Quantum photoeffects associated with photons of wavelength less than 0.8 micron lead to higher bulk and surface diffusion coefficients. We have exploited this fundamental property in designing rapid isothermal processing (RIP) systems for shallow junction formation in silicon. A detailed comparative study of diffusion, metallization and CVD with and without high energy photons has been carried out. The results show that microscopic defects, cycle time and processing temperature is lower than what can be achieved byconventional methods is realized by using photons in the ultra violet (UV) and vacuum ultra violet (VUV) spectrum.

We performed a theoretical study of the microscopic arrangement of the C atoms in Si1–x–yGexCy, alloys using the Sankey local-orbital density-functional formalism. Our first-principles calculations show that in the dilute limit, the lattice constant in Si1–yCy, alloys decreases much more (by about 2/3) with concentration than predicted by Vegard's law. We show thatthis result is consistent with Raman and infrared experiments and with previous empiricalcalculations. As the C concentration increases, the interactions between the C atoms become important: Interstitial C becomes more abundant and increases the lattice constant, and the substitutional C atoms array under appropriate growth conditions. The effect of this ordering on the lattice constant is small, but it can be seen in the Raman spectra of layer-by-layer grown samples.

The formation of cavity microstructures in silicon following helium implantation (10 or 40 keV; 1×1015, l×1016 and 5×1016 cm−2) and annealing (800 °C) is investigated by means of Transmission Electron Microscopy (TEM), Rutherford Backscattering Spectrometry and Channeling (RBS/C), and Elastic Recoil Detection (ERD). The processes of cavity nucleation and growth are found to depend critically on the implanted He concentration. For a maximum peak He concentration of about 5×1020 cm−3 the resulting microstructure appears to contain large overpressurized bubbles whose formation cannot be accounted by the conventional gas-release model and bubble-coarsening mechanisms predicting empty cavities. The trapping of Fe and Cu at such cavity regions is studied by Secondary Ion Mass Spectrometry (SIMS).

Cavities, formed in Si by hydrogen implantation and subsequent annealing, can provide ideal gettering sites for metal impurities. In this study, we have observed large differences in the accumulation of Cu at cavities depending on whether Cu was introduced into Si during cavity formation or into wafers with pre-formed cavities. The observed behaviour is consistent with a high flux of Si interstitials emitted during cavity formation which induce the dissolution of Cu3Si and the enhanced transport of Cu to cavities. In further studies, boron implantation was carried out into wafers containing pre-formed cavities and transient enhanced diffusion (TED) of boron was suppressed duringsubsequent annealing.

High-energy, self-ion implantation has been used to form deep gettering layers in Si. Subsequently samples have been contaminated with Cu and subjected to heat treatment. The residual defects act as gettering centres for Cu. The decoration of defects byCu making them detectable by secondary ion mass spectromety analysis. Metastable defect complexes have been detected which, because of their small size, are not directly detectable by other analytical techniques such as transmission electron microscopy and MeV-particle channeling. These defects are probably of interstitial type and have been found mainly midway between the sample and the projected ion range, i.e. around Rp/2. The gettering ability of these small defect complexes may largely exceed that of the post-anneal damage at the projected i.e range, Rp. The results obtained demonstrate that by means of metal gettering the formation, growth and dissolution of very small defect complexes in ion-implanted Si can be studied.

External gettering techniques trap removed impurities in regions of a semiconductor which contain defects or in which the solubility is enhanced like in phosphorus diffused layers or in Al-Si alloy. If a fast diffuser like nickel is used in place of Al to obtain an alloy or a compound close to the surface, contamination and decontamination can occur resulting of the indiffusion and removal of nickel atoms. P type FZ samples were partly covered by nickel (nickel òn part 1 and not on part 2) and then annealed at 950°C for 1h.

As expected, after nickel indiffusion the diffusion length of minority carriers decreases but the values are always higher in part 1 than in part 2. Scanning infrared microscopy detects precipitates in part 2 and not in part 1. The results are explained assuming that the nickel-silicon compound formed at the surface serves as a diffusion source as well as an external self-gettering region.

Interstitial copper (CUi) drift induced transient capacitance signals of a Schottky barrier are analysed after copper indiffusion at various temperatures and thermal quenching. Against the general expectation we show that most Cu impurities remain interstitially dissolved. Consequently, such transient ion drift analyses can be used to detect copper in silicon with a detection limit of the order of 1011cm−3. We applied the method to study Cu contamination of silicon samples after chemical cleaning and high temperature annealing. After a short dip into a copper contaminated aqueous solution and subsequent annealing at 1000C for 10 minutes, a strong correlation is found between the bulk copper concentration and the copper density in the solution. A density as low as 50 atomic parts per billion is found to be sufficient to induce a bulk copper concentration of about 5.1013cm−3.

As devices are scaled down and their performance is enhanced, the role of oxygen in silicon remains critical. The precipitation of oxygen away from the device region leads to desirable gettering of deleterious impurities and is a useful technique. In the device region, however, oxygen has recently been shown to take part in the annealing of dopant implants and segregate to damaged regions. This may result in the stabilization of damage structures and the attraction of unwanted impurities to sensitive regions of the device. In this work we examine these phenomena and their potential impact on device performance. For example, oxygen has been observed to segregate at two positions following a 1000°C, one hour anneal of a 2 MeV, 3.5×1015/cm2 phosphorus implant into CZ p-Si. The oxygen concentration is greatly enhanced above its initial level of 1018 O/cm3, both in the end-of-range region and at approximately half that depth, near the peak of the predicted damage profile. Iron deliberately introduced at the surface prior to annealing, getters out strongly at these locations No significant redistribution of carbon can be detected. The results assume significance in view of the increasing use of high energy implantation for both epi-layer replacement and retrograde well formation.

It is critical to understand the behavior of metallic impurities in polycrystalline silicon used for solar cells. These impurities significantly increase the minority carrier recombination rate and, in turn, degrade cell performance. Impurity gettering is a commonly used method to remove these impurities from the material, however, past work has suggested that impurity release from structural defects drastically limits the gettering process. Presently, there is only a limited understanding of impurity release from structural defects. In this work, a correlation between structural defects and the location of metal impurities in as-grown material is established and the release of nickel and copper from structural defects in polycrystalline silicon was studied in as-grown material and after sequential thermal treatments which dissolve the impurities into the silicon matrix. Synchrotron-based x-ray fluorescence impurity mapping with spatial resolution of ≈ 1μm, was used to determine impurity distributions after each thermal treatment.

Aluminium-silicon alloying is applied to dislocation-containing FZ silicon samples in order to verify if this external gettering technique is able to remove impurities trapped by these defects. Samples were scratched (front surface), bent and annealed at 750°C for 6h in order to create dislocation arrays. Standard samples and also nickel contaminated samples were investigated by light beam induced current (LBIC) mapping technique before and after alloying at 900°C for 4h with a 1 μm thick aluminium layer deposited on the backside. The LBIC technique detects the features of the dislocation array in agreement with X ray topographs. In the standard samples, the contrast is relatedto an inadvertent contamination of the samples and disappears after gettering. In the contaminated samples, dislocation contrast is higher, the aluminium treatment attenuates strongly the dislocations electrical activity but not so neatly as for the standard samples. This result is explained by the possible microprecipitation of nickel at dislocations.

Si was implanted with fluorine and boron ions. Boron gettering by fluorine was studied. By varying the implantation energy of F+ ions, three types of samples were made in which Rp(F)<Rp(B), Rp(F)∼Rp(B) and Rp(F)>Rp(B). There are three special regions in which the redistribution of Boron atoms during annealing differ. The first one is the near surface region in which the Boron concentration exceeds the value of Boron solid solubility at the annealing temperature. In this region, B atoms are intensively redistributed and have two peaks of concentration. The positions of these peaks coincide with the position of the peaks on the fluorine concentration profile. In the second specific region, the boron concentration is less than the value of solid solubility in Si and higher than the value of the intrinsic carrier concentration in Si at the annealing temperature. In this region, the B concentration was identical in all samples, and it exponentially decreased to the value of 5×108 cm−3. The slope of the boron concentration in the third region dictates the depth of p-n junction. A possible mechanism of boron gettering by fluorine is discussed.

We study the leading mechanisms of kink mobility of 30° and 90°-partial dislocations in the glide set {111} of silicon. The calculations are performed using a new empirical potential for Si, which has been shown to give an accurate description of core properties of partial dislocations [1], to study mechanisms of kink formation and migration. In the case of 30°-partial dislocation, two kinds of kinks are identified, left and right, which have different structures, formation energies, and mobilities.

We report theoretical investigations of the chemical trends in the electronic properties of substitutional gold-interstitial transition-metal complexes in silicon. The results show that the stable pairs in trigonal symmetry are formed by a covalent mechanism which includes, besides Au and TM impurities, also the Si neighbors, rather than being derived from interactions between two electrostatically bound point charges.

Determination of transport mechanisms and energetics in amorphous silicon presents an interesting modeling challenge. Transport in amorphous silicon films is likely to involve energetically distributed traps and mobile species, as in the case of hydrogen and electron diffusion. Detailed kinetic models using discrete energy levels have been developed, however, the density of states of the diffusing species in amorphous silicon is likely to be continuous and distributed, due to the amorphous nature of the films. We have developed a mean-field, equilibrium based model which utilizes a continuous density of states for the diffusing species. The transport in amorphous silicon is modeled as a function of a gradient in the quasi-chemical potential, rather than concentration, of the diffusing species. The model is applicable when the local equilibration processes are fast relative to the transport process. This approach is extremely numerically efficient, as well as flexible, allowing for modeling of tracer experiments, such as deuterium diffusion in a-Si:H films, and possible changes in density of states with time, temperature, and diffusing species concentration. We demonstrate the utility of the model by simulating hydrogen evolution from a-Si:H films.

Fe precipitation on bulk stacking faults in Czochralski-grown silicon are investigated by means of the electron-beam-induced-current (EBIC) technique and other techniques. It is found that Fe impurities only precipitate on Frank partial dislocations bounding stacking faults when the specimen is cooled slowly; however, they precipitate on both Frank partials and fault planes when the specimen is cooled fast. It is explained that small oxygen precipitates on fault planes, together with Frank partials, serve as the gettering centers for Fe impurities in the fast cooled specimen.

Highly doped (∼1018 to 1021cm−3) polycrystalline Si1-xGex films, crystallized from amorphous (a) state at relative low temperatures, are prospective materials in a variety of applications, such as liquid-crystal displays, solar cells and integrated thermoelectric sensors on large-area glass substrates. Since the nature of the grains in the crystallized film defines properties such as carrier mobility, the nucleation and growth process of the a-SiGe films is of fundamental interest. We have studied the crystallization of undoped and highly doped (B or Ga) amorphous SiGe films. The films were deposited by RFCVD or molecular beam on oxidized (001)Si and for TEM study on cleaved NaCl. The incubation time and grain growth rate were studied by means of in situ TEM using a heating stage. The crystallization process in undoped SiGe followed Avrami relationship. An average grain size between 0.1 and 2μm was observed. However, the highly p-doped (with B or Ga) SiGe films crystallized to a stable nanocrystalline structure (grain size <10nm). The process of the a-SiGe crystallization is explained on the basis of self-diffusion. During the first stage, the nucleation of crystals is accompanied with nonequilibrium vacancy generation at the amorphous/crystalline interface. During the second stage, the growth of crystals takes place by vacancy outdiffusion which is hindered by B and Ga interaction with vacancies.