Showing papers on "Ion implantation published in 1970"

A model for the formation of amorphous Si by ion bombardment

Abstract: The effective annealing of ion implantations in Si is aided by the formation of continuous amorphous layer. The amorphous layer regrows epitaxially at 500–600°C and incorporates the dopant in an electrically active, uncompensated form. A phenomenological model is proposed which, with adjustable parameters, accounts for the variation of the critical dose required to produce a continuous amorphous layer by ion bombardment with ion, target, temperature, and, with minor additional assumptions, dose rate.

Esr and optical absorption studies of ion‐implanted silicon

Abstract: The g value, line shape, and linewidth of an ESR signal in Si layers which have been damaged by ion implantation of Si, P, or As at room temperature are found to be identical to those of the electron states observed in amorphous Si films prepared by rf sputtering. Interference phenomena observed in the optical absorption spectra allow a determination of the depth to which the Si has been damaged by the energetic heavy ions. These two techniques together provide a new tool for investigating lattice disorder in ion‐implanted Si layers.

Spatial distribution of energy deposited into atomic processes in ion-implanted silicon

Abstract: A method is given for calculating the spatial distribution of the production of a quantity, Q, averaged over many ions incident randomly on a solid for any energy dependent interaction between the ions and target atoms. The method is basically a two step method. First, the spatial distribution of the ions in the solid is followed as the ions lose energy. Then, at each intermediate energy the spatial distribution of Q-production is obtained and the result is integrated over the range of intermediate energies assumed by the ions. Saturation effects are ignored in the procedure so that explicit consideration must be given to saturation effects when applying the method to high dose cases. Annealing and diffusion effects are also ignored, and the method is restricted in applicability to experimental conditions where annealing and diffusion are unimportant. Results of calculations by this method are presented of the depth distribution of energy ultimately deposited into atomic processes for Li7, B11, C...

Channeling study of boron‐implanted silicon

Abstract: The channeling technique has been used to determine the lattice location of boron implanted into silicon by using the B11 (p,α) nuclear reaction. Approximately 30% of the boron lies on substitutional sites after a room‐temperature implantation of 3×1015/cm2. The substitutional content decreases with annealing temperature up to 700 °C and then increases at higher annealing temperature. This explains a reverse annealing behavior observed in the carrier concentration. Nearly all of the boron lies on substitutional sites after annealing at 1100 °C. The nonsubstitutional boron atoms do not occupy the normal tetrahedral interstitial sites. For annealing temperatures up to 500 °C they appear to lie along atomic rows, but they do not lie midway between row lattice sites. After higher‐temperature annealing their location appears less well‐defined.

Analysis of disorder distributions in boron implanted silicon

Abstract: Channeling effects on the yield of large angle backscattered energetic H+ or He+ were used to investigate properties of the disorder in Si samples previously implanted with B11 ions at substrate temperatures of -150°C to -50°C. Exposure of the implanted samples to the analyzing beam at room temperature caused a large reduction in the amount of disorder measured. A successive layer removal technique by anodic oxidation and stripping was used to obtain a depth scale for 1.8 MeVHe+ analyses of lightly disordered samples. The results of additional experiments indicate that for 1 MeV He+ in Si the aligned stopping power is about 80 per cent of the random stopping power. Measured and calculated values of the minimum yield X R for the substrate underlying amorphous layers suggested a plural scattering treatment of dechanneling of the analyzing beam using previously measured values of the critical angle. The scattering center distributions calculated from this treatment of dechanneling are presented for ...

The lattice location of boron ions implanted into silicon

Abstract: Channeling‐effect measurements have been used to investigate the lattice location of boron atoms implanted into silicon at an energy of 56 keV and with doses in the interval 1014–1015 ions/cm2. Measurements have been made as a function of implantation temperature and subsequent anneal treatment. The effect of post‐bombardment with different doses of 680‐keV protons has also been investigated.

Technique used in Hall effect analysis of ion implanted Si and Ge

Abstract: Hall effect and sheet resistivity measurements are commonly used to evaluate ion implanted layers. We describe the general principles, sources of error, and factors influencing the measurements. Implantation conditions, sample preparation and measurement technique used for van der Pauw geometry are discussed. Considerations important in annealing, layer removal and measurement temperature are also presented.

Damage produced by ion mplantation in silicon

Abstract: Silicon specimens were irradiated with Ne+, B+, P+ and Sb+ ions. The energy was 80kV, and the doses in the range 1013 to 1016 ions/cm2. The resulting structural damage was investigated by both scanning and transmission electron microscopy. The SEM electron channelling pattern method was used to determine the variation of gross damage with ion type, dose, depth below the surface, and annealing treatment. The TEM method was used to study the detailed nature of the defects formed by the annealing. These included dislocations, rod-like structures, micro-twins, semi-polycrystalline material, misoriented zones, etc. The results are briefly compared with known electrical properties of implanted silicon, and some correlations noted.

Ion implantation depth distributions: energy deposition into atomic processes and ion locations

Abstract: A useful method of calculating the energy/unit depth deposited in atomic processes by energetic ions in solids is presented. The calculated energy density is shown to correlate well with previous Monte Carlo calculations of the vacancy concentration resulting from ion bombardment and recent experimental measurements of the depth distribution of ion damage. The method also provides the depth distribution of ions in the solid as a function of their energy during the stopping process. This information would allow, for example, calculation of the location and rate of various energy‐dependent interactions between incident ions and host atoms.

INFRARED STUDIES OF THE CRYSTALLINITY OF ION-IMPLANTED Si.

Abstract: The crystallinity of ion-implanted silicon has been investigated using ion mass and ion fluence dependences of divacancy formation as measured by the characteristic 1.8 μ absorption band. Room temperature, nonchanneled implants of 400-keV B11, Zn64, and Sb121 ions were performed to maximum fluences of 1014 ions/crn2 for Sb and Zn and to 2 × 1015 ions/cm2 for B. The results are interpreted on the basis of ion energy spent in atomic processes per unit volume, e, within the implanted layer. For e ≤ 1019 keV/cm3 the energy to form a divacancy (1.5 ± 0.5 keV) is nearly ion independent. Maxima appear in the divacancy densities at ∼1013 Sb ions/cm2 and ∼2 × 1013 Zn ions/cm2 where e ≤ 1020 keV/cm3. The divacancy density for B implantation did not exhibit a distinct maximum at E = 1020 keV/cm3, but continued to increase with fluence. The B results are attributed to defect motion because divacancies are observed beyond the calculated depth for energy deposition after a high fluence B implant. In addition t...

Methods of manufacturing a semiconductor device

Abstract: A method of making a semiconductor device is described which combines ion implantation with another process of forming an impurity-containing semiconductor region. In particular, a surface-adjoining region of a semiconductor is formed in such manner that the portion of that region adjacent the surface is formed by a process other than ion implantation, whereas a surface remote or buried portion of that region which defines a P-N junction is formed by ion implantation. This combination of steps yields improved semiconductor devices.

The ejection of atomic particles from ion bombarded solids

Abstract: Considerable interest has developed in the last decade in the study of atomic collisions in solids, particularly single crystal solids. This has been heightened by the observation of channelling and the interest in industrial aspects of ion implantation. In the following article, we present an up-to-date account of the work being done on the particles ejected from a single crystal, following ion bombardment. These particles include sputtered atoms and ions, scattered atoms and ions and secondary electrons. An attempt has been made to indicate the similarities and differences, particularly between the energy and angular distributions of these ejected particles. The major link involving the dependence of these distributions of all particles on crystallography is stressed.

Properties of ion implanted silicon, sulfur, and carbon in gallium arsenide

Abstract: Work is described in which chromium-doped semi-insulating gallium arsenide has been successfully doped n-type with ion implanted silicon and sulfur, and p-type with ion implanted carbon. A dilute chemical etch has been employed in conjunction with differential Hall effect measurements to obtain accurate profiles of carrier concentration and mobility vs. depth in conductive implanted layers. This method has so far been applied to silicon-and sulfur-implanted layers in both Cr-doped semi-insulating GaAs and high purity vapor grown GaAs. In the case of sulfur implants, a strong diffusion enhancement has been observed during the annealing, presumably due to fast-diffusing, implantation-produced damage. Peak doping levels so far obtained are about 8 × 1017 electrons/cm3 for silicon implants and 2 × 1017 electrons/cm3 for sulfur implants. Mobility recovery has been observed to be complete except in regions near the surface which are heavily damaged by the implantation.

ion-implanted junctions and conducting layers in SiC

Abstract: n-type conducting layers have been formed in n- and p-type hexagonal SiC and n-type cubic SiC by implanting ions from column V of the periodic table; N, P, Sb, or Bi. The implantations were made at room temperature and with energies ranging from 5 to 300 keV. The implanted layers have been evaluated by van der Pauw-Hall effect and sheet resistivity measurements and by scanning electron microscopy for anneal temperatures ranging from 1100 to 1800°C. Type conversion of the implanted layers to n-type has been observed after a 750°C anneal. Considerable lattice reordering is suggested from the observed carrier mobility values after annealing at 1600°C. We have attempted to form p-type layers in n-type SiC by implanting Re, B, Al, Ga, and Tl. Application of anneal procedures identical to those used to form n-type layers have not resulted in measurable p-type layers. The p-n junctions formed by the donor-implanted layers have been evaluated as a function of anneal temperature. After annealing at 1200°C...

Correlation of electron microscope studies with the electrical properties of boron implanted silicon

Abstract: A number of 1ω cm n-type silicon substrates have been implanted with boron at room temperature. Samples from these substrates were annealed at predetermined temperatures between 600°K and 1400°K and four-probe electrical measurements and transmission electron micrographs were taken. An attempt has been made to correlate the two studies and it is possible to explain the observed annealing behaviour in terms of precipitation during recrystallization, and migration of defects.

Anomalously high yields of elastically scattered 12C-ions from Zr, Hf, Tl, and Hg atoms implanted into silicon

Abstract: Silicon single crystals were implanted under various conditions with Zr, Hf, Tl, and Hg. The yield of scattered 1.8 MeV 12C-ions was studied as a function of the angle of incidence. Around the (111) direction the normal channeling dip was observed. Around the (110) direction either a narrow peak or a narrow peak superimposed on a channeling dip was found. The occurence of the peak is interpreted as being caused by scattering from impurity atoms located in the tetrahedral interstitial holes in the silicon lattice. For a beam entering along a (111) direction these atoms are screened, but they are fully exposed to a beam entering parallel to a (110) direction. The implications of this effect to earlier measurements of the lattice location of impurity atoms in silicon and germanium will be discussed. A similar effect has recently been observed at the University of Aarhus.(6)

Anneal behavior of defects in ion-implanted GaAs diodes

Abstract: Doping of semiconductors by implantation of high-energy ions creates lattice damage which in general must be removed by annealing to form good qualityp-n junctions. Implantation of zinc or cadmium ions inton-type gallium arsenide substrates held at 400°C produces ap-n junction after the samples are annealed at elevated temperature (≥500°C for zinc, ≥600°C for cadmium). However, the resulting junctions are not abrupt; they contain a semiinsulating (I) layer and have ap-i-n structure. The thickness of the semiinsulating layer changes with annealing. For example, an implant of 1.3×1015 per sq cm, 20 kv, Zn+ ions produced a junction with an I layer of 28 μ thickness after annealing for 10 min at 600°C. An identically implanted sample, annealed for 10 min at 900°C, had an I layer thickness of 120 μ. A similar increase in I layer thickness with annealing was observed for samples implanted with 20 kv Cd+ ions at 400°C. Implantation of Zn+ and Cd+ ions into GaAs substrates held at room temperature produced junctions with much thinner I layers after annealing than those observed for the 400°C implants. The formation of the semiinsulating layer in the junction and its subsequent variation of thickness with annealing are attributed to deep diffusion of defects during the implantation or subsequent anneal, which produces compensation to the depth where the concentration of defects equals the substrate impurity concentration. The compensating centers are thought to be arsenic vacancy-substrate dopant atom complexes.

Apparatus for bombarding a target with ions

Abstract: An electromagnetic separator adapted for ion implantation on an industrial production scale has its beam current stabilized and a mechanism within the target chamber for automatically moving targets through the ion beam according to a predetermined scanning pattern.

Fabrication of insulated gate field-effect transistors involving ion implantation

Abstract: An insulated gate field-effect transistor is made which utilizes both Schottky barrier connections and ion-implanted zones. The resultant structure incorporates source and drain zones, which are formed by ion implantation and whose spacing is fixed by the gate electrode, and source and drain electrodes which make ohmic connection to the implanted source and drain zones and rectifying connections to unimplanted material.

Depth Distribution of Divacancies in 400-keV O + Ion-Implanted Silicon

Abstract: The integral depth distribution for divacancies produced in silicon at room temperature by 400‐keV O+ion implantation has been measured. The divacancy distribution was determined from repeated measurements of the characteristic 1. 8μ absorption band following successive anodizations and strippings of the implanted layer. Most of the divacancies are located between 4500 and 12 000 A with a half‐value at ∼ 7500 A and a concentration of ∼ 4 × 1019 cm−3 near the center of the distribution. The measured integral depth distribution for the ion‐produced divacancies is proportional within experimental error to theoretical calculations by Brice for the integral depth distribution of ion energy spent in atomic processes.

Evidence for vacancy motion in low temperature ion-planted Si

Abstract: Recent results have reported a strong implantation-temperature dependence between 125°K and room temperature for lattice disorder produced by 200-keV B implantation into Si. Using our previous annealing model incorporating implant temperature and dose rate, we have calculated a characteristic defect annealing time at the threshold of the crystalline to amorphous transition at 173°K for 1 μA/cm2 and find that it agrees very closely with that for neutral vacancy annealing. In addition, we have made measurements of the 1.8μ infrared absorption band, characteristic of the Si divacancy, following room temperature and 85°K implants of 400-keV B11 or Sb121 ions. Very few divacancies are observed immediately after 85°K implants, but annealing growth of divacancies occurs between 150 and 300°K yielding a density almost equal to that for the same ion Ruence at 300°K. These results strongly suggest that below 300°K neutral vacancy motion and trapping control both the implantation-temperature dependence of l...

Electrical properties of Cd, Zn and S ion-implanted layers in GaAs

Abstract: The electrical properties of cadmium, zinc, and sulfur ion-implanted layers in gallium arsenide have been measured by the van der Pauw-Hall technique. Ion implantation was performed with the substrates held at room temperature. The dependence of sheet resistivity, surface carrier concentration, and mobility on ion dose and on post-implantation anneal temperature was determined. In the case of 60 keV Cd+ ions implanted into n-type substrates, a measurable p-type layer resulted when samples were annealed for 10 minutes at a temperature in the range 600—900°C. After annealing at 300—900°C for 10 minutes, 100 per cent electrical activity of the Cd ions resulted for ion doses ≤ 1014/cm2. The properties of p-type layers produced by implantation of 85 keV Zn+ ions were similar to those of the 60 keV cadmium-implanted layers, in that no measurable p-type behavior was observed in samples annealed below a relatively high temperature. However, in samples implanted with 20 keV Zn+ ions a p-type layer was obs...

High-power millimeter wave IMPATT oscillators with both hole and electron drift spaces made by ion implantation

Abstract: CW powers of 640 mW at 50 GHz have been obtained from double-drift region IMPATT diodes. This result represents the highest product of CW power times frequency squared obtained to date from any IMPATT diode. The diodes are p+pnn+structures and have both hole and electron drift spaces. The systematic fabrication (by ion implantation) and the evaluation of the dc and millimeter wave characteristics are presented.

Ion bombardment induced phase transformations and inert gas mobility in the semiconductors Ge, Si, and GaAs

Abstract: The present study contributes some new aspects to the general understanding of the ion implantation behaviour of 3 common semiconductor materials, and of diffusion processes in these materials. Single crystals of Si, Ge, and GaAs were bombarded with Kr- or Xe-ions at energies of 40 or 500 keV and doses between 1011 and 2 × 1016 ions/cm2. Gas release measurements and Rutherford scattering of 1 MeV He+-ions combined with channeling were used to study bombardment damage (amorphization) and inert gas diffusion. At low bombardment doses (1011 ions/cm2) and energy (40 keV), no damage was observed and the gas release was compatible with volume diffusion resembling Group I and VIII behaviour. Hence, the pre-exponential terms, D 0, were low (range 10-5±1 cm2 sec−1) and the activation enthalpies, Δ H, were much lower than those of self-diffusion or of diffusion of Group III and V elements. The Δ H's for gas diffusion followed the relation Δ H = (1.05±0.1) × 10−3 Tm eV with the melting point, Tm , in °K. Th...

PHOTOELECTRONIC PROPERTIES OF ION-IMPLANTED CdS.

Abstract: The photoelectronic properties of CdS successively implanted with 1.0‐, 0.5‐, and 0.3‐MeV P+ ions versus annealing cycles were studied. Diodes thus formed were highly photosensitive, with photoconductive gains between 102 and 104. Yellow electroluminescence was observed at 77°K. Short annealing creates centers with a sheet resistance thermal activation energy of 0.13 eV. Prolonged annealing creates states 0.75–0.8 eV below the band edge.

The distribution of damage produced by ion implantation of silicon at room temperature

Abstract: Experiments designed to determine the damage distribution produced by energetic heavy ions in Si are described. For low ion doses (1011 to 1013 cm−2), the location of the damage peak was determined by changes, which were produced by ion damage, in the electrical properties of thin (0.6 μ), uniformly doped Si layers as a function of depth. The ratio of the peak position in the damage distribution to the peak position in the ion distribution was determined to be approximately 0.6 ± 0.1 for Si29 (150 keV), P31 (70, 140, 200 keV), B11 (60 keV), and As75 (280 keV). A comparison of carrier removal rates and the number of displaced lattice atoms previously reported from back-scattering experiments with He ions indicates that the nature of damage produced by Si29 and B11 are different. In the former case, cluster damage (amorphous disordered regions) appears to be an important form of radiation damage, while in the latter case, isolated defects are the dominant form of radiation damage for room temperatu...

Ion implantation in InAs and InSb

Abstract: Ion implantation in InAs and InSb with sulfur and zinc ions has been used to fabricate p-n junction diodes which have been characterized for their infrared detector properties. Planar mosaic infrared detectors have been produced in both materials with good characteristics. Blackbody detectivities (D∗BB) of 2 × 109 cm Hz1/3/ watt and 4 per cent uniformity have been measured for InAs. Similarly, InSb has shown D∗BB values of 1.3 × 1010 cm Hz1/2/watt and 49 per cent uniformity between elements in an array. Experimental range-energy data for zinc in InSb has been obtained between 0.2 and 1.8 MeV and compared with predicted values from the LSS theory. Theory predicts a deeper range than experimental values indicate; however, the differences are sufficiently small to make the curves useful for device design. The slopes of the curves indicate that a large component of nuclear stopping predominates in this energy region.