The solar electricity including solar photovoltaics (PV) is expected to contribute to the primary energy with a share of approximately 20% and 70% in 2050 and 2100, respectively. It respectively occupies in the total energy of the world, according to the recommendation (World Energy Vision 2100) by the German Advisory Council on Global Change [1]. According to “Sky Scenario” reported by Shell [2], the cumulative capacity for the power systems based on PV in the world is reached to 22 TW by 2050. However, the number is only 600 GW in 2019. The fact suggests that the importance of further installation of the PV based power systems, and the importance of further development of science and technology and deployment of PV. Especially, development of PV-powered vehicles applications is desirable and very important for creation of new clean energy infrastructure based on PV [3] [4]. In order to realize PV-powered vehicles, development of high-efficiency, low-cost, light weight, 3-dimensional curved and colorful solar cell modules is necessary.

This paper especially presents the importance of developing highly-efficient and low-cost solar cells and modules for automotive applications by showing efficiency impact on PV-powered vehicles in Section 2 and cost impact on them in Section 4. This paper also shows analytical results for efficiency potential of various types of solar cells in order to provide knowledge for selecting candidates of high-efficiency solar cell modules for automotive applications as described in Section 3. In Section 5, our approaches to PV-powered vehicle applications by using III-V triple-junction cell modules, static low concentrator III-V triple-junction solar cell modules and III-V/Si partial concentrator solar cell modules, and III-V/Si tandem cells are also reported.

Even in the transport sector, reducing CO₂ emission is a critical challenge for contributing to sustainable development, because the proportion of CO₂ emission from the road transport in the overall energy-related CO₂ emission in Japan, USA, and the world are 15.9%, 27.8%, and 16.9%, respectively [5]. Figure 1 shows the CO₂ emission per 1 km driving for various types of vehicles in Japan [6]. Although BEV (Battery-powered Electric Vehicle) has advantage for less CO₂ emission compared to ICE (Internal Combustion Engine vehicle), FCV (Fuel Cell-powered vehicle), and HEV (Hybrid Electric Vehicle), it is essential to achieve further reduction in CO₂ emission. Figure 1 shows that the PV can make a significant improvement on CO₂ emission.

According to survey reports [7] [8] of 5000 people and vehicles, effectiveness of the PV-EV (PV-powered EV), a PV-EV with 1 kW rated-power-PV and a 4 kWh rated battery capacity would reduce 12% of CO₂ emission compared to the HEV for 12 years.

In addition, the development of infra-structures such as battery charging stations for BEV and hydrogen stations for FCV is delayed. Figure 2 shows changes in cumulative registration number of Nissan LEAF (BEV) and number of the installed quick chargers [9] [10]. Therefore, development of PV-powered vehicles

is very useful to overcome the gap between registration number of BEV and charging stations.

Figure 3 shows the conversion efficiency of PV modules impact on PV-powered vehicle applications estimated from the survey reports [7] [8]. This figure implies that R&D on PV is essential in order to introduce PV-powered vehicles as a clean and usable major vehicle for the market. Especially, development of high-efficiency solar cells and modules is essential for PV-powered vehicles.

Figure 4 shows the prediction of projected cumulative number of PV-powered vehicles according to the NEDO’s Interim Report by “PV-Powered Vehicle Strategy Committee” [3]. According to the NEDO’s report, new broader PV market with more than 10 GW and 50 GW in 2030 and 2050, respectively are expected to be established. Cumulative PV capacity for PV-powered vehicles will be 50 GW and 0.4 TW in 2030 and 2050, respectively.

According to a statistics by Ministry of Japan [11] [12], approximately 70% of the family car runs less than 30 km per day in Japan. Standard EV runs approximately 9 km per kWh electricity, but after weight saving from 1400 kg to 600 kg, the rate is expected to increase to 17 km/kWh [13]. Divided the range of 30 km by 17 km/kWh, the necessary electricity will be 1.76 kWh/day. Namely, the average annual energy yield that is required for the light-weight family car powered by sunlight will be 642 kWh/year which is not an impossible value as well as a promising when we use a high-efficiency PV with efficiency of larger than 30% enables the society that majority of the family cars run by the solar power and without supplying fossil fuels. Thus, it is important for us to develop high-efficiency and low-cost solar cells and modules for automotive applications.

Figure 5 shows the required module efficiency as a function of the solar cell module area for the solar cell modules to achieve 800 W rated-output power [13]. The figure suggests that a promising way to realize the installation of PV on a family car is to create a small size module with 800 W rated output power which can be installed on roof area. The PV module can be realized by using III-V compound multi-junction solar cells which have conversion efficiency larger than 33% under 1 sun illumination.

Therefore it can be said that the development of a highly-efficient PV module with an efficiency of larger 30% is important to realize PV-powered vehicles. Figure 6 shows the recorded conversion efficiency of various types of single-junction solar cells along with their extrapolations [14]. The data were fitted with the Goetzberger function [15]:

η ( t ) = η l i m i t [ 1 − exp ( t 0 − t ) / c ] , (1)

where η(t) is the time-dependent efficiency, η_limit is the practical limiting efficiency, t₀ is the year for which η(t) is zero, t is the calendar year, and c is a characteristic

development time. Fitting of the curve was done with three parameters which are given in Table 1. The analysis shows that the improvement of the conversion efficiencies for each cell is converging or will converge soon, which is mainly bounded by the Shockley-Queisser limit [16].

Table 2 summarizes the reported world-record conversion efficiencies [14] [17] [18] [19] [20] [21] for various solar cells and Japanese contributions. As shown in Table 2, Japanese team has contributed to development of high-efficiency solar cells greatly as a result of Japanese national PV R&D programs such as the NEDO projects.

In this part, analytical results for efficiency potential of various solar cells are presented. We developed a model for the analysis [14] [17] for comparing the sources of efficiency loss of different types of solar cells. This model only attributes the efficiency loss to non-radiative recombination and resistance loss, which is a reasonable assumption because most solar cells have a minimal optical loss. The external radiative efficiency (ERE), which is the ratio of radiatively recombined carriers against all recombined carriers, is used for the quantification of the non-radiative recombination loss. In other words, we have ERE-1 at Shockley-Queisser limit [16]. The EREs of state-of-the-art solar cells have been reported in the publications such as references [18] [19] [20] [21]. In this work, the EREs of various solar cells were estimated by the following equation [22]:

V o c = V o c : r a d + ( k T / q ) ln ( ERE ) , (2)

where V_oc is the measured open-circuit voltage, k is Boltzmann constant, T is the

absolute temperature, and q is the elementary charge. V_oc:rad is the radiative open-circuit voltage. We use the reported V_oc;rad values extracted from [23] [24] in the analysis. The offsets of E_g/q – V_oc,rad used in the calculations are 0.23 V for CIGS, CdTe, a-Si and OPV, 0.25 V for perovskite, 0.26 V for Si, 0.28 V for III-V compounds [17] [24]. The second term on the right-hand side of Equation (2) is denoted as V_oc;nrad because it associates to the voltage-loss due to non-radiative recombination. In the case of multi-junction tandem solar cells, we define average ERE (ERE_ave) by using average Voc loss:

Σ ( V o c , n − V o c , r a d , n ) / n = ( k T / q ) ln ( ERE ave ) , (3)

where n is the number of junction.

The resistance loss of a solar cell is estimated solely from the measured fill factor. The ideal fill factor FF₀, defined as the fill factor without any resistance loss, is estimated by [25]

F F 0 = ( v o c − ln ( v o c + 0.71 ) ) / ( v o c + 1 ) , (4)

where v_oc is

v o c = V o c / ( n k T / q ) . (5)

The measured fill factors can then be related to the series resistance and shunt resistance by the following equation [25]:

F F ≈ F F 0 ( 1 − r s ) ( 1 − r s h − 1 ) ≈ F F 0 ( 1 − r s − r s h − 1 ) = F F 0 ( 1 − r ) , (6)

where r_s is the series resistance, and r_sh is the shunt resistance normalized to R_CH. The characteristic resistance R_CH is defined by [25]

R C H = V o c / J s c , (7)

r is the total normalized resistance defined by r = r s + r s h − 1 .

In the calculation, highest values [21] obtained were used as J_sc.

By assuming no optical loss, we can project the efficiency of various solar cells at different EREs. Figure 7 shows calculated and obtained one-sun efficiencies of

various solar cells. In summary, crystalline Si solar cells have a potential efficiency of 28.8% with total normalized resistance r of 0.02 by improvements in ERE from around 1% to 30%. GaAs has a potential efficiency of 30.0% with r of 0.02 by improvements in ERE from 22.5% to 40%. III-V triple-junction and penta-junction cells have potential efficiencies of 42% and 46% with r of 0.02 by improvements in ERE from 3% to 30% and from 1% to 30 %, respectively. CIGSe, CdTe and perovskite cells have potential efficiencies of 26.5% with r of 0.02 by improvements in ERE from around 1% to 10%.

Table 3 summarizes efficiency potential and efficiencies achieved of various candidate solar cells for PV-powered EV applications. The first request is a high-efficiency PV panel on the roof of the car. Because of the space limitation, it should be a high-efficiency panel. III-V compound multi-junction (tandem) cells and III-V/Si tandem cells are one candidate. The CPV (concentrator PV) is very attractive for saving the cost of the cell. Cars move quickly, and appearance is essential. Trackers were thought to challenge to implement. One of our choices is a static concentrator customized to the automobile.

Regarding III-V multi-junction solar cells, efficiency of multi-junction solar cells increases with increase in number of junction as shown in Figure 8. However, it is thought that triple-junction is optimal number of junction because in the case of higher number of multi-junction solar cells of more than quad-junctions, external radiative efficiency (ERE) decreases with increase in number of junction as shown in Figure 8. Fill factor for multi-junction solar cells is also decreases with increase in number of junction as shown in Figure 9.

In addition to high-efficiency and low-cost, development of PV modules with 3-dimesional curvature, color variation and good temperature coefficient is necessary. Figure 10 shows changes in temperature coefficients of various solar cells as a function of open-circuit voltage of solar cells [26] [27] in comparison of calculated values. Calculated values of relative temperature coefficients TC_rel. of solar cells were semi-empirical equation [28].

T C r e l . = ( 1 − n E g / q V o c ) / T = ( 1 − 1.1 E g / q V o c ) / T , (8)

where n is the diode ideality factor. Because the III-V triple-junction solar cells have lower temperature coefficients of 0.09% - 0.15%/˚C, compared to that (0.24%/˚C) mono crystal Si solar cells as shown in Figure 10, the higher efficiency solar cells are thought to be attractive for PV-powered vehicle applications.

Cost reduction of high-efficiency solar cell modules is also very important for PV-powered vehicle applications. Figure 11 shows cost impact on PV-powered vehicle applications estimated from the survey reports [7] [8]. This figure implies that R&D on high-efficiency and low-cost PV is essential in order to introduce PV-powered vehicles as a clean and usable major vehicle for the market.

Although III-V multi-junction solar cells have an extremely high conversion efficiency with efficiencies of 39.2% under 1-sun and 47.1% under concentration, which is suitable for this application, cost reduction is necessary to realize the concept as shown in Figure 5.

Figure 12 shows module efficiency and module cost for III-V multi-junction (MJ) solar cell modules, conventional flat Si PV solar cell modules, PV modules developed for Toyota PRIUS released in 2010 and those for Toyota New PRIUS released in 2017, and module efficiency and module cost targets of PV-EV [29] [30]. Because module price estimated [29] are about $30/W for Toyota Prius released in 2010 and $12/W for New Prius released in 2017, respectively, it is thought that target of module cost may be less than $10/W.

Figure 13 shows a comparison of module cost for III-V tandem (MJ), III-V/Si tandem (MJ), and concentrator III-V/Si tandem solar cell modules as a function of module production volume [31]. The III-V/Si tandem solar cell modules have low-cost potential if technology development, expansion of application areas and an increase in output volume is made.

Therefore, Si tandem solar cells by combining Si and other materials such as III-V compound, II-VI compound, chalcopyrite, perovskite and so forth are desirable for realizing super high-efficiency and low cost. Recently, Si tandem solar cells have paid considerable attention because of high-efficiency and low-cost potential.

Efficiency potential of various Si tandem solar cells is also analyzed [32]. The similar procedure and values described above were used. Figure 14 shows calculated 1-sun efficiency of III-V/Si triple-junction and dual-junction tandem solar cells, perovskite/Si dual-junction tandem solar cells, CdZnTe/Si dual-junction

tandem solar cell and GaAs nano-wire/Si dual-junction tandem solar cell as a function of ERE and r_s + 1/r_sh.

At present, the III-V/Si triple-junction and dual-junction tandem solar cells have shown higher efficiency with 35.9% [33] and 32.8% [33] compared to perovskite/Si dual-junction tandem solar cells with efficiencies of 28.0% [21], CdZnTe/Si dual-junction tandem solar cell with an efficiency of 16.8% [34] and GaAs nano-wire/Si dual-junction tandem solar cell with an efficiency of 11.4% [35]. Such an efficiency difference is thought to be differences in material quality. For example, the external radiative efficiency values are 2% - 4% for III-V/Si tandem cells, 1% - 2% for perovskite/Si tandem cells, 3.5 × 10⁻² % for CdZnTe/Si tandem cells and 1 × 10⁻³ % for GaAs nano-wire/Si tandem cells. Therefore, material quality is critical for further improvements in the performance of Si tandem solar cells. Although efficiency (35.9%) [33] of 4-terminal mechanical stacked InGaP/GaAs/Si triple-junction tandem solar cells is close to that (37.9%) [36] of InGaP/GaAs/InGaAs triple-junction cells, resistance loss is higher as shown in Figure 14. Resistance loss for the perovskite/Si tandem cells, CdZnTe/Si tandem cells and GaAs nano-wire/Si tandem cells are much higher compared to the III-V/Si tandem solar cells as s shown in Figure 9.

In this section, recent Japanese activities of III-V/Si tandem solar cells are presented. Table 4 shows a summary of Japanese activities of InGaP/GaAs/Si tandem solar cells. The authors have demonstrated 28.2% [32] [37] with mechanically stacked InGaP/GaAs/Si triple-junction solar cell. Osaka City University also demonstrated 25.5% efficiency InGaP/GaAs/Si triple-junction solar cells

with 0.25 cm² area by using surface-activated bonding (SAB) [38]. Sharp Corporation has also attained 33.0% efficiency [39] with 3.604 cm² InGaP/GaAs/Si triple-junction solar cells by using the mechanical stack. Most recently, AIST has demonstrated a novel interconnect layer made of palladium nanoparticle (PdNP) between III-V and silicon that could be both optically transparent and electrically conductive, allow two-terminal configuration. A 30.8% efficiency [40] under 1-sun with InGaP/AlGaAs/Si triple-junction solar cells fabricated by using PdNP array-mediated bonding was demonstrated.

Table 5 shows that the progress of practical phase and development phase PV-powered vehicles [41]. Toyota Motor Co. released Prius equipped with 56 W multi-crystalline Si solar cell modules and new Prius PHV with Si HIT solar cells modules with 180 W into markets in 2010 and 2017, respectively. Other auto companies also have developed PV-powered vehicles, and some of them have announced that they will start selling the PV-powered vehicles in 2020 and 2021.

Most recently, Toyota has developed test car by using Sharp’s high-efficiency In GaP/GaAs/InGaAs triple-junction solar cell modules (output power of 860 W, average cell efficiency of 34.9%) under the NEDO support [42] as shown in Figure 15. The daily driving range of 44.5 km is expected to be realized by using solar energy. Data collection and analysis are under driving test by using the Toyota Prius PHV test car.

We recently proposed a new static low concentrator which does not require tracking system for the automotive application. It is effective to reduce the cost of PV module by utilizing a concentrator because it can be reduced the total solar cell area. Since the installing a tracker on a vehicle roof is difficult, a static concentrator is suitable for the automotive application. Static concentrators that combined with III-V compound solar cells have many other advantages: 1) applicability of mounting on a three-dimensional curved surface, 2) robustness to partial shading, and 3) a shark-skin structure has advantage for aero-dynamics property. A high Voc can be attained thanks to the robustness to partial shading, allows a bypass diode to be equipped with each cell in the gap between cells in the module, and facilitates size reduction of the lens aperture.

Figure 16 shows the calculated results of the trends of annual power yield by III-V/Si tandem modules and crystalline Si modules as a function of concentration ratio [43] [44]. The crystalline Si module cannot be attained the target yield even with a considerable (4 m²) roof area., while III-V/Si tandem module and static low concentrator module up to 4-suns concentration factor can generate the target value, which indicating that the cells are promising candidate for automotive applications.

We recently demonstrated aperture efficiency of 27.6% with the static low concentrator module combined with III-V based triple-junction solar cell [43] [44]. Figure 17 shows that the module has 99 solar cells (33 cells were connected in series, and three strings were connected in parallel) which were mounted on the circuit board. The low concentrated lens was aligned on each cell and silicone (sealing resin) was filled in the gap between the cell and lens. The module in which area is 41.2 cm² has a conversion efficiency of 27.6% under on-axis 1 sun illumination. The module efficiency can be improved by reducing optical and electrical losses of the concentrator, such as reflection at the interface of Air/PMMA (3.9%), reflection at PMMA/Silicone (0.7%), diode loss (0.5%), voltage mismatch (0.3%), and current mismatch (7.1%). It is expected that the module efficiency is reached to 30% by reducing the mismatching loss and optical losses.

Figure 18(a) and Figure 18(b) show the schematic drawing of a new partial CPV with static low concentrator concept [45]. The concentrator “partially” focuses on a component of the sunlight energy toward high-efficiency triple-junction cells; the remaining portions are captured using low-cost Si cells. This concept can eliminate the inactive area and improve the conversion efficiency of the module by harvesting global sunlight in a full module aperture.

Figure 18(c) shows the prototype III-V/Si partial LCPV submodule with an area of 32 × 40 mm² [46]. The prototype flexible submodule composes of a 4 × 5 silicone lens array (each area 8 × 8 mm²), 4 × 5 InGaP/GaInAs/Ge triple-junction cells (each area 4.05 × 4.57 mm²), and a crystalline Si cell which is located underneath of the III-V cells. This submodule is a 4-terminal mechanical stack in which the silicone lens encapsulates the triple-junction solar cells atop the Si cell via a thin transparent glass substrate. Because module thickness should be as low

as possible for the automotive applications, the thickness of the assembled module was designed to approximately 6.0 mm. The measurement results show that the prototype module can potentially achieve up to 27.3% annual module efficiency, as projected from the actual daily submodule efficiency of 20.8% assuming state-of-the-art record cell efficiency.

Development of high-efficiency solar cell modules and new application fields are significant for the further development of photovoltaics (PV) and creation of new clean energy infrastructure based on PV. Especially, development of PV-powered EV applications is desirable and very important for this end. Efficiency impact on reducing CO₂ emission and increase in driving distance in PV-powered vehicles was shown in this paper. Development of high-efficiency solar cell modules with an efficiency of more than 30% is essential for PV-powered vehicle applications.

This paper presented analytical results developed by the authors for efficiency potential of various solar cells for choosing candidates of high-efficiency solar cell modules for automobile applications. As a result of the analysis, static low concentration PV and Si tandem solar cells such as III-V/Si tandem solar cells are thought to be some of their candidates. This paper also overviewed efficiency potential and recent activities of various Si tandem solar cells such as III-V/Si, II-VI/Si, chalcopyrite/Si, perovskite/Si and nanowire/Si tandem solar cells. Present status of Si tandem solar cells such as 35.9% for III-V/Si triple-junction, 32.8% for III-V/Si dual-junction, 28.0% for perovskite/Si dual-junction tandem solar cells was overviewed. The III-V/Si tandem solar cells are expected to have significant potential for various applications because of high efficiency with efficiencies of more than 42% under 1-sun AM1.5 G, lightweight and low-cost potential. As a result of further development of Si tandem solar cell modules and creation of PV-powered vehicle market, cost reduction of modules for PV-powered vehicles will be realized. Recent results for our 28.2% efficiency and Sharp’s 33% mechanically stacked InGaP/GaAs/Si triple-junction solar cells were also presented. Some approaches such as InGaP/GaAs/InGaAs triple-junction solar cell powered vehicle, static low concentrator III-V triple-junction cell modules and low-concentration static III-V/Si partial concentrator cell modules were also presented.

In summary, the authors have shown that solar modules for PV-powered vehicles have great ability to reduce CO₂ emission. Development of high efficiency (>30%), colored, and flexible modules is essential. Static concentrator PV, Si tandem PV and colored PV are attractive candidates for future car applications.

The authors thank members of the NEDO, METI and Toyota Motor Co., Sharp Co. and Mr. T. Sato, Ms. M. Hasegawa, NEDO, Prof. N. Yamada, Nagaoka Univ. Tech., Prof. T. Hirota, Waseda Univ., Dr. K. Komoto, Mizuho, Inf. Res. Inst., and members of the NEDO’s PV-powered Vehicle Strategy Committee for their cooperation and providing fruitful information.

The authors declare no conflicts of interest regarding the publication of this paper.

Yamaguchi, M., Masuda, T., Araki, K., Sato, D., Lee, K.-H., Kojima, N., Takamoto, T., Okumura, K., Satou, A., Yamada, K., Nakado, T., Zushi, Y., Yamazaki, M. and Yamada, H. (2020) Role of PV-Powered Vehicles in Low-Carbon Society and Some Approaches of High-Efficiency Solar Cell Modules for Cars. Energy and Power Engineering, 12, 375-395. https://doi.org/10.4236/epe.2020.126023