Silicon Wafer Solar Cells Research Articles

Efficiency Potential of Rear Heterojunction Stripe Contacts Applied in Hybrid Silicon Wafer Solar Cells

A 2-D numerical analysis evaluating the silicon solar cell efficiency potential of rear-side heterojunction stripe contacts embedded in an AlOx/SiNx passivation stack is presented. Simulated implied efficiencies of correspondingly passivated wafers and device efficiencies of corresponding hybrid solar cells are stated and compared with more conventional solar cell device architectures (i.e., n-PERT and full-area hybrid heterojunction cells). One-dimensional (full-area) and 2-D (stripe contact) lifetime samples are investigated as experimental input for simulation calibration. The influence of the rear-contact area fraction and pitch on both surface recombination and series resistance are investigated. The solar cell efficiency potential (i.e., the implied efficiency) of rear heterojunction stripe contacts is calculated using measured and simulated effective carrier lifetime and series resistance data. For a given (measured) injection-dependent lifetime of the heterojunction and dielectric passivation layers used, the optimum contact geometry (contact area fraction and pitch) is predicted. Assuming either a 1-ms effective lifetime for an ideal a-Si:H(i)/μc-Si:H(p) repassivated lifetime sample (i.e., assuming no laser-induced Si crystal damage due to the contact opening process) or an experimentally observed 300-μs lifetime for an a-Si:H(i)/μc-Si:H(p) repassivated lifetime sample, the calculated 1-sun efficiency potential of rear heterojunction stripe contacts (with optimized rear contact geometry) is 25.8% and 22.9%, respectively. The simulated efficiencies of corresponding hybrid heterojunction cells (having a conventional diffused front-contact system) are 22.7% and 21.4%, respectively. This is significantly higher than the calculated efficiency potential of more conventional solar cell architectures, such as n-PERT cells (20.6%) and full-area hybrid heterojunction solar cells (20.7%).

Uncertainty analysis in contact resistivity measurements of crystalline silicon solar cells

The achievement of obtaining low contact resistivity between silver electrodes and the emitter layer of a crystalline silicon solar cell has been a constant issue for the silver paste developers in crystalline silicon solar cell industries. However, it has been found that there are factors that significantly distort the accurate characterization of contact resistivity. One group of such factors have been identified to be not related with the physical traits of silver paste nor the size of a crystalline silicon solar cell wafer. These factors include mathematical formulations to calculate contact resistivity, the preparation method for specimens, probing methods, current source settings, and environmental conditions. On the other hand, the other group of such factors are directly related with the formulation method and composition of silver paste, and the size of a crystalline silicon solar cell wafer where specimens for the resistance measurements are prepared. Through this study, the impact of each factor on resistance measurement using a three-point probe method for the calculation of contact resistivity is analyzed.

Study of hydrogen influence and conduction mechanism of amorphous indium tin oxide for heterojunction silicon wafer solar cells

The optoelectrical properties of the indium tin oxide (ITO) films are investigated at low‐power and low‐temperature conditions for heterojunction silicon wafer solar cell applications. ITO films deposited at 100 W in hydrogen‐diluted argon ambient have the best optoelectrical quality, giving transparent and conductive (carrier mobilities in the 50–60 cm2/Vs range) films. X‐ray diffraction analysis shows that the ITO films deposited at 100 W are primarily of amorphous (or quasi‐amorphous) nature, regardless of the deposition conditions. Based on several other characterisation methods, it is found that the high electron mobility achieved by the hydrogenated ITO films is mainly due to the reduction of scattering centres. A few In2O3 samples are prepared at similar deposition conditions to investigate the electrical conduction mechanism of ITO films. Due to the low doping efficiency of tin atoms in amorphous ITO, the main source of the free electron charge carriers are the oxygen vacancies.

Determination of metal contact recombination parameters for silicon wafer solar cells by photoluminescence imaging

In order to reliably extract metal–silicon interface recombination parameters to aid optimization of silicon wafer solar cell designs, a test metallization pattern with regions of varying front metal contact fractions is screen printed to create special test cells, alongside solar cells with a standard H-pattern print. Both the test pattern and H-pattern cells are analyzed using intensity-dependent photoluminescence imaging (Suns-PL), and the H-pattern cells are additionally probed at each busbar to monitor their open-circuit voltage (via Suns-Voc measurements). The resultant data are analyzed in two ways, the first being a simple four parameter graphical fitting to the Suns-PL plots of the regions of interest, and the second being a detailed finite element method (FEM) based simulation and numerical fitting. By accounting for the lateral balancing currents in the FEM based simulation, test patterns with imaginary isolation across the edges of the mini cells could be used, which were the easiest to fabricate. The FEM method is more rigorous and is able to replicate simultaneously the Suns-PL characteristics of both test pattern and H pattern cells, using a common set of recombination parameters.

Screen-printed masking of transparent conductive oxide layers for copper plating of silicon heterojunction cells

Replacing expensive silver with inexpensive copper for the metallisation of silicon wafer solar cells can lead to substantial reductions in material costs associated with cell production. Copper metallisation is especially applicable to hydrogenated amorphous/crystalline silicon heterojunction cells since the transparent conductive oxide (TCO) layer in such cells is expected to provide an adequate barrier to prevent cell degrading copper diffusion into silicon. For copper plating on heterojunction cells it is necessary to mask the TCO surface to define the grid electrode. In this paper we investigate screen-printed masking of TCO surfaces to define copper-plated electrodes of heterojunction cells. A masking process is developed and various masking and plating aspects are evaluated. The focus of these investigations is on the characterisation of metal–TCO interfaces and on determining the influence of textured silicon wafer surfaces on the masking process. As a proof of concept, heterojunction cells are fabricated with copper-plated front contacts defined by screen-printed TCO masking. The copper-plated heterojunction cells achieve conversion efficiencies comparable to reference heterojunction cells with evaporated front contacts defined by photolithography.

Comparative study of amorphous indium tin oxide prepared by pulsed-DC and unbalanced RF magnetron sputtering at low power and low temperature conditions for heterojunction silicon wafer solar cell applications

Comparative study of indium tin oxide (ITO) was performed between films prepared by pulsed direct current (PDC) and radio frequency (RF) magnetron sputtering methods at low power and low temperature conditions. The best achieved film resistivity is in the range of 3–6 × 10−4 Ω cm with the highest electron carrier mobility of ∼60 cm2/V. The properties of the surface layer and the bulk material were analysed in details by means of X-ray photoelectron spectroscopy (XPS). Although there is a little difference in elemental compositions, it is revealed that the ITO's properties are directly related to the chemical bonding conditions. Owing to the fast film growth rate of PDC sputtering method, second phase Sn3O4 is formed which impedes the optical performance of the ITO films. Our observations suggest that the darkening of ITO is most likely due to the existence of Sn3O4 second phases that trapped inside of the film bulk. Besides material properties, we have demonstrated that the PDC sputtering method introduces more particle bombardment damage, while for RF sputtering the damage is mainly contributed by ultraviolet radiation emission. The major sputtering induced damage can be recovered by 15 min thermal annealing at 200 °C in air.

0.4% absolute efficiency increase for inline-diffused screen-printed multicrystalline silicon wafer solar cells by non-acidic deep emitter etch-back

Emitter formation is one of the most critical and crucial process steps in the fabrication of standard silicon wafer solar cells. Typically the photovoltaic industry uses tube based phosphorus diffusion, using phosphorus oxychloride as the dopant source. Alternately, a low-cost inline diffusion using phosphoric acid as a dopant source can be used. However, proper process conditions must be used to meet solar cell energy conversion efficiencies obtained by tube diffusion. In this work, we present the application of a non-acidic homogeneous emitter etch-back process – the ‘SERIS etch’ – for inline-diffused emitters in order to raise the efficiency of multicrystalline silicon (multi-Si) wafer solar cells. We apply both light and heavy emitter etch-backs on inline-diffused emitters with sheet resistance (Rsq) values in the 40–60Ω/sq range to achieve emitters with a target Rsq of ~70Ω/sq. The emitter surface reflectance and doping uniformity are maintained even after an etch-back that results in a Rsq change of ~30Ω/sq. An average cell efficiency gain of 0.4% (absolute) is reported for cells with heavy etch-back when compared to the as-diffused non-etch-back screen-printed full-area aluminum back surface field solar cells and efficiencies up to 17.9% are achieved. Besides, best lot of the etch-back inline-diffused cells shows a 0.2% (absolute) efficiency gain over the standard tube-diffused cells. These results show that the ‘SERIS etch’ etch-back process can enable higher-efficiency industrial inline-diffused multi-Si wafer solar cells.

Investigation of laser ablation on boron emitters for n‐type rear‐junction PERT type silicon wafer solar cells

Abstractn‐type silicon wafer solar cells are receiving increasing attention for industrial application in recent years, such as the n‐type rear‐junction Passivated Emitter Rear Totally‐diffused (PERT) solar cells. One of the main challenges in fabricating the n‐PERT solar cells is the opening of the rear dielectric for localized contacts. In this work laser ablation is applied to locally ablate the rear dielectric. We investigate the laser damage to the emitter at the laser‐ablated regions using the emitter saturation current density, J0e,laser, extracted by two approaches. J0e,laser is observed to be injection dependent due to high J02 recombination caused by laser damage to the space charge region. By using the optimized laser ablation parameters, n‐PERT solar cells with an efficiency of up to 21.0% are realized. Copyright © 2015 John Wiley & Sons, Ltd.

Full loss analysis for a multicrystalline silicon wafer solar cell PV module at short‐circuit conditions

AbstractWe present an extended analysis to quantify losses in a module under short‐circuit conditions. The presented method includes an analysis of the area‐related losses in a full‐sized module and introduces the concept of “effective area coverage.” The effective area coverage accounts for the amount of light lost on a certain type of area and is different from the actual area coverage, if light trapping area occurs. An example for this effect is light reflected from the backsheet, trapped in the glass cover and reaching the solar cell. Based on our analysis, the effective solar cell area is increased by 5% absolute by this effect. Furthermore, we used an extension of Basore's model for modules in the loss analysis and used it to quantify losses in one exemplary, encapsulated multicrystalline silicon wafer solar cell. These losses are absorption in the rear reflector (corresponding to a loss of 3.5 mA/cm2 on the active solar cell area), collection losses (3.4 mA/cm2) and free carrier absorption (0.18 mA/cm2). Finally, the presented analysis allowed us to calculate a breakdown of all losses in a full‐sized module under short‐circuit conditions. A list of all considered losses sorted by the corresponding current is presented at the end. Copyright © 2015 John Wiley & Sons, Ltd.

The effect of front pyramid heights on the efficiency of homogeneously textured inline-diffused screen-printed monocrystalline silicon wafer solar cells

The majority of industrial monocrystalline silicon (c-Si) wafer solar cells are alkaline textured (at least the illuminated surface) to reduce reflection and increase absorption of incident light. Therefore, understanding the influence of front pyramid heights on the solar cell parameters is essential for further improving cell efficiency. In this work we report the impact of pyramid height on the performance of inline-diffused c-Si solar cells. Three alkaline texture processes with potassium silicate additives are optimised to result in homogeneous coverage of pyramids. By modifying the process, surface textures with small (∼5 μm maximum), medium (∼6 μm maximum) and large (∼8 μm maximum) pyramid heights are formed. The impact of pyramid size on cell parameters is experimentally studied using industrial-grade 156-mm pseudo-square p-type Czochralski wafers. It is found that within the pyramid size range studied here, there is no significant variation in effective minority carrier lifetime, reflectance, open-circuit voltage or short-circuit current. However, fill factor and hence efficiency is significantly impacted by pyramid size. While cells in all three groups demonstrate high fill factor (>79%), it is shown that an average fill factor gain of up to 1% absolute can be achieved by using the best-suited texture process.

The performance of thin industrial passivated emitter and rear contacts solar cells with homogeneous emitters

With the continuous development of sawing technology for silicon wafers, the use of thin silicon wafers for solar cells has become an effective method to decrease the cost of industrial silicon solar cells. This study focused on the solar cells based on the concept of a structure with a homogeneous emitter and passivated rear surface (PERC solar cells). Fabricated using normal production equipment, the performance of PERC solar cells with thicknesses of 80, 100, 120, and 160 μm was compared, and the differences were discussed based on the measurements of optical, electrical, and mechanical properties. The highest efficiency was recorded for a solar-cell thickness of 120 μm. By optimizing the Al paste and firing condition, we obtained PERC solar cells with a thickness of 80 μm, and there has been about a 0.2% reduction in cell efficiency compared to cells with a thickness of 160 μm. Good tortuosity of thin silicon cell opens a wider range of applications.

Extraction of Surface Recombination Velocity at Highly Doped Silicon Surfaces Using Electron-Beam-Induced Current

In this paper, we demonstrate that the surface recombination velocity of electrons S <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n0</sub> at highly p <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">+</sup> -doped silicon surfaces can be quantified using the electron-beam-induced current (EBIC) technique. First, the 3-D electron-beam sample interaction is simulated using CASINO Monte-Carlo software in order to generate the 3-D carrier generation profile. Subsequently, this carrier generation profile is used in Sentaurus Technical Computer-Aided Design device modeling, and the EBIC response is simulated as a function of S <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n0</sub> . The simulation results show a near-perfect match with the EBIC measurements obtained on a passivated and depassivated p <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">+</sup> -emitter of n-type silicon wafer solar cells. In addition, localized Sn0 extraction on an area of 30 × 30 nm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> is presented, clearly illustrating the advantage of EBIC as an electron-beam-based characterization technique compared with optical techniques.

Nanoetching process on silicon solar cell wafers during mass production for surface texture improvement.

Major challenge in nanotechnology is to improve the solar cells efficiency. This can be achieved by controlling the silicon solar cell wafer surface structure. Herein, we report a KOH wet etching process along with an ultrasonic cleaning process to improve the surface texture of silicon solar cell wafers. We evaluated the KOH temperature, concentration, and ultra-sonication time. It was observed that the surface texture of the silicon solar wafer changed from a pyramid shape to a rectangular shape under edge cutting as the concentration of the KOH solution was increased. We controlled the etching time to avoid pattern damage and any further increase of the reflectance. The present study will be helpful for the mass processing of silicon solar cell wafers with improved reflectance.

Investigation of the short-circuit current increase for PV modules using halved silicon wafer solar cells

It is well established that using halved silicon wafer solar cells in a photovoltaic (PV) module is an efficient way to reduce cell-to-module resistive losses. In this work we have shown that PV modules using halved cells additionally show an improvement in their optical performance, resulting in a higher current generation. We attribute this increase in current to gains in light reflected from the backsheet area. An optical model is presented that quantitatively determines the influence of the backsheet on the short-circuit current of a PV module. We find that, for an accurate prediction, several factors have to be taken into account, including the geometry of the module, the backscattering properties of the backsheet and the illumination spectrum. Particularly the angularly and spectrally resolved scattering properties of the backsheet are shown to have a large impact on the current generation. Furthermore, light beam induced current (LBIC) measurements are used to test the backscattering properties of the backsheet and also the influence of the illumination spectrum. LBIC measurements are also used to verify the simulation results, giving good agreement. Thus the design of a PV module can be optimized by simulation. A standard full-size cell module and a halved-cell module with optimized cell spacing are fabricated. Compared to the standard module, the half-cell module is shown to have 4.60% more power (315.3 vs. 329.8W), 1.46% higher fill factor (75.5 vs. 76.6%), and 3.08% more current (9.08 vs. 9.36A).

Accurate extraction of the series resistance of aluminum local back surface field silicon wafer solar cells

Aluminum local back surface field (Al-LBSF) silicon wafer solar cells are currently intensively investigated in the photovoltaic community and are expected to enter mass production in the near future. In this work we show that this solar cell architecture can pose significant challenges in the determination of the series resistance at the maximum power point. We also show that some of the traditional methods for extracting the series resistance of these cells result in a severe underestimation, due to injection dependent saturation current densities. By using a combination of electro- and photo-luminescence images, we demonstrate that the series resistance of Al-LBSF solar cells can be accurately determined.

Two-dimensional numerical simulation of boron diffusion for pyramidally textured silicon

Multidimensional numerical simulation of boron diffusion is of great relevance for the improvement of industrial n-type crystalline silicon wafer solar cells. However, surface passivation of boron diffused area is typically studied in one dimension on planar lifetime samples. This approach neglects the effects of the solar cell pyramidal texture on the boron doping process and resulting doping profile. In this work, we present a theoretical study using a two-dimensional surface morphology for pyramidally textured samples. The boron diffusivity and segregation coefficient between oxide and silicon in simulation are determined by reproducing measured one-dimensional boron depth profiles prepared using different boron diffusion recipes on planar samples. The established parameters are subsequently used to simulate the boron diffusion process on textured samples. The simulated junction depth is found to agree quantitatively well with electron beam induced current measurements. Finally, chemical passivation on planar and textured samples is compared in device simulation. Particularly, a two-dimensional approach is adopted for textured samples to evaluate chemical passivation. The intrinsic emitter saturation current density, which is only related to Auger and radiative recombination, is also simulated for both planar and textured samples. The differences between planar and textured samples are discussed.

Influence of random pyramid surface texture on silver screen-printed contact formation for monocrystalline silicon wafer solar cells

Most industrial monocrystalline silicon wafer (mono-Si) solar cells are metallised by screen printing. Silver (Ag) pastes are commonly used to form electrodes to phosphorus-doped n+ Si. However, there is still ambiguity about the influence of the random-pyramid surface texture on Ag screen-printed contacts for mono-Si solar cells. We present an experimental study to investigate this influence using screen-printed p-type mono-Si cell groups fabricated with a controlled variation in the alkaline surface texturisation process. It is observed that cell groups fabricated on Czochralski (Cz) wafers with smaller pyramids achieve higher average fill factor (FF) and lower average specific contact resistance than cell groups fabricated on Cz wafers with larger pyramids. To explain these observations, the pyramid texture height distributions are characterised statistically and the distribution statistics are correlated to electrical solar cell measurements and microstructure investigations of the Ag/n+ Si contact interface. Microstructure investigations reveal that most Ag crystallite growth is concentrated around the upper part of the pyramids and hence pyramid density is identified as an important parameter influencing contact formation. The influence of average pyramid height and pyramid height uniformity within a pyramid texture height distribution is also clarified with regards to contact formation. It is further observed that direct Ag crystallite contacts to the bulk Ag metallisation are not a prerequisite for achieving high FF (>80%). Based on the study, guidelines are developed for tailoring random-pyramid surface textures to optimise Ag screen-printed contact formation to n+ Si for mono-Si solar cells.

Influence of rear located silver nanoparticle induced light losses on the light trapping of silicon wafer-based solar cells

Management of the light losses associated with silver nanoparticle integrated plasmonic back reflectors in silicon wafer solar cells is critical to realize performance enhancement. The light losses, including the intrinsic absorption loss from silver nanoparticles and the additional absorption loss induced by the void plasmons in the aluminum reflectors, are quantitatively studied for cells with different front surface morphologies. The study reveals that silver nanoparticles are effective to enhance the photocurrent in cells with planar front surface, while the absorption enhancement can be significantly offset by the plasmonic losses in the textured cells, contributing to marginal or even decreased photocurrent.

On the nature and removal of saw marks on diamond wire sawn multicrystalline silicon wafers

Clearly visible saw marks are a significant barrier to commercial use of diamond wire sawn multicrystalline silicon wafers for solar cells. Two types of saw marks on the diamond-cut multicrystalline silicon wafers are identified—the millimeter scale round-run fringes caused by round-running of the saw wires, and the micron scale scratches caused by scribing of the diamond tips. The latter consists of smooth and shiny grooves covered by a thin layer of amorphous phase. The micro-roughness of diamond-cut wafers is actually ~25% less than that of the conventional slurry-cut wafers. The reason for the visibility of the round-run fringes to naked eyes, and for the relatively rough appearance of diamond-cut wafers, is the visual enhancement from the shiny scratches. Therefore, the key to remove the round-run fringes is to roughen the smooth grooves, as flattening the very slightly sloped fringe zones is very difficult due to lack of chemical contrast over them. Acid-etching texturization cannot remove the saw marks on the diamond-cut silicon wafers. Alkaline-etching can only remove the saw marks on grains near (001) orientation. A vapor blast etching method has been attempted. The preliminary result is encouraging—complete removal of the saw marks has been achieved, along with a good surface texture, which reduces the light reflectivity to 19%.

Polarized photoluminescence imaging analysis around small-angle grain boundaries in multicrystalline silicon wafers for solar cells

We have shown the effectiveness of polarized photoluminescence imaging for analyzing the structural and spectroscopic properties of small-angle grain boundaries (SA-GBs) in multicrystalline Si. The dislocation-related deep-level emission band at approximately 0.79 eV at room temperature was found to be polarized, whereas the band-edge emission did not show the polarization effect. The anisotropy of the 0.79 eV band was classified into two groups depending on the tilt and twist characteristics of SA-GBs determined by the electron backscatter diffraction measurement.