January 31, 2020.
The origin of solar cell was dated back to around 1839 when the French physicist Alexandre Edmond Becquerel first experimentally demonstrated the photovoltaic effect in an electrode in conductive solution exposed to light. Around 50 years later, in 1883, Charles Fritts – an American inventor developed a solar cell using selenium on a thin layer of gold to form a device giving less than 1% power conversion efficiency (PCE). Solar cell technology received a tremendous development and improvement in materials research when the need of more energy increases around 1975 as shown in the development of PCE chart in Figure 1. The solar cells made from crystalline silicon (Si) seem to have higher PCE than others. At present, the Si wafers used in solar cell industry come from the same source as those used in chip industry. There are many steps required to purify the Si to be used in high efficiency devices. The thickness of Si wafer used in the photovoltaic panel is approximately 0.2 mm or less. This is still a lot of starting material of Si to be used in the devices. Thus, researches in thin film semiconducting materials (with thickness approximately 0.002 – 0.005 mm) gained attentions to be used as photon absorber layers for solar cell. These materials are, for examples, CdTe or CuIn1-xGaxSe2 (CIGS). In this report, we will pay our attention to CIGS which is a semiconductor whose band gap energy can be adjusted between 1.0 – 1.7 eV depending upon the amount of Ga. As of 2020, the world record PCE of CIGS is approximately 23% (lab scale devices). It can be noticed in Figure 1 that it took around 40 years until the PCE of these devices are respectable. However, there is one kind of solar cell known as perovskite solar cells (PSCs – see black line in Figure 1) that it has a very rapid development of PCE that reaches 25% in 7 – 8 years. PSCs received a lot of attentions in world-wide research including in Thailand. There are more than 10 research groups in many university and institutions, including those supported by Thailand Center of Excellence in Physics, in Thailand working in various areas of PSCs. However, some problems in PSCs still persist, for example, (i) starting materials are both organic and inorganic, (ii) fabrication processes are mostly in glovebox that the control of oxygen and humidity is required, (iii) instability of PCE, (iv) devices are sensitive to temperature and humidity and (v) large band gap energy (1.5 – 1.6 eV) – lost in long wavelengths, etc.
Those solar cells described above are single junction devices whose theoretical limit of PCE known as Shockley – Quiesser limit does not exceed 30%. The top PCE of Si and CIGS has been stable for quite a while. To increase 0.5 – 1.0% of PCE in these devices is considerably difficult. It needs some breakthrough in the fabrication or material processing in order to achieve higher PCE. One can, however, bring the strong points of either Si or CIGS together with PSC in the tandem structure or multi-junction solar cell, the PCE of Si or CIGS can be enhanced. In the tandem structure, the cell on the top (top cell), such as PSC, is the solar cell with larger band gap, while the cell at the bottom (bottom cell), such as Si or CIGS is the solar cell with smaller band gap as shown in Figure 2. The light with longer wavelength passes to the bottom cell, while the shorter wavelength is captured by the top cell. Thus the tandem solar cell tends convert the light with wider range of the light spectrum than that in the single junction solar cell.
Fig. 1 Power conversion efficiency development of various solar cells. 
Fig. 2 Two tandem structure solar cells; (a) two junctions four terminals, (b) two junctions two terminals. 
In this article, we only focus on the bottom cell made from CIGS. The CIGS thin film solar cells generally comprise 5 layers of materials on the soda-lime glass (SLG) substrate with the structure of SLG-substrate / Mo / CIGS / CdS / i-ZnO / ZnO(Al) / Al-grid as shown in Figure 3. The typical fabrication process of CIGS solar cells requires the use of vacuum deposition chambers. The Mo layer acting as the back electrical contact is deposited by the DC magnetron sputtering. The typical thickness of this layer is around 500 – 600 nm. It is then followed by the deposition of CIGS of approximately 2 micron (μm) thick. The deposition of CIGS layer is done by thermal co-evaporation of Cu, In, Ga and Se under ultra-high vacuum (10-10 Torr). Next, a 50 nm thick of CdS is coated on the CIGS surface by chemical bath deposition method. Then, 50 nm thick and 200 nm thick of i-ZnO and ZnO(Al) are deposited by RF magnetron sputtering, respectively. Finally, 2 μm of Al-grid is thermal evaporated as the front metal contact for the device. The example of CIGS thin film solar cell fabricated at the Semiconductor Physics Research Laboratory, Chulalongkorn University with PCE in the range of 17% is shown in Figure 4. The current-density (J) – voltage (V) curve is illustrated in Figure 5.
Fig. 3 Structure of CIGS solar cell.
Fig. 4 Photograph of 3cm x 3cm CIGS solar cells.
Fig. 5 Current-density – Voltage curves of CIGS solar cells (blue line) and perovskite solar cells (red line). Both solar cells were fabricated at Semiconductor Physics Research Laboratory, Chulalongkorn University.
For the perovskite solar cell or PSC, it is made from the materials in the form of ABX3 and acts as the photon absorber layer of the solar cell. A is generally organic material such as methylammonium (MA), B is a metal such as Pb or Sn and X is halide such as Cl, Br or I. The common material widely used in the fabrication of PSC is MAPbI3. The structure of PSC consists of an electron transport material (ETM, n-type) and hole transport material (HTM, p-type) with the perovskite material in between the ETM and HTM layers. The FTO coated glass is commonly used as the substrate for the PSCs.
Fig. 6 Scanning electron micrograph showing cross-section image of various thin film layers in perovskite solar cell (n-i-p structure).
Fig. 7 Photographs of perovskite solar cells on 3cm x 3cm substrates.
There are two main structures of PSCs depending the on the order of ETM and HTM layers. For example, when the HTM layer is on top of the perovskite layer and the ETM layer is underneath the perovskite layer, this is known as the direct structure or n-i-p structure. On the other hand, when the HTM and ETM layers are switched, it is known as the invert structure or p-i-n structure. For the MAPbI3 material, the band gap energy is approximately 1.5 – 1.6 eV. When the PSC is shone with the visible light with the wavelength in range 400 – 700 nm, electron – hole pairs are created. Electrons and holes are swept to the ETM and HTM, respectively, to the electrical contacts of the device. In general, the spin coating technique  is commonly used for the fabrication of lab-scale PSCs with the substrate size around 2 – 3 cm on a side. The active area of the device is approximately less than 1 cm2. The cross-section image of the n-i-p structure PSC as seen by the scanning electron microscope is illustrated in Figure 6. The MAPbI3 PSCs fabricated at the Semiconductor Physics Research Laboratory, Chulalongkorn University with PCE in the range of 18 – 19% is shown in Figure 7 whose J-V curve is compared with that of CIGS in Figure 5.
Fig. 8 Structure of CIGS – PSC tandem solar cell. 
For the design of CIGS – PSC tandem solar cell, the PSC is the top cell while the CIGS is the bottom cell as shown in Figure 8. This structure is the two-junction and two-terminal tandem structure. The fabrication process of the CIGS bottom cell is similar to what is previously described above without Al-grid. The p-i-n type PSC is then deposited on top of the CIGS. In this case, the HTM is made of PEDOT:PSS and the ETM is made of PCBM. The fabrication of PSC is still possible by the spin-coating technique that the underlying CIGS layer can withstand. The ZnO nanoparticle is coated on top of PCBM prior to the deposition of ZnO(Al) or AZO by the RF magnetron sputtering. The front contact can be made of either Al or Ag. However, the design and fabrication of tandem solar cell is complicated in the sense that the solar cell parameters of the top and bottom cells such as open-circuit voltage (Voc), short-circuit current density (Jsc), fill factor (FF) can be vastly different. The challenge in the design is to match these parameters of the top and the bottom cells such that the loss can be minimized.
In summary, the tandem structure solar cell, either CIGS – PSC or Si – PSC is one of the ways to help increasing the PCE of the existing technology of CIGS or Si solar cells. The is made possible by the help of great advancement in research and development of perovskite materials and fabrication processes compared to other semiconducting materials used for solar cells. However, challenges still remain for PSCs such as sensitivity to humidity and oxygen during fabrication, scalability to large area, stability of PSC under real environment, Pb component in the perovskite materials, etc. These problems, thus, leave us room for research for this material for the future technology.
 Robert F. Service, Science 347, 225 (2015).
 Kim, H., Lee, C., Im, J. et al., “Lead Iodide Perovskite Sensitized All-Solid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%”, Sci Rep 2, 591 (2012) doi:10.1038/srep00591.
 Yoon Hee Jang, Jang Mi Lee, Jung Woo Seo, Inho Kim and Doh-Kwon Lee, “Monolithic tandem solar cells comprising electrodeposited CuInSe2 and perovskite solar cells with a nanoparticulate ZnO buffer layer”, J. Mater. Chem. A, 2017, 5, 19439.
Asst. Prof. Dr. Sojiphong Chatraphorn
Semiconductor Physics Research Laboratory
Department of Physics, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand
Tel. 02-218-7550, FAX. 02-253-1150