Boundary Recombination Velocity Research Articles

Effective diffusion length of multicrystalline solar cells

Previous definitions of effective diffusion length of a multicrystalline semiconductor are reviewed. A new definition is proposed, which is based on the comparison of the current collected from the base of a multicrystalline solar cell and a single-crystal cell for uniform carrier generation, and holds for any sample thickness. An analytical expression for is obtained from the solution of the three-dimensional equation for the charge collection probability inside a grain with square cross section. This expression is used to numerically study the dependence of on the bulk diffusion length, grain size and grain boundary recombination velocity, and to make a comparison with previous specifications of . The results indicate, in particular, that is the value that would be determined by the surface photovoltage method. The case of a cell with grains of different properties is briefly discussed and an average is defined. It is also shown that can be used to calculate the contribution of the base to the reverse saturation current of the cell.

Optimum two-dimensional short circuit collection efficiency in thin multicrystalline silicon solar cells with optical confinement

The two-dimensional short-circuit AM1.5 collection efficiency is studied in thin multicrystalline silicon solar cells with optical confinement. The collection efficiency is calculated by linking an optical analytical generation profile with the two-dimensional collection probability in pn junction solar cells. The calculations are carried out for variable grain boundary recombination velocity, cell thickness, grain width, diffusion length, and back surface recombination velocity. The role of optical confinement leading to a strong dependence of the collection efficiency on the cell thickness in very thin cells is confirmed. The optimum cell thickness for maximum collection efficiency increases in cells with low back reflection or poor back surface passivation. Also, the optimum thickness in very thin cells increases significantly with increasing the diffusion length. It is also found that the effect of grain boundary recombination is predominant if the cell thickness is larger than the diffusion length and if the diffusion length is larger than half the grain width, especially, in cells with unpassivated grain boundaries. On the other hand, back surface recombination dominates the response in cells with unpassivated back surface if the thickness is smaller than or comparable to the diffusion length.

Modeling of the recombination at grain boundaries in preferentially doped polysilicon solar cells

Making use of results developed in an earlier paper and according to the theory of Oualid et al., the photovoltaic parameters of a preferentially doped N + P solar cell are studied as a function of the density of interface states ( N t) at the grain boundaries. The results are compared to the ones established with respect to the grain boundary recombination velocity V s (without considering grain boundary recombination theory). The results show that in the case of small recombination at the grain boundaries ( N t < 10 12 cm −2) the variation of the photovoltaic parameters with respect to the density N t is similar to that obtained with the velocity V g. The results also show that, when the density N t is superior to this value, the decrease of the optimum base doping giving the maximum efficiency of the cell with respect to the grain width (in a log-log plot), is no longer linear. This result contradicts the one of Dugas and Oualid.

Anomalously high collection probability in thin film polycrystalline silicon solar cells from numerical modeling

The impact of grain size on the average collection probability of fine grained (&amp;lt;10 μm) polycrystalline silicon has been investigated using a numerical device simulator. This model predicts that the collection probability can improve dramatically once the depletion regions around adjacent grain boundaries overlap. Thus, contrary to most analytical models, the collection probability is not monotonic with grain size. The collection probability has a local maximum at a grain size determined by the grain boundary trap state density. The magnitude of the improvement is dependent upon the grain boundary recombination velocity.

Photovoltaic properties and high efficiency of preferentially doped polysilicon solar cells

Making use of new solutions for the photocurrent and dark current, developed earlier, the photovoltaic properties of a preferentially doped n + p solar cell are examined taking into consideration Shockley-Read-Hall recombination, Auger recombination and heavy doping effects. Numerical calculations have been carried out for the analysis of the effects of the doping concentration, grain size, preferential doping penetration depth along the grain boundaries, and grain boundary recombination velocity on the photovoltaic parameters of the cell. The results reveal that, in order to obtain a high cell efficiency, the emitter doping level should not exceed 5 × 10 19 cm −3 and the base must be doped at an optimum level. It is found that the optimum base doping depends on the grain width and grain boundary recombination velocity and is practically unchanged with preferential doping. The results also show that a conversion efficiency exceeding 20% under AM1 conditions is obtained, when the grain is preferentially doped, and when the grain boundaries and the top and back contact surface are passivated.

Open-circuit voltage decay in polycrystalline silicon solar cells

Theoretical expressions for the open-circuit voltage decay induced by an electrical (dark current) or optical (photocurrent) excitation in a base-dominated polycrystalline silicon solar cell are derived. These two conventional transient methods which are of forward current induced open-circuit voltage decay (FCVD) and photoinduced open-circuit voltage decay (PVD) are investigated through the influence of grain size and grain boundary recombination velocity.

3D modelling of a reverse cell made with improved multicrystalline silicon wafers

Abstract The spectral response, short-circuit photocurrent and conversion efficiency of a reverse cell made with multicrystalline silicon wafers have been computed taking into account different values of base thickness, grain size, grain boundary recombination velocity, front and back surface recombination velocities and minority carrier diffusion length in the grains. The results are compared to those of a conventional multicrystalline cell computed under the same conditions. It is shown that the reverse cell structure, which simplifies the technological problems, particularly those related to the conventional emitter and front grid, could be a good alternative solution in the case of improved thin wafers.

Comments, with reply, on "A scanning electron- or light-beam-induced current method for determination of grain boundary recombination velocity in polycrystalline semiconductors" by C.A. Dimitriadis

An asymptotic analysis of beam-induced-current profiles of grain boundaries in polycrystalline solar cells presented in the above-titled paper by C.A. Dimitriadis (ibid., vol.32, p.1761-5, Sept. 1985) is corrected and extended to the case of arbitrary interface recombination velocity. Dimitriadis replies that the original analysis is correct within the first-order approximation.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Grain boundary effects in the ebic response of thin-film solar cells

Abstract The diffusion length of minority carriers in a CdS-CdTe thin film polycrystalline junction device was measured using Electron Beam Induced Current. The EBIC response did not resemble the approximately exponential shape predicted by current theory. Rather it consisted of a broad region located approximately at the junction but consisting of multiple peaks and notched tails. Accordingly, a general model of the EBIC response was developed which takes into account surface recombination, finite generation volume, and grain boundary recombination to account for the features seen in the data. The important role of grain boundary recombination velocity can be seen in the model.

Grain Boundary effects in polycrystalline silicon solar cells

The polycrystalline silicon solar cell is divided into four regions: (a) skin (emitter) grain region, (b) skin grain boundary region, (c) grain base region and (d) grain boundary base region. Since the grain base region, having a much larger area than other regions, is the dominant photoabsorbing region in silicon solar cells, the photocurrent and dark current contributions from this region have been computed using two-dimensional analysis with the help of Green's function technique. The contributions from the other regions, being much smaller, have been computed using the usual one-dimensional analysis. The dependence of the short-circuit current and open-circuit voltage of these cells on the different parameters such as grain size, grain boundary potential barrier, grain boundary recombination velocity etc., have also been investigated taking into account the variation of the diffusion length in the grain boundary region. The control of important parameters for the optimization of polycrystalline solar cells is discussed.

Exact analysis of a three-dimensional cylindrical model for a polycrystalline solar cell

Analytical expressions for the short-circuit current densities and the spectral response were derived for a thin-film polycrystalline n/p junction solar cell. A three-dimensional model was used for a typical grain. It has the shape of a columnar cylinder. The spectral response and the total short-circuit current density were then evaluated for different values of grain size, grain boundary recombination velocity, and film thickness. The calculations aimed at studying some possible enhancements in the spectral response and the short-circuit current. Such enhancements can result from increasing the grain size or from reducing the grain-boundary recombination velocity. These two parameters can more or less be controlled experimentally through recrystallization and passivation of grain boundary states, respectively.

Kinetics of oxygen diffusion to the grain boundary in oxygen-rich directionally cast silicon bicrystals

Abstract The electron beam induced current mode of a scanning electron microscope has been used to measure the minority carrier recombination velocity at the depletion layer edge of the grain boundary as a function of annealing time (30 – 240 min) and temperature (600 – 900 °C) in oxygen-rich directionally cast silicon bicrystals. The minority carrier recombination velocity at the grain boundary has been found to increase significantly with increasing annealing time and temperature. The grain boundary barrier potential has been determined from the zero-bias capacitance as a function of annealing time and temperature. A theoretical model has been proposed whereby the diffusivity of oxygen in silicon can be computed from the increase in grain boundary interface trap density, calculated from the grain boundary recombination velocity and barrier potential. The diffusitity D ( cm 2 s −1 ) of oxygen in silicon was found to be D=0.27exp -2.56eV kT in the temperature range 600 – 900 °C.

Effective minority carrier diffusion length of polycrystalline silicon solar cells

By means of a modification of the model proposed by Koliwad and Daud, the effective diffusion length L nef was computed in terms of intragranular diffusion length L 0 , grain size g and grain boundary recombination activity γ. The theoretical computations were compared with experimental measurements of effective electron diffusion length L nef and grain boundary recombination velocity s m . It is shown that the theoretical and experimental variations of L nef with g depend mainly on the quality of grains and of the activity of grain boundaries.

Transient photoconductivity analysis near a grain boundary of polycrystalline semiconductors

The effect of grain boundary recombination on the photoconductance transient response, after excitation with a delta function light pulse near a grain boundary, is studied here. An analytical expression is derived for this decay which can be readily computed numerically. The photoconductance turns out to be a non-exponential function of time even at large decay times. A practical method, based on such non-exponential decay curves obtained with the light beam positioned away or near a grain boundary, is developed to deduce accurate value of the bulk lifetime and the grain boundary recombination velocity.

A two-dimensional model for thin film polycrystalline semiconductor P-N junction solar cells

The influence of the grain boundary recombination velocity and grain size on the spectral response and performance of a solar cell under air mass 1 conditions was calculated on the basis of an idealized two-dimensional model describing solar cells made from polycrystalline silicon material with predominantly columnar orientation. The problem of carrier flow within an individual grain was analysed by solving the diffusion equation using boundary conditions. The grain boundaries were considered as recombination planes characterized by a grain boundary recombination velocity which was assumed not to depend on excitation. Detailed numerical calculations were done for the effect of grain size on V oc , FF and efficiency.

Quantitative SEM/BIC studies of carrier recombination in silicon bicrystals

The electron-beam-induced-current technique (SEM/BIC) is applied in silicon bicrystals in order to study both the bulk diffusion length and the grain boundary recombination velocity. Different methods of evaluation are applied and the results are compared. The effect of various parameters on the quantitative analysis is discussed, in particular (i) the observation conditions (accelerating voltage and injection level), (ii) the angle between the grain boundary and the top surface and (iii) variation of the minority carrier diffusion length.

A model of the dependence of photovoltaic properties on effective diffusion length in polycrystalline silicon

In polycrystalline silicon, an effective diffusion length is often defined in order to take into account grain boundary recombination. The definition given here is associated with the surface photovoltage method which is generally used to determine the diffusion length in the base of a solar cell. The effective diffusion length and photovoltaic properties are calculated from the minority carrier distribution within a grain, which is determined by means of a three-dimensional model. The relation between effective diffusion length and photovoltaic properties has been calculated for various values of the material parameters (grain size, bulk diffusion length, effective grain boundary recombination velocity and doping level). It has been found that photovoltaic properties depend not only on the effective diffusion length but also on all the material parameters. It is demonstrated that measurement of the effective diffusion length can provide a sufficiently accurate estimate of the efficiency of a polycrystalline silicon solar cell, using an equivalent monocrystalline cell, but on the condition that the grain size is greater than ten times the average diffusion length within the grains. If this condition is not met, a three-dimensional model is required to obtain satisfactory accuracy.

Evaluation of grain boundary recombination velocity in polycrystalline silicon from the spectral response of Schottky-barrier solar cells

A method based on the spectral response of polycrystalline semiconductor photovoltaic cells for evaluation of the grain boundary recombination velocity (S GB ) is presented. The wavelength of the maximum of the spectral response, for a polycrystalline semiconductor with known grain size, depends on S GB . By comparison of the experimental with the theoretical wavelength of the maximum of the spectral response, it is possible to determine the grain boundary recombination velocity.

Determination of diffusion length and grain boundary recombination velocity by laser excitation

A new technique is developed here for accurately evaluating the diffusion length and the grain-boundary recombination velocity in a semiconductor bicrystal from photovoltage measurements taken by scanning of a laser "infinite line" parallel to the grain boundary. The present method can also be applied in semiconductor bicrystals of small diffusion length, in contrast to a previous scanning laser spot technique, which is limited by the spot size. Experimental application shows that the values of diffusion length and grain boundary recombination velocity measured with the present method are in very good agreement with the values measured in previous work in similar materials.

Influence of grain boundary recombination velocity and grain size on the minority carrier lifetime in polycrystalline semiconductors

The effective minority carrier lifetime in polycrystalline semiconductors is determined in terms of the grain size and the recombination velocity at the grain boundaries. The effect of the grain boundaries is to reduce the effective lifetime when the grain size is smaller than few diffusion lengths. By comparison of the experimental with the theoretical results it is possible to determine the grain boundary recombination velocity. Excellent agreement is observed between theory and experiment in polysilicon.