Plasticity of misoriented (001) GaAs surface

Abstract

In the field of semiconductor studies, surface plasticity has proved to be of great importance as it may be determinant to optoelectronic performance of III–V heterostructures. In fact, it is well known that performance of the devices is dramatically affected by threading dislocations that extend through the active layers. The mechanism by which such dislocations appear in mismatched structures is still a controversial topic. However, the use of a compliant substrate made of a thin film twist bonded on a bulk substrate induced much improvement of the structural quality of mismatch structures [1]. Using microand nanometer scale indentation techniques has proved to be a very powerful tool for investigating surface plasticity. The case of a (001) face of GaAs (common orientation used for heteroepitaxy) has been intensively investigated during the past two decades [2–9]. However, some studies made on single crystals of Au and Ni-based superalloys showed that nano-mechanical behavior can be changed whether the material is tested in step-free zones or in the vicinity of steps [10, 11]. In particular, mean applied stress at yield was determined to be about 30–45% lower near a surface step in Au [10] while hardness was found to be higher on plateaus than in the vicinity of steps in Ni-based superalloys [11]. Here, we have investigated the plastic behavior of a (001) vicinal surface (4◦ intentionally misoriented). Such a surface presents steps whose density increases with the misorientation. Nanoindentation as well as transmission electron microscopy (TEM) were used to investigate the mechanical behavior of such a misoriented surface. (001) oriented Czochralski-grown single crystals of undoped gallium arsenide have been used. One of them was misoriented by 4◦ around the [110] direction. Assuming steps of height a/2 = 0.28 nm, the distance between two steps can be estimated to be about 4 nm. The other crystal was non-intentionally misoriented (0 ± 0.1◦, distance between steps >160 nm). From here onward, we shall refer to the two types of samples by 4◦ and 0◦ respectively. Both samples were deformed at room temperature in air by a Berkovitch diamond pyramid indentor using a NHT machine from CSEM (Switzerland). The tests were performed at room temperature in the force-control mode of the machine. Indentation arrays were made for subsequent relocalization in the TEM samples. The maximum load was varied between 0.3 and 50 MN. 54 indents were made in each of the two samples and were observed subsequently by TEM. The calibration procedure suggested by Oliver and Pharr [12] was used to correct for the load-frame compliance of the apparatus and the imperfect shape of the indenter tip. To prepare TEM planview thin foils, the undeformed side of the samples (back side) was mechanically and chemically thinned with a bromine-methanol solution until sufficiently thin to transmit the electron beam. Fig. 1 shows the loading and unloading curves obtained for the 4◦ and the 0◦ samples under the same experimental conditions. During the loading, both reversible and non-reversible deformations were observed. During the unloading, the reversible (elastic) deformation was recovered, leaving a residual deformation hr (about 160 nm in Fig. 1). It should be noted that for all the loads used here, the lateral size of the contact between the indentor and the sample (under maximum load) is always more than the distance between adjacent surface steps. For example, under a maximum load of Fmax = 1 MN, the contact depth was determined to be hc = 48 nm so that the lateral size of the contact (about ∼7 hc = 340 nm) is about 100 times the inter step distance. The unloading curves were analyzed using the method proposed by Oliver and Pharr [12]. The (Vickers equivalent) hardness was found to decrease slightly with the maximum load applied on the indentor and was determined to be about 8.1 ± 0.1 GPa and