cells are generally referred to as Interdigitated Back-Contact (IBC) cells. In Figure 4, a cross-section of a typical high-efficiency IBC solar cell is shown. IBC cells hold the world-record efficiency for c-Si solar cells, and this record efficiency has this year increased to 26.6% for a 180 cm² cell¹, which is very close to the predicted upper limit for c-Si solar cells of 29.4 %³. In comparison, for solar cells where the positive and negative contacts are placed on opposite sides, the record energy-conversion efficiency is currently 25.7% for a 4 cm² cell.⁴

Figure 4. Typical photovoltaic module comprising 60 (multi)crystalline silicon solar cells.

Although IBC solar cells thus have the advantage that shading by the metal grid is prevented, the patterning of both positive and negative contacts on the rear side typically makes manufacturing these cells more complicated and hence more expensive. So, even though they are the most efficient c-Si solar cells, they are typically not (yet) the most cost-effective ones. Therefore, on most roofs you can still see the grey lines on the cells (look carefully!). Nevertheless, IBC solar cells are commercially available.

Figure 5. Top left: Square multicrystalline (also referred to as polycrystalline) silicon solar cell. The silver metal lines conduct the electricity out of the cell. The multicrystalline silicon consists of distinct crystals which different orientations, which can be clearly identified in this photograph. n-Si is negative type silicon, which only conducts negative electrical charges and forms the negative contact of the solar cell. In a similar way, p-Si forms a positive contact. Top right: Single crystalline silicon solar cell as used in the Stella Vie. Bottom left: Schematic cross-section of the current industrial standard crystalline silicon solar cell termed Al-BSF (aluminum back surface field) comprising metal fingers at the front surface, a surface texture and anti-reflection coating. Bottom right: Schematic cross-section of a high-efficiency interdigitated back-contact (IBC) crystalline silicon solar cell.

In addition, it can be observed that the cells inside Stella Vie are not completely square (in the field, such cells are referred to as pseudo-square). Although this looks like a small detail, it actually hints at the origin of the c-Si base material used. The solar cells of Stella Vie are (like many other high-efficiency solar cells) made from one single crystal, referred to as monocrystalline. The single crystal is made by pulling and turning a seed crystal from a silicon melt. A long, round rod or ingot of one – nearly perfect – silicon crystal is sawn into 100−180 μm thin wafers (about the thickness of a sheet of paper) which are later processed further to form the solar cells. Other c-Si solar cells are made in a cheaper way, by pouring the silicon melt into a crucible. This results in a big square block of so-called multicrystalline silicon (also known as polycrystalline silicon), which later is sawn into small (perfectly) square wafers to make solar cells. In Fig. 5 you can clearly see that the square multicrystalline solar cell consist of multiple small crystals. At the boundaries between the different crystalline facets, however, electrical losses can occur due to imperfections in the material. For this reason monocrystalline solar cells generally can reach a higher efficiency. Even so, multicrystlalline Si can still reach a conversion efficiency of 21.9%, and this efficiency is also continuously improving (this record stems from February 2017).

A final point which becomes immediately clear from looking at the cells in Stella Vie is their very dark appearance. This dark blue, nearly black appearance is markedly different from the blue multicrystalline cell in Fig. 5. Obviously, a solar cell should absorb as much sunlight as possible to convert the light into electricity. If all sunlight is absorbed, an object appears black. So, transparent solar cells are probably not a very good idea, because then the sunlight is not absorbed. Nature does not do an excellent job of collecting sunlight either. Plants for instance reflect green light, which is why they appear green. However, plants are not evolved (or designed) to create electricity! In general, black solar cells are thus better than blue ones. Several ‘tricks’ are typically used to enhance the light absorption in c-Si solar cells. Most prominently, the solar cell surface is textured which means that the front surface consist of small pyramids which reduce reflection. Furthermore, an anti-reflection coating is also applied, as can be seen in the electron microscopy image of Fig. 6. Note that the thickness of this anti-reflection coating layer is typically 75 nanometers, which is ~1000 times thinner than a piece of paper.

Figure 6. Electron microscopy image of the top surface of a c-Si solar cell, which comprises a random pyramid texture and a 75-nanometer thick anti-reflection coating to enhance the light absorption.

In conclusion, the solar cells in Stella Vie are highly-efficient as they comprise interdigitated back-contacts, monocrystalline silicon and effective light trapping.

 Connection to solar cell research at the TU/e

High efficiency solar cells like those in the Stella Via are the product of decades of dedicated research by industry and academia. At the Plasma & Materials Processing (PMP) group at the TU/e led by prof. Erwin Kessels, research has been ongoing for about 20 years to improve the efficiency c-Si photovoltaics and to reduce its cost. This research has thus far led to several innovations in industry. One of these successes was the upscaling of the so-called Expanding Thermal Plasma method for the preparation (deposition) of anti-reflection coatings on crystalline silicon solar cells. After the development of this method in the lab, it was subsequently scaled up and commercialized for high volume manufacturing. This was done by the Dutch company OTB Engineering, which became later OTB Solar, then Roth & Rau B.V. and now Meyer-Burger B.V. The name of the high-throughput system was the DEPx (see Fig. 7) and over 100+ systems were sold. The DEPx is still sold today but now under the name FLEx XL. You can read more about it in another post. Interestingly, the silicon nitride anti-reflection coatings deposited by this method developed at the TU/e was applied by Sunpower in their interdigitated back contact (IBC) cells – the cells used on Stella Vie – over many years. The fact that Sunpower bought DEPx systems from OTB Solar in the mid 00’s for their production lines in the Philippines can be learned from several resources in the internet. Dick Swanson, the founder of Sunpower, has listed OTB Solar in his presentations whereas Ron Kok, the old owner of OTB Solar, has mentioned it in a public report as well. Well, one might wonder whether the cells on Stella Vie have still an anti-reflection coating deposited by the TU/e technology. We don’t know for sure, but as the site in the Philippines has been closed, it  is unlikely.

Figure 7. The DEPx system developed by OTB Solar in the early 00’s. The DEPx employs the Expanding Thermal Plasma method, which was invented and developed in the Plasma & Materials Processing (PMP) group of the TU/e. More than 100 systems have been sold for the deposition of anti-reflection coatings on crystalline silicon solar cells. In the meantime, the company has become part of Meyer-Burger and the system has been renamed as FLEx XL.

The research of the PMP group has led to further innovations in solar cell technology as well. One with an even greater impact than the silicon nitride anti-reflection coatings is the implementation of aluminium oxide (Al₂O₃) thin film “passivation layers” in the so-called PERC solar cells. This new industrial high-efficiency standard cell concept has an estimated production capacity of 37 GW in 2017 (estimated total capacity is 100 GW). But let’s come back to these Al₂O₃ layers⁵ and the method of atomic layer deposition (which is used to prepare them) in a future blog.

What’s next on the horizon for c-Si solar cells technology? With 5 postdoctoral researchers working in this area, the PMP group is very active in several national research projects together with industry to realize the next innovations in solar cell technology. The aim is to simplify solar cell manufacturing and to bring the efficiency of (commercial) crystalline silicon solar cells closer to the upper limit of 29.4%. To achieve this, the research focus at the TU/e has now shifted to so-called passivating contacts, which consist of thin films placed between the c-Si and the metal contacts and serve to reduce the electrical losses at these contacts and which currently limit the performance of high-efficiency solar cells. These films can consist of metal oxides as well as materials like polycrystalline silicon. And guess what? The rumor is that such passivating contacts with polycrystalline silicon are also employed in the Sunpower cells of Stella Vie.

Figure 8. Solar cell researchers of the Plasma & Materials Processing group working in the NanoLab@TU/e cleanroom at the Eindhoven University of Technology. Photo Courtesy: Bart van Overbeeke.

References

1. Yoshikawa, K. et al. Silicon heterojunction solar cell with interdigitated back contacts for a photoconversion efficiency over 26%. Nat. Energy 2, 17032 (2017).
2. Yamaguchi, T., Ichihashi, Y., Mishima, T., Matsubara, N. & Yamanishi, T. Achievement of More Than 25 % Conversion Efficiency With Crystalline Silicon Heterojunction Solar Cell. IEEE J. Photovoltaics 4, 1433–1435 (2014).
3. Richter, A., Hermle, M. & Glunz, S. W. Reassessment of the Limiting Efficiency for Crystalline Silicon Solar Cells. IEEE J. Photovoltaics 3, 1184–1191 (2013).
4. Press release Fraunhofer ISE, “New World Record Efficiency of 25.7% for Both Sides-Contacted Monocrystalline Silicon Solar Cell”, 06^th of March 2017, https://www.ise.fraunhofer.de/en/press-media/news/2017/new-world-record-efficiency-of-25-point-7-percent-for-both-sides-contacted-monocrystalline-silicon-solar-cell.html
5. Hoex, B., Heil, S. B. S., Langereis, E., Van De Sanden, M. C. M. & Kessels, W. M. M. Ultralow surface recombination of c-Si substrates passivated by plasma-assisted atomic layer deposited Al2O3. Appl. Phys. Lett. 89, 42112 (2006).
6.  A full review on the current research can be found at: B. Macco, B.W.H. van de Loo, W.M.M. Kessels, ‘Atomic Layer Deposition for High-Efficiency Crystalline Silicon Solar Cells’, a chapter in ‘Atomic Layer Deposition in Energy Conversion Applications’, Wiley, ed. By J. Bachmann, 2017.