Junction depth, diffusion

Diffusivities of various elements ate determined experimentally. Dopant profiles can be determined. The junction depth can be measured by chemically staining an angle-lapped sample with an HE/HNO mixture. The -type region of the junction stains darker than the n-ty e region. The sheet resistivity can also be measured using a four-point probe measurement. These two techniques ate used for process monitoring. 

Integrated circuit technology requires a reduction in device dimensions. The junction depth, where the donor and acceptor concentrations are equal, is set between 10 and 30 nm. These shallow junction requirements restrict ion implantation technology - not on the implantation procedures themselves, but on the subsequent diffusion of the implanted species during thermal annealing. 

The implantation of MeV Si ions into a Si substrate can also suppress boron-enhanced diffusion normally associated with a high boron concentration layer (Shao et al. 2003). Junction depths of 20 nm were achieved in samples implanted with 0.5 keV B ion at a dose of 1015 cm-2 following a 1,000°C thermal anneal. 

A potential gr< dient can be created by forming an interface or junction with a semiconducting material. Metal-semiconductor, p-n semiconductor, and semiconductor-electrolyte interfaces have been used in the construction of photovoltaic cells.25 27 The interface in a p-n junction photovoltaic cell can be constructed by doping the surface of an n or p-type semiconductor with atoms that invert the semiconductor type. These atoms are then thermally diffused into the host semiconductor to an optimal depth. Diffusion rates... 

The Fe was diffused, from a spin-on glass film, onto n-type wafers at 700 to 900C. The diffusivities, as determined using junction-depth and conductivity techniques, could be explained in tenns of a model which assumed the existence of exhaustible diffusion sources. It was found that the diffusivity was described by ... 

By using p-n junction-depth and electrical conductivity measurements, a study was made of diffusion in single crystals. At 1250C, the diffusivity was equal to 3.3 x 10"12cm2/s and, at 1100 to 1250C, the results could be described by ... 

Electrical conductivity and p-n junction-depth measurements were used to study the diffusion of K (introduced by ion implantation) in crystals with very low O and N concentrations. At 500 to 800C, the diffusivity obeyed ... 

Another important diffusion parameter is junction depth, Xj. It represents the depth (i.e., value of j ) at which the diffusing impurity concentration is just equal to... 

FigMre 5.10 Schematic concentration profiles taken after (1) predeposition and (2) drive-in diffusion treatments for semiconductors. Also shown is the junction depth, Xj. 

A recent result on Al-implanted 4H-SiC, which is in sharp contrast to the common opinion that amorphization should be avoided, has been reported by E. Kalinina from the Ioffe Institute in St. Petersburg [124]. She showed that very good activation could be achieved by RT implantation of Al at doses in excess of 10 cm . This completely amorphized material is shown to be fairly well recrystallized by an RTA process that also causes some of the Al to diffuse into the low-doped n-type epilayer, forming the p n junction at a larger depth than the highly defective implanted area. Nearly ideal forward IV-characteristics were also shown for current densities of several kA cm . Even if the stability of such heavily implanted devices may be questioned this result shows that there is still a long way to go before a fully optimized implantation technology is at hand. 

FIGURE 1 shows a typical SIMS profile of Mg in GaN. The Mg concentration was uniformly distributed at a growth temperature of 750°C [8], The concentration of the Mg is 2 - 3 x 101 cm" near the surface and is the same at a depth of 0.4 pm towards the substrate. At the junction between the GaN Mg and undoped GaN buffer, the Mg concentration increases to 1 x 1019 cm 3 and is seen to diffuse into the buffer layer. This Mg peak is probably caused by enhanced diffusion of Mg, associated with defects and dislocations generated at the layer/substrate interface, towards the substrate. A wide chemical doping range in GaN Mg of 1 x 1017to 1 x 1019 cm 3 was obtained. 

It is seen from Table 5.1 that the values of the conversion efficiency in bilayer solar cells also is quite low. As mentioned in the introduction it is difficult to dissociate excitons in the conducting polymers. The Donor/Acceptor (D/A) junction between the polymer and the fullerene is rectifying and can be used for designing photovoltaic cells or photodetectors. In this bilayer cell also the conversion efficiency is low. The cause of the low efficiency is that the charge separation occurs only at the D/A interface that results low collection efficiency. The diffusion length of the exciton is a factor 10, lower than the typical penetration depth of the photon. 

Fig. 8.4. Profile of light intensity at the semiconductor electrolyte junction. W is the width of the depletion layer and L, is the hole diffusion length. The penetration depth of the light...

On the other hand, by using more than one ion-implantation step followed by drive-in with or without a subsequent epitaxy step, adjacent etch-stop regions of different depths can be achieved [24, 25]. Shallower junctions can be created either by a shallow diffusion or by an epitaxial step. 

The depth of the recrystallised layer is very small, of the order of 0.01 -0.1 pm. Thus, a possible way to eliminate its effect on p-n junction properties is to perform a drive-in diffusion of aluminium from the epitaxial layer. Unfortunately, the diffusivity of aluminium from an epitaxial layer is extremely slow in SiC. The diffusivity is 3 - 5 orders of magnitude lower than that observed for diffusing aluminium from the vapour phase [70]. The authors of [69] had to employ very high diffusion temperatures, over 2500 °C. The anneal produced a shift of the p-n junction into the crystal bulk and the electrical properties were substantially improved. However, this could not provide the elimination of the weak points of the junctions. The characteristics of the p-n junctions were worse than those with the recrystallised layer removed by sublimation etching. In addition, the surface evaporation and graphitisation at temperatures above 2500 °C severely reduces the reproducibility of the results. 

The lithium drifting process, developed by Pell, consists of two major steps (1) formation of an n-p junction by lithium diffusion, and (2) increase of the depletion depth by ion drifting. 

---