Crack-free GaN/Si(111) epitaxial layers grown with InAlGaN alloy as compliant interlayer by metalorganic chemical vapor deposition

A new method is demonstrated to be effective in reducing mismatch-induced tensile stress and suppressing the formation of cracks by inserting InAlGaN interlayers during the growth of GaN upon Si (1 1 1) substrate. Compared with GaN film without quaternary interlayer, GaN layer grown on InAlGaN compliant layers shows a five times brighter integrated PL intensity and a (0 0 0 2) High-resolution X-ray diffraction (HRXRD) curve width of 18 arcmin. Its chi(min), derived from Rutherford backscattering spectrometry (RBS), is about 2.0%, which means that the crystalline quality of this layer is very good. Quaternary InAlGaN layers, which are used as buffer layers firstly, can play a compliant role to endure the large mismatch-induced stress and reduce cracks during the growth of GaN epitaxy. The mechanisms leading to crack density reduction are investigated and results show that the phase immiscibility and the weak In-N bond make interlayer to offer tenability in the lattice parameters and release the thermal stress. (c) 2005 Elsevier B.V. All rights reserved.

Crack-free InAlGaN quaternary alloy films grown on Si(111) substrate by metalorganic chemical vapor deposition

Crack-free In0.08Al0.25Ga0.67N quaternary films, with and without thick (> 1.5 mum) high-temperature-GaN (HTGaN) interlayer, have been grown on Si(1 1 1) substrates by a low-pressure metalorganic chemical vapor deposition (MOCVD) system. Mole fractions of In and Al in quaternary alloy layers are determined by Energy dispersive spectroscopy (EDS) and Rutherford backscattering spectrometry (RBS), which are recorded as similar to8% and similar to25-27%, respectively. High-resolution X-ray diffraction (HRXRD) and room temperature photoluminescence (RT-PL) results evidence the film's single crystal structure and the existence of local In- and/or Al-rich regions. Compared with GaN film grwon on Si(1 1 1) substrate, no crack is observed in the quaternary ones. Two explanations are proposed. First, mismatch-induced strain is relaxed significantly due to gradual changes of In concentration. Second, the weak In-N bond is likely to break when the sample is cooled down to the room temperature, which is expected to favor the releasing of thermal stress. (C) 2004 Elsevier B.V. All rights reserved.

Spatial distribution of deep level defects in crack-free AlGaN grown on GaN with a high-temperature AlN interlayer

The deep level luminescence of crack-free Al0.25Ga0.75N layers grown on a GaN template with a high-temperature grown AlN interlayer has been studied using spatially resolved cathodoluminescence (CL) spectroscopy. The CL spectra of Al0.25Ga0.75N grown on a thin AlN interlayer present a deep level aquamarine luminescence (DLAL) band at about 2.6 eV and a deep level violet luminescence (DLVL) band at about 3.17 eV. Cross-section line scan CL measurements on a cleaved sample edge clearly reveal different distributions of DLAL-related and DLVL-related defects in AlGaN along the growth direction. The DLAL band of AlGaN is attributed to evolve from the yellow luminescence band of GaN, and therefore has an analogous origin of a radiative transition between a shallow donor and a deep acceptor. The DLVL band is correlated with defects distributed near the GaN/AlN/AlGaN interfaces. Additionally, the lateral distribution of the intensity of the DLAL band shows a domainlike feature which is accompanied by a lateral phase separation of Al composition. Such a distribution of deep level defects is probably caused by the strain field within the domains. (c) 2006 American Institute of Physics.

Crack control in GaN grown on silicon (111) using In doped low-temperature AlGaN interlayer by metalorganic chemical vapor deposition

The effects of In doped low-temperature (LT) AlGaN interlayer on the properties of GaN/Si(111) by MOCVD have been investigated. Using In doping LT-interlayer can decrease the stress sufficiently for avoiding crack formation in a thick (2.0 mu m) GaN layer. Significant improvement in the crystal and optical properties of GaN layer is also achieved. In doping is observed to reduce the stress in AlGaN interlayer measured by high-resolution X-ray diffraction (HRXRD). It can provide more compressive stress to counteract tensile stress and reduce crack density in subsequent GaN layer. Moreover, as a surfactant, indium is observed to cause an enhanced PL intensity and the narrowed linewidths of PL and XRD spectra for the LT-interlayer. Additionally, the crystal quality of GaN layer is found to be dependent on the growth parameters of underneath In-doped LT-AlGaN interlayer. The optimal parameters, such as TMIn flow rate, TMAl flow rates and thickness, are achieved to obtain nearly 2.0 mu m thick crack free GaN film with advanced optical and crystal properties. (c) 2005 Elsevier B.V. All rights reserved.

High responsivity ultraviolet photodetector based on crack-free GaN on Si (111)

A Schottky-based metal-semiconductor-metal photodetector is fabricated on 1 mu m-thick, crack-free GaN on Si (I 11) substrate using an optimized AlxGal-xN/AlN complex buffer layer. It exhibits a high responsivity of 4600A/W at 366nm which may be due to both a crack-free sample and high internal gain. The relationship between responsivity and bias voltage is also investigated. The experiment results indicate that the responsivity increases with the bias voltage and shows a tendency to saturate. (c) 2007 WILEY-VCH Verlag GmbH &Co. KGaA, Weinheim.

MOCVD growth of high quality crack-free GaN on Si(III) substrates

High quality crack free GaN epilayers were grown on Si(111) substrates. Low temperature AlN interlayer grown under low V/III ratio was used to effectively eliminate the formation of micro-cracks. It is found that tensile stress in the GaN epilayer decreases as the N/Al ratio decreases used for AlN interlayer growth. The high optical and structural qualities of the GaN/Si samples were characterized by RBS, PL and XRD measurements. The RT-PL FWHM of the band edge emission is only 39.5meV The XRD FWHM of the GaN/Si sample is 8.2arcmin, which is among the best values ever reported.

Crack Patterns in Ceramic Plates after Quenching

The crack patterns generated in a real ceramic plate and in a plate stacked by ceramic slabs under quenching are experimentally studied. The results here reveal that there are some distinct differences between the two crack patterns. The reasons that caused the differences are the size and boundary effects of the slabs. These crack patterns are very useful to understand the failure mechanisms of ceramic materials in thermal shock.

Dynamic behaviors of mode III interfacial crack under a constant loading rate

In an earlier study on intersonic crack propagation, Gao et al. (J. Mech. Phys. Solids 49: 2113-2132, 2001) described molecular dynamics simulations and continuum analysis of the dynamic behaviors of a mode II dominated crack moving along a weak plane under a constant loading rate. The crack was observed to initiate its motion at a critical time after the onset of loading, at which it is rapidly accelerated to the Rayleigh wave speed and propagates at this speed for a finite time interval until an intersonic daughter crack is nucleated at a peak stress at a finite distance ahead of the original crack tip. The present article aims to analyze this behavior for a mode III crack moving along a bi-material interface subject to a constant loading rate. We begin with a crack in an initially stress-free bi-material subject to a steadily increasing stress. The crack initiates its motion at a critical time governed by the Griffith criterion. After crack initiation, two scenarios of crack propagation are investigated: the first one is that the crack moves at a constant subsonic velocity; the second one is that the crack moves at the lower shear wave speed of the two materials. In the first scenario, the shear stress ahead of the crack tip is singular with exponent -1/2, as expected; in the second scenario, the stress singularity vanishes but a peak stress is found to emerge at a distance ahead of the moving crack tip. In the latter case, a daughter crack supersonic with respect to the softer medium can be expected to emerge ahead of the initial crack once the peak stress reaches the cohesive strength of the interface.

Plastic zone in front of mode I crack in phenolphthalein polyether ketone

The plastic zone size and crack opening displacement of phenolphthalein polyether ketone (PEK-C) at various conditions were investigated. Both of them increase with increasing temperature (decreasing strain rate), i.e. yield stress steadily falls. Thus, the mechanism increasing the yield stress leads to increased constraint in the crack tip and a corresponding reduction in the crack opening displacement and the plastic deformation zone. The effect of the plastic deformation on the fracture toughness is also discussed.

COMPARISON OF THE CRACK-GROWTH METHOD AND THE CRACK STRESS WHITENING ZONE METHOD FOR THE FRACTURE-TOUGHNESS DETERMINATION OF PHENOLPHTHALEIN POLY(ETHER KETONE)

Fracture toughness values of phenolphthalein poly(ether ketone) (PEK-C) at 190 degrees C were determined by two different methods, i. e. the conventional crack growth method and the crack stress whitening zone method, which show consistent results. This indicates that the crack stress whitening zone method can be used to determine the crack initiation of some polymers for which the blunting line concept is unsuitable.

J-INTEGRAL STUDIES OF CRACK INITIATION AND CRACK TIP-BLUNTING OF PHENOLPHTHALEIN POLYETHER KETONE

The J-integral is applied to characterize the fracture initiation of phenolphthalein polyether ketone (PEK-C) for which the concepts of linear elastic fracture mechanics (LEFM) are inapplicable at high temperatures for reasonably-sized specimens due to ex

Fracture mechanics of piezoelectric materials

This paper presents an analysis of crack problems in homogeneous piezoelectrics or on the interfaces between two dissimilar piezoelectric materials based on the continuity of normal electric displacement and electric potential across the crack faces. The explicit analytic solutions are obtained for a single crack in an infinite piezoelectric or on the interface of piezoelectric bimaterials. For homogeneous materials it is found that the normal electric displacement D-2, induced by the crack, is constant along the crack faces which depends only on the remote applied stress fields. Within the crack slit, the perturbed electric fields induced by the crack are also constant and not affected by the applied electric displacement fields. For bimaterials, generally speaking, an interface crack exhibits oscillatory behavior and the normal electric displacement D-2 is a complex function along the crack faces. However, for bimaterials, having certain symmetry, in which an interface crack displays no oscillatory behavior, it is observed that the normal electric displacement D-2 is also constant along the crack faces and the electric field E-2 has the singularity ahead of the crack tip and has a jump across the interface. Energy release rates are established for homogeneous materials and bimaterials having certain symmetry. Both the crack front parallel to the poling axis and perpendicular to the poling axis are discussed. It is revealed that the energy release rates are always positive for stable materials and the applied electric displacements have no contribution to the energy release rates.

Dislocation theory of the fracture criterion for anisotropic solids

A dislocation theory of fracture criterion for the mixed dislocation emission and cleavage process in an anisotropic solid is developed in this paper. The complicated cases involving mixed-mode loading are considered here. The explicit formula for dislocations interaction with a semi-infinite crack is obtained. The governing equation for the critical condition of crack cleavage in an anisotropic solid after a number dislocation emissions is established. The effects of elastic anisotropy, crack geometry and load phase angle on the critical energy release rate and the total number of the emitted dislocations at the onset of cleavage are analysed in detail. The analyses revealed that the critical energy release rates can increase to one or two magnitudes larger than the surface energy because of the dislocation emission. It is also found elastic anisotropy and crystal orientation have significant effects on the critical energy release rates. The anisotropic values can be several times the isotropic value in one crack orientation. The values may be as much as 40% less than the isotropic value in another crack orientation and another anisotropy parameter. Then the theory is applied to a fee single crystal. An edge dislocation can emit from the crack tip along the most highly shear stressed slip plane. Crack cleavage can occur along the most highly stressed slip plane after a number of dislocation emissions. Calculation is carried out step by step. Each step we should judge by which slip system is the most highly shear stressed slip system and which slip system has the largest energy release rate. The calculation clearly shows that the crack orientation and the load phase angle have significant effects on the crystal brittle-ductile behaviours.

Molecular dynamics simulation of the role of dislocations in microcrack healing

The molecular dynamics method is used to simulate microcrack healing during heating or/and under compressive stress. A centre microcrack in Cu crystal would be sealed under compressive stress or by heating. The role of compressive stress and heating in crack healing was additive. During microcrack healing, dislocation generation and motion occurred. When there were pre-existing dislocations around the microcrack, the critical temperature or compressive stress necessary for microcrack healing would decrease, and, the higher the number of dislocations, the lower the critical temperature or compressive stress. The critical temperature necessary for microcrack healing depended upon the orientation of the crack plane. For example, the critical temperature for the crack along the (001) plane was the lowest, i.e. 770K.

Numerical Analysis of Rainfall Infiltration in the Slope with a Fracture

With the finite volume method, a 2D numerical model for seepage in unsaturated soil has been established to study the rainfall infiltration in the fractured slope.The result shows that more rain may infiltrate into the slope due to existing fracture and then the pore pressure rises correspondingly. Very probably, it is one of the crucial factors accounting for slope failure. Furthermore a preliminary study has been conducted to investigate the influence of various fracture and rainfall factors such as the depth, width and location of a crack, surface condition, rainfall intensity and duration. Pore pressure and water volumetric content during the transient seepage are carefully examined to reveal the intrinsic mechanism.