Laplace transform deep level transient spectroscopy: new insight into defect microscopy

Abstract

It is demonstrated that a resolution of deep level transient spectroscopy (D LTS) can be improved using the Laplace transform method for the emission rate analysis. Considerable confidence in this approach was gained through numerous tests carried out on two numerical algorithms used for the calculations as well as through measurements of a selection of well characterised point defects in various semiconductors. For each of these defects conventional DLTS gives broad featureless lines, whereas Laplace DLTS reveals afine structure in the emission process producing the spectra.

MST/3321     

Characterization of deep level defects in Si irradiated with MeV Ar+ ions using constant capacitance time analyzed transient spectroscopy

An isothermal spectroscopic technique called time analyzed transient spectroscopy (TATS) in the constant capacitance (CC) mode has been used to characterize electrically active defects in the MeV Ar⁺ implanted silicon. The problems associated with high defect density and the presence of damaged region in the as-implanted material are overcome by CC-TATS method. The CC-TATS spectra of the as-implanted sample shows two positive peaks and an attendant negative peak. Two distinct traps have also been identified using thermally stimulated capacitance method modified to operate in constant capacitance mode. Variable pulse width measurements using CC-TATS show exponential capture kinetics in contrast to extremely slow capture observed in conventional deep level transient spectroscopy (DLTS) experiment. The results indicate that trapping behaviour is due to point-like defects associated with extended defects such as dislocation and stacking fault.     

Deep levels in GaN studied by deep level transient spectroscopy and Laplace transform deep-level spectroscopy

In this study we present the results of investigations on Schottky Au-GaN diodes by means of conventional DLTS and Laplace DLTS methods within the temperature range of 77–350 K. Si-doped GaN layers were grown by Molecular Beam Epitaxy technique (MBE) on sapphire substrates. DLTS signal spectra revealed the presence of four majority traps: two hightemperature and two low-temperature peaks. Using LDLTS method and Arrhenius plots the activation energy and capture cross sections were obtained. For two high-temperature majority traps they are equal to E1 = 0.65 eV, σ₁ = 8.2 × 10^?16cm² and E2 = 0.58 eV, σ₂ = 2.6 × 10^?15 cm² whereas for the two low-temperature majority traps E3 = 0.18 eV, σ₃ = 9.72 × 10^?18 cm² and E4 = 0.13 eV, σ₄ = 9.17 × 10^?18 cm². It was also found that the traps are related to point defects. Possible origin of the traps was discussed and the results were compared with the data found elsewhere [1–5].     

Majority carrier capture rates for transition metal impurities in germanium

The carrier capture cross-sections of deep levels in semiconductors are important parameters determining the properties of devices. Therefore not only the electrical levels of the transition metal impurities in germanium have been identified with DLTS but also the capture cross-sections have been directly determined by means of the DLTS amplitude as a function of the DLTS filling pulse duration. Due to the accurate fitting method the temperature dependence of these cross-sections could be measured. For the electron traps the empirical rule for phonon-assisted capture was observed, while for the hole traps a cross-section inversely proportional to the temperature was measured. Comparing the observed values it might be predicted that Ni and Co are the most efficient lifetime killers in p-type germanium, while Cr and Hf are the least efficient.     

Defect introduction in Ge during inductively coupled plasma etching and Schottky barrier diode fabrication processes

We have etched Sb-doped n-type (111) oriented Ge by inductively coupled plasma (ICP), using argon, and subsequently studied the defects that this process introduced as well as the effect of this etching on Schottky barrier diode quality. Deep level transient spectroscopy (DLTS) revealed that ICP etching introduced only one prominent defect, EP_0.31, in Ge with a level at 0.31 eV below the conduction band. The properties of this defect are different to those of defects introduced by other particle-related processing steps, e.g. sputter deposition and electron beam deposition, that each introduces a different set of defects. DLTS depth profiling revealed the EP_0.31 concentration was a maximum (3.6 × 10¹³ cm^− 3) close to the Ge surface and then it decreased more or less exponentially into the Ge. Annealing at 250 °C reduced the EP_0.31 concentration to below the DLTS detection limit. Finally, current-voltage (I-V) measurements as a function of temperature revealed that the quality of Schottky contacts fabricated on the ICP-etched surfaces was excellent at − 100 K the reverse leakage current at − 1 V was below 10 ¹³ A (the detection limit of our I-V instrumentation).     

Deep level transient spectroscopy study for the development of ion-implanted silicon field-effect transistors for spin-dependent transport

A deep level transient spectroscopy (DLTS) study of defects created by low-fluence, low-energy ion implantation for development of ion-implanted silicon field-effect transistors for spin-dependent transport experiments is presented. Standard annealing strategies are considered to activate the implanted dopants and repair the implantation damage in test metal-oxide-semiconductor (MOS) capacitors. Fixed oxide charge, interface trapped charge and the role of minority carriers in DLTS are investigated. A furnace anneal at 950 °C was found to activate the dopants but did not repair the implantation damage as efficiently as a 1000 °C rapid thermal anneal. No evidence of bulk traps was observed after either of these anneals. The ion-implanted spin-dependent transport device is shown to have expected characteristics using the processing strategy determined in this study.     

Quantitative study of isoelectronic copper pairs by photoluminescence and deep level transient spectroscopy

Abstract

The correlation between the Cu induced deep level at E_v + 0.1 eV and the appearance of luminescent Cu pairs, leading to a characteristic line spectrum with the most intense Cu no phonon line at 1.014 ev, has been examined using deep level transient spectroscopy and photoluminescence. Samples of floating zone (FZ) grown p-Si with concentrations of the 0.1 eV Cu centre varying from 10¹¹ to about 10¹⁴ cm^-3 were obtained by a Cu contamination treatment without quenching. The concentration of the luminescent Cu pairs was estimated by analysing the dependence of the Cu line intensity on the excitation power in the transition region to saturation. The saturation intensities, which are proportional to the concentration of luminescent Cu centres, show a linear dependence on the 0.1 eV deep centres with good correlation, suggesting that the same Cu induced centre is detected by deep level transient spectroscopy and photoluminescence.     

Amorphization, recrystallization and end of range defects in germanium

The controlled doping of germanium by ion implantation is a process which requires basic research before optimization. For this reason, we have experimentally studied by transmission electron microscopy both the kinetics of amorphization and of recrystallization of Ge during ion implantation (Ge, P and B) and further annealing. As in Si, the crystalline to amorphous phase transition occurs through the linear accumulation of damage with the dose until a certain threshold is reached above which the material turns amorphous. We show that the Critical Damage Energy Density (CDED) model can be used in germanium to predict the existence, position and extension of amorphous layers resulting from the implantation of ions for almost all mass/energy/dose combinations reported here and in the literature. During annealing, these amorphous layers recrystallize by solid-phase epitaxy following an Arrhenius-type law which we have determined. We observe that this regrowth results in the formation of extended defects of interstitial type. During annealing these defects evolve in size and density following an Ostwald ripening mechanism which becomes non-conservative (defects “evaporate”) as the temperature is increased to 600 °C. These results have important implications for the modeling of diffusion of implanted dopant in Ge. Transient diffusion may also exist in Ge, driven by an interstitial component usually not evidenced under equilibrium conditions.     

Extended deep level defects in Ge-condensed SiGe-on-Insulator structures fabricated using proton and helium implantations

SiGe-on-Insulator (SGOI) structures were created using the Ge condensation method, where an oxidation process is performed on the SiGe/Si structure. This method involves rapid thermal chemical vapor deposition and H⁺/He⁺ ion-implantations. Deep level defects in these structures were investigated using deep level transient spectroscopy (DLTS) by varying the pulse injection time. According to the DLTS measurements, a deep level defect induced during the Ge condensation process was found at 0.28 eV above the valence band with a capture cross section of 2.67 × 10^− 17 cm², two extended deep levels were also found at 0.54 eV and 0.42 eV above the valence band with capture cross sections of 3.17 × 10^− 14 cm² and 0.96 × 10^− 15 cm², respectively. In the SGOI samples with ion-implantation, the densities of the newly generated defects as well as the existing defects were decreased effectively. Furthermore, the Coulomb barrier heights of the extended deep level defects were drastically reduced. Thus, we suggest that the Ge condensation method with H⁺ ion implantation could reduce deep level defects generated from the condensation and control the electrical properties of the condensed SiGe layers.     

Characterization of deep levels in high electron mobility transistor by conductance deep level transient spectroscopy

The influence of a substrate voltage on the dc characteristics of an AlGaN/GaN high electron mobility transistor (HEMT) on silicon (111) substrate is profited to investigate traps that are located between the substrate and the two-dimensional electron gas (2DEG) channel. The transient of the drain current after applying a negative substrate voltage is evaluated in the temperature range from 77 to 600 K. With this method, known as Conductance Deep Level Transient Spectroscopy (CDLTS), majority deep levels with activation energy of 61 meV as well as minority carrier traps at 74 meV and capture cross-section respectively 2.56 × 10^− 15 cm², 2.1 × 10^− 15 cm² are identified. Finally, the correlation between the anomalies observed on the output characteristics and defects is discussed.     

Charge-based deep level transient spectroscopy of B-doped and undoped polycrystalline diamond films

The undoped and B-doped polycrystalline diamond thin film was synthesized by hot filament chemical vapor deposition and microwave plasma, respectively. The structural characterization was performed by scanning electron microscopy, X-ray diffraction and Raman spectroscopy. The electrical properties of synthesized diamond layer were characterized by dc-conductivity method and charge deep level transient spectroscopy. The B-doped diamond layers show higher sp²/sp³ ratios in comparison with that of undoped layers what can have an essential influence on the localized density of states associated with shallow hydrogen acceptor states what is reflected in the values of activation energies which reached the values of 38 meV for B-doped and 55 meV for undoped diamond layers, respectively. The existence of deep level traps, as, for example, associated with B-related acceptors, was not observed.     

Deep level transient spectroscopy on charge traps in high-k ZrO2

The deep trap properties of high-dielectric-constant (k) ZrO₂ thin films were examined by deep level transient spectroscopy (DLTS). The hole traps of a ZrO₂ dielectric deposited by sputtering were investigated in a MOS structure over the temperature range, 375 K-525 K. The potential depth, cross section and concentration of hole traps were estimated to be ∼ 2.5 eV, ∼ 1.8 × 10^− 16 cm² and ∼ 1.0 × 10¹⁶ cm^− 3, respectively. DLTS of ZrO₂ dielectrics can be used to examine the threshold voltage shift (?V_th) during the operation of SONOS-type flash memory devices, which employ high-k materials.     

A compact microcomputer-based deep level transient spectroscopymeasurement system

A specifically designed system for deep-level transient spectroscopy (DLTS) measurement is described. It is compact and fully automated, can measure all the spectra in just one temperature scan, and permits simultaneous analysis of two devices. The user sets up the measurement conditions with a reduced number of commands. This makes the system very flexible and versatile. A self-scaled gain output amplifier has been incorporated to provide the maximum DLTS signal amplitude compatible with the input range of the analog-to-digital (A/D) converter. This self-scaling allows the detection of traps with a large amplitude range. Up to 90 sampling times can be selected for each transient. To improve the signal-to-noise ratio (SNR), as many as 32768 scans may be accumulated for averaging. Improvements as high as 45 dB have been used to characterize deep centers in AlGaAs/GaAs devices grown by liquid-phase and molecular-beam epitaxies     

Deep level transient spectroscopy: instrumentation inducedanomalous characteristics

The authors attempted to clarify, compare, and contrast some previously unpublished difficulties encountered in the operation of deep-level transient spectroscopy (DLTS) systems, and describe corrective measures where applicable, including instrumentation and deep level anomalies which may affect capture cross section designations, such as pulse squaring and degeneracy factor effects. Thermal lag between sample and sensor has been demonstrated to cause the apparent peak temperatures of deep levels to be in error by several degrees. In addition, analysis of deep level data from a single ΔC-vs.-T plot has been achieved from half-peak-height points of the graph, for use where no other data are available from the sample under test. Trap depth determination from a single thermal scan of fragile Schottky barriers was considered     

Investigation of polarization fatigue induced deep level transient spectroscopy in lead zirconate titanate thin film

Deep level transient spectroscopy (DLTS) in pulsed laser deposited lead zirconate titanate (Pb(Zr_0.52Ti_0.48)O₃ thin film grown on LSCO (La_0.5Sr_0.5)O₃ bottom electrode and LaAlO₃ substrate has been investigated. The transient capacitance and the deep level energy inside the energy band gap are correlated with the number of bipolar switching cycles. It was found that both of them increase with the number of switching cycles.     

SEM study of defects in PVD hard coatings

Hard coating defects are produced by foreign particle contamination on substrate surface before and during coating, or due to arcing. In this work, CrN, TiAlN and CrN/TiAlN multilayer hard coatings were prepared by thermoionic arc ion plating deposition system BAI 730 (Balzers) and by sputter deposition in CC800 (CemeCon). We investigated the concentration of defects, its size and structure after tool steel substrate surface pretreatment (polishing, ion etching) as well as after deposition by means of atomic force microscopy (AFM) and scanning electron microscopy (SEM).     

Honeycomb voids due to ion implantation in germanium

For future semiconductor devices, germanium layers are very attractive due to their high carrier mobility with ion implantation remaining the dominant method for forming pn junctions. Yet, implantation of heavy ions above a critical dose causes inadmissible surface roughness and formation of voids. To understand the main factors of influence, a comprehensive study on void formation was performed with different ions (BF₂, P, Al, Ga, Ge, As, Sb) implanted at various doses, dose rates, and energies. It was found that the dose is the most important parameter for void formation. The critical dose was determined to be 2 · 10¹⁵ cm^− 2 for As, 2 · 10¹⁵ cm^− 2 for Ga, and 5 · 10¹⁴ cm^− 2 for Sb, respectively. For ions with lower mass (BF₂, P, Al), no or only negligible surface roughening was observed.     

Divacancies at room temperature in germanium

We have studied the annealing of vacancy defects in neutron and proton irradiated germanium. After neutron irradiation the Sb-doped samples were annealed at 473, 673 and 773 K for 30 min. The positron lifetime was measured as a function of temperature (30 - 295 K). A lifetime component of 330 ps with no temperature dependence is observed in as irradiated samples, identified as the positron lifetime in a neutral divacancy. The average positron lifetime in the samples annealed at 473 K has a definite temperature dependence, suggesting that the divacancies become negative as the crystal recovers and the Fermi level moves upward in the band gap. Proton irradiation of germanium at 37 K with subsequent room temperature annealing also resulted in a similar lifetime component 315 ps, in good agreement with the neutron irradiation experiment.     

Hydrogen-related defects in crystalline semiconductors: a theorist''s perspective

Hydrogen is a common impurity in all semiconductors. Although it is sometimes deliberately introduced, hydrogen often penetrates into the crystal during device processing. It interacts with broken or weak covalent bonds, such as those found at extended and localized defect centers. The main results of these covalent interactions are shifts of energy levels out of (or into) the gap and new optical activity (infrared absorption and Raman scattering). The shifts in energy levels lead to the passivation (or activation) of the electrical activity of various centers. Hydrogen can also interact with the perfect crystal and with itself, sometimes leading to the formation of extended structures known as platelets. Finally, H also acts as a catalyst, dramatically enhancing the diffusivity of interstitial oxygen in Si. The consequences of these interactions are substantial changes in the electrical and optical properties of the crystal, and in the lifetime of charge carriers. The thermal stability of the complexes containing hydrogen varies from room temperature up to several hundreds of degrees Celsius, and the diffusion of H is trap-limited up to rather high temperatures. Hydrogen normally exists in more than one configuration and charge state in semiconductors. A range of experimental and theoretical techniques have been used to investigate the rich properties of hydrogen in semiconductors, and several extensive reviews focusing mostly on the experimental side of these issues have been published in the past five years. The present review focuses mostly on the theoretical work performed in this field. However, the most recent experimental results are also discussed, and the current understanding of hydrogen interactions in semiconductors summarized.     

Deep level transient spectroscopy of anisotropic semiconductor GaTe

Deep level transient spectroscopy (DLTS) was carried out on single crystals of the layered chalcogenide p-GaTe using Schottky barriers parallel and perpendicular to the layer planes to study the possible anisotropy of the defect levels. Deep levels with the same energies (0·28 eV and 0·42–0·45eV) have been found in both directions with concentrations ranging from 10¹³cm⁻³ to 10¹⁴ cm⁻³ and capture cross-sections from 10⁻¹⁵cm² to 10⁻¹⁷cm². The difference in the spectra obtained from the two planes and the possible reason for the deep level energies being independent of crystal orientation are discussed.