Self-shape-transformable 3D tessellated bifacial crystalline Si solar cell module enabling extra energy gain through intervals and an integrated actuator

Abstract

The previously reported strategies for photovoltaics (PVs) have been one-size-one-fits-all methods under a watt per cost concept, which is changing in favor of a customized-fit strategy under an energy yield per watt concept to generate more energy per installed photovoltaic (PV) device and thereby expand applications of PVs. Bifacial solar cells have attracted attention for possible implementation of this new PV concept. In this study, as a novel approach for bifacial technology in urban environments, we propose an automated solar trackable and self-shape-transformable three-dimensional (3D) tessellated bifacial module that can be used with any curved surface and can change shape according to the angle of incidence (AOI) of light to maximize the power output without external assistance. The proposed self-shape-transformable 3D tessellated bifacial module with 2 mm intervals between the cells and with an installed reflector achieved a 47% enhancement in the power output compared with that of a flat module under 1 sun, 1.5 a.m. conditions. At high AOI values, the power output was greater than that of a flat module, owing to solar illumination AOI sensitive shape changes and the effect of bifacial energy production with a controlled reflector. This indicates that the proposed module has strong potential for use in urban environments.

---

Introduction

The development of photovoltaic (PV) solar energy harvesting technology has become a topic of interest, along with increasing renewable energy sharing, in response to climate change and the emerging energy crisis. The PV industry is expanding as the use of photovoltaics (PVs) in various applications and in various environments continues to increase. With respect to the application of PVs in various environments, from the utility scale using Giga-watts to the internet of things that integrate PVs using Milli-watts, and considering the broader market, customized-fit PVs that maximize the energy yield per watt concept are on the rise for applications that consume low amounts of energy instead of one-size-fits-all PVs that minimize the cost per watt concept for applications that consume a high amount of energy.^1,2 In addition, small-scale customized-fit PVs, functional and aesthetic factors should be tailored to the specific application.^3–6

Many countries have urban environments in which buildings and skyscrapers occupy a small geographic area. In the urban environment, the light that PVs receive as directional incident light and omnidirectional incident light is composed of reflected and scattered light from reflectors and diffused light. Therefore, extensive research is being devoted to increasing the annual energy yield by utilizing various incident light sources available in space-restricted installation areas. To enhance the annual energy yield, bifacial PV technology has received considerable attention because of its reduced levelized cost of energy.^7–10 Numerous studies involving simulations, experiments, and predictions related to bifacial technology have been reported. Owing to this attention and the potential of bifacial PV, technical research considering application in the urban environment is needed.

To address the problems of a limited installation area and omnidirectional incident light in urban environments, the curved PV^11–14 and flexible PV^15,16 concepts have been introduced and developed; a solar tracking system has also been proposed to increase the energy yield. It is assumed that a curved array panel is installed on a curved surface, such as in the case of a building-integrated PV (BIPV) or a car roof. However, rigid encapsulation materials, such as ethylene vinyl acetate with a cover glass and the stress on the curved Si wafer and soldered strip parts during the deformation process can result in problems. In the case of flexible PVs, which represent a viable alternative to using easily breakable crystalline Si solar cells, an etching process has been introduced with interdigitated back contacts. However, this approach is limited with respect to the interconnection and for expanding the scale of the modules. Furthermore, an integrated solar tracking system^17–19 for maximizing the energy output is complex and requires additional maintenance, including additional machinery parts, and a wide swing area that is difficult to achieve in a space-limited area, such as an urban environment.

To overcome these limitations, here we introduce a self-shape-transformable three-dimensional (3D) tessellated bifacial module that is not restricted to certain environments. It can be used for any application, including those involving a curved surface, and can be custom-fit. With the integration of an actuator (i.e., a shape-memory alloy (SMA)^20–22 that reverts to its memorized shape in response to heat), the self-shape-transformable 3D tessellated bifacial module automatically changes shape in response to the position of the sun at all angles of incidence (AOIs) without an external energy supply, such as a motor.

Using this technology, we can achieve a 47% enhancement in the production of electricity over a period of 1 d compared with a conventional fixed flat panel, with a greater energy yield at low AOIs as a result of the transformed tessellated module shape. In addition, we proposed an all-parallel electrical configuration to avoid shading and mismatch problems. The proposed approaches, which are applied to the bifacial PV field for the first time in the present work, led to a good performance, and various geometries are presented in simple 3D shapes. This work represents the first step towards the development of the concept of 3D tessellated structures with optimized reflector materials or geometries, which have a strong potential in the PV technology field.

Results and discussion

An energy harvesting system that can be installed in small, restricted areas in an urban environment, which represents difficult conditions for the installation of a large-area energy harvesting system such as a plant, is desirable. To utilize a PV system in an urban environment, light environments, such as vertical directional light and omnidirectional light composed of reflected, scattered, and diffused light, should be considered. To meet this need, previous studies have focused on new module designs that can enhance energy production over the course of a day, such as 3D structures, as alternatives to typical flat modules considering the effect of the AOI.^23–26 For the design of the novel module, a casting encapsulation process with silicone materials (polydimethylsiloxane, PDMS) was used to design a 3D structure, such as that reported previously by our group. Cut Si solar cells were connected with a flexible wire and then encapsulation was conducted on each of the tessellated cells. Therefore, it is possible to implement a 3D structure after the encapsulation process. In addition, there are several advantages to using PDMS to replace ethylene-vinyl acetate (EVA) and the cover glass, including the ability to transmit short-wavelength light, trap incident light, and overcome the browning effect.

In the present study, to diversify the module shape, we used a tessellation structure. The basic tessellation structure is a compact arrangement of units with certain shapes and connecting units. A primary feature of the tessellation structures is that they can be adapted to any curved or folded surface by adjusting their unit-to-unit spacing. The introduction of a tessellation structure onto a solar cell module enables its installation on any surface and its implementation into 3D-shaped modules (Fig. 1a). We fabricated a tessellation-structured module using bifacial solar cells, in which we controlled the interval between the unit cells using a reflection plate to attain additional energy production by increasing the bifacial gain through the spaces between the cells. In addition, to maximize the energy yield during the course of a day, we also integrated self-shape transformation technology by integrating an actuator onto the surface of a solar cell module. An SMA was used as an actuator that recovered its memorized shape in response to an increase in temperature. The heat from the surface of the solar cells under solar irradiation was transferred to the actuator, where it induced a shape change in the SMA to its memorized shape through a phase transition. In the present research, we used a flat memorized shape. The product energy can be maximized by maximizing the area of the cells exposed to incident light as a result of a change to the memorized flat shape in response to a change in the position of the sun (Fig. 1b). In addition, with the incorporation of a reflector into the self-shape-transformation module, an additional energy gain was achieved by increasing the bifacial gain.

Fig. 1 Schematic diagrams showing (a) the basic plane tessellation structure for the developed 3D and 3D tessellated bifacial modules with uniform intervals and a reflector, and (b) the concept of the self-shape-transformable 3D tetragon tessellated bifacial module as the module operates over the course of a day.

First, to confirm the bifacial gain effect through the gaps between unit cells, the performance of the fixed 3D tessellated module was evaluated according to its angle of incidence (AOI). Twelve rectangular unit cells with a size of 31 × 10 mm were arranged at regular intervals and fixed by attachment to a flexible backstrip. Rods were attached to both ends of the tessellated module so that 3D modules with various shapes could be installed in a space-limited installation area.

The length of the flat tessellated module depends on the interval between its unit cells (Fig. S1†). We fabricated 3D arch and 3D tetragon tessellated bifacial modules in a space-restricted installation area using a flat tessellated module with identical intervals. In the 3D arch or 3D tetragon tessellated bifacial module, under vertical directional incident light, most of the energy is produced through the exposed front side cells; the bifacial gain can be increased by incident light that passes through the intervals and is scattered at the reflector (left side of Fig. 2a). As an urban environment scatters or absorbs a substantial amount of omnidirectional incident light, we evaluated the performance of the module on the basis of its AOI. In the case of omnidirectional incident light, we expected an enhanced bifacial gain as a result of direct illumination of the back side of the cell and by incident reflected and scattered light on the reflector through the unit-cell intervals (right side of Fig. 2a). The performance of the 3D tessellated bifacial module was confirmed to depend on the interval between cells, which was varied from 1 to 2 to 2.5 mm.

Fig. 2 (a) Schematic diagram of the 3D arch tessellated bifacial module under vertical directional incident light (left side) and under omnidirectional incident light composed of reflected, scattered, and diffused light (right side) from the reflector through the intervals between units. Photographs of the 3D arch tessellated bifacial module under vertical directional incident light (left side) and under omnidirectional incident light at an AOI of 70° (right side), with intervals of (b) 1 mm and (c) 2 mm. Photographs of the 3D tetragon tessellated bifacial module under vertical directional incident light (left side) and under omnidirectional incident light at an AOI of 70° (right side), with intervals of (d) 1 mm and (e) 2 mm. The conversion efficiency of (f) the 3D arch tessellated bifacial module and (g) the 3D tetragon tessellated bifacial module. (h) Specific power over the course of a day considering the installation area of the 3D tetragon tessellated bifacial module with various intervals with or without a reflector. Specific power was measured under 1 sun and 1.5 a.m. conditions.

In the case of a simple 3D arch tessellated bifacial module, a longer flat module has a higher arch shape (Fig. S2†), which reduces the exposed area of the front side of the cells, thereby decreasing the specific power under vertical directional incident light. In the case of the 3D arch tessellated bifacial module with 1- or 2 mm intervals, the exposed area of the front side of the cells is similar (Fig. 2b and c). Thus, the specific power under the vertical incident light leads to approximately the same results. However, under omnidirectional incident light, the module with 2 mm intervals generates a greater specific power than that with 1 mm intervals under both the nonreflected and the reflected conditions (Fig. 2f). By introducing even a small space between the unit cells, extra bifacial gain is attained according to the AOI by utilizing the incident light that passes through the unit-cell intervals.

As another simple 3D tessellated bifacial module, we fabricated a tetragonal tessellated bifacial module. The shape of this module resembles a pot, and we expected an increase in yield energy as a result of the enhanced reflection effect of incident light that passes through the unit-cell intervals (Fig. S3†). The 3D tetragon tessellated bifacial module with 1 mm intervals exhibits a trapezoid shape with a large angle between the two sides (Fig. 2d). Under the vertical directional incident light, the specific power differs because of different exposures of the front cells according to the trapezoid shape or rectangular shape corresponding to 2 mm (Fig. 2e) or 2.5 mm (Fig. S4†) intervals. In the case of omnidirectional incident light, the reflection effect through the intervals was maximized, even though the 3D tetragon tessellated bifacial module with 2- or 2.5 mm intervals generated a high specific power (Fig. 2g). Owing to the pot-like shape of the tetragons, at a 2 mm interval, the use of reflected, scattered, and diffused light by the module was maximized by the reflection of incident light through the intervals between the unit cells.

In the 3D tessellated bifacial module with a slightly larger interval between cells, extra energy was obtained from the incident light passing through the intervals between the unit cells. Under the nonreflected condition, the specific power increased substantially as a result of the direct illumination of the back side of the cells, as shown in Fig. S5† (3D arch) and Fig. S6† (3D tetragon). Owing to the effect of reflection, the bifacial gain was increased by the incident light passing through the unit-cell intervals. Thus, almost the entire back side of the cells could produce energy. In the case of the 3D arch tessellated bifacial module, energy production from the reflected, scattered, and diffused light increased by 5.4% compared with the nonreflected conditions in the same space-restricted installation area and with the same module. If the interval between cells is increased to 2.5 mm, incident light passing through the intervals demonstrates the opposite effect of escaping through the other side of the intervals, resulting in a diminished performance at an AOI of 70°, as shown in the photographs shown in Fig. S5.† A comparison of the energy produced during the course of a day shows that, at low incident light intensity (i.e., an AOI of 30° or greater), the 3D arch tessellated bifacial module demonstrates a better performance than a flat module that was also considered and consisted of 6-cells installed in the same area as the other type of 3D tessellated bifacial module and had a flattened shape owing to being placed on the ground (Fig. S7†). To determine the detailed performance of the module, the relationship between the current density and voltage is shown in Fig. S8† according to the vertical directional incident light (AOI 0°) and AOI 50°. The normalized accumulated specific power in a space-restricted installation area (Fig. S9†) shows that the 3D arch tessellated bifacial module exhibits 41% greater energy production than the flat module because of the reflection effect. In the case of the 3D tetragon tessellated bifacial module, because of the reflection effect, energy production is improved by 8.09% (interval of 2 mm) or 7.97% (interval of 2.5 mm) compared with the nonreflected conditions in the same restricted installation area and with the same module (Fig. S10†). We expect that, with further development, the pot-shaped module could demonstrate greater energy production than the simple 3D tessellated bifacial module. The energy production of the 3D tetragon tessellated bifacial module during the course of a day is shown in Fig. 2h. Similar to the 3D arch tessellated bifacial module, at a low incident light intensity and an AOI of 27° or greater, the 3D tetragon tessellated bifacial module demonstrates a better performance than a flat module. The normalized accumulated specific power results (Fig. S11†) show that, in the same restricted installation area, the 3D tetragon tessellated bifacial module shows 37% greater energy production than a flat module with the reflection effect.

For the same flat tessellated module with a 2 mm interval between cells, the 3D shape can be varied from an arch to a dome shape (Fig. 3a). Even with the same interval between cells in the same space-restricted installation area, the modules with different 3D shapes demonstrate different results for the bifacial gain. Depending on the 3D shape, the front side cell area is exposed directly to incident light and the back side cells are exposed to reflected, scattered, and diffused light from the reflector (bottom photographs shown in Fig. 3a) when the AOI is 40°. The performance of various 3D tessellated bifacial modules is compared in Fig. 3b. The 3D hexagon and dome shapes demonstrate the best specific power results under vertical directional incident light and also demonstrate a high reflection effect for both vertical and omnidirectional incident light. However, under high-AOI conditions, the specific power decreases substantially because of the small exposed area of the front cells. The module with a 3D tetragon shape shows low specific power under vertical directional incident light. However, at low AOIs from 20° to 70°, it shows a strong reflection effect. In contrast, for the modules with a 3D arch and triangle shapes, although the reflection effect is not large, the exposed area of their front side cells under incident light is large at high AOIs, resulting in a high performance. The energy production of the modules with various 3D tessellated bifacial geometries observed over the course of a day is shown in Fig. 3c. Irrespective of the geometry, the 3D tessellated modules show a better performance than a flat module. Considering the features of an urban environment, we expect that the energy yield can be maximized by optimizing the module geometry, the interval through which the incident light passes through the spaces between cells, and the reflector materials or geometry of the 3D tessellated bifacial modules. The accumulated specific power results (Fig. 3d) show that, when the reflection effect is disregarded, the overall energy production of the 3D arch and triangle tessellated bifacial modules is higher because these modules have a large exposed front cell area. If the reflection effect is considered, the 3D tetragonal-, hexagonal-, and dome-shaped tessellated bifacial modules with pot-like shapes that trap incident light demonstrate an excellent ability to utilize incident light that passes through the intervals between cells (Fig. S12†).

Fig. 3 (a) Photographs of various 3D tessellated bifacial modules under vertical directional incident light (top) and under omnidirectional incident light at an AOI of 40° (bottom), with an interval of 2 mm and without a reflector. (b) Conversion efficiency, (c) specific power over the course of a day, and (d) the accumulated specific power considering the installation area of a 3D tessellated bifacial module with various 3D shapes with or without a reflector under 1 sun and 1.5 a.m. conditions.

As shown in Fig. 1b, to maximize the energy production by increasing the exposed front cell area while maintaining or increasing the reflection effect, we fabricated a 3D tessellated bifacial module with 2 mm intervals that can self-transform according to the position of the sun. For this technology, we used an SMA actuator that transforms into its memorized shape when subjected to heat. At both ends of the tessellated bifacial module, the SMA was attached using heat-radiating double-sided tape. The incident light on the surface of the cells increased their temperature, and the heat generated by the increase in surface temperature was transferred to the entire SMA area through heat-radiating double-sided tape (Fig. 4a). The SMA surface temperature was measured to confirm whether the surface heat on the cells was evenly transferred to the surface of the SMA during the experiment (Fig. 4b). Measurements of the temperature from the center of the cell to the center of the 10 mm interval between two cells (Fig. S13†) show that the temperature at the center of the interval was 39.5 °C and that the temperatures of the other parts were greater than 41.5 °C. Thus, the heat was equally transferred to the entire SMA area. To confirm the possibility that the shape of the SMA changed in response to the heat transferred from the solar cell surface, the transition temperature of the actuator was measured through differential scanning calorimetry (DSC) analysis (Fig. S14†). The austenite start temperature (A_s) was 20.78 °C, the austenite finishing temperature (A_f) was 31.21 °C, and the martensite start temperature (M_s) was 7.74 °C during cooling. Thus, the SMA demonstrates an elastic character at temperatures less than 20.78 °C (A_s) and, at higher temperatures, the SMA stiffens and returns to its memorized shape. At temperatures greater than 31.21 °C, the SMA completely transforms into its memorized shape. Thus, the heat transferred through the heat-radiating double-sided tape is sufficient to transform the SMA integrated into the tessellated bifacial module to its memorized shape under light irradiation. The self-shape-transformation to the memorized shape was confirmed using three bifacial cells connected at 10 mm intervals and subjected to 1 sun illumination (Fig. S15 and Movie S1†).

Fig. 4 (a) Schematic diagram of the heat transfer mechanism of heat generated at the surface of a cell, passed through the heat-radiating double-sided tape, and transferred to the SMA. (b) Surface temperature of the SMA attached to the surface of the bifacial cells with an interval of 10 mm using heat-radiating double-sided tape. The location of the measurement is shown in Fig. S8.† (c) Photographs of the front side (left) and back side (right) of a tessellated bifacial module with fixed rods at the edge part, which enable the module to rotate freely enabling transformation of its shape. Schematic diagrams of (d) the elastic backbone strip with an embedded microspring that can deform and recover to its original shape reversibly, and (e) the 3D tetragon tessellated bifacial module moving reversibly in response to the SMA transformation and the elastic behavior of the backbone. (f) Photographs of completely transformed 3D tetragon (left) and arch (right) tessellated bifacial modules under omnidirectional illumination at an AOI of 40°.

For the SMA-integrated 3D tessellated bifacial module (Fig. 4c), to follow the shape transformation during the change of the SMA to its memorized shape, rods were attached to both ends of the module, allowing it to rotate freely, thereby enabling the 3D tessellated bifacial module itself to change shape within the space-restricted installation area. Under the nonilluminated state (dark state), the 3D tessellated bifacial module should recover to its original 3D module shape. Thus, a microspring that can deform and recover to its original shape was embedded in the soft silicone elastic in the strip form. This strip was attached as a backbone to the back side of the 3D tessellated bifacial module (Fig. 4d and S16†). A self-shape-transformable 3D tessellated bifacial module with the SMA, rod, and elastic backbone-embedded microspring was designed and fabricated. The 3D tessellated bifacial module automatically and freely changed shape according to the position of the sun illuminating the space-restricted installation area, and the original shape was recovered in the dark state (Fig. 4e). Fig. 4f shows the final transformed 3D tetragon (left side) and arch (right side) tessellated bifacial modules at an AOI of 40°. Both the 3D tessellated modules transformed automatically in response to a change in the illumination position (Fig. S17, Movies S2 and S3†), and the exposed area of the front side cells increased as a result of the shape transformation. In the case of 3D tetragon tessellated bifacial modules, the module shape transformed, maintaining a light-trapping structure similar to a dome, therefore we also expected to observe contributions from the effect of reflection.

To confirm the performance of the self-shape-transformable 3D tessellated bifacial module, we measured the power output of the SMA-integrated module as a function of the AOI. In the case of the 3D tetragon tessellated bifacial module, under vertical directional incident light, the tetragon shape was transformed to the pot-like shape with a low height, thus, the specific power and the effect of reflection significantly increased compared with those associated with the original shape (Fig. 5a and S18†). In the case of the 3D arch tessellated bifacial module, the power output increased compared with that of the original module performance as a result of the expansion of the exposed front cell area as the front parts became flattened by the integrated SMA (Fig. 5b and S19†). The results showing the energy yield over the course of a day also indicate the high performance under omnidirectional incident light (i.e., the light conditions encountered in an urban environment) (Fig. 5c and d). The accumulated specific power results (Fig. 5e and f) show that the power output of the 3D tetragon tessellated bifacial module and the 3D arch module increased by 45.4% and 47.1%, respectively, compared with that of the flat module. Owing to the integrated SMA actuator, the power output of the 3D tetragon tessellated bifacial module and the 3D arch module increased by 8.69% and 5.95%, respectively, compared with the outputs of the corresponding modules with the integrated SMA. The energy yield achieved with the simple 3D tessellated bifacial module was amplified by simply controlling the intervals between the cells and installing a reflector.

Fig. 5 (a) and (b) Conversion efficiency, (c) and (d) specific power over the course of a day, and (e) and (f) accumulated specific power considering the installation area of the self-shape-transformable 3D tetragon (left) and arch (right) tessellated bifacial modules with an integrated SMA actuator with and without a reflector under 1 sun, 1.5 a.m. conditions.

Self-shape-transformable 3D tessellated bifacial modules can enhance the energy yield and generate additional energy by reflection compared with the energy yield and energy output of a flat module. This proposed concept can be applied to any curved surface as a custom-fit device, and the power output can be maximized through integration of a self-shape-transformation function with an actuator. Thus, the proposed method is applicable to both building integrated PVs (BIPVs) and portable-device-integrated PVs. Utilizing this concept, all-parallel connections are used to overcome shading and mismatch problems associated with 3D geometries. The approach proposed here enables the development of solar cells with a 3D structure for enhanced energy yields and improved aesthetics.

Conclusions

In this study, a novel concept for a self-shape-transformable 3D tessellated bifacial module was introduced as a new paradigm for customized-fit solar cells with a high energy yield. A tessellation structure was introduced for compatibility with any curved surface, and the proposed module could be fabricated with 3D shapes. A fixed 3D arch tessellated bifacial module with a 2 mm interval between the unit cells showed a 41% greater energy production and a fixed 3D tetragon tessellated bifacial module with 2 mm intervals and showed an enhanced energy production of 37% compared with a flat module in a space-restricted installation area. In addition, to maximize the energy output, an SMA actuator was integrated into a 3D tessellated bifacial module, in which the SMA transformed to its memorized shape under heat stimulus. A 3D tessellated bifacial module, which includes an SMA actuator, rod, and elastic backbone-embedded microspring was designed and fabricated. This module could freely change shape in response to a change in the position of the sun illuminating it and could recover its original shape under the dark state in a space-restricted installation area. The power output of the 3D tetragon tessellated bifacial module and the 3D arch module increased by 45.4% and 47.1%, respectively, compared with that of a flat module. Upon integration of the SMA actuator, the power output of the 3D tetragon tessellated bifacial module and the 3D arch module increased by 8.69% and 5.95%, respectively, compared with the power output of the corresponding modules with the original fixed shape.

Experimental details

Fabrication of the self-shape-transformable tessellated bifacial module

For the tessellated bifacial module, bifacial passivated emitter rear cells (Bifacial-PERC; LWM5BB-BiFi-SE-223, Lightway) were used after being cut into rectangular shapes (31 × 10 mm) using a fiber laser. In the case of a static 3D tessellated bifacial module, 12 tessellated units were arranged with various intervals (1, 2 and 2.5 mm) on a flexible backbone (polyethylene terephthalate (PET) film). For the flat module, six tessellated units with a flattened shape were arranged with 1 mm intervals and placed on the ground. For the dynamic 3D tessellated bifacial module and self-shape-transformable tessellated bifacial module, six tessellated units were arranged with uniform 2 mm intervals on the backbone strip fabricated using a 3 mm-wide PDMS (Sylgard 184 A/B, Heesung STS) strip with an embedded microspring (fabricated from Cu wire by MicroSpring Company). Bifacial tessellated unit cells were interconnected using an all-parallel connection with commercial Cu wire (100 μm) and were soldered using Pb-free wire (HSE-02-SR34, Heesung Material) and a soldering iron (FX-951, Hakko). After the electrical connections were completed, the SMA components (Nitinol Flat Wires, Kellogg's Research Labs) were attached to the surface of the tessellated bifacial module using heat-radiating double-sided tape (LG Chemicals). After fabrication of the tessellated bifacial module with intervals between the cells, the module was encapsulated via a casing process using PDMS (Sylgard 184 A/B, Heesung STS). The cured and completely fabricated tessellated bifacial modules were formed into various 3D shapes by fixing the rods in a frame simulating a space-restricted installation area. This installation area is the base area described in Fig. S20† and it was applied to all types of 3D tessellated bifacial modules including the flat module.

Characterization

DSC analysis (Q600, TA Instruments) was conducted from 0 °C to 100 °C at a heating/cooling rate of 10 °C min⁻¹ with the sample under a N₂ atmosphere to characterize the thermal properties of the SMA. The temperatures of the surface of the SMA on the bifacial solar cell and in the intervals between cells were measured using a thermometer (mini Logger GL840, Graphtec). For measurement of the photovoltaic performance, we first calibrated a solar simulator (Sun 2000, 1000 W Xe source; Abet Technologies; 2400 Keithley source meter) using a KG-3 filter and reference cell certified by the National Renewable Energy Laboratory and set the simulator to 1 sun, 1.5 a.m. For the photovoltaic performance as a function of the AOI, a 3D tessellated bifacial module was fixed on the stage of an angle controller, in which the stage was tilted in increments of 10° by a motor that controlled the tilting angle.

Author contributions

This research was supported. M. J. Y. and S. I. C. suggested the self-transformation module concept by utilizing bifacial solar cell and shape memory alloys as actuators, conducted experimental details and wrote the manuscript. Y. H. S. contributed to the analysis of the performance of self-tracking tessellated solar cell module and provided feedback for module encapsulation and connection. D. Y. L. contributed to the analysis of the shape formation of the shape memory alloy and surface temperature of the silicon solar cell. All authors have reviewed the manuscript and agreed regarding its submission.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was supported by the Korea Electrotechnology Research Institute (KERI) Primary Research Program through the National Research Council of Science & Technology (NST) funded by the Ministry of Science and ICT (MSIT) (No. 21A01007).

References

1. C. Erban, KWH output is not the main objective of photovoltaic, presented at 26th European Photovoltaic Solar Energy Conference, EU PVSEC, Hamburg, Germany, 2011.
2. W. C. Sinke, Renewable Energy, 2019, 138, 911–914.
3. C. Ballif, L. Perret-Aebi, S. Lufkin and E. Rey, Nat. Energy, 2018, 3, 438–442.
4. B. P. Jelle, Energies, 2016, 9, 21.
5. T. E. Kuhn, C. Erban, M. Heinrich, J. Eisenlohr, F. Ensslen and D. H. Neuhaus, Energy Build., 2020, 110381.
6. P. K. Nayak, S. Mahesh, H. J. Snaith and D. Cahen, Nat. Rev. Mater., 2019, 4, 269–285.
7. R. Guerrero-Lemus, R. Vega, T. Kim, A. Kimm and L. Shephard, Renewable Sustainable Energy Rev., 2016, 60, 1533–1549.
8. ITRPV, International Technology Roadmap for Photovoltaic 2020, 2020.
9. R. Kopecek and J. Libal, Energies, 2021, 14, 2076.
10. T. S. Liang, M. Pravettoni, C. Deline, J. S. Stein, R. Kopecek, J. P. Singh, W. Luo, Y. Wang, A. G. Aberle and Y. S. Khoo, Energy Environ. Sci., 2019, 12, 116–148.
11. A. Aronescu and J. Appelbaum, Energies, 2021, 14, 2245.
12. S. Cheng, Curved photovoltaic surface optimization for BIPV: an evolutionary approach based on solar radiation simulation. 2009.
13. F. Kassaei, R. Rafiei and F. Torabi, Environ. Prog. Sustainable Energy, 2021, 40, e13513.
14. S. Neven-du Mont, M. Heinrich, A. Pfreundt, C. Kutter, A. Tummalieh and H. Neuhaus, Energy Yield Modelling of 2D and 3D Curved Photovoltaic Modules, presented at Presented at the 37th European PV Solar Energy Conference and Exhibition, 2020.
15. N. El-Atab, W. Babatain, R. Bahabry, R. Alshanbari, R. Shamsuddin and M. M. Hussain, ACS Appl. Mater. Interfaces, 2019, 12, 2269–2275.
16. N. El-Atab, N. Qaiser, W. Babatain, R. Bahabry, R. Shamsuddin and M. M. Hussain, MRS Commun., 2020, 10, 391–397.
17. A. Amelia, Y. Irwan, I. Safwati, W. Leow, M. Mat and M. S. A. Rahim, Management Science and Engineering, 2020, 767, 012052.
18. C. Lee, P. Chou, C. Chiang and C. Lin, Sensors, 2009, 9, 3875–3890.
19. H. Mousazadeh, A. Keyhani, A. Javadi, H. Mobli, K. Abrinia and A. Sharifi, Renewable Sustainable Energy Rev., 2009, 13, 1800–1818.
20. G. Florian, A. R. Gabor, C. Nicolae, A. Rotaru, C. A. Marinescu, G. Iacobescu, N. Stănică, S. Degeratu, O. Gîngu and P. Rotaru, J. Therm. Anal. Calorim., 2019, 138, 2103–2122.
21. H. Jin, E. Dong, G. Alici, S. Mao, X. Min, C. Liu, K. Low and J. Yang, Bioinspiration Biomimetics, 2016, 11, 056012.
22. H. Rodrigue, W. Wang, B. Bhandari, M. Han and S. Ahn, Int. J. Precis. Eng., 2014, 1, 153–156.
23. M. J. Yun, Y. H. Sim, S. I. Cha and D. Y. Lee, Prog. Photovoltaics, 2020, 28, 372–381.
24. M. J. Yun, Y. H. Sim, D. Y. Lee and S. I. Cha, RSC Adv., 2020, 10, 12415–12422.
25. M. J. Yun, Y. H. Sim, D. Y. Lee and S. I. Cha, RSC Adv., 2020, 10, 34837–34846.
26. S. Olalla, J. M. Manuel, M. Tiago, C. João, N. Daniela, A. Hugo, F. Elvira and M. Rodrigo, Sol. Energy, 2020, 196, 92–98.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1se01205c

---