Epitaxial layer thickness

When the epitaxial layer thickness is quite high, typically of the order of one micrometre, we can apply the simple criteria discussed in Chapter 3 to determine the layer parameters from the rocking curve. The effective mismatch can be determined by direct measurement of the angular splitting of the substrate and layer peaks and the differential of the Bragg law. This simple analysis catmot be applied when the layer becomes thin, typically less than about 0.25 //m, where, even for a single layer, interference effects become extremely important. We consider these issues in section 6.2 below. 

Figure 9. Epitaxial layer thickness as a function of growth time for the growth of GaAs (, A) and Al0.4Ga0.eAs ( , V). (Reproduced with permission from reference 84. Copyright 1982 Elsevier.)...

Figure 11. Calculated epitaxial layer thickness as a function of time for equilibrium, step-cooling, and supercooling temperature programs. (Reproduced with permission from reference 94. Copyright 1984 American Institute of...

Fig. 9.3. (a) Linear increase of epitaxial layer thickness grown from 1,050°C on (111) substrate with 0.5°C min-1 cooling rate using tin or indium as solvent, (b) Thickness versus temperature drop for various cooling rate for tin solvent... 

SLS thickness is thicker than the critical thickness [12]. It is suggested from the experimental results that the dislocations generated at SLS are bended by TCA2, resulting in the low dislocation density. Until now, the low etch pit density on the order of 10 cm has been obtained using SLS and TCA for the total epitaxial layer thickness of more than 3.5 pm [40-43]. Few papers have been reported on the growth of GaAs on Si, with the dislocation density of 10 cm at the epitaxial layer thickness of less than 3 pm. 

The electronic quality of the layer was determined measuring the bulk minority carrier lifetime on a free-standing Si epitaxial layer passivated with a corona charged (Schofthaler et al. 1994) thermal oxide. From a measured Hall mobility of the holes of 186 cm A s, the bulk lifetime of (0.27 0.08) ps, a diffusion length of 11 pm for the electrons in the p-type Si could be estimated. The diffusion length of 11 pm was about double the epitaxial layer thickness of 5.8 pm. Therefore, the lAD method seemed suitable for the production of epitaxial Si solar cells (Krinke et al. 1999). 

By moving the sample across the opening in the mask or over the illuminated area in a micro-accessory, it is possible to obtain a reflection profile of the sample. If much of this work is to be done, a platform with micrometer screws for positioning the sample can be substituted for the sample holder. This technique is especially useful in the semiconductor industry, where obtaining an epitaxial layer-thickness profile may be required. 

An instance of Process Settings comprises of aU Parameters which can be used in order to adjust the result. In the discussed example, only Process Duration was chosen. By increasing the Process Duration, the epitaxial layer thickness is getting higher, or reverse. The Process Control Settings shall be discussed in httle more depth in the sequel. In this example, the used instance of Process Control Settings is MySPCForEpiThicknessOfS pm which comprises of the information which is needed to parameterize statistical process control (SPC) in the context of gauging layer thicknesses. 

For construction of suitable samples molecular beam epitaxy was selected, the method of choice for the production of complicated epitaxial layer systems with different materials. As substrates Si wafer material (about 20x20 mm-, thickness 1 mm) and SiO, discs (diameter 30 mm, thickness 3 mm) were used. Eight layered structures (one, two and three layers) were built up with Al, Co, and Ni, with an indicated thickness of 70 nm, each. 

RBS and channeling are extremely useful for characterization of epitaxial layers. An example is the analysis of a Sii-j Gejc/Si strained layer superlattice [3.131]. Four pairs of layers, each approximately 40 nm thick, were grown by MBE on a <100> Si substrate. Because of the lattice mismatch between Sii-jcGe c (x a 0.2) and Si, the Sii-j Ge c layers are strained. Figure 3.51 shows RBS spectra obtained in random and channeling directions. The four pairs of layers are clearly seen in both the Ge and Si... 

The short penetration depth of UV/blue photons is the reason that frontside CCD detectors have very poor QE at the blue end of the spectrum. The frontside of a CCD is the side upon which the polysilicon wires that control charge collection and transfer are deposited. These wires are 0.25 to 0.5 /xm thick and will absorb all UV/blue photons before these photons reach the photosensitive volume of the CCD. For good UV/blue sensitivity, a silicon detector must allow the direct penetration of photons into the photosensitive volume. This is achieved by turning the CCD over and thinning the backside until the photosensitive region (the epitaxial layer) is exposed to incoming radiation. 

Similarly, the (111) GaAs substrate could be used to achieve epitaxial growth of zinc blende CdSe by electrodeposition from the standard acidic aqueous solution [7]. It was shown that the large lattice mismatch between CdSe and GaAs (7.4%) is accommodated mainly by interfacial dislocations and results in the formation of a high density of twins or stacking faults in the CdSe structure. Epitaxy declined rapidly on increasing the layer thickness or when the experimental parameters were not optimal. 

A second application of current interest in which widely separated length scales come into play is fabrication of modulated foils or wires with layer thickness of a few nanometers or less [156]. In this application, the aspect ratio of layer thickness, which may be of nearly atomic dimensions, to workpiece size, is enormous, and the current distribution must be uniform on the entire range of scales between the two. Optimal conditions for these structures require control by local mechanisms to suppress instability and produce layer by layer growth. Epitaxially deposited single crystals with modulated composition on these scales can be described as superlattices. Moffat, in a report on Cu-Ni superlattices, briefly reviews the constraints operating on their fabrication by electrodeposition [157]. 

From a practical point of view, the optical detection of possible X—H bonds in hydrogenated samples is performed at LHeT as a better sensitivity is obtained at this temperature because the features are sharper than the ones observed at ambient. The sensitivity of Fourier Transform Spectroscopy (FTS) allows usually a normal incidence geometry of the optical beam. Two kinds of samples are generally used in the hydrogenation studies. The first are thin epitaxial layers (1 to 5 in thickness) with dopant concentrations in the 1017-102° at/cm3 range on a semi-insulating... 

Silicon-based pressure sensors are amongst the most common devices making use of this process. A thin low-n-doped epitaxial layer on the wafer determines an etch stop depth and thus the thickness of e.g. the pressure sensor membrane. 

In this chapter we discuss the measurement and analysis of simple epitaxial stractures. After showing how to select the experimental conditions we show how to derive the basic layer parameters the composition of ternaries, mismatch of quaternaries, misorientation, layer thickness, tilt, relaxation, indications of strain, curvature and stress, and area homogeneity. We then discuss the hmitations of the simple interpretation. 

We now assume that we have properly recorded rocking curves available, and that a substrate with a single epitaxial layer >0.5 //m thick (and less than, say, 5 m) is measured. This will result in two peaks, one each from the substrate and layer. The same analysis will apply to multiple peaks (other than those from superlattices) provided that they are well separated so that interference effects are minimised. The basic parameters are derived as follows, with the symmetric reflection used unless otherwise specified. The examples are given for (001) substrates and layers, but are quite general (with the caution that if the Poisson ratio is required, the value appropriate to the crystal orientation should strictly be used). 

The scattered intensity is proportional to the volume of the crystal. This implies that the scattering from a thin epitaxial layer, large in area compared with the beam diameter, will be proportional to the layer thickness. 

An important featnre to note in double-axis topography experiments is that when the beam area is large, the measnred rocking curve widths are not necessarily intrinsic. For example, mismatched epitaxial layers curve substrate wafers by an amonnt which depends on the degree of mismatch and layer thickness. Topographs of snch curved wafers show bands of diffracted intensity. 

The films are epitaxial in the sense that the lattice constant is intermediate between those of copper and nickel. As indicated above, that modulated strain is probably responsible for the increased hardness. Other authors (5) have tried to explain similar effects by stating that the layers were specifically oriented. Our example (6) demonstrates that these considerations must be reexamined since it was possible to achieve the effect in a crystalline multilayer deposited on an amorphous nickel-phosphorus underlayer. It appears that layer thickness is the important parameter here. 

---