Dr Radovan Kopecek is one of the founders of ISC Konstanz. He has been working at the institute as a full time manager and researcher since January 2007, and is currently the head of the Advanced Solar Cells department.

Since the 15th century we know that the earth is round and since the 16th century we have accept that the earth is orbiting around the sun. However, the heliocentric system was not easy to accept by many disbelievers.

When ISC Konstanz started the bifiPV workshop in Konstanz in 2012, similar contrary wind was blowing from the monofacial PV community - laughing and disbelieving of the power of albedo. After 7 years this is now very different. The “albedists” have gained popularity and bifacial technology is becoming bankable.

So yes!! Finally, bifaciality will win because bifacial technology is an evolutionary consequence of rear passivated solar cell technologies and long-lasting double glass module technologies [1, 2].

Since the US recently dropped import taxes on bifacial PV modules [3], there is no doubt that the US will become a dominant bifacial PV country and push this technology to the next level – mostly combined with HSAT (horizontal single axis trackers).

From 2020, at least 16GW of new PV systems are expected to be installed in the US [4] of which 12GW is expected to be utility-scale, while the rest will be comprised of commercial and residential installations. Under the current conditions, the vast majority of systems will be bifacial in nature, regardless of whether the rear side will operating. Indeed, JinkoSolar has recently confirmed lately that according to their assessment, monofacial modules will be relegated to the history books [5].

The recent bifiPV2019 workshop in Amsterdam on September 16 and 17 proved to be a huge success. A good brief summary is given in reference [6].

With this blog we want to summarize the findings and discuss the most important still open points regarding technology and economy. All the presentations can now be downloaded on the bifiPV2019 web-site.

The next bifiPV2020 Workshop will be together with SiliconPV/nPV2020 next year on April 2-3 in Hangzhou, China.

Summary of bifiPV2019

bifiPV2019 in Amsterdam was extremely well attended with around 200 registrations from which 160 people were able to participate; 40 remained on the waiting list!

About 25% of the visitors were from the US, comprised of institutes, characterization labs and EPCs as the US bifaciality community and projects are growing fast. From Asia most of the participants were from major solar cell and module manufacturers such as Q CELLS, LONGi Solar, Jolywood, Canadian Solar and JinkoSolar.

European-based attendees were mostly system owners and institutes that are involved in energy yield modelling as well as solar cell and modules producers, such as SoliTek and 3SUN. Hopefully more will participate in the future.

In the following we will summarize the most important technical and economic issues which remain open and are under discussion and development.

Solar cell and module technology

The most discussed question these days in the solar cell arena is what is the limit of standard/bifacial industrial PERC (Passivated Emitter Rear Cell) technology and when n-type concepts will penetrate the market at commercial and competitive volumes?

This was discussed in Session 1 of bifiPV2019. The question actually remains open as PERC average efficiencies in production are around 22.5%, which makes it extremely difficult for n-type PERT and other concepts to “overtake” in spite of their higher bifacial factor.

IBC, nPERT-“TOPCon” and HJT solar cells are all bifacial concepts which have been developed by many institutes and solar cell producers such as SPIC/ISC Konstanz (ZEBRA), FhG ISE (TOPCon), Trina Solar (TOPCon), Jolywood/IMEC (TOPCon), Tempress/ECN>TNO (TOPcon), INES (HJT), EPFL/CSEM (HJT) REC/Meyer Burger (HJT).

The goal of all these institutions is to bring their technologies to the bifacial market, while also being cost competitive with PERC, at least at the LCOE (Levelized Cost of Electricity) level. The race is still ongoing, due to the extremely fast moving target PERC has been setting.

In addition, greater attention has been given to tandem solar cells consisting of a c-Si bottom cell and a Perovskite top cell. It would seem that 2, 3 and 4 terminal structures are possible, reaching efficiencies of around 30%. In order not to miss the bifacial trend, tandem solar cells must have a bifacial nature as well - as it was presented by Gianluca Coletti at EUPVSEC 2019 in Marseille and at bifiPV2019 in Amsterdam.

Therefore, we plan to start a new workshop similar to nPV and bifiPV. TandemPV will be launched in 2020 in Konstanz to connect R&D with industry participants.

Modules

PV module technology for bifacial cells has rapidly become a no-brainer. Bifacial modules are very similar to monofacial modules only with 3 adjustments [7]. Differences include the need for junction boxes to be placed next to the cells but not to shadow the rear side of the solar cells. In order not to cannibalize the front side power of the module, reflective ceramics are used in-between the cells – “simulating” the white backsheet of a monofacial module. As there is more current density flowing in the module the by-pass diodes selection is different than monofacial requirements.

However, the module geometry and module components are very similar to monofacial modules. The trend for bifacial modules is towards half-cut cells, double glass modules and therefore also clearly towards POEs (Polyolefins) - also because of lowering PID (Potential Induced Degradation) and other degradation mechanisms, which in bifacial modules can occur not only from the front side but also for the rear cells.

Degradation in bifacial modules

It should be noted that both sides of a PERC cell can be affected by PID (Potential Induced Degradation) and LeTID (Light and elevated Temperature Induced Degradation) [8]. In addition, more advanced cell devices can be affected by more complex degradation. The graph below shows a production inspection for a large utility-scale PV project in South America. Bifacial PERC solar modules were tested here on LID, PID and LeTID [9]. These tests reveal that bifacial modules “open one more side for degradation” and should to be tested accordingly.

We can conclude from this graph that for this PERC technology the front side degradation was taken care of however the rear side is still causing some problems as it is showing notable degradation levels. This can be solved at cell level with the known strategies for front side cells. In addition, the use of good quality POE encapsulant is essential for bifacial modules for many reasons - no acetic acid formation, suppression of degradations and no yellowing in regions with high UV irradiation.

Standards

What is very often misleading for many scientists is that the efficiency of a bifacial solar cell is actually lower than its monofacial counterpart, as efficiency is defined as power out divided by power in- and the fact that the rear-side cell is inherently less efficient. This means that the module efficiency is an average between front and rear side. But of course, the power is higher as the rear side cell enables more light conversion, overall.

For this reason it is has been proposed and discussed in technical circles that the PV industry should think in terms of efficiencies only – something like “equivalent efficiency power” and “equivalent energy efficiency” to be informed and educated to bifaciality, especially scientists involved in the development of new solar cell concepts.

However, these descriptions can sometimes be misleading and we should rather instead, separate between efficiency, power and energy.

For bifacial modules to be highly bankable, or being able to sell the advantages of the technology, new standards are needed in addition to the STC module measurement. This has been discussed since the first bifacial workshop bifiPV2012 in Konstanz.

However, only since January 2019, have we the first standard in place [10] that has been published as IEC standard TS 60904-1-2:2019. This standard, highlights that for measuring bifacial module output they should be illuminated from both sides or the front side illumination increased according to its bifacial factor as described in the standard.

In such a way, e.g. bifi100 (assumption of 100W/m2 from the rear) and bifi200 (200W/m2) Pmpp measurements can identify how such a module compares to the Pmpp at STC of a monofacial counterpart to have a better selling point for bifacial modules.

Our comment on this is that in the end, full bifacial performance modelling for energy yield prediction is needed to understand the full potential of the module used, as many factors play a role in the yield levels. In addition, the one-sided measurement option of the new standard for bifacial devices is expected to become obsolete when more complicated devices such as bifacial tandem cell modules enter the market.

Systems

Good development has been done in adapted construction for bifacial modules [11]. For utility scale and flat roofs slanted tilt and vertical constructions are used.

As HSAT bifacial systems are the future for reaching the lowest LCOEs in the PV utility-scale market, there is a huge competition from mounting system companies that are supplying trackers. Currently, the hottest topic being discussed to whether to use 1V or 2V systems. Both have their advantages and disadvantages. Not only related to costs but electrical gains have to be considered in this equation as well as maintenance and subjects such as wind loads.

Bifacial performance modeling

Session 6 of bifiPV2019 was dedicated to bifacial performance modelling. Here the most important task is to compare complex simulations with experimental data. The good news here is that there is an increasing level of real field data being generated and analyzed, even from large systems and also for vertical installations, as well as equator oriented fixed tilt and with HSAT systems.

The bad news is that not many companies are willing to share the real field data. Therefore, the institutes involved in modeling have their own cooperation and generating data from their own test systems in order to improve their simulation models.

In general, it can be stated that commercial software tools such as PVsyst, most of them using a 2D view factor concept for optical modelling, seem to generate accurate predictions of the yearly energy yield for some bifacial standard configurations such as fixed tilt PV system.

Whereas, for bifacial tracking the accuracy seems to be further away from the level already achieved for monofacial systems. Furthermore, up to now, vertical bifacial installations cannot be modelled by many of the software platforms available on the market and for this configuration there is a particular lack of model validation with field data.

Therefore, many institutes – as well as some companies - are developing their own simulation tools like ECN with ‘Big Eye’, NREL with SAM and ISC Konstanz with MoBiDiG.

Some of these are using 2D or 3D-view factor based optical modelling, while for others ray tracing is used to model the irradiance in the plane of array on front and rear side of the modules. All these activities aim for reaching the highest accuracy of energy yield predictions also for non-standard bifacial PV systems while keeping the computational effort as low as possible.

The main challenge in this field, apart from the scientific and technical expertise required, is the need for relevant field data. Relevant in this case means: highly accurate measurements, including all required parameters monitored on a PV system with a commercially relevant configuration. Only by validating the results from the simulation models with such field data is the prediction accuracy of the models proven – and further improved – in a way that the trust of investors will reach the same level as for monofacial PV systems.

Realistic bifacial gain and lowest LCOE

There have been so many bifacial gains reported, notably in tests sites that everyone who is not an expert in this area can easily be confused. From 0% gains to even more than 100% gains have been reported but actually things are quite simple as you just have to look closer at what is being reported: e.g. what is the reference, what is the albedo, are people talking about stand-alone modules or a large system, a one day measurement or yearly average, in order to understand why the reported values are not in line with what you would expect or heard from someone else.

At the beginning let us mention that maximizing bifacial gain is not what is reasonable but maximizing the electricity output in relation to CAPEX and OPEX, which means to minimize the LCOE in costs/kWh.

The bifacial gain is however a nice instrument to illustrate the contribution of the rear side of the module, independent of the module characteristics, installation and albedo of the ambient in comparison with a monofacial system installation.

In the above graph you can observe the many possible installations for bifacial systems and schematic power generation curves, in comparison with their monofacial counterpart. So, for the vertical installations (red curve) the bifacial gain can be more than 100% which is maybe interesting to realize, but as it makes no sense to install monofacial modues this way this comparison is useless for economic considerations.

If you want to estimate bifacial yield in a system these are the 4 most important factors:

? Module′s bifacial factor (higher bifacial factor higher bifacial gain)

? Installation geometry (higher installation height and lower shadowing from the rear leads to higher bifacial gain)

? Albedo (higher albedos lead to higher bifacial gains)

? Latitude (higher latitude lead to more diffuse light and higher bifacial gains)

A nice summary of what can be reached in large systems in comparison to monofacial reference systems is given by PI Belin in their white paper [13] and perfectly matches what we observe in large systems, we have reliable numbers from.

A 5% gain is what you get on the lowest side, whereas 45% power gain can be reached even in large systems when including large tracking gains combined with high albedos, optimized row spacing and good installation conditions.

Recently, LONGi has published their bifacial gain numbers from own installations confirmed by TüV SüD as well [14]. Interestingly- as observed also by other companies and institutes- LONGi confirmed that bifacial modules can operate even at lower temperatures as their monofacial reference modules.

Albedo enhancing measures are also a very interesting topic in bifiPV and is being discussed intensively in the community.

bifiPV2020 and future of bifacial PV

Bifacial PV is now completely unchained since the drop of import taxes in the US [3] and since then it is becoming increasingly bankable for investors. The relaxing of import taxes will be challenged by a few companies producing modules in the US. However, the bifacial PV boom in the US cannot be stopped by that anymore.

The newest report from Wood Mackenzie Power [15] shows that in H1 2019 more than 5GW of bifacial PV systems have been installed. We expect that in 2025 bifacial PV will represent half of the utility scale market worldwide.

We are looking forward to meet you in Hangzhou to discuss the bright future of bifacial PV in April 2020.

This guest blog was co-written with Dr. Joris Libal, ISC Konstance.

References

[1] R. Kopecek and J. Libal 2018, Towards large-scale deployment of bifacial photovoltaics, Nature Energy volume 3, pp. 443–446

[2] Libal, J. & Kopecek, R. (eds.) 2018, Bifacial Photovoltaics: Technology, Applications and Economics. Institution of Engineering and Technology, ISBN: 978-1-78561-274-9.

[3] https://www.pv-tech.org/news/bifacial-modules-now-exempt-from-trumps-trade-tariffs

[4] https://www.woodmac.com/research/products/power-and-renewables/us-solar-market-insight/

[5] https://www.pv-magazine.com/2019/08/29/jinko-the-days-of-mono-will-soon-be-behind-us/

[6] https://www.pv-magazine.com/2019/09/19/a-common-language-for-bifacial-pv/

[7] http://bifipv-workshop.com/fileadmin/layout/images/bifiPV/Bifacial_module_technology_KOPECEK_Plenary_EUPVSEC_2018.pdf

[8] https://www.pv-tech.org/guest-blog/is-letid-degradation-in-perc-cells-another-degradation-crisis-even-worse-th

[9] https://www.linkedin.com/in/vicente-parra-garc%C3%ADa-716b761a/detail/recent-activity/

[10] https://pvpmc.sandia.gov/pv-research/bifacial-pv-project/bifacial-pv-characterization-and-rating-standards/

[11] H. Nussbaumer et al. PV systems with lowest LCOE using bifacial modules: State-of-the-art systems and components, PV Power Tech, February 2019, p.16

[12] I. Shoukry, Bifacial Modules – Simulation and Experiment, Master Thesis at the University Stuttgart, 2015

[13] https://www.pi-berlin.com/news/191-white-paper-bifacial-pv-technology.html

[14] https://www.pv-magazine.com/2019/09/30/longi-unveils-results-of-bifacial-module-test/

[15] https://www.greentechmedia.com/articles/read/bifacial-solar-market-set-to-grow-tenfold-by-2024