Search

We experimentally proved an effective use of UV-ozone oxide layer in junction passivation and current tunneling applications of crystalline silicon (c-Si) solar cells. The UV-ozone generated oxide layer can improve the passivation quality of aluminum oxide(AlO x ) when inserted between the silicon wafer surface and atomic layer deposited(ALD) AlO x . When tested on junctions that moderately diffused by phosphorus and boron (sheet resistance R sh ~110 Ω/□), the UV-ozone oxide and AlO x stack resulted in a J 0 no more than 12fA/cm2. The same oxide layer is also considered as a passivating contact material. When applied as an interlayer between diffused silicon surface and aluminum contact, the passivated contact structure realized a contact resistivity (ρ c ) of ~ 1mΩ-cm2 and ~ 25mΩ-cm2 for boron and phosphorus passivating contact structures, respectively, with moderately doped diffusions.

Silicon solar cells incorporating double-sided pyramidal texture are capable of superior light trapping over cells with front-side only texture. However, increased surface area, roughness and exposed crystal planes of textured surfaces not only causes increased recombination, but also makes cells susceptible to shunting through pinholes in the dielectric at the sharp peaks and valleys of the textured pyramids. A polyimide film as an insulating interlayer film is investigated to circumvent the tradeoff between improved light trapping, increased recombination and increased shunt paths. When applied at the rear of the interdigitated back contact silicon solar cell structure, the polyimide film provides an excellent electrical insulation (> 1000 MΩ of insulation resistance) and increases photocurrent (~ 1.1 mA/cm2) owing to an increased rear internal reflectance. The polyimide is also compatible with metal annealing of passivating dielectrics such as silicon nitride. Optical simulation and experimental results are combined in a 3D semiconductor simulation (Quokka) to quantify the possible gain of implementing the double-sided texture in high efficiency silicon solar cells.

Carrier lifetime degradation of reactive ion etch-processed silicon samples are investigated. Two types of carrier recombination: reversible and irreversible degradations induced by reactive ion etching (RIE) are identified. Irreversible carrier recombination is due to surface damage created by the RIE process that propagates a few microns deep into the silicon substrate. Reversible carrier recombination, on the other hand, is found to be caused by radiation damage when RIE etches only into the silicon oxide, and nitrogen annealing can restore the degraded carrier lifetime. A recombination-free RIE process is then developed in combination with a passivation stack consisting of silicon dioxide and silicon nitride layers. This improved RIE process is applied to the development of high efficiency silicon solar cells resulting in a conversion efficiency exceeding 24%.

We investigate the light trapping in Si wafers that are textured with conventional random pyramids on their front surface and rounded random pyramids on their rear. It is well established that rounding the pyramids leads to better surface passivation, but whether or not it improves light trapping depends on the cell structure. In this paper, we apply ray tracing, spectrophotometry, and photoluminescence spectroscopy (PLS) to understand and quantify how rounding the rear pyramids might affect the light trapping in back-contact solar cells. We describe how rounding the pyramids leads to two competing optical effects: 1) reduced absorption in the rear films and 2) reduced scattering from the rear texture. The first effect improves light trapping whereas the latter degrades it. We show how the influence of each effect depends on wavelength and how they can be discerned (but not easily quantified) in reflectance curves. With PLS measurements, we conclude that for our sample structure and etch solution, the generation current is approximately constant for etch durations less than ∼60 s, and decreases significantly as the etch duration increases. Thus, by limiting the duration of the rounding etch, superior surface passivation can be attained without degrading the light trapping.

Introduction Treating ventricular tachycardia (VT) in patients with cardiac sarcoidosis (CS) is challenging as patients present in different phase of the disease (inflammatory, scar, or sometimes both). A customized approach to treatment is required for better outcomes. We describe our experience in the management of VT in CS based on the phase of the disease. Methods and results Patients were considered to have myocardial inflammation if there was an increased myocardial 18fluorodeoxy glucose (FDG) uptake in PET-CT scan of the chest (n = 14). These patients were treated with antiarrhythmic drugs (AADs) and immunosuppression. Patients with scar related VT (without active inflammation) were managed with AADs and underwent radiofrequency ablation (RFA) if unresponsive to drug therapy (n = 4). Patients previously treated for CS who presented with VT and evidence of reactivation (abnormal FDG uptake) after a quiescent period of 6 months were treated with intensified immunosuppression alongside AADs (n = 3/14). Patients with myocardial inflammation responded well to immunosuppression. Patients with drug resistant VT in the scar phase responded well to RFA. Four patients in the inflammatory group had recurrence of VT during follow-up of whom 3 were found to have disease reactivation. Intensified immunosuppression suppressed VT in all 3 patients. In 1 patient, VT recurrence was found to be scar related and required RFA for control. Conclusion Tailoring therapy for VT in CS according to the phase of disease results in good clinical outcome and avoids unnecessary immunosuppression.

The focus of this work is on the characterization of passivated crystalline Si (c-Si) surfaces subjected to various cleaning sequences involving UV ozone (UVo) treatment and HF-dip. A combination of photoconductance decay (PCD) measurements and high-resolution transmission electron microscopy (HRTEM) studies were used to obtain a deeper insight into passivation mechanisms of UVo and its origin at the nano-scale.

It is long recognized that the effective surface clean is critical for the increased performance of solar cell and semiconductor devices. In this contribution, we introduced the effectiveness of crystalline silicon surface clean by a simple ultraviolet-ozone (UVo) process by comparing it against the industry standard RCA and UV assisted deionized water (DiO 3 ) techniques. Despite being simple, UV-ozone cleaning results in an effective surface passivation quality that is comparable to both RCA and DiO 3 clean, i.e., saturation current density (J 0 ) of 7 fA/cm2 compared to 5 fA/cm2 and 8 fA/cm2. In addition to the surface clean, we presented that both UVo and DiO 3 oxides can be used as a highly-quality chemical passivation to the crystalline silicon substrate, but with UVo oxide offering an improved passivation than DiO 3 oxide. Incorporating the UV}o oxide in between the interface of silicon and aluminum oxide or silicon nitride reduces J 0 by $\gt 50$%, compared to the interface without the UVo oxide.

The contact resistivity of evaporated Al on doped silicon is examined for a range of process conditions common to the fabrication of laboratory silicon solar cells. The effects of silicon surface preparation prior to evaporation, sintering temperature, the use of a shutter, and evaporation power are investigated. The presented evaporation conditions yielded the lowest published contact resistivity between Al- and phosphorus-doped Si over a large range of doping concentration. It is also demonstrated that a contact resistivity below 10−6 Ω·cm 2 can be achieved without sintering. Three-dimensional simulations are utilized to compare the obtained results for evaporated Al contacts with those for passivated contacts.

Passivation quality of silicon nitride (SiNx), aluminum oxide (AlOx) and a stack of AlOx/SiNx has been investigated in the presence and absence of a thin silicon oxide (SiOx) layer formed using ozonated deionized water. Lifetime measurements show $\approx 3$ ms effective carrier lifetime $(\tau_{eff})$ for the stack of AlOx/SiNx in the presence of the oxide. Low saturation current density $(Jo)$ and interfacial trap density $(D_{it})$ confirm and explain the high $\tau_{eff}$ for this sample. The stack of AlOx/SiNx can offer excellent surface passivation because of both field-effect passivation and chemical passivation. Note that the presence of oxide also shows a crucial impact in achieving a good passivation.

The interdigitated back contact (IBC) solar cells developed at the Australian National University have resulted in an independently confirmed (Fraunhofer Institut fur Solare Energiesysteme (ISE) CalLab) designated-area efficiency of 24.4 ± 0.7%, featuring short-circuit current density of 41.95 mA/cm2, open-circuit voltage of 703 mV and 82.7% fill factor. The cell, 2 × 2 cm2 in area, was fabricated on a 230 µm thick 1.5 Ω cm n-type Czochralski wafer, utilising plasma-enhanced chemical vapour deposition (CVD) SiNx front-surface passivation without front-surface diffusion, rear-side thermal oxide/low-pressure CVD Si3N4 passivation stack and evaporated aluminium contacts with a finger-to-finger pitch of 500 µm. This paper describes the design and fabrication of lab-scale high-efficiency IBC cells. Characterisation of optical and electronic properties of the best produced cell is made, with subsequent incorporation into 3D device modelling used to accurately quantify all losses. Loss analysis demonstrates that bulk and emitter recombination, bulk resistive and optical losses are dominant and suggests a clear route to efficiency values in excess of 25%. Additionally, laser processing is explored as a means to simplify the manufacture of IBC cells, with a confirmed efficiency value of 23.5% recorded for cells fabricated using damage-free deep UV laser ablation for contact formation. Meanwhile all-laser-doped cells, where every doping and patterning step is performed by lasers, are demonstrated with a preliminary result of 19.1% conversion efficiency recorded. Copyright © 2014 John Wiley & Sons, Ltd.

We investigate the versatility of anodically grown silicon dioxide (SiO2) films in the context of process durability and exceptional surface passivation for high efficiency (>23%) silicon solar cell architectures. We show that a room temperature anodic oxidation can achieve a thickness of ~70 nm within ~30 min, comparable to the growth rate of a thermal oxide at 1000 °C. We demonstrate that anodic SiO2 films can mask against wet chemical silicon etching and high temperature phosphorus diffusions, thereby permitting a low thermal budget method to form patterned structures. We investigate the saturation current density J0 of anodic SiO2/silicon nitride stacks on phosphorus diffused and undiffused silicon and show that a J0 of

Silicon remains the material of choice for photovoltaics because of its abundance, non-toxicity, high and stable cell efficiencies, the maturity of production infrastructure and the deep and widespread level of skill available in relation to silicon devices. Rapidly decreasing module prices mean that area-related balance of systems costs are an increasing proportion of photovoltaic systems price. This places a premium on efficient cells. In recent years there have been large improvements in mass production of high quality wafers, the ability to handle thin wafers, maintenance of high minority carrier lifetimes, surface passivation, minimisation of optical losses, device characterisation and in other areas. Many of these improvements are viable in mass production. The upper limit of silicon solar cell efficiency is 29%, which is substantially higher than the best laboratory (25%) [1] and large-area commercial (24%) [2] , [3] cells. Cell efficiencies above 25% appear to be feasible in both a laboratory and commercial environment. Such a cell will have minimal bulk recombination due to a combination of a thin substrate with a very high minority carrier lifetime; superb surface passivation; small-area electrical contacts consistent with low contact recombination, free carrier absorption and contact resistance; excellent optical control through the use of texturing, antireflection coatings and rear surface reflectors; low edge recombination assisted by the use of thinner wafers, larger cells and edge passivation; and sufficient metal coverage to minimise resistive losses. This paper will survey current work in high- performance silicon solar cell design and fabrication, and discuss approaches to efficiency improvements.

Interdigitated back-contact (IBC) solar cells developed in the past two years have efficiencies in the range 24.4%-25.6% As high as these efficiencies are, there are opportunities to increase them further by improving on the light trapping. Silicon solar cells incorporating double-sided pyramidal texture are capable of superior light trapping than cells with texture on just the front. One of the principle losses of double-sided pyramidal texture is the light that escapes after a second pass through the cell when the facet angles are the same on the front and rear. This contribution investigates how this loss might be reduced by changing the facet angle of the rear pyramids. A textured pyramid rounding is introduced to improve the light trapping. The reduction in surface recombination that rounding the facets introduces is also evaluated. With confocal microscopy, spectrophotometry and ray tracing, the rounding etch time required to yield the best light trapping is investigated. With photoconductance lifetime measurements, the surface recombination is found to continue to decrease as the rounding time increases. The spectrophotometry and ray tracing suggests that the double sided textured samples featuring rounded rear pyramids have superior light trapping to the sample with a planar rear surface. The high-efficiency potential of rounded textured pyramids in silicon solar cells is demonstrated by the fabrication of 24% efficient back-contact silicon solar cells.

Reactive Ion Etching (RIE) is used in the fabrication of some types of solar cells to achieve a highly directional etch. However, cells fabricated using RIE have lower than expected efficiency, possibly caused by increased carrier recombination. Characterisation of the carrier lifetime in solar cells was conducted using the quasi steady state photoconductance (QSSPC) measurement technique. Substantial effective lifetime degradation was observed for silicon samples processed by RIE. Lifetime degradation for samples where RIE etches into silicon is found to be permanent, while for samples where RIE etches only on dielectric layers of SiO2 grown on the wafer, the lifetime degradation is found to be reversible. The reversible degradation in RIE-processed samples is associated with radiation damage. By reducing the proportion of a wafer exposed to RIE, the degradation of the effective lifetime of RIE-etched silicon samples can be minimised, and the performance of silicon solar cells can be improved significantly. Copyright © 2010 John Wiley & Sons, Ltd.

Rubidium (Rb) is explored as an alternative cation to use in a novel multication method with the formamidinium/methylammonium/cesium (Cs) system to obtain 1.73 eV bangap perovskite cells with negligible hysteresis and steady state efficiency as high as 17.4%. The study shows the beneficial effect of Rb in improving the crystallinity and suppressing defect migration in the perovskite material. The light stability of the cells examined under continuous illumination of 12 h is improved upon the addition of Cs and Rb. After several cycles of 12 h light–dark, the cell retains 90% of its initial efficiency. In parallel, sputtered transparent conducting oxide thin films are developed to be used as both rear and front transparent contacts on quartz substrate with less than 5% parasitic absorption of near infrared wavelengths. Using these developments, semi-transparent perovskite cells are fabricated with steady state efficiency of up to 16.0% and excellent average transparency of ≈84% between 720 and 1100 nm. In a tandem configuration using a 23.9% silicon cell, 26.4% efficiency (10.4% from the silicon cell) in a mechanically stacked tandem configuration is demonstrated which is very close to the current record for a single junction silicon cell of 26.6%.

The process of making Interdigitated Back Contact (IBC) solar cell is implemented by a novel simplified etch-back technique, while aiming for no compromise on high-efficiency potentials. Simplified etch-back creates localized heavy and light phosphorus and boron diffusions simultaneously. This process also leaves localised heavy diffusions to be approximately a micron higher than neighbouring light diffusion regions. In comparison to the IBC solar cells that ANU developed to date [1], key advantages of this technique feature reduction in cell process steps; requires only two diffusions to create p, p+, n and n+ diffusions; no high-temperature oxidation masking steps required as diffusion barriers; independent optimization of contact recombination, lateral carriers transport and surface passivation; and potential higher silicon bulk lifetime and reduced contamination due to low thermal budget. Based on the etch-back technique, the total saturation current density deduced from the test structures for the IBC cell is below 30 fA/cm2.

A procedure to quantify the optical loss mechanisms in back-contact solar cells is presented. It incorporates recent developments in optical simulation that yield rapid and precise results. The procedure includes spectrophotometry, ellipsometry, quantum efficiency measurements, ray tracing, and the thin-film matrix method. The paper shows how experiments and simulation can be combined to quantify reflection from the front surface, absorption in the antireflection coatings, non-ideal light trapping, and free-carrier absorption—all in terms of a ‘lost generation current’. The procedure is demonstrated on a back-contact solar cell with single-layer and double-layer antireflection coatings. It is extendable to other cell structures.

In this paper a detailed discussion of different options of silicon solar cell design meant to be used in conjunction with a tandem cell stack is presented. Conventional single junction (SJ), horizontally-stacked (HS) and vertically-stacked (VS) silicon solar cell approaches are discussed. The design considerations of each individual silicon solar cell approach (SJ, HS and VS) are detailed, and benefits and drawbacks of these cells are presented. Expected conversion efficiencies of SJ, HS and VS silicon solar cells under a condition — illuminated with a light spot of 1.9 mm diameter through 2 W/cm2 intensity under the infrared light spectrum — are deduced for the purpose of selecting a suitable design for the tandem stack.

Various types of shunt occur in solar cells, including process errors, defects, tunnel shunts between adjacent opposing diffusions and pinholes in insulating dielectric layers used to separate opposite-polarity regions. We have found that boron diffusions into small windows in dielectric layers generate pinholes in the layers following the removal of borosilicate glass (BSG) after the diffusion. These “boron-spots” lie close to the edge of the diffusion windows. If a phosphorus diffused region underlies the dielectric then subsequent metallisation can short circuit the two regions.

Miniature silicon solar cells (8 × 2.0 mm2) are being fabricated for use in tandem-cell concentrator systems. Several factors combine to make the achievement of high efficiency problematical. These include surface, bulk and edge recombination. The latter is relatively important because the surface area of the edge of a small cell is a large fraction of the total surface area. Surface recombination in the cells is caused by the loss of passivating hydrogen beneath a conformal LPCVD SiN x coating, induced by high temperature annealing. Bulk carrier lifetime degradation mechanisms that we have encountered include silicon crystal damage induced by laser scribing of the cells, which affects a relatively large proportion of the volume of the cell. The Quasi-steady state photoconductance (QSSPC) measurement technique was used for the carrier lifetime degradation study. Firstly, a detailed study was undertaken of the carrier lifetime degradation due to the loss of hydrogen in conformally deposited LPCVD silicon nitride grown samples and the effect of hydrogenation on these layers, when subjected to various anneal times and temperatures. Additionally, LPCVD nitride was studied to determine whether it can be used as a layer that can prevent or resist possible contamination, induced by prolonged high temperature anneals. Secondly, a comparison was made between reference samples and samples that were exposed to laser scribing and dicing to determine whether laser scribing is suitable for the shaping of miniature silicon solar cells. Finally, cells with different pn junction designs were fabricated and tested to study edge recombination.

This paper presents the development of small (2.5×8.0mm2) silicon solar cells, to be used in a six-junction tandem device. PC1D, numerical modeling and quasi steady state photoconductance (QSSPC) measurement were used to predict the targeted efficiency of silicon solar cells. Early batch of cells had problems of shunting, series resistance and high carrier recombination. Various techniques - junction isolation, pin-hole analysis, diffusion drive-in, light-induced plating, lifetime degradation studies and implied-Voc - were used to improve the performance of the solar cells.

Reactive Ion Etching (RIE) is observed to cause substantial effective carrier lifetime degradation in silicon wafers. Degradation of lifetime is permanent for samples where RIE etches into silicon, while the lifetime degradation is temporary for samples where RIE etches only dielectric layers of SiO 2 grown on the wafer. The degradation of the effective lifetime of RIE-etched silicon samples can be minimized by exposing only a few percent of the wafer to the etch.

In this paper, a discussion is made of the design of silicon cells to be used in a six-junction tandem solar cell structure as part of the Very High Efficiency Solar Cell (VHESC) program. Minority carrier recombination at surfaces and in the volume, internal quantum efficiency, resistance losses, free carrier parasitic absorption, optical reflection, light trapping, and light absorption must be traded off against each other. Modelling was used to analyse the various parameters and produce estimates of short circuit current, fill factor and open-circuit voltage of the cell. In addition, quasi-steady-state photoconductance measurements to analyse carrier recombination and emitter saturation current (Joe) as well as to predict the open-circuit voltage of solar cell is presented. For metallisation of such small solar cells, alternate methods of making contact such as light-induced plating and electrolyte plating in addition to evaporating metal on the contacts were explored and employed. Numerical resistive loss modelling was made to calculate the optimum metal thickness achieved by light-induced and electroplating to minimise resistive losses. Experiments were conducted to determine the proper plating rate by light-induced and electrolyte plating. Cells were fabricated by standard silicon processing techniques followed by testing of IV curves using current-voltage flash-tester to achieve the target efficiency.

A major objective for photovoltaic conversion is to develop high efficiency solar cells. Many approaches are under investigation - Multiple Junction Solar Cell, Multiple Spectrum Solar Cell, Multiple Absorption Path Solar Cell, Multiple Energy Solar Cell, and Multiple Temperature Solar Cells [1]. The Multiple Junction Solar Cell approach based on a six-junction tandem solar cell has been adopted to achieve conversion efficiency of greater than 50% in the VHESC program sponsored by DARPA [2]. In six-junction tandem solar cells, individual solar cells are stacked on one another and each solar cell absorbs the best-matched slice of the solar spectrum. Silicon is one of the cells in the tandem stacks, and absorbs photon energy of 1.42 – 1.1 eV. The role of the silicon cell is to convert 7% of the light incident on the tandem stack into electricity. Other cells in the stack contribute the balance of the electricity. Key design parameters for the silicon cells are that it should have dimensions of 2.5 × 8 mm2 and it needs to transfer light with energy of less than 1.1ev to the underlying solar cells. In this paper, discussion is made of the design of the silicon cell. Minority carrier recombination at surfaces and in the volume, internal quantum efficiency, resistance losses, free carrier parasitic absorption, optical reflection, light trapping, and light absorption must be traded off against each other. PC1D modeling is used to analyze the various parameters and produce estimates of short circuit current, fill factor and open-circuit voltage of the cell [3]. In addition, characterization of solar cell by photoconductance measurement to analyze carrier recombination and emitter saturation current as well as to predict the open-circuit voltage of solar cell [4, 5] is presented. Discussion of cell fabrication process followed by I–V testing is presented. Completed solar cells were tested in ANU using an in-house fabricated current-voltage flash tester [6] under AM1.5D.

N-type all-back-contact (ABC) silicon solar cells incorporating a simple oxide-nitride passivation scheme are presently being developed at the Australian National University. Having already achieved promising efficiencies with planar ABC cells [1], this work analyses the cell performance after integrating a surface texturing step into the process flow. Although the textured cells have significantly lower front surface reflection, the measured short-circuit current density is actually lower than that of the planar cells. Photoconductance decay data indicate the presence of high carrier recombination at the textured surface of the ABC cells, which are deposited with a stack of thermal oxide and LPCVD nitride. Further examination confirmed that high carrier recombination is due to stress induced by the LPCVD nitride on the peaks and valleys of the textured surface. Use of PECVD nitride instead of LPCVD nitride as an antireflection layer avoids the degraded carrier lifetime caused by the textured surface. Therefore, PECVD nitride should be a good substitute for constructing the oxide-nitride stacks of our future ABC cells.

The collaboration between the Solar Energy Research Institute of Singapore (SERIS), Trina Solar and ANU is progressing well, and ANU has already developed all-back-contacted (ABC) silicon wafer cells with best one-sun efficiencies of 21.2% and 22.1% on FZ material, when measured with the aperture areas of 16 cm2 (includes busbars) and 13 cm2 (excludes busbars) respectively. This paper presents the continuing development of ABC cells targeting the efficiency of 23.5% on 16-cm2 cell area. Further developments such as optimising front surface field (FSF), rear diffusion, anti-reflection coating (ARC), and incorporation of lithographically aligned metal contacts were undertaken on the ABC cells. Phosphorus diffusion of the FSF was made lighter from the sheet resistance of 190 Ω/□ to 240 Ω/□, resulting in the reduction of the saturation current density ( Joe ) of the FSF by 22 fA/cm2. The optimised thickness of anti-reflection coating (ARC) PECVD SiNx further reduces the average reflectance across the wavelength range of 300 to 1200 nm by about 4%. Incorporation of aligned metal contacts and heavier rear phosphorus diffusion has contributed to the reduction in the total series resistance by 0.08 Ωcm2. The above optimised improvements have increased the efficiency of the champion ABC cell by 0.5% absolute. In addition, we present further refinements in areas of texturing; FSF passivation; electrical shading loss in terms of cell pitch, bus- bar and base doping; and metallisation to aim for the 16-cm2 ABC cells with the conversion efficiency > 22% in the near term.

Surface passivation is a key process to achieve high-efficiency in silicon solar cells. In this paper we applied UV-ozone treatment to achieve high-quality passivation on both planar and textured, n-type and p-type wafers.