Table of contents

Passivated contacts for solar cells can be realized using a variety of differently formed ultra-thin tunnel oxide layers. Assessing their interface properties is important for optimization purposes. In this work, we demonstrate the ability to measure the interface defect density distribution D_it(E) and the fixed interface charge density Q_f for ultra-thin passivation layers operating within the tunnel regime (<2 nm). Various promising tunnel layer candidates [i.e., wet chemically formed SiO_x, UV photo-oxidized SiO_x, and atomic layer deposited (ALD) AlO_x] are investigated for their potential application forming electron or hole selective tunnel layer passivated contacts. In particular, ALD AlO_x is identified as a promising tunnel layer candidate for hole-extracting passivated contact formation, stemming from its high (negative) fixed interface charge density in the order of −6 × 10¹² cm⁻². This is an order of magnitude higher compared to wet chemically or UV photo-oxidized formed silicon oxide tunnel layers, while keeping the density of interface defect states D_it at a similar level (in the order of ∼2 × 10¹² cm⁻² eV⁻¹). This leads to additional field effect passivation and therefore to significantly higher measured effective carrier lifetimes (∼2 orders of magnitude). A surface recombination velocity of ∼40 cm/s has been achieved for a 1.5 nm thin ALD AlO_x tunnel layer prior to capping by an additional hole transport material, like p-doped poly-Si or PEDOT:PSS.

We have fabricated Si-based photonic nanostructures with submicron sizes by the maskless wet etching of Ge quantum dot (QD) multilayers and demonstrated that the photonic nanostructures result in the enhanced optical absorption in the near-infrared light owing to light trapping. In this study, the optical properties of Si-based photonic nanostructures with surface morphology randomness were calculated by the finite-difference time-domain (FDTD) method. The obtained results indicate that as the degree of randomness increased, the absorption in a near-infrared light range enhanced, suggesting that the enhancement of optical absorption in the near-infrared light by photonic nanostructures is due to the randomness of the nanostructures.

The hot carrier (HC) solar cell is one of the most promising advanced photovoltaic concepts. It aims to minimise two major losses in single junction solar cells due to sub-band gap loss and thermalisation of above band gap photons by using a small bandgap absorber, and, importantly, collecting the photo-generated carriers before they thermalise. In this paper we will present recent development of the two critical components of the HC solar cell, i.e., the absorber and energy selective contacts (ESCs). For absorber, fabrication and carrier cooling rates in potential bulk materials — hafnium nitride, zirconium nitride, and titanium hydride are presented. Results of ESCs employing double barrier resonant tunneling structures Al₂O₃/Ge quantum well (QW)/Al₂O₃ and Al₂O₃/PbS quantum dots (QDs)/Al₂O₃ are also presented. These results are expected to guide further development of practical HC solar cell devices.

InGaAs and GaNAs were selected as components of a multiple quantum well (MQW) with a free-barrier conduction band (FB-CB) in which the quantum confinement for electrons was eliminated. Since the calculation demonstrated that the energy gap of a strain-balanced FB-CB InGaAs/GaNAs MQW could be decreased to 1.2 eV with lattice matching to Ge, this structure was expected as a potential absorber of the middle cell of a three-junction solar cell based on the Ge bottom cell. Additionally, the InGaAs/GaNAs MQW could mitigate detrimental impacts of the short lifetime of GaNAs because of the preferential existence of holes in InGaAs, and it can realize more efficient carrier transport than bulk GaInNAs. The time-resolved photoluminescence (TRPL) results demonstrated that the InGaAs/GaNAs MQW cell provided a significantly longer lifetime than the GaInNAs thin-film cell. The open-circuit voltage of the InGaAs/GaNAs MQW cell was superior to that of the GaInNAs thin-film cell.

The impact of ohmic shunts on the current–voltage characteristics of single junction solar cells is well understood. Yet, for monolithic dual-junction solar cells, the effects of shunts have far less been investigated, especially if there is a current mismatch between the two sub cells. In this work, we investigate theoretically and experimentally how current–voltage characteristics of a monolithic GaAs/GaAs tandem solar cell depend on shunts in the top or the bottom sub cell, when either of this sub cells is current limiting. The open-circuit voltage of the device is observed to transition to the open-circuit voltage of the non-shunted sub cell as the shunting increases. In the same way, the fill factor is found to be more significantly affected when ohmic shunts occur in the current-limiting sub cell. Finally, as the current-limiting sub cell is shunted, the short-circuit current of the device is observed to transition to the short-circuit current of the non-current-limiting sub cell. These results allow identifying the shunted sub cell and enable characterization of ohmic shunts under current limiting conditions in dual-junction solar cells.

The effect of substrate orientation on strain relaxation mechanisms of an InGaAs layer grown on vicinal GaAs substrates was investigated by in situ X-ray diffraction (XRD). The crystallographic tilt and indium segregation in the InGaAs layer were altered depending on the miscut direction and angle. In the case of the substrate tilted 6° toward the [110] direction, one type of misfit dislocations was formed preferentially rather than other types, especially in the rapid relaxation phase. While in the case of the substrate tilted 6° toward the $[1\bar{1}0]$ direction, no anisotropies during relaxation were observed. The present finding indicates that the appropriate use of vicinal substrates may lead to a novel method of improving the crystal quality of heterolayers.

We have developed a compact solar-pumped laser (µSPL) employing an off-axis parabolic mirror with an aperture of 76.2 mm diameter and an yttrium aluminum garnet (YAG) ceramic rod of φ1 mm × 10 mm doped with 1% Nd and 0.1% Cr as a laser medium. The laser oscillation wavelength of 1.06 µm, just below the optical absorption edge of Si cells, is suitable for photoelectric conversion with minimal thermal loss. The concept of laser beam power feeding to an electric vehicle equipped with a photovoltaic panel on the roof was proposed by Ueda in 2010, in which the electricity generated by solar panels over the road is utilized to drive a semiconductor laser located on each traffic signal along the road. By substituting this solar-electricity-driven semiconductor laser with a solar-pumped laser, the energy loss of over 50% in converting the solar electricity to a laser beam can be eliminated. The overall feasibility of this system in an urban area such as Tokyo was investigated.

Carrier-selective contacts have recently gained significant interest in the photovoltaic community. Apart from their minority and majority carrier properties, their thermal stability is also important from an application viewpoint. In this paper, we present a detailed study of the thermal stability of WO_x, which is a promising hole-selective contact for silicon wafer solar cells. The film properties are studied after a post deposition annealing in the 200 to 800 °C temperature range. Fourier infrared transmission and X-ray diffraction measurements indicate that WO_x films remain amorphous for annealing temperatures below 300 °C. For higher annealing temperatures, the film crystallises and a reduction in oxygen content is observed after 800 °C post deposition annealing. The resistance of the test structure Al/Si(p)/WO_x/Al decreases rapidly at 600 °C. A minimum resistance of ∼32 mΩ·cm² was achieved after annealing at 700 °C. Photoluminescence imaging indicates that the minority carrier recombination significantly increases for annealing temperatures above 600 °C.

The absorber of the hot carrier solar cell (HCSC) needs to have a considerably reduced hot carrier thermalisation rate, in order to maintain the photo-generated hot carriers for enough time such that they can be extracted. The slow carrier cooling effect is predicted in materials in which the phononic band gap is sufficiently large to block the Klemens decay. Binary compounds with a large mass ratio between the constituent elements are likely to have large phononic band gap. Titanium hydride is one of these binary compounds that has the potential to become an absorber of the HCSC. Whilst a large phononic gap has been observed in stoichiometric TiH₂, it has not been experimentally confirmed for hydrogen deficient TiH_x (where x < 2). In this article, we report the phonon density of states of TiH_1.65 measured using inelastic neutron scattering and presented to clearly show the phononic band gap. We also present the carrier thermalisation process of a TiH_x (1< x <2) thin film by transient absorption, and estimate the carrier cooling time in this material.

Here, we demonstrate the use of an ultrathin TiO₂ film as a passivating carrier-selective contact for silicon photovoltaics. The effective lifetime, surface recombination velocity, and diode quality dependence on TiO₂ deposition temperature with and without a thin tunneling oxide interlayer (SiO₂ or Al₂O₃) on p-type crystalline silicon (c-Si) are reported. 5-, 10-, and 20-nm-thick TiO₂ films were deposited by thermal atomic layer deposition (ALD) in the temperature range of 80–300 °C using titanium tetrachloride (TiCl₄) and water. TiO₂ thin-film passivation layers alone result in a lower effective carrier lifetime compared with that with an interlayer. However, SiO₂ and Al₂O₃ interlayers enhance the TiO₂ passivation of c-Si surfaces. Further annealing at 200 °C in N₂ gas enhances the surface passivation quality of TiO₂ tremendously. From these findings, design principles for TiO₂–Si heterojunction with optimized photovoltage, interface quality, and electron extraction to maximize the photovoltage of TiO₂–Si heterojunction photovoltaic cells are formulated. Diode behaviour was analysed with the help of experimental, analytical, and simulation methods. It is predicted that TiO₂ with a high carrier concentration is a preferable candidate for high-performance solar cells. The possible reasons for performance degradation in those devices with and without interlayers are also discussed.

Transparent conductive oxides (TCOs) have been widely used in various optoelectronic devices. Among these TCOs, indium–tin oxide (ITO) is the most commonly used TCO material. However, owing to the scarcity of indium, there exists a strong need to replace ITO with an alternative transparent conductive coating. A TCO/metal/TCO-based multilayer structure has been considered as one promising candidate. In this work, several Al-doped ZnO (AZO) AZO/Ag/AZO samples were prepared with different Ag thicknesses. The AZO/Ag/AZO structure allows a low sheet resistance of around 10 Ω/sq and a visible transmission above 80% achieved with an overall thickness of ∼110 nm. The optimisation of front AZO thickness helps to reduce reflection via destructive interferences. We demonstrated that the adhesion strength of the stacks can be improved by modifying top AZO deposition conditions. The adhesive tape test confirms good film adhesion (i.e., peel-off strength) to the glass substrate. The annealing studies confirm good thermal stabilities of the fabricated sandwich structure.

In this work, we investigate the impact of light illumination on crystalline silicon surfaces passivated with inline atomic layer deposited aluminum oxide capped with plasma-enhanced chemical vapor deposited silicon nitride. It is found that, for dedicated n-type lifetime samples under illumination, there is no light induced degradation (LID) but enhanced passivation. The lifetime increase happened with a much faster speed compared to the lifetime decay during dark storage, resulting in the overall lifetime enhancement for actual field application scenarios (sunshine during the day and darkness during the night). In addition, it was found that the lifetime enhancement is spectrally dependent and mainly associated with the visible part of the solar spectrum. Hence, it has negligible impact for such interfaces applied on the rear of the solar cells, for example p-type aluminum local back surface field (Al-LBSF) cells.

We present an approach to model the distribution of solar cell efficiencies achieved in production lines based on numerical simulations, metamodeling and Monte Carlo simulations. We validate our methodology using the example of an industrial feasible p-type multicrystalline silicon "passivated emitter and rear cell" process. Applying the metamodel, we investigate the impact of each input parameter on the distribution of cell efficiencies in a variance-based sensitivity analysis, identifying the parameters and processes that need to be improved and controlled most accurately. We show that if these could be optimized, the mean cell efficiencies of our examined cell process would increase from 17.62% ± 0.41% to 18.48% ± 0.09%. As the method relies on advanced characterization and simulation techniques, we furthermore introduce a simplification that enhances applicability by only requiring two common measurements of finished cells. The presented approaches can be especially helpful for ramping-up production, but can also be applied to enhance established manufacturing.

Bifacial photovoltaic (PV) modules make optimal use of diffuse and ground-reflected light. The gain in energy yield depends on both the local climatic conditions and the PV system layout. These determine the additional irradiance on the rear of the PV panels. The rear response of the (laminated) solar cell(s) determines how much additional energy this rear irradiance generates. Based on our experiments and simulations, the main parameters that determine the bifaciality factor of solar cells with a front side junction are the rear metal coverage, the base resistivity and the diffusion profile on the rear. These will be evaluated and discussed in this paper. Front-junction solar cells with low base resistivity have a lower short circuit current when illuminated from the rear due to enhanced recombination in the BSF. Stencil printed rear metallization yields a higher bifaciality factor compared to screen printed by reducing the metal coverage and consumption and maintaining the front side efficiency. For our optimized 239 cm² bifacial cell we estimate that the output with 20% contributed by the rear side is equivalent to that of a 24.4% efficient monofacial cell.

In this work, the effect of an ambient plasma treatment powered by compressed dry air on the passivation quality of silicon wafers coated with intrinsic amorphous silicon sub-oxide is investigated. While long-time storage deteriorates the effective lifetime of all samples, a short ambient plasma treatment improves their passivation qualities. By studying the influence of the plasma treatment parameters on the passivation layers, an optimized process condition was identified which even boosted the passivation quality beyond its original value obtained immediately after deposition. On the other hand, the absence of stringent requirement on gas precursors, vacuum condition and longtime processing makes the ambient plasma treatment an excellent candidate to replace conventional thermal annealing in industrial heterojunction solar cell production.

The effect of positive fixed charge on the recombination rate at SiN_x-passivated p⁺ surfaces is studied in this work. It is shown that a high positive fixed charge on a low defect density, passivated doped surface can result in a near injection level independent lifetime in a certain injection level range. This behaviour is modelled with advanced computer simulations using Sentaurus TCAD, which replicates the measurements conditions during a photoconductance based effective minority carrier lifetime measurement. The resulting simulations show that the shape of the injection level dependent lifetime is a result of the surface recombination rate, which is non-linear due to the surfaces moving into inversion with increasing injection level. As a result, the surface recombination rate switches from being limited by electrons to holes. Equations describing the surface saturation current density, J_0s, during this regime are also derived in this work.

We investigate a novel doping method, catalytic impurity doping (Cat-doping), for application to the fabrication of silicon heterojunction (SHJ) solar cells. Thin n- or p-type doped layers can be formed on intrinsic amorphous Si (a-Si) films by exposing P- or B-related radicals generated by the catalytic cracking of phosphine (PH₃) or diborane (B₂H₆) gas molecules. The passivation quality of underlying a-Si films can be maintained both for phosphorus (P) and boron (B) Cat-doping if we carefully choose the appropriate substrate temperature during Cat-doping. We confirm the rectifying and photovoltaic properties of an SHJ solar cell containing a B Cat-doped layer as a p-type a-Si emitter. These findings suggest the applicability of Cat-doping to SHJ solar cells.

This paper investigates TeO₂, one of the front Ag paste additives, to understand its role in low contact and gridline resistances for screen-printed Si solar cell. It is concluded that TeO₂ aids the reduction of molten glass frit viscosity during contact co-firing. This in turn, leads to uniform flow of molten glass frit, both in the gridline bulk and interface of gridline and SiN_x. Therefore, the uniform wetting and etching of SiN_x and consequently larger contact area of metal to Si compared to its counterpart without TeO₂. Hence, the current transport mechanism from Si to gridline can be said to be both direct and tunneling. The Raman spectra showed a blue shift in the phase of the TeO₂ after contact co-firing in the gridline bulk confirming a crystalline γ-TeO₂.

Excellent c-Si tunnel layer surface passivation has been obtained recently in our lab, using atomic layer deposited aluminium oxide (ALD AlO_x) in the tunnel layer regime of 0.9 to 1.5 nm, investigated to be applied for contact passivation. Using the correspondingly measured interface properties, this paper compares the theoretical collection efficiency of a conventional metal–semiconductor (MS) contact on diffused p⁺ Si to a metal–semiconductor–insulator–semiconductor (MSIS) contact on diffused p⁺ Si or on undoped n-type c-Si. The influences of (1) the tunnel layer passivation quality at the tunnel oxide interface (Q_f and D_it), (2) the tunnel layer thickness and the electron and hole tunnelling mass, (3) the tunnel oxide material, and (4) the semiconductor capping layer material properties are investigated numerically by evaluation of solar cell efficiency, open-circuit voltage, and fill factor.

Amorphous SiO_x was prepared by plasma enhanced chemical vapor deposition (PECVD) to form SiO_x/tungsten-doped indium oxide (IWO) double anti-reflective coatings for silicon heterojunction (SHJ) solar cell. The sheet resistance of SiO_x/IWO stacks decreases due to plasma treatment during deposition process, which means thinner IWO film would be deposited for better optical response. However, the comparisons of three anti-reflective coating (ARC) structures reveal that SiO_x film limits carier transport and the path of IWO–SiO_x–Ag structure is non-conductive. The decrease of sheet resistance is defined as pseudo conductivity. IWO film capping with SiO_x allows observably reduced reflectance and better response in 300–400 and 600–1200 nm wavelength ranges. Compared with IWO single ARC, the average reflection is reduced by 1.65% with 70 nm SiO_x/80 nm IWO double anti-reflective coatings (DARCs) in 500–1200 nm wavelength range, leading to growing external quantum efficiency response, short circuit current density (J_sc), and efficiency. After well optimization of SiO_x/IWO stacks, an impressive efficiency of 23.08% is obtained with high J_sc and without compromising open circuit voltage (V_oc) and fill factor. SiO_x/IWO DARCs provide better anti-reflective properties over a broad range of wavelength, showing promising application for SHJ solar cells.

Combined with advanced crystal growth technology and reduced dislocation densities, the higher tolerance to metal contamination of n-type silicon makes n-type cast-grown silicon a potential option for low cost high quality substrates for solar cells. Using a combination of photoconductance based lifetime testing and photoluminescence imaging, we have investigated the carrier lifetime in wafers from the bottom, middle, and top parts of a n-type high-performance multicrystalline (HPM) silicon ingot, and wafers from n-type mono-like silicon ingots after each high temperature solar cell processes, including after boron diffusion, phosphorus diffusion, and hydrogenation. Although boron diffusion leads to a degradation of the sample lifetime, phosphorus diffusion and hydrogenation is effective at recovering the lifetime in the intra-grain region and at the grain boundaries respectively. Quasi-steady-state photoconductance (QSSPC) measurements show that the arithmetic average lifetime of HPM silicon wafers and mono-like silicon wafers can reach up to 1.8 and 3.3 ms respectively for a process sequence including a boron diffusion, with corresponding implied open circuit voltage of about 720 mV. If the boron diffusion can be avoided, average lifetimes up to 3.0 and 6.6 ms can be achieved respectively, highlighting the excellent potential of n-type cast-grown materials.

In this work, we demonstrated an enhanced surface passivation of epitaxially grown boron-doped Si emitters by replacing thermal SiO₂ as a passivation layer employed in a 15.9% efficient 21-µm Si solar cell (88 cm²) on stainless steel with a remote-plasma atomic layer deposition (ALD) of an Al₂O₃ film. A thin Al₂O₃ film deposited by remote-plasma ALD was very effective at reducing the emitter saturation current density (J_0e) of epitaxial p⁺-emitter to 16.2 fA/cm², compared to the J_0e of 184.9 fA/cm² by thermal SiO₂ films. This reduction in J_0e enables an increase in an implied open-circuit voltage (iV_oc) from 630 to 688 mV. Quokka simulation shows that about a 1.1% absolute efficiency increase in the calculated baseline efficiency of a 15.1% of the ultrathin Si solar cell can be achievable by enhancing emitter surface passivation without changing the concentration in either the epi-emitter or epi-base. Finally, our results show that a high efficiency of 17.3% can be reached from the calculated baseline efficiency of 15.1% using the optimized conditions of an epitaxially grown emitter in combination with increasing the base doping concentration and improved base recombination lifetime.

Silicon nitride (SiN_x) synthesised by low-temperature plasma enhanced chemical vapour deposition (PECVD) is the most extensively used antireflection coating for crystalline silicon solar cells because of its tunable refractive index in combination with excellent levels of surface and bulk passivation. This has attracted a significant amount of research on developing SiN_x films towards an optimal electrical and optical performance. Typically, recipes are first optimised in lab-scale reactors and subsequently, the best settings are transferred to high-throughput reactors. In this paper, we show that for one particular, but widely used, PECVD reactor configuration this upscaling is severely hampered by an important experimental artefact. Specifically, we report on the unintentional deposition of a dual layer structure in a dual mode AK 400 plasma reactor from Roth & Rau which has a significant impact on its surface passivation performance. It is found that the radio frequency (RF) substrate bias ignites an unintentional depositing plasma before the ignition of the main microwave (MW) plasma. This RF plasma deposits a Si-rich intervening SiN_x layer (refractive index = 2.4) while using a recipe for stoichiometric SiN_x. This layer was found to be 18 nm thick in our case and had an extraordinary impact on the Si surface passivation, witnessed by a reduction in effective surface recombination velocity from 22.5 to 6.2 cm/s. This experimental result may explain some "out of the ordinary" excellent surface passivation results reported recently for nearly stoichiometric SiN_x films and has significant consequences when transferring these results to high-throughput deposition systems.

We report on the progress for the understanding of carrier-induced degradation (CID) in p-type mono and multi-crystalline silicon (mc-Si) solar cells, and methods of mitigation. Defect formation is a key aspect to mitigating CID. Illuminated annealing can be used for both mono and mc-Si solar cells to reduce CID. The latest results of an 8-s UNSW advanced hydrogenation process applied to industrial p-type Czochralski PERC solar cells are shown with average efficiency enhancements of 1.1% absolute from eight different solar cell manufacturers. Results from three new industrial CID mitigation tools are presented, reducing CID to 0.8–1.1% relative, compared to 4.2% relative on control cells. Similar advanced hydrogenation processes can also be applied to multi-crystalline silicon passivated emitter with rear local contact (PERC) cells, however to date, the processes take longer and are less effective. Modifications to the firing processes can also suppress CID in multi-crystalline cells during subsequent illumination. The most stable results are achieved with a multi-stage process consisting of a second firing process at a reduced firing temperature, followed by extended illuminated annealing.

The surface passivation performance of atomic layer deposited ultra-thin aluminium oxide layers with different thickness in the tunnel layer regime, i.e., ranging from one atomic cycle (∼0.13 nm) to 11 atomic cycles (∼1.5 nm) on n-type silicon wafers is studied. The effect of thickness and thermal activation on passivation performance is investigated with corona-voltage metrology to measure the interface defect density D_it(E) and the total interface charge Q_tot. Furthermore, the bonding configuration variation of the AlO_x films under various post-deposition thermal activation conditions is analyzed by Fourier transform infrared spectroscopy. Additionally, poly(3,4-ethylenedioxythiophene) poly(styrene sulfonate) is used as capping layer on ultra-thin AlO_x tunneling layers to further reduce the surface recombination current density to values as low as 42 fA/cm². This work is a useful reference for using ultra-thin ALD AlO_x layers as tunnel layers in order to form hole selective passivated contacts for silicon solar cells.

We demonstrate an independently confirmed 25.0%-efficient interdigitated back contact silicon solar cell with passivating polycrystalline silicon (poly-Si) on oxide (POLO) contacts that enable a high open circuit voltage of 723 mV. We use n-type POLO contacts with a measured saturation current density of J_0n = 4 fA cm⁻² and p-type POLO contacts with J_0p = 10 fA cm⁻². The textured front side and the gaps between the POLO contacts on the rear are passivated by aluminum oxide (AlO_x) with J_0AlOx = 6 fA cm⁻² as measured after deposition. We analyze the recombination characteristics of our solar cells at different process steps using spatially resolved injection-dependent carrier lifetimes measured by infrared lifetime mapping. The implied pseudo-efficiency of the unmasked cell, i.e., cell and perimeter region are illuminated during measurement, is 26.2% before contact opening, 26.0% after contact opening and 25.7% for the finished cell. This reduction is due to an increase in the saturation current density of the AlO_x passivation during chemical etching of the contact openings and of the rear side metallization. The difference between the implied pseudo-efficiency and the actual efficiency of 25.0% as determined by designated-area light current–voltage (I–V) measurements is due to series resistance and diffusion of excess carriers into the non-illuminated perimeter region.

This paper presents a comprehensive assessment of the electronic properties of an industrially grown p-type high performance multicrystalline silicon ingot. Wafers from different positions of the ingot are analysed in terms of their material quality before and after phosphorus diffusion and hydrogenation, as well as their final cell performance. In addition to lifetime measurements, we apply a recently developed technique for imaging the recombination velocity of structural defects. Our results show that phosphorus gettering benefits the intra-grain regions but also activates the grain boundaries, resulting in a reduction in the average lifetimes. Hydrogenation can significantly improve the overall lifetimes, predominantly due to its ability to passivate grain boundaries. Dislocation clusters remain strongly recombination active after all processes. It is found that the final cell efficiency coincides with the varying material quality along the ingot. Wafers toward the ingot top are more influenced by carrier recombination at dislocation clusters, whereas wafers near the bottom are more affected by a combination of their lower intra-grain lifetimes and a greater density of recombination active grain boundaries.

The effects of indium tin oxide (ITO) films on the performance of heterojunction silicon wafer solar cells is investigated, using heterojunction (HET) solar cell precursors. Different ITO deposition conditions are used, which result in significant differences in the performance of HET solar cells. It is found that HET solar cells with ITO films deposited at room temperature exhibit severer sputter damage, while those with substrate heating show less damage. Besides the ITO deposition temperature, the sputtering gas ambient is also investigated. The hydrogen gas used in the ITO deposition can greatly affect the interface properties between the ITO film and the amorphous silicon layers. The champion solar cell fabricated under the optimum ITO deposition conditions (a deposition temperature of 150 °C with optimal gas concentration) shows a conversion efficiency of 19.7%.

We examine thermally evaporated MoO_x films as a full-area rear contact to crystalline p-type Si solar cells for efficient hole-selective contacts. Prior to front- and rear-metallization, the implied open-circuit voltage (iV_oc) is evaluated to be 646 mV with implied fill factor (iFF) of 82.5% for the tunnel SiO_x/MoO_x rear contacted cell structure with the passivated emitter on the textured surface, showing it is possible to achieve an implied 1-sun efficiency of 20.8%. Numerical simulation reveals that the electron affinity (χ) of the MoO_x material strongly influences the performance of the MoO_x contacted p-Si cell. Simulated band diagrams show that the values in χ of the MoO_x layer must be sufficiently high in order to lower junction recombination, indicating that the highest efficiency of 21.1% is achievable for a high χ of 5.6 eV of MoO_x films and back surface recombination velocity of <100 cm/s at p-Si/MoO_x.

Flash spectral imaging of full area (156 mm by 156 mm) silicon solar wafers and cells is realized in a setup integrating pseudo-monochromatic LEDs over the wavelength range of 370 to 1050 nm and a high-resolution monochrome camera. The captured information allows the computation of sample reflectance as a function of wavelength and coordinates, thereby constituting a spectral reflectance map. The derived values match that obtained from monochromator-based measurements. Optical inspection is then based on the characteristic reflectance of surface features at optimally contrasting wavelengths. The technique reveals otherwise hidden stains and anti-reflection coating (ARC) non-uniformities, and enable more selective visualization of grains in multicrystalline Si wafers. Optical contrast enhancement of metallization significantly improves accuracy of metal detection. The high effective resolution of the monochrome camera also allows fine metallization patterns to be measured. The rapid succession of flash-and-image-capture at each wavelength makes the reported optical metrology technique amenable in photovoltaic manufacturing for solar wafers/cells sorting, monitoring and optimization of processes.

This paper presents amorphous silicon deposited at temperatures below 200 °C, leading to an excellent passivation layer for boron doped emitter and phosphorus doped back surface field areas in interdigitated back contact solar cells. A higher deposition temperature degrades the passivation of the boron emitter by an increased hydrogen effusion due to lower silicon hydrogen bond energy, proved by hydrogen effusion measurements. The high boron surface doping in crystalline silicon causes a band bending in the amorphous silicon. Under these conditions, at the interface, the intentionally undoped amorphous silicon becomes p-type conducting, with the consequence of an increased dangling bond defect density. For bulk amorphous silicon this effect is described by the defect pool model. We demonstrate, that the defect pool model is also applicable to the interface between amorphous and crystalline silicon. Our simulation shows the shift of the Fermi energy towards the valence band edge to be more pronounced for high temperature deposited amorphous silicon having a small bandgap. Application of optimized amorphous silicon as passivation layer for the boron doped emitter and phosphorus doped back surface field on the rear side of laser processed back contact solar cells, fabricated using four laser processing steps, yields an efficiency of 23.3%.

A back-contact amorphous-silicon (a-Si)/crystalline silicon (c-Si) heterojunction is one of the most promising structures for high-efficiency solar cells. However, the patterning of back-contact electrodes causes the increase in fabrication cost. Thus, to simplify the fabrication of back-contact cells, we attempted to form p-a-Si/i-a-Si/c-Si and n-a-Si/i-a-Si/c-Si regions by the conversion of a patterned area of p-a-Si/i-a-Si/c-Si to n-a-Si/i-a-Si/c-Si by plasma ion implantation. It is revealed that the conversion of the conduction type can be realized by the plasma ion implantation of phosphorus (P) atoms into p-a-Si/i-a-Si/c-Si regions, and also that the quality of passivation can be kept sufficiently high, the same as that before ion implantation, when the samples are annealed at around 250 °C and also when the energy and dose of ion implantation are appropriately chosen for fitting to a-Si layer thickness and bulk c-Si carrier density.

Back-contact silicon heterojunction solar cells with an efficiency of 22% were manufactured, featuring a simple aluminium metallisation directly on the doped amorphous silicon films. Both the open-circuit voltage and the fill factor heavily depend on the parameters of the annealing step after aluminium layer deposition. Using numerical device simulations and in accordance with the literature, we demonstrate that the changes in solar cell parameters with annealing can be explained by the formation of an aluminium silicide layer at temperatures as low as 150 °C, improving the contact resistance and thus enhancing the fill factor. Further annealing at higher temperatures initialises the crystallisation of the amorphous silicon layers, yielding even lower contact resistances, but also introduces more defects, diminishing the open-circuit voltage.

We analyze the formation kinetics of the boron–oxygen defect in compensated n-type upgraded metallurgical-grade (UMG) silicon solar cells. Through time-resolved open-circuit voltage measurements, we explore the influence of temperature, forward bias, and light intensity on the formation kinetics of the defect. Our results confirm that the boron–oxygen defect forms more slowly in compensated n-type silicon than in p-type silicon. We present evidence which suggests that the slower kinetics in n-type silicon may be due to a lower frequency factor for defect formation.

Solar cell simulators based on light emitting diodes (LED) have the potential to achieve a large potential market share in the next years. As advantages they can provide a short and long time stable spectrum, which fits very well to the global AM1.5g reference spectrum. This guarantees correct measurements during the flashes and throughout the light engines' life span, respectively. Furthermore, a calibration with a solar cell type of different spectral response (SR) as well as the production of solar cells with varying SR in between two calibrations does not affect the correctness of the measurement result. A high quality 21 channel LED solar cell spectrum is compared to former study comprising a standard modified xenon spectrum light source. It is shown, that the spectrum of the 21-channel-LED light source performs best for all examined cases.

We have achieved a record high cell efficiency of 20.29% for an industrial 6-in. p-type monocrystalline silicon solar cell with a full-area aluminum back surface field (Al-BSF) by simply modifying the cell structure and optimizing the process with the existing cell production line. The cell efficiency was independently confirmed by the Solar Energy Research Institute of Singapore (SERIS). To increase the cell efficiency, for example, in four busbars, double printing, a lightly doped emitter with a sheet resistance of 90 to 100 Ω/□, and front surface passivation by using silicon oxynitride (SiON) on top of a silicon nitride (SiN_x) antireflection layer were adopted. To optimize front side processing, PC1D simulation was carried out prior to cell fabrication. The resulting efficiency gain is 0.64% compared with that in the reference cells with three busbars, a single antireflection coating layer, and a low-sheet-resistance emitter.

The wafer bonding technique has received much attention for its ability to form highly efficient multijunction solar cells composed of lattice-mismatched materials. Here, we investigated the cause of open-circuit voltage (V_oc) loss in the mechanically stacked InGaP/GaAs//InGaAsP/InGaAs solar cells composed of subcells with unequal areas. The external quantum efficiency and current–voltage characteristics of the proposed solar cells were thoroughly investigated. We clarify that a part of V_oc loss is caused by the measurement artifact that originates from the aperture mask shadow and inherent low shunt resistance of the InGaAsP/InGaAs subcell. We demonstrated that modification of the InGaAsP/InGaAs subcell area resulted in enhanced V_oc.

Cu-doping effects and a CdS_xTe_1−x mixed crystal layer in CdS/CdTe solar cells were investigated on the basis of the photoluminescence (PL) of the CdS/CdTe junction using excitation lights incident on the glass substrate side (junction PL) with various excitation wavelengths. In the Cu-doped CdS/CdTe solar cells, broad emissions at 910–950 nm, which were probably caused by donor–acceptor pair (DAP) emission between Cu_Cd acceptors and Cl_Te donors, were observed. The intensity of the junction PL markedly increased owing to the Cu doping. This result suggests that the intensity of junction PL is relevant to the conversion efficiency of CdTe solar cells. Furthermore, the PL peak energy increased with increasing excitation wavelength. This result indicates that the CdS_xTe_1−x mixed crystal layer is formed in the CdS/CdTe interface, and that the S composition decreased from the CdS/CdTe interface to the rear.

Certified efficiency of 22.3% has been achieved for Cu(In,Ga)(Se,S)₂ solar cell. Compared to our previous record cell with 20.9% efficiency, the major breakthrough is due to the increased V_oc, benefited from potassium treatment. A lower reverse saturation current and a longer carrier collection length deduced from electron-beam induced current indicate that the degree of carrier recombination at the heterojunction and depletion region for the 22.3% cell is lower. Further characterizations (capacitance–voltage profiling, temperature-dependent V_oc, Suns–V_oc) and analysis indicate that the recombination coefficients at all regions were reduced, especially for the interface and depletion regions. Device simulation was performed assuming varying defect densities to model the current–voltage curve for the 22.3% cell. The best model was also used to estimate the achievable V_oc if defect densities were further reduced. Furthermore, by using higher bandgap Cd-free buffer layers, a higher J_sc was achieved which gives an in-house solar cell efficiency of 22.8%. Recombination analysis on the 22.8% cell indicates that the interface recombination is further reduced, but the recombination coefficients at the depletion region was higher, pointing out that further improvement on the depletion region recombination could help to achieve a higher V_oc and therefore an efficiency beyond 23%.

An organic material, 2,2',7,7'-tetrakis(N,N-di-p-methoxyphenylamine)9,9'-spirobifluorene (spiro-OMeTAD), is generally used as a hole conductor of perovskite solar cells (PSCs), but spiro-OMeTAD is much more expensive than other materials used in PSCs. In this study, we have prepared PSCs with a cost-effective CuI hole transport layer by spin coating. The merit of using spin coating for CuI is good compatibility with other steps, such as spin coating of a TiO₂ electron transport layer and a perovskite active layer. The CuI-based PSC recorded power conversion efficiencies of η = 2.22% (max) on the day of production and η = 6.52% (max) after the 20 days of production. Moreover, the CuI-based PSC had a smaller hysteresis than the spiro-based PSC, suggesting that CuI is a highly promising alternative hole conductor for PSCs.

In this work we have studied the feasibility of highly symmetric guanidinium (GA; CH₆N₃⁺) cation in perovskite based solar cells. It is an alternative to the methyl ammonium (MA; CH₃NH₃⁺) or formamidinium (FA; CH(NH₂)²⁺) ions commonly utilized in the perovskite solar cells, may bring additional advantages due to its triad rotational symmetry and zero dipole moment (μ). We noticed that due to steric factors, GA preferably forms two-dimensional (2D) layer, where GA₂PbI₄ is favored monoclinic structure (E_g = 2.5 eV) and the obtained device efficiency is η = 0.45%. This study provides a guideline for designing guanidinium based perovskite absorbers for solar cell devices. In addition, through further compositional engineering with mixed organic cations using GA may enhance the device performance.

Single-junction GaAs solar cells were grown by metalorganic vapor-phase epitaxy (MOVPE) at various input V/III ratios. All growth parameters other than V/III ratio were carefully controlled for an accurate comparison. Nearly identical cell performance characteristics including short-circuit current density (J_sc) and open-circuit voltage (V_oc) indicate that cell performance is independent of V/III ratio. To determine the relationship between the electrically measured V_oc and V/III ratio in a more precise manner, photoluminescence (PL) was applied as a potent optical measurement tool, which does not depend on device processing and contacting issues. We also evaluated the projected cell performance under low-concentration sunlight by electroluminescence (EL) analysis. Similarly to electrical measurement, optical measurement showed no obvious degradation owing to a low V/III ratio. This study strongly demonstrates that low-cost high-efficiency GaAs solar cells can be realized by MOVPE using a low V/III ratio.

To improve the superlattice (SL) solar cell performance, we carried out an accurate estimation of transition energies and miniband widths and focused on understanding of the optical properties of the SL structure using piezoelectric photothermal (PPT), photoreflectance (PR), and photoluminescence (PL) methods. Solar cell structure samples with different barrier thicknesses from 2.0 to 7.8 nm in quantum wells were prepared. From the PR and theoretical calculation, the formation of a miniband was confirmed. The PL peak showed a redshift and a decrease in signal intensity with decreasing barrier thickness, which were explained by carrier separation as a consequence of electron transportation through the miniband without recombination. The PPT signal intensities of the SL were still large even for the 2.0-nm-barrier-thickness sample. It is conceivable that the multiple-phonon emission during carrier transport through the miniband was detected. The usefulness of multidimensional investigation by using the above three methods is clearly demonstrated.

The growth of ternary InGaP alloys is often susceptible to atomic ordering, which leads to an anomalous bandgap reduction as well as the formation of antiphase boundaries (APBs). The effect of substrate miscut on the performance of lattice-matched In_0.52Ga_0.48P solar cells grown on GaAs(001) substrates by solid-source molecular beam epitaxy (SS-MBE) is investigated. A B-type miscut enhanced single-variant atomic ordering even with SS-MBE, resulting in a bandgap (E_g) reduction from 1.87 eV for an alloy grown on an exact substrate to 1.85 eV for that grown on the substrate miscut 6° toward (111)B. Conversely, an A-type miscut suppressed the formation of atomic ordering, resulting in the E_g widening of the alloy grown on the substrate miscut 6° toward (111)A to 1.89 eV. With regard to solar cell performance, InGaP solar cells grown on A-type miscut substrates enhanced the open-circuit voltage (V_OC) and W_OC (= E_g/q − V_OC) because of the low degree of atomic ordering. Large improvements in W_OC and efficiency to 0.58 V and 10.93%, respectively, were obtained for the cell grown on the substrate miscut 2° toward (111)B. A reduction in the number of APBs due to single-variant atomic ordering was related to this latter result.

Ag(In,Ga)Se₂ (AIGS) is one of the promising candidates for the top cell absorber in the tandem structure. However, the conversion efficiency of AIGS solar cells is still lower than that required for the top cell. In this study, to improve the conversion efficiency of AIGS solar cells, we controlled the conduction band offset (CBO) at the buffer layer/ZnO and buffer layer/AIGS interfaces. The reduction in interface recombination at the CdS buffer layer/AIGS interface was achieved by introducing a ZnS(O,OH) buffer layer instead of a CdS buffer layer, although the fill factor (FF) decreased markedly because the CBO at the ZnS(O,OH)/ZnO interface prevented the electron flow under a forward bias. We found that the introduction of a CdS/ZnS(O,OH) hybrid buffer layer is efficient in controlling the CBO at both the buffer layer/AIGS and buffer layer/ZnO interfaces and improving the solar cell conversion efficiency.

The leakage of liquid electrolytes is crucial for the application of dye-sensitized solar cells (DSCs). The ion transport in solid imidazolium iodides used as the quasi-solid state electrolyte (EL) of DSCs was investigated. The imidazolium EL, whose melting point is ca. 100 °C, was chosen to fully penetrate in porous TiO₂ by the heat permeation method. The apparent diffusion coefficients (D) of I₃⁻ in porous TiO₂ are on the order of 1,3-dimethly-imidazolium (DM⁺) > 1-ethyl-3-methylimidazolium > 1,2-dimethyl-3-propylimidazolium > 1-butyl-2,3-dimethylimidazolium. D increases with decreasing imidazolium cation size. Therefore, the effective distance between I⁻ and I₃⁻ was assumed to decrease with decreasing cation size. The transport-limited current of DM⁺ at the TiO₂ thickness of 20 µm is ca. 30 mA·cm⁻². The result suggests that the I₃⁻ transport in DM⁺ can compensate for the large short circuit current (30 mA·cm⁻²).

Several GaAs tunneling junctions (TJs) and p–i–n single junction solar cells grown using a planetary metalorganic vapor phase epitaxy (MOVPE) reactor utilizing various dopant species including Zn, C, S, and Te were investigated. The incorporation of Te atoms into GaAs was approximately two orders larger than that of S atoms. Although only 30% of Te atoms could be electrically activated, a carrier concentration of 10¹⁹ cm⁻³ was achieved. Highly C-doped GaAs was successfully obtained by decreasing the growth temperature and increasing the amount of H₂ carrier gas in order to prevent the predecomposition of CBr₄ dopant gas. A hole concentration of about 10²⁰ cm⁻³ was realized with a growth temperature of 450 °C. The C–Te-doped GaAs TJ exhibited the best ohmic tunneling behavior with a resistivity of 12.5 mΩ·cm², while the others had diode characteristics. The GaAs solar cell grown with the Zn–S dopant showed the highest conversion efficiency ascribed to a longer minority carrier lifetime.

In this work, potassium fluoride (KF)-treated Cu(In,Ga)Se₂ (CIGS) thin films were rinsed in ammonia and water solutions before buffer layer (CdS) deposition and the effects of rinsing on photovoltaic properties were investigated. X-ray photoelectron spectroscopy (XPS) and secondary ion mass spectrometry (SIMS) measurements revealed that sodium atoms out-diffused at the surface region during KF deposition. Water and ammonia rinsing processes of KF-treated CIGS thin films reduced alkali metals from the surface. However, sodium at the Cu-depleted surface layer remained at a high concentration, suggesting the occupation of Cu vacancies with sodium atoms. On the other hand, ammonia rinsing removed the Cu-poor region from the surfaces of KF-treated CIGS thin films affecting the growth (or nucleation) of the CdS layer. The surface coverage of the CdS layer deposited on the ammonia-rinsed KF-treated CIGS thin film was inferior to than that of water-rinsed samples, resulting in the poor cell performance due to an increased interface recombination.

Solar cells operate under various light intensities. In addition, in a tandem solar cell configuration, the bottom cell receives a spectrum that is filtered by the top cell. The performance of the bottom cell at low light intensity is thus of great importance. In this paper, we study various types of silicon solar cells under various light intensities and filtered spectra. We find that, as expected, the short-circuit current varies linearly with the illumination intensity, regardless of the input spectrum. However, we also observe that for a constant number of incident photons but different long-pass filters (i.e., different photogeneration profiles), the short-circuit current density of the solar cells reduces with increasing cut-on wavelength of the used filter. According to our analysis, the loss in short-circuit current can be partly explained by the strong wavelength dependence of the external quantum efficiency at long wavelengths (1050–1200 nm) as well as the different carrier generation profiles under filtered spectra.

Material and structural parameters may affect the efficiency of a tandem solar cell differently from the way they do in a single-junction solar cell. We fabricated a III–V/Si four-terminal tandem solar cell and developed an opto-electronic model simulating this device. The optical properties were simulated with the transfer matrix method, while the electrical properties were simulated using the numerical device simulator PC1D. For this simulated tandem structure, we determined the parameters which have the largest potential impact on the device efficiency. A sensitivity analysis of the impact of these parameters on the device efficiency was also performed. In addition, to reduce the cost of the tandem solar cells, we identified the parameters that do not require tight control during the manufacturing process. The Si cell was simulated both as a single-junction cell and as the bottom cell of a tandem device. Finally, we determined those device parameters that are more critical in a tandem configuration than in a single-junction configuration.

Alkali treatment effects on Cu(In,Ga)Se₂ (CIGS) solar cells deposited on polyimide-coated soda lime glass (PI-coated SLG) were investigated. CIGS on PI-coated SLG shows Na diffusion from the substrate, which should be controlled to obtain high efficiencies. Further incorporation of Na was achieved by enhancing diffusion from the substrate or by external incorporation using post-deposition treatment (PDT) methods. Both methods lead to a high efficiency of approximately 15%. Moreover, aside from Na, K was also incorporated by KF-PDT, resulting in efficiency improvement from 12% for an untreated CIGS to more than 18% at the maximum substrate temperature of 450 °C, which is comparable to CIGS deposited at higher temperatures using the same equipment. It was also found that the alkali concentration of CIGS deposited on PI-coated SLG shows almost the same behavior as that of a film deposited on a rigid glass, suggesting that the deposition technique for CIGS on the rigid glass can be applied to flexible substrates.

Atomic layer deposition of Zn(O,S) is an attractive dry and Cd-free process for the preparation of buffer layers for chalcopyrite solar modules. As we previously reported, excellent cell and module efficiencies were achieved using absorbers from industrial pilot production. These absorbers were grown using a selenization/sulfurization process. In this contribution we report on the interface engineering required to adapt the process to sulfur-free multi source evaporated absorbers. Different approaches to a local sulfur enrichment at the heterojunction have been studied by using surface analysis (XPS) and scanning transmission electron microscopy. We correlate the microstructure and element distribution at the interface with device properties obtained by electronic characterization. The optimized completely dry process yields cell efficiencies >16% and 30 × 30 cm² minimodule efficiencies of up to 13.9% on industrial substrates. Any degradation observed in the dry heat stress test is fully reversible after light soaking.

In perovskite solar cells with an organic inorganic hybrid metal halide perovskite crystalline semiconductor as the active layer, the properties of the n-type semiconductor scaffold, the materials used, and the morphology, wettability, and surface reactivity of the cells are important decisive factors affecting the overall device efficiency of the perovskite solar cells. We control the orientation of anatase titania nanosheets by a self-assembly technique to create the ordered mesoporous scaffolds with ordered voids. Differences between nanosheet orientations in each mesoporous scaffold indicate differences in the photoelectric properties of CH₃NH₃PbI₃ perovskite crystals embedded in each scaffold. Although each scaffold consists of the same anatase TiO₂ nanosheets, the properties of the solar cells are affected by the oxide scaffold nanomorphology, which determines the growth orientation of CH₃NH₃PbI₃ perovskite crystals that affects the solar cell properties.

To develop polycrystalline thin-film tandem solar cells, a SrCuSeF/In₂O₃:Sn (ITO) bilayer film was studied. The transparent p-type conductive SrCuSeF layer was deposited by pulsed laser deposition (PLD), and the n-type conductive ITO layer was deposited by RF sputtering. The SrCuSeF/ITO bilayer film showed ohmic I–V characteristics. A tunnel junction between the p-type SrCuSeF and n-type ITO layers was successfully formed because the p-type SrCuSeF and the n-type ITO layers had sufficiently high carrier concentrations. The SrCuSeF/ITO bilayer film was applied as the back contact of a CdS/CdTe solar cell. The photovoltaic performance of the CdS/CdTe solar cell depends considerably on the thickness of the SrCuSeF layer. The CdTe solar cell with a back contact of the SrCuSeF layer with a thickness of 34 nm and the ITO layer with a thickness of 200 nm showed a high conversion efficiency of 14.3% (V_OC = 804 mV, J_SC = 27.5 mA/cm², and FF = 0.65). The conversion efficiency was much higher than that of the CdTe solar cell with the SrCuSeF single-layer back contact (11.6%) and that of the CdTe cell with the ITO single-layer back contact (2.75%).

A full size TESSERA shade tolerant module has been made and was tested under various shadow conditions. The results show that the dedicated electrical interconnection of cells result in an almost linear response under shading. Furthermore, the voltage at maximum power point is almost independent of the shadow. This decreases the demand on the voltage range of the inverter. The increased shadow linearity results in a calculated increase in annual yield of about 4% for a typical Dutch house.

Some photovoltaic module technologies use toxic materials. We report long-term leaching on photovoltaic module pieces of 5 × 5 cm² size. The pieces are cut out from modules of the four major commercial photovoltaic technologies: crystalline and amorphous silicon, cadmium telluride as well as from copper indium gallium diselenide. To simulate different environmental conditions, leaching occurs at room temperature in three different water-based solutions with pH 3, 7, and 11. No agitation is performed to simulate more representative field conditions. After 360 days, about 1.4% of lead from crystalline silicon module pieces and 62% of cadmium from cadmium telluride module pieces are leached out in acidic solutions. The leaching depends heavily on the pH and the redox potential of the aqueous solutions and it increases with time. The leaching behavior is predictable by thermodynamic stability considerations. These predictions are in good agreement with the experimental results.

Reducing levelized cost of electricity (LCOE) is important for solar photovoltaics to compete against other energy sources. Thus, the focus should not only be on improving the solar cell efficiency, but also on continuously reducing the losses (or achieving gain) in the cell-to-module process. This can be achieved by choosing the appropriate module material and design. This paper presents a detailed and systematic characterization of various photovoltaic (PV) module materials (encapsulants, tabbing ribbons, and backsheets) and an evaluation of their impact on the output power of silicon wafer-based PV modules. Various characterization tools/techniques, such as UV–vis (reflectance) measurement, external quantum efficiency (EQE) measurement and EQE line-scan are used. Based on the characterization results, we use module materials with the best-evaluated optical performance to build "optimized modules". Compared to the standard mini-module, an optical gain of more than 5% is achievable for the "optimized module" with selected module materials.

Bifacial cells are conventionally measured using gold-plated chuck, which is conductive and reflective. This measurement setup does not portray the actual operating conditions of the bifacial cells in a module. The reflective chuck causes an overestimation of the current due to the cell transmittance for the infrared light. The conductive chuck creates a shorter current flow path in the rear side of the cell and causes an over inflation of the fill factor measurement. In this study, we characterize and quantitatively analyze the difference between the bifacial cell measurements on different mounting chucks and calculate the cell-to-module (CTM) loss. To characterize the optical behavior of the bifacial cell and module, we perform external quantum efficiency, reflectance and transmittance measurements. The electrical behavior of the bifacial cell is studied using in-house developed software Griddler. Using Griddler, we calculate the difference in the fill factor of the bifacial cell due to the measurement using a conductive and non-conductive chuck, and estimate the corresponding CTM resistive losses.

The purpose of this study is to develop a method of estimating the electric power from various photovoltaic technologies with high precision. The actual outdoor performance of eight kinds (12 types) of photovoltaic (PV) modules has been measured since January 2012 in order to verify the precision of the method. Using ambient climatic datasets including solar irradiance, module temperature, and solar spectrum, the performance of these PV modules is corrected to the performance under standard test conditions (STC), which should be constant ideally. The results indicate that the performance of bulk crystalline silicon (c-Si) and copper indium gallium diselenide (CIGS) PV modules can be estimated with high precision (approximately less than ±2%). However, the estimation precision of thin-film Si and cadmium telluride (CdTe) PV modules is low because of the initial light-induced degradation and seasonal variation due to metastability.

The performance of photovoltaic (PV) modules deteriorates with time due to outdoor exposure. We investigated the time-dependent changes in PV modules and evaluated the amount of power generated during their lifetime. Once a year, the exposed modules were removed and measured under standard test conditions using a solar simulator. Their outputs were measured indoors and normalized to nominal values. In addition, the relationship between the indoor measurement and the energy yield for thin-film PV modules will be reported. In CIGS PV modules, the normalized maximum power (P_MAX) and performance ratio (PR) differ with the type of module. The P_MAX and PR of CdTe PV modules significantly decrease after outdoor exposure for three years. These results help to determine the characteristics of the time-dependent changes in the P_MAX of PV modules due to outdoor exposure.

Precise outdoor measurement of the current–voltage (I–V) curves of photovoltaic (PV) modules is desired for many applications such as low-cost onsite performance measurement, monitoring, and diagnosis. Conventional outdoor measurement technologies have a problem in that their precision is low when the solar irradiance is unstable, hence, limiting the opportunity of precise measurement only on clear sunny days. The purpose of this study is to investigate an outdoor measurement procedure, that can improve both the measurement opportunity and precision. Fast I–V curve measurements within 0.2 s and synchronous measurement of irradiance using a PV module irradiance sensor very effectively improved the precision. A small standard deviation (σ) of the module's maximum output power (P_max) in the range of 0.7–0.9% is demonstrated, based on the basis of a 6 month experiment, that mainly includes partly sunny days and cloudy days, during which the solar irradiance is unstable. The σ was further improved to 0.3–0.5% by correcting the curves for the small variation of irradiance. This indicates that the procedure of this study enables much more reproducible I–V curve measurements than a conventional usual procedure under various climatic conditions. Factors that affect measurement results are discussed, to further improve the precision.

Sensors with wireless communication can be powered by photovoltaic (PV) devices. However, using solar power requires thoughtful design of the power system, as well as a careful management of the power consumption, especially for devices with cellular communication (because of their higher power consumption). A design approach can minimize system size, weight, and/or cost, while maximizing device performance (data transmission rate and persistence). In this contribution, we describe our design approach for a small form-factor, solar-powered GPS tracker with cellular communication. We evaluate the power consumption of the device in different stages of operation. Combining measured power consumption and the calculated energy-yield of a solar cell, we estimate the battery capacity and solar cell area required for 5 years of continuous operation. We evaluate trade-offs between PV and battery size by simulating the battery state of charge. The data show a trade-off between battery capacity and solar-cell area for given target data transmission rate and persistence. We use this analysis to determine the combination of solar panel area and battery capacity for a given application and the data transmission rate that results in minimum cost or total weight of the system.

Thailand is an agricultural country, with rice, sugar, and cassava as the major export products. Production of rice, sugar cane, and cassava entails agricultural activities that give rise to significant airborne dusts. In this work, five photovoltaic (PV) units (one solar rooftop and four power plants) are selected for the study. From the study of dust accumulation on glass surface located near rice farms, it was found that opaque areas due to the deposition of dust are 11–14% after 1–2-week exposure. As a consequence, PV system performance is affected. Performance ratio was calculated to determine these effects. Overall results reveal that during the dry and hot seasons, dust deposition significantly affects the performance ratio. The performance ratio reduces by 1.6–3% for 1-month dust accumulation and reduces by 6–8% for 2-month dust accumulation. After cleaning the dust accumulated, the performance ratio greatly increases, resulting in the increase in the energy output by 10%. This increase provides economic and cost benefits of PV cleaning. The performance ratio is not significantly changed during the rainy season, which PV modules are relatively clean as the dust is washed away by rain. It was also found that most of the solar power plants in Thailand still rely on manual cleaning of PV modules with washing water followed by wiping. However, only one power plant, employs a machine for cleaning, resulting in lower cleaning costs.

The fabrication of an axial-type microwire-structure silicon (Si-MW) solar cell has been studied, and it is compared with the radial-type Si-MW solar cell. In this study, a number of microwires were formed on a p-type Si wafer and an n-layer were fabricated on the top region of microwires to form p–n junctions. After that, Si-MWs were filled with an Al_1−xO_x film, deposited by thermal atomic layer deposition, and silica particles, respectively. Then, the Al_1−xO_x film and silica particles on the n-layer were removed by chemical mechanical polishing to form an electrical contact with the electrode. As a result, we obtained a conversion efficiency of 8.2% from the axial-type Si-MW solar cell. In addition, compared with a radial-type Si-MW solar cell, a higher quantum efficiency was obtained in the short-wavelength region (300–600 nm) for the axial-type Si-MW solar cell.