Search

Obtaining a lateral S-parameter image scan from positrons implanted into semiconductor devices can be a helpful research tool both for localizing device structures and in diagnozing defect patterns that could help interpret function. S-parameter images can be obtained by electromagnetically rastering a variable energy positron beam of small spot size across the sample. Here we describe a general hardware and software architecture of relatively low cost that has recently been developed in our laboratory which allows the whole sub-surface S-parameter image of a sample or device to be obtained in real time. This system has the advantage over more conventional sequential scanning techniques of allowing the operator to terminate data collection once the quality of the image is deemed su‐cient. As an example of the usefulness of this type of imaging architecture, S-parameter images of a representative sample are presented at two difierent positron implantation energies.

Temperature-dependent Hall (TDH) measurements and confocal micro-Raman spectroscopy have been used to study the free carrier spatial distribution and scattering mechanism in unintentionally doped GaN film grown on the sapphire substrate with the method of metalorganic chemical vapor deposition. Both the TDH data and the depth-profiled Raman spectra agreed with the existence of a nonuniform spatial distribution of free carriers in the GaN film with a highly conductive layer of ∼1 μm thickness near the GaN sapphire boundary. With the consideration of this parallel conduction channel adjacent to GaN sapphire boundary, detailed analysis of the TDH mobility data suggests that a relatively high concentration of nitrogen vacancies exists and nitrogen vacancy scattering has an important influence on limiting the electron mobility in the bulk film of the present GaN sample.

Two-dimensional angular correlation of annihilation radiation (2D-ACAR) spectra have been taken for 1019 cm−3 phosphorus-doped Si in the as-grown state after having been subjected to 1.8 MeV electron fluences of 1×1018 and 2×1018 cm−2. Positron annihilation lifetime spectroscopy confirms, in accordance with previous works, that positrons are saturation trapping into (VSi:P) pair defect (E-center) monovacancy sites in the electron irradiated samples. In the as-grown case, the positron–electron autocorrelation functions along the [111] and [1-10] directions, obtained through Fourier transformation of the 2D-ACAR data, reveal zero-crossings that deviate only slightly from the lattice points, in a manner consistent with positron–electron correlation effects. Conversely, in the spectra of the irradiated samples, the zero-crossing points are observed to move outward further by between 0.15 and 0.50 A. This displacement is associated with positron annihilation with electrons in localized orbitals at the defect s...

Current–voltage (I–V) characteristics of macroscopic Schottky diodes fabricated on different GaN templates grown by metalogranic chemical vapor deposition on sapphire substrates were investigated. The number of dislocations under the Au Schottky contact was determined by atomic force microscopy combined with hot H3PO4 etching and the screw dislocations in the GaN films were found to have a strong influence on the reverse leakage current of the Au/n-GaN Schottky diodes. The leakage current is increased when high-density screw dislocations exist under the Au Schottky contact. A model based upon the presence of dislocations at the Au/GaN interface has been used to explain this behavior. It has been proposed that these dislocations result in the lowering of the barrier height in the localized regions, and thus significantly affect the reverse I–V characteristics of the Schottky diodes.

A generalized least-square method with Tikonov–Miller regularization and non-negativity constraints has been developed for deconvoluting two-dimensional coincidence Doppler broadening spectroscopy (CDBS) spectra. A projected Newton algorithm is employed to solve the generalized least-square problem. The algorithm has been tested on Monte Carlo generated spectra to find the best regularization parameters for different simulated experimental conditions. Good retrieval of the underlying positron–electron momentum distributions in the low momentum region is demonstrated. The algorithm has been successfully used to deconvolute experimental CDBS data from aluminum.

Deep level transient spectroscopy has been employed to study the deep level defects introduced in n-type 6H-SiC after neutron irradiation. Deep levels situated at E-C-0.23, E-C-0.36/0.44, E-C-0.50, and E-C-0.62/0.68 eV have been detected in the temperature range of 100-450 K, which have been identified with the previously reported deep levels ED1, E-1/E-2, E-i, and Z(1)/Z(2), respectively. Thermal annealing studies of these deep levels reveal that ED1 and E-i anneal at a temperature below 350degreesC, the Z(1)/Z(2) levels anneal out at 900degreesC, while the intensity of the E-1/E-2 peaks is increased with annealing temperature, reaching a maximum at about 500-750degreesC, and finally annealing out at 1400degreesC. The possible nature of the deep levels ED1, E-1/E-2, E-i, and Z(1)/Z(2) are discussed in the context of their annealing behavior. Upon further annealing at 1600degreesC, four deep levels labeled NE1 at E-C-0.44 eV, NE2 E-C-0.53 eV, NE3 E-C-0.64 eV, and NE4 E-C-0.68 eV are produced. Evidence is given that these levels are different in their origin to E-1/E-2 and Z(1)/Z(2). (C) 2003 American Institute of Physics.

In our recent report, [Xu , Appl. Phys. Lett. 76, 152 (2000)], profile distributions of five elements in the GaN/sapphire system have been obtained using secondary ion-mass spectroscopy. The results suggested that a thin degenerate n(+) layer at the interface is the main source of the n-type conductivity for the whole film. The further studies in this article show that this n(+) conductivity is not only from the contribution of nitride-site oxygen (O-N), but also from the gallium-site silicon (Si-Ga) donors, with activation energies 2 meV (for O-N) and 42 meV (for Si-Ga), respectively. On the other hand, Al incorporated on the Ga sublattice reduces the concentration of compensating Ga-vacancy acceptors. The two-donor two-layer conduction, including Hall carrier concentration and mobility, has been modeled by separating the GaN film into a thin interface layer and a main bulk layer of the GaN film. The bulk layer conductivity is to be found mainly from a near-surface thin layer and is temperature dependent. Si-Ga and O-N should also be shallow donors and V-Ga-O or V-Ga-Al should be compensation sites in the bulk layer. The best fits for the Hall mobility and the Hall concentration in the bulk layer were obtained by taking the acceptor concentration N-A=1.8x10(17) cm(-3), the second donor concentration N-D2=1.0x10(18) cm(-3), and the compensation ratio C=N-A/N-D1=0.6, which is consistent with Rode's theory. Saturation of carriers and the low value of carrier mobility at low temperature can also be well explained. (C) 2001 American Institute of Physics.

Undoped and Fe-doped liquid encapsulated Czochralski (LEC) InP has been studied by Hall effect, current–voltage (I–V), and infrared absorption (IR) spectroscopy. The results indicate that a native hydrogen vacancy complex donor defect exists in as-grown LEC InP. By studying the IR results, it is found that the concentration of this donor defect in Fe-doped InP is much higher than that in undoped InP. This result is consistent with the observation that a much higher concentration of Fe2+ than the apparent net donor concentration is needed to achieve the semi-insulating (SI) property in InP. By studying the I–V and IR results of Fe-doped InP wafers sliced from different positions on an ingot, the high concentration of Fe2+ is found to correlate with the existence of this hydrogen complex. The concentration of this donor defect is high in wafers from the top of an ingot. Correspondingly, a higher concentration of Fe2+ can be detected in these wafers. These results reveal the influence of the complex defect o...

Variable-temperature current–voltage has been used to study the conduction properties of Fe-doped semi-insulating (SI) InP in the as-grown and annealed states. It is found that the trap-filling (TF) process disappears gradually with lengthening of annealing time. This phenomenon is explained by the decrease of the concentration of the empty Fe deep level (Fe3+) that is caused by the thermally induced donor defect formation. The TF process cannot be observed in annealed undoped and long-time annealed Fe-doped SI InP material. The breakdown field of annealed undoped and Fe-doped SI InP is much lower than that of as-grown Fe-doped InP material. The breakdown field decreases with decreasing of temperature indicating an impact ionization process. This breakdown behavior is also in agreement with the fact that the concentration of the empty deep level in annealed InP is lowered.

We report changes in variable-energy positron annihilation spectroscopy measurements on undoped hydrogenated amorphous silicon films after light soaking. The change, seen predominantly in the high momentum band of the annihilation radiation, is not reversed by thermal annealing. We suggest, following recent models of the Staebler-Wronski effect, that light exposure induces hydrogen trapped in vacancylike defects to become mobile in the Si network. The observations place constraints on models of hydrogen motion fitting macroscopic Staebler-Wronski effect kinetics and may help to achieve a definitive description of metastability in a-Si:H.

Fe-doped liquid encapsulated Czochralski InP has been annealed between 500 and 900 °C for different durations. The electrical property of annealed InP has been studied by temperature-dependent Hall measurement. Defects in annealed Fe-doped InP have been detected by room-temperature transient photocurrent spectroscopy. Upon annealing, the change of electrical property in this material is indicative of the formation of a high concentration of defects. The formation process of these thermally induced defects is discussed.

The fluence-dependent properties and the annealing behavior of electron-irradiation-induced deep levels in n-type 6H–SiC have been studied using deep-level transient spectroscopy (DLTS). Sample annealing reveals that the dominant DLTS signal at EC−0.36 eV (labeled as E1 by others) consists of two overlapping deep levels (labeled as ED3L and ED3H). The breakup temperature of the defect ED3L is about 700 °C. The ED3H center together with another deep level located at EC−0.44 eV (so-called E2) can withstand high-temperature annealing up to 1600 °C. It is argued that the involvement of the defect ED3L is the reason that various concentration ratios of E1/E2 were observed in the previous work. The revised value of the capture cross section of the deep-level ED3H has been measured after removing ED3L by annealing. A deep level found at EC−0.50 eV is identified as a vacancy–impurity complex since it was found to have a lower saturated concentration and weak thermal stability. Two other deep levels, EC−0.27 eV an...

Positron-lifetime spectroscopy has been used to investigate the electric-field distribution occurring at the Au\char21{}semi-insulating GaAs interface. Positrons implanted from a ${}^{22}\mathrm{Na}$ source and drifted back to the interface are detected through their characteristic lifetime at interface traps. The relative intensity of this fraction of interface-trapped positrons reveals that the field strength in the depletion region saturates at applied biases above 50 V, an observation that cannot be reconciled with a simple depletion approximation model. The data, are, however, shown to be fully consistent with recent direct electric-field measurements and the theoretical model proposed by McGregor et al. [J. Appl. Phys. 75, 7910 (1994)] of an enhanced ${\mathrm{EL}2}^{+}$ electron-capture cross section above a critical electric field that causes a dramatic reduction of the depletion region's net charge density. Two theoretically derived electric field profiles, together with an experimentally based profile, are used to estimate a positron mobility of $\ensuremath{\sim}95\ifmmode\pm\else\textpm\fi{}35{\mathrm{cm}}^{2}{\mathrm{}\mathrm{V}}^{\mathrm{\ensuremath{-}}1}{\mathrm{}\mathrm{s}}^{\mathrm{\ensuremath{-}}1}$ under the saturation field. This value is higher than previous experiments would suggest, and reasons for this effect are discussed.

Positron-lifetime and infrared-absorption spectroscopies have been used to investigate the compensation defects that render undoped n-type liquid encapsulated Czochralski-grown InP semi-insulating under high-temperature annealing. The positron measurements, carried out over the temperature range of 25--300 K, reveal in the as-grown material a positron lifetime of $282\ifmmode\pm\else\textpm\fi{}5 \mathrm{ps}$ which we associate with either the isolated indium vacancy ${V}_{\mathrm{In}}^{3\ensuremath{-}}$ or related hydrogen complexes. The shallow donor complex ${V}_{\mathrm{In}}{\mathrm{H}}_{4},$ responsible for much of the n-type conductivity and the strong infrared absorption signal at 4320 nm, is ruled out as a significant trapping site on the grounds that its neutral state is present at too low a concentration. After annealing at $950\ifmmode^\circ\else\textdegree\fi{}\mathrm{C},$ in conjunction with the disappearance of the ${V}_{\mathrm{In}}{\mathrm{H}}_{4}$ infrared-absorption signal, trapping into ${V}_{\mathrm{In}}$-related centers is observed to increase slightly, and an additional positron trapping defect having a lifetime of 330 ps appears at a concentration of $\ensuremath{\sim}{10}^{16} {\mathrm{cm}}^{\ensuremath{-}3},$ indicating divacancy trapping. These results support the recent suggestion that the ${V}_{\mathrm{In}}{\mathrm{H}}_{4}$ complex present in as-grown InP dissociates during annealing, forming ${V}_{\mathrm{In}}{\mathrm{H}}_{n}^{(3\ensuremath{-}n)\ensuremath{-}}(0l~nl~3)$ complexes and that the recombination of ${V}_{\mathrm{In}}$ with a phosphorus atom results in the formation of EL2-like deep donor ${\mathrm{P}}_{\mathrm{In}}$ antisite defect, which compensates the material. It is suggested that the divacancy formed on annealing is ${V}_{\mathrm{In}}{V}_{\mathrm{P}},$ and that this defect is probably a by-product of the ${\mathrm{P}}_{\mathrm{In}}$ antisite formation.

In this paper, positron annihilation measurements have been carried out on a-Si: H thin films deposited by plasma-enhanced chemical vapor deposition (PECVD) at high and low rates by means of the variable energy positron beam Doppler-broadening technique. The depth profiles of microvoids in the films grown under different conditions have been determined. We found a smaller void fraction in the surface region of all films compared to the bulk, and a smaller void fraction in low rate than in high growth rate films. By plotting S and W parameters in the (S, W) plane, we have shown that the vacancies in all of the high-rate and low-rate deposited intrinsic samples, and in differently doped low-rate samples are of the same nature, although there appears to be a higher density of defects in the boron than phosphorus doped films. The depth profiles of the microvoid-like defects in the a-Si: H films are extracted by use of the vepfit program.

Variable energy positron beam and positron annihilation lifetime experiments have been carried out to study the DX center in Cd0.8Zn0.2Te:Cl at 50 K. A short positron effective diffusion length of 275±25 A and a large intensity of 79.0%±0.3% for the long lifetime component indicate a strong trapping effect at DX centers. A trapping rate of κ=1.53±0.05×109 s−1 and a positron lifetime of 335±2 ps at the DX center were obtained. The concentration of DX centers is found to be 5.9±0.7×1016 cm−3, which is in good agreement with the results obtained using Hall effect and thermo-electric effect measurements.

Interface trap generation in nMOS transistors during both stressing and post-stress periods under the conditions of oxide field (dynamic and dc) stress with FN injection is investigated with charge pumping technique. In contrast to the post-stress interface trap generation induced by hot carrier stress which is a logarithmical function of post-stress time, the post-stress interface trap generation induced by oxide-field stress with FN injection first increases with post-stress time but then becomes saturated. The mechanisms for the interface trap generation in both stressing and post-stress periods are described.

The interfacial electric field established under different reverse bias conditions in Au and Ni on semi-insulating GaAs junctions has been studied by means of a low energy positron beam. The technique used is that of monitoring the positron drift to the interface through changes in the annihilation radiation lineshape as a function of incident positron beam energy at different reverse biases. The data show a small but clear electric field drift of positrons towards the interface that increases more rapidly at low voltages (less than 50 V) which at higher biases tends towards saturation. This confirmation of electric field saturation adds further weight to the picture of an electric field enhanced electron capture cross section for the ionized EL2 defect. Electric field values extracted from the data are compared with results from other techniques and suggest that enhanced electron capture is already occurring at the relatively low built-in fields (∼1 kV cm−1) found at the unbiased junction, with a rapid i...

The positron annihilation characteristics have been calculated taking the electron-positron correlation in the generalized gradient approximation (GGA). The calculated electron-positron momentum distributions in Si along the [110] direction in the GGA scheme agree very well with the experiment. The comparison of anisotropies of the momentum distributions along different crystal directions with the theory shows that only the GGA scheme gives the exact values. The enhancement factor for the valence electrons in the electron-positron momentum density is found to be weakly dependent on the momentum. The positron lifetimes in group IV, III-V, and II-VI semiconductors agree very well with the previous calculations and the experiment. {copyright} {ital 1997} {ital The American Physical Society}

Step quenching from the solid solution temperature is a common treatment to enhance the shape memory properties in many copper-based shape memory alloys (SMAs). However, the martensite stabilisation problem is more serious in step quenched Cu Al Ni Mn Ti (CANTiM) SMAs. After step quenching, martensite in CANTiM cannot reverse transform to its parent phase. In contrast, fast martensite stabilisation is still present. However, martensite stabilisation can be removed by post-quench ageing at 100°C. In rapid quenching from high temperature, a large number of defects such as vacancies are usually produced. The martensite stabilisation due to fast quenching is attributed to the pinning effect of these excessive vacancies. After post-quench ageing, the concentration of vacancies is lowered and thus the pinning effect is reduced. This paper reports the behaviour of vacancies at different cooling rates in different quenching medium, and the effect of post-quench ageing. The evolution of vacancies in the alloy is studied using positron annihilation techniques. In contrast to the martensite stabilisation due to ageing at higher temperatures, the martensite stabilisation due to fast quenching rate can be removed by a carefully chosen low temperature ageing programme.

We have investigated the structural and defect characteristics of GaAs and AlxGa1−xAs grown at low substrate temperature (250 °C) by molecular beam epitaxy. Using x-ray diffraction we have observed an increase in lattice parameter for all as-grown layers, with the AlxGa1−xAs layers showing a smaller expansion than the GaAs layer. However, infrared absorbtion measurements revealed that the concentration of neutral arsenic antisite defect, [AsGa]0, was not significantly affected by aluminum content (x), with only a small reduction for x=0.36. Positron beam studies showed that the low temperature layers had a higher concentration of vacancy-related defects (∼1017 cm−3) than the semi-insulating substrate, with the AlxGa1−xAs layers having the highest values. After annealing (600 °C, 15 min) the lattice constants relaxed to those of conventionally grown material and [AsGa]0 was reduced in all cases, with the smallest reduction occurring for the x=0.36 layer, indicating that the Al atoms strengthen the lattice ...

The electron subband energies and wave functions in an interdiffusion‐induced AlxGa1−xAs/GaAs/AlxGa1−xAs single quantum well are calculated in the presence of the dc electric field using the finite difference method. The mean lifetimes are obtained from the time‐dependent probability of tunneling of the wave packet out of the well by the applied electric field. The effect of the applied electric field on the subband energies in the well is the same as in the as‐grown square quantum well when the interdiffusion length is below 20 A . In the well with higher diffusion length the barrier height reduces so that the wave function tunnels out of the well. The linear and nonlinear intersubband absorption coefficients and the change in the real part of the index of refraction are calculated with the applied electric field at 100 kV/cm and without the field in both the as‐grown square well and the diffusion modified well with the interdiffusion length at 20 A.

An energy-tunable monoenergetic positron beam was used to study positron diffusion in the space-charge region of an Au/GaAs(SI) (semi-insulating) Schottky contact, where the electric field reaches \ensuremath{\sim}${10}^{5}$ V ${\mathrm{cm}}^{\mathrm{\ensuremath{-}}1}$ by reverse biasing the diode. An analytical solution of the time-dependent positron drift-diffusion model under an electric field was obtained for the case of a semi-infinite body with a capturing boundary, and explains the experimental results well. A positron diffusion coefficient of 1.8\ifmmode\pm\else\textpm\fi{}0.2 ${\mathrm{cm}}^{2}$ ${\mathrm{s}}^{\mathrm{\ensuremath{-}}1}$, and a positron mobility of 70\ifmmode\pm\else\textpm\fi{}10 ${\mathrm{cm}}^{2}$ ${\mathrm{V}}^{\mathrm{\ensuremath{-}}1}$ ${\mathrm{s}}^{\mathrm{\ensuremath{-}}1}$ in GaAs(SI) at 300 K, were obtained independently. This result is consistent with the Einstein relation. The dependence of the positron current density at the Au/GaAs interface on the electric field shows that GaAs(SI) is a possible candidate for the fabrication of the field-assisted positron moderator. \textcopyright{} 1996 The American Physical Society.

Acceptors in liquid encapsulated Czochralski-grown undoped gallium antimonide (GaSb) were studied by temperature dependent Hall measurement and positron lifetime spectroscopy (PLS). Because of its high concentration and low ionization energy, a level at EV+34meV is found to be the important acceptor responsible for the p-type conduction of the samples. Two different kinds of VGa-related defects (lifetimes of 280ps and 315ps, respectively) having different microstructures were characterized by PLS. By comparing their annealing behaviors and charge state occupancies, the EV+34meV level could not be related to the two VGa-related defects.

A deep level transient spectroscopy (DLTS) study has been performed on a GaN n+-p junction fabricated by implanting Si into a Mg-doped p-type GaN epilayer. A high concentration of a deep level defect has been revealed within the interfacial region of the junctions by the unusual appearance of a minority peak in the majority carrier DLTS spectra. The deep level defect appears to be an electron trap at EC-0.59 eV in the p-side region of the junction and has tentatively been attributed to the VN–Mg complex. The high concentration of this electrically active deep level defect in the depletion layer of the Si-implanted GaN n+-p junction diodes suggests the need for further investigations.

Beryllium implantation into p-type 6H–SiC and subsequent thermal annealing have been performed. The deep level defects induced by this beryllium-implantation process have been investigated using deep level transient spectroscopy. Four deep levels labeled BEP1, BEP2, BEP3, and BEP4 were detected in the temperature range 100–500 K. The prominent hole trap BEP1 with an energy level at EV+0.41 eV agrees well with the acceptor level of beryllium determined by Hall effect measurements. The remaining hole traps with energy levels at EV+0.60 eV, EV+0.76 eV, and EV+0.88 eV, respectively, are proposed to be beryllium-implantation induced defects or complexes.

Spectroscopic ellipsometric measurements for Pt/n‐Si samples with different thickness of Pt films have been performed. The thickness of the Pt films determined with the three‐phase model (air/Pt/Si) changes with the wavelength λ while that with the four‐phase model (air/Pt/interface layer/Si) remains unchanged, showing the existence of an interface layer. At the same time, the apparent optical dielectric constants of the interface layer as a function of λ are also obtained. A calculation based on the effective medium theory is carried out to simulate the optical dielectric data of the interface layer. Some structural information of the interface layer is obtained from the calculation.

A ±100 V square wave applied to a Au/semi-insulating SI–GaAs interface was used to bring about electron emission from and capture into deep level defects in the region adjacent to the interface. The electric field transient resulting from deep level emission was studied by monitoring the positron drift velocity in the region. A deep level transient spectrum was obtained by computing the trap emission rate as a function of temperature and two peaks corresponding to EL2 (Ea=0.81±0.15 eV) and EL6 (Ea=0.30±0.12 eV) have been identified.

Unlike its conventional applications in lattice defect characterization, positron annihilation lifetime technique was applied to study temperature-dependent deep level transients in semiconductors. Defect levels in the band gap can be determined as they are determined by conventional deep level transient spectroscopy (DLTS) studies. The promising advantage of this application of positron annihilation over the conventional DLTS is that it could further extract extra microstructure information of deep-level defects, such as whether a deep level defect is vacancy related or not. A demonstration of EL2 defect level transient study in GaAs was shown and the EL2 level of 0.82±0.02 eV was obtained by a standard Arrhenius analysis, similar to that in conventional DLTS studies.

Arsenic doped ZnO and ZnMgO films were deposited on SiO2 using radio frequency magnetron sputtering and ZnO–Zn3As2 and ZnO–Zn3As2–MgO targets, respectively. It was found that thermal activation is required to activate the formation of p-type conductivity. Hall measurements showed that p-type films with a hole concentration of ∼ 1017 cm−3 and mobility of ∼ 8 cm2 V−1 s−1 were obtained at substrate temperatures of 400–500 °C. The shallow acceptor formation mechanism was investigated using x-ray photoelectron spectroscopy, positron annihilation, low temperature photoluminescence, and nuclear reaction analysis. The authors suggest that the thermal annealing activates the formation of the AsZn-2VZn shallow acceptor complex and removes the compensating hydrogen center.

n-type semiconducting InP is changed into p-type semiconducting by short time annealing at 700 °C. Further annealing for a longer time leads to a second conduction-type conversion changing the material back to n type again but with a much higher resistivity. These conduction conversions indicate the formation of both acceptor and donor defects and the progressive variation of their relative concentrations during annealing.

1.7 MeV electron irradiation-induced deep levels in p-type 6H–SiC have been studied using deep level transient spectroscopy. Two deep hole traps are observed, which are located at EV+0.55 eV and EV+0.78 eV. They have been identified as two different defects because they have different thermal behaviors. These defects at EV+0.55 eV and EV+0.78 eV are annealed out at 500–200 °C, respectively, and are different from the main defects E1/E2, Z1/Z2 observed in electron irradiated n-type 6H–SiC. This indicates that new defects have been formed in p-type 6H–SiC during electron irradiation.

The autocorrelation functions of positron—electron pairs in InP with indium vacancies in different charge states are calculated in this work. It is found that the autocorrelation function can be used to identify the charge states of vacancies in solids. In the case of perfect lattice the autocorrelation function oscillates on the lattice with the periodicity of the lattice and decays gradually up to about the third layer of the lattice points. In the case of neutral vacancies, the central peak expands significantly, but the nearest peak and dip are greatly reduced. In the case of negative charge states of vacancies, the central peak, however, contracts slightly compared to the case of the perfect lattice, and the nearest peak and dip and the next-nearest peak nearly disappear.

Two Ga-acceptor levels, located at EV+0.31 eV and EV+0.37 eV, respectively, have been observed in the gallium implantation manufactured p+n diodes using deep level transient spectroscopy. The behavior of the implanted gallium is very similar to that of implanted aluminum, except that the positions of the introduced levels are different. This result strongly supports the recent model, which was used to explain the discrepant results between boron and aluminum implantation induced deep levels. Besides the two acceptor levels, a thermally stable electron trap is also observed and has been tentatively attributed to a Ga-related complex.

The concentration of hydroen-indium vacancy complex VInH4 in liquid encapsulated Czochralski undoped and Fe-doped n-type InP has been studied by low-temperature infrared absorption spectroscopy. The VInH4 complex is found to be a dominant intrinsic shallow donor defect with concentrations up to similar to 10(16) cm(-3) in as-grown liquid encapsulated Czochralski InP. The concentration of the VInH4 complex is found to increase with the compensation ratio in good agreement with the proposed defect formation model of Walukiewicz [W. Walukiewicz, Phys. Rev. B 37, 4760 (1998); Appl. Phys. Lett. 54, 2094 (1989)], which predicts a Fermi-level-dependent concentration of amphoteric defects. (C) 1998 American Institute of Physics, [S0003-6951(98)04435-0].

Deep traps in the boron extended tail region of ion implanted 6H–SiC pn junctions formed during annealing have been studied using deep level transient spectroscopy. Dramatically high concentrations of ∼1016 cm−3 of the D center have been observed through the unusual appearance of minority peaks in the majority carrier spectra. No evidence is found for any shallow boron acceptor in this region, but an induced hole trap Ih at EV+0.46 eV is found under cold implantation conditions. These results support the picture of the extended tail, rich in boron-vacancy complexes such as the D center, which forms as a result of vacancy enhanced indiffusion. The dominance of the electrically active D center in the depletion layer of the technologically important SiC pn junction diode suggests the need for further research in this area.

Fourier transform infrared spectroscopy measurements have been carried out on liquid-encapsulated Czochralski-grown undoped InP wafers, which reproducibly become semi-insulating upon annealing in an ambient of phosphorus at 800–900 °C. The measurements reveal a high concentration of hydrogen complexes in the form VInH4 existing in the material before annealing in agreement with recent experimental studies. It is argued that the dominant and essential process producing the semi-insulating behavior is the compensation produced by an EL2-like deep donor phosphorus antisite defect, which is formed by the dissociation of the hydrogen complexes during the process of annealing. The deep donor compensates acceptors, the majority of which are shallow residual acceptor impurities and deep hydrogen associated VIn and isolated VIn levels, produced at the first stage of the dissociation of the VInH4 complex. The high concentration of indium vacancies produced by the dissociation are the precursor of the EL2-like phosp...

ZnO nanorod arrays were fabricated using a hydrothermal method. The nanorods were studied by scanning electron microscopy, photoluminescence (PL), time-resolved PL, X-ray photoelectron spectroscopy, and positron annihilation spectroscopy before and after annealing in different environments and at different temperatures. Annealing atmosphere and temperature had significant effects on the PL spectrum, while in all cases the positron diffusion length and PL decay times were increased. We found that, while the defect emission can be significantly reduced by annealing at 200 degrees C, the rods still have large defect concentrations as confirmed by their low positron diffusion length and short PL decay time constants.

Neutron irradiation induced defects and their effects on the carrier concentration of GaN epilayers are investigated with Raman scattering and X-ray diffraction techniques. Relative to the as-grown sample, the neutron-irradiated samples exhibit a clear variation in the position and lineshape of the A1(LO)-mode Raman peak as well as in the full-width at half-maximum height (FWHM) of the XRD rocking curves. Careful curve fitting and adequate calculations give the carrier concentrations of the irradiated GaN. It is found that the defects induced by neutron irradiation act as carrier trap centres which capture the electron carriers so that the carrier concentration of the irradiated GaN is reduced

6H‐SiC samples subjected to He‐implantation and e−‐irradiation (Ee=0.2MeV–1.7MeV) were investigated by deep level transient spectroscopy (DLTS). E1/E2 were identified in the He‐implanted and the e−‐irradiated samples with Ee⩾0.3MeV. Considering the minimum e− energy required to displace the atoms in the lattice, the E1/E2 creation was related to the C‐atom displacement. Similar to previous reports, the peak intensity and the capture cross sections of E1/E2 anomalously varies from samples to samples. It was shown that these anomalies were due to the presence of a DLTS peak overlapping with the E1/E2 signals.

A simple and inexpensive circuit to facilitate the direct measurement of capture cross section, when synchronized with a deep level transient spectroscopy system, is described. It avoids the most commonly encountered problem of loading and distortion of the bias (trap filling) pulses of nanosecond duration in the capture cross‐section measurement. The capacitance meter, whose internal circuitry is responsible for the distortion, is connected and disconnected from the rest of the apparatus with the help of simple and low‐cost reed relay switches featuring high operating speed and low contact resistance. Sharp bias pulses as small as 30 ns can successfully be applied to the sample with no observable distortion. Finally, a representative measurement is shown to demonstrate the simplicity and high performance of the circuit.

In this paper we are going to present the first real-time S-parameter positron imager. This is a useful tool in solid state technology for mapping the lateral defect types and concentrations on a material sample. This technology has been developed for two major categories of researchers, the first being those that have a focused low energy positron beam and second those that do not. Here we describe the design and implementation of a real-time automated scanning system that rasters a sample surface with a 0.5mm diameter positron source (or beam focus) so as to give an S-parameter image of a sample. The source (or beam) rasters across a region of a semiconductor sample in rectilinear motion while gamma ray energies E γ are processed using a standard HP Ge spectroscopy system and a 14 bit nuclear ADC. Two other ADCs are used to obtain the x, y coordinate data corresponding to each event by storing voltage pulses from the x & y stepper motor drives (or saddle coil currents) gated with the event pulses. Using these event data triplets (x, y, Eγ) the S-parameter is computed in real time for each pixel region and is used to refresh a color image display on the screen coordinates. Optimal use is made of processing time and the system resources. This user-friendly system is efficient for producing high resolution S-parameter images of the sample. (patent pending 2003)

Deep level transient spectroscopy (DLTS) was used to study deep level defects in He-implanted n-type 6H-SiC samples. Low dose He-implantation (fluence ∼2×1011 ions/cm2) has been employed to keep the as-implanted sample conductive so that studying the introduction and the thermal evolution of the defects becomes feasible. A strong broad DLTS peak at 275K-375K (called signal B) and another deep level at EC-0.50eV were observed in the as-implanted sample. The intensity of the peak B was observed to linearly proportional to the logarithm of the filling pulse width, which is a signature for electron capture into a defect related to dislocation. After annealing at 500°C, the intensity of peak was significantly reduced and the remained signal has properties identical to the well known Z1/Z2 deep defects, although it is uncertain whether the Z1/Z2 exist in the as-implanted sample or it is the annealing product of the dislocation-related defect. The E1/E2 defect (EC-0.3/0.4eV) was not presence in the as-implanted sample, but was observed after the 300°C annealing.

For nickel on the chemically clean surface of undoped semi‐insulating GaAs at room temperature, an upward surface band bending of 0.062 eV and a barrier height of 0.690 eV have been observed by the photovoltage and the internal photoemission techniques, respectively. The observed surface band bending is in excellent agreement with its predicted value, and the observed barrier height also agrees very well with its value from the very careful analysis of reversed I‐V data. It has been determined that the interfacial Fermi level lies at 0.690 eV below the GaAs conduction band minimum at the interface. The interfacial Fermi level is found to coincide with the energy level of the EL2 native defect, indicating the importance of the EL2 in the Fermi level pinning at the interface.

Positron lifetime spectroscopy was employed to study the as-electron-irradiated (10 MeV, 1×1018 cm-2) n-type 6H silicon carbide sample in the measuring temperature range of 15 K to 294 K. Isochronal annealing studies were also performed up to the temperature of 1373 K by carrying out the room temperature positron lifetime measurement. Negatively charged carbon vacancies and VCVSi divacancy were identified as the major vacancy type defects induced by the electron irradiation process. The concentration of the VCVSi divacancy was found to decrease dramatically after the 1973 K annealing.

P-rich, In-rich and stoichiometric undoped InP melts have been synthesised by phosphorus in-situ injection method. InP crystal ingots have been grown from these melts by the liquid encapsulated Czochralski method (LEC). Samples from these ingots grown from various stoichiometric melts have been characterized by Hall effect and Fourier Transform Infrared (FT-IR) spectroscopy measurements respectively. The Hall effect measurement results indicate that the net carrier concentration of P-rich undoped InP is higher than that of In-rich and stoichiometric undoped InP materials. FT-IR spectroscopy measurements reveal that there are intensive absorption peaks which have been due to a hydrogen related indium vacancy complex V/sub In/H/sub 4/. By comparing FT-IR spectra, it is found that P-rich InP material has the most intensive absorption peak of V/sub ln/H/sub 4/, while In-rich InP material has the weakest absorption peak.

Positron lifetime, Photoluminescence and Hall measurements were performed to study undoped p-type gallium antimonide materials. A 314ps lifetime component, attributed to $V_{Ga}$ related defect, was identified in the positron lifetime measurement. In the PL measurement, a $778meV$ and a $797meV$ peaks were observed. Isochronal annealing studies were performed and at the temperature of $300^{o}C$, both the 314ps positron lifetime component and the two PL signals disappeared, which gives a clear and strong evidence for their correlation. However, the hole concentration ($\sim 2\times 10^{17}cm^{-3}$) was observed to be constant throughout the whole annealing temperature range up to $500^{o}C$. Contradictory to general belief, this implies, at least for samples with annealing temperatures above $300^{o}C$, the Ga vacancy is not the acceptor responsible for the p-type conduction., 3 pages, 3 figures