Supersaturated levels of interstitial oxygen in Czochralski silicon can lead to the formation of oxide precipitates. Although beneficial from an internal gettering perspective, oxygen-related extended defects give rise to recombination which reduces minority carrier lifetime. The highest efficiency silicon solar cells are made from n-type substrates in which oxide precipitates can have a detrimental impact on cell efficiency. In order to quantify and to understand the mechanism of recombination in such materials, we correlate injection level-dependent minority carrier lifetime data measured with silicon nitride surface passivation with interstitial oxygen loss and precipitate concentration measurements in samples processed under substantially different conditions. We account for surface recombination, doping level, and precipitate morphology to present a generalised parameterisation of lifetime. The lifetime data are analysed in terms of recombination activity which is dependent on precipitate density or on the surface area of different morphologies of precipitates. Correlation of the lifetime data with interstitial oxygen loss data shows that the recombination activity is likely to be dependent on the precipitate surface area. We generalise our findings to estimate the impact of oxide precipitates with a given surface area on lifetime in both n-type and p-type silicon.

Abstract not Available.

Transient and quasi-steady-state photoconductance methods were used to measure minority carrier lifetime in p-type Czochralski silicon processed in very clean conditions to contain oxide precipitates. Precipitation treatments were varied to produce a matrix of samples, which were then characterised by chemical etching and transmission electron microscopy to determine the density and morphology of the precipitates. The lifetime component associated with the precipitates was isolated by preventing or factoring out the effects of other known recombination mechanisms. The lifetime component due to unstrained precipitates could be extremely high (up to ~4.5ms). Recombination at unstrained precipitates was found to be weak, with a capture coefficient of ~8 x 10-8cm3s-1at an injection level equal to half the doping level. Strained precipitates and defects associated with them (dislocations and stacking faults) act as much stronger recombination centres with a capture coefficient of ~3 x 10-6cm3s-1at the same level of injection. The lifetime associated with strained precipitates increases with temperature with a ~0.18eV activation energy over the room temperature to 140°C range. The shape of the injection level dependence of lifetime was similar for all the specimens studied, with the magnitude of the lifetime being dependent on the precipitate density, strain state and temperature, but independent of precipitate size.

Transient and quasi-steady-state photoconductance methods were used to measure minority carrier lifetime in ∼10 Ω cm p-type Czochralski silicon processed in very clean conditions to contain oxide precipitates. The nucleation and growth times for precipitation were varied to produce 35 samples, which were then characterised by chemical etching and transmission electron microscopy to determine the density and morphology of the precipitates. The effects of other known recombination mechanisms (band-to-band, Coulomb-enhanced Auger, iron-related, and boron-oxygen related) were factored out to isolate the lifetime component associated with the precipitates as accurately as possible. In the samples processed to contain mainly unstrained precipitates, it was shown that the lifetime component due to the precipitates could be extremely high (up to ∼4.5 ms). Recombination at unstrained precipitates is weak and it is estimated that the capture coefficient lies between 3 × 10−8 cm3 s−1 and 1.3 × 10−7 cm3 s−1 at an injection level corresponding to half the doping level. Strained precipitates act as strong recombination centres with a capture coefficient of ∼1 × 10−6 cm3 s−1 at the same level of injection. For the samples investigated, the effective capture coefficient is increased by a factor of ∼3 to 4 when other extended defects (such as dislocations and stacking faults) accompany the strained precipitates. The shape of the injection level dependence of lifetime was similar for all the specimens studied, with the magnitude of the lifetime being dependent on the precipitate density and strain state but approximately independent of precipitate size.

The Optical Precipitate Profiler (OPP) and similar instruments are establishing themselves as routine techniques for measuring bulk micro-defect density (BMD) and denuded zone depths in semiconductor wafers. This task has traditionally been done by means of the cleave-and-etch technique in which defects are counted on a cleaved edge of an etched wafer using an optical microscope. The OPP makes this task faster, simpler and more accurate. In addition it can perform the measurement on a whole wafer. We have used the OPP to measure defect density and scattering intensities from a set of silicon wafers with different precipitate sizes. The wafers were all given the same nucleation heat treatment but the precipitates were then grown at 1000°C for 1, 2, 4, 8, 16 and 32 hours. The size distributions and densities of the precipitates in each wafer were determined by Transmission Electron Microscopy (TEM). In a second experiment wafers with precipitate densities varying over four orders of magnitude as determined by cleaveand-etch were also measured by the OPP. From these experiments we have estimated that the OPP can detect defects with an equivalent radius of 16 ± 8 nm and larger. We have also found that the scattering voltage (strength) depends on the precipitate radius, r 3 , as approximately r 3 and that there is an upper number density limit above which the OPP is unable to measure the precipitate density accurately. For our setup this upper density limit was about 1.5 - 2.0 × 10 10 defects/cm 3 as measured by cleave-and-etch.

We present a numerical model for the calculation of the electro-optical properties (energy diagram, density of states, gain ...) of a semiconductor quantum dash active material including the presence of the wetting layer (WL) as a uniform film formed at the bottom of the self-assembled quantum dashes. The numerical results show that the WL significantly changes the density of states and reduces the gain from the excited state confined in the dashes.

Systematic measurements aimed at quantifying the contribution of precipitates to the infrared absorption of silicon are presented here, carried out on wafers with the same initial interstitial oxygen (Oi) concentration subjected to three-step treatments. To precisely determine the intensity of the precipitate-related bands, the measurements were performed at liquid He temperature. The results demonstrate the possibility of accounting for the contribution of precipitates in the standard quantitative determination of Oi in silicon at room temperature.

This work investigates the determination of oxygen precipitate density in silicon using non-destructive methods, based essentially on electrical and optical techniques. Regarding electrical techniques, minority carrier diffusion length measurements by means of surface photo voltage or electrolytical metal tracer methods to determine the density of oxygen precipitates in p-type silicon wafers is evaluated, taking advantage of the activity of oxygen precipitates as recombination centres for minority carriers. Regarding optical techniques, the scanning infra-red microscope is considered, which can detect oxygen precipitates in silicon exploiting their effectiveness as light scattering centres. These techniques are compared with the traditional etching technique, considering samples in a wide range of precipitate density (5×106–1011cm−3) and submitted to a wide variety of thermal treatments. Correlations are discussed with the support of transmission electron microscopy, X-ray diffraction and infrared spectroscopy observations, which provide complementary information on size, morphology and matrix stress of the oxygen precipitates.

A study of the optical response in the infrared (IR) spectral region of oxygen precipitates in silicon is performed at liquid helium temperature. The measurements concern a set of samples with the same density of precipitates of different dimensions and a second set of samples with different densities of precipitates of equal dimension. Two main shapes of precipitates are detected in both sets, with different behavior in terms of volume fractions occupied by each of them in the sample as a function of the thermal treatments performed.

Quantitative measurements of the oxygen precipitate rate as a function of annealing were made in Czochralski-grown silicon wafers that contained different initial concentrations of oxygen. All wafers were annealed at 1000 °C for 15 min to ensure that the initial cluster-size distributions were identical in all samples of the same composition prior to the multi-step annealing treatments used for the precipitation studies. The experimental data are compared with numerical predictions for time-dependent nucleation within the classical theory of nucleation. Quantitative agreement is obtained between the measured and calculated densities of oxygen precipitates for nucleation temperatures greater than 600 °C, but only over a narrow range of oxygen composition. Below 600 °C, the measured density for all samples is orders of magnitude larger than is predicted from the model. Further, the measured data show an anomalously small temperature dependence for the induction time for nucleation that does not scale with the diffusion coefficient, as expected from the classical theory of nucleation. Fundamentally, the classical theory of nucleation cannot explain the time-dependent nucleation of oxygen precipitates for temperatures below 650 °C. A possible reason is given.

SPV technique has been used to control the minority carrier diffusion length changes induced by an annealing at 1150 degrees C with respect to starting [Oi], annealing time and gas ambient. Oxygen precipitation significantly occurs if [Oi] is higher than a threshold value and it is coupled with both a diffusion length reduction and a weak dependence on annealing ambient. Indication of both an oxygen precipitate coalescence process during the annealing and a possible link between oxygen precipitate size and its efficiency as minority carrier recombination centres are observed.

A new kind of silicon wafer and a new class of materials engineering techniques for silicon wafers is described. This wafer, called the “Magic Denuded Zone” or MDZ wafer, is produced through the manipulation of the vacancy concentration and, in particular, vacancy concentration depth profiles in the wafer prior to the development of oxygen precipitates in subsequent heat treatments. The result is a wafer with ideal oxygen precipitation behavior for use in all types of integrated circuit applications. The methods used to prepare such wafers combine Frenkel pair generation with injection and the use of surface sinks. Simulations of the vacancy profiles produced by these techniques are presented and discussed. It is shown that within the range of vacancy concentration accessible by these techniques (up to ca. 10 13 cm −3 ) the rate and oxygen concentration dependence of oxygen clustering can be substantially modified. Such techniques can be used to precisely engineer unique and desirable oxygen-related defect performance in silicon wafers both in terms of distribution and rate of defect formation. One result of the application of such techniques is an ideally precipitating silicon wafer in which the resulting oxygen precipitate profile (denuded zone depth and bulk density of precipitates) is independent of the concentration of oxygen of the wafer, the details of the crystal growth process used to prepare the wafer and, to a very large extent, the details of thermal cycles used to process the wafer into an electronic device. Optimal, generic and reliable internal gettering performance is achieved in such a wafer.