With precision in In\(_{x}\)Ga\(_{1-x}\)N ratio so desired, it is no surprise that molecular beam epitaxy (MBE) was been investigated as a mechanism of synthesis. Operating in an ultra-high vacuum, MBE involves the reaction of high-temperature molecules as ’molecular beams’ with a crystalline surface\cite{Cho_1975}. This means that, similarly to HVPE, the operator has full control over the relative amounts of the inputs– but the high vacuum reduces the likelihood of any unhelpful competing reactions.
Illustrative of the basic procedure is the work of Hu et al\cite{Hu_2007}, who created InGaN/GaN nanocolumn crystals with an InN interlayer, demonstrating complete control of all InGaN’s components. Grown on a silicon substrate, Hu and company used InN as an interlayer due to the small lattice constant difference. Once the base layer of InN had been laid down, it was possible to grow a remarkable InGaN/GaN heterostructure through periodic deposits of InGaN and GaN. The resulting SEM images revealed a flowered structure with helpful internal quantum efficiency properties.
Applying this powerful synthesis technique to nanopillars, Vajpeyi et al. employed nitrogen radiofrequency-plasma source MBE (RF-MBE) to deposit In\(_{x}\)Ga\(_{1-x}\)N directly onto a silicon surface\cite{Vajpeyi_2009}. Varying the substrate temperature revealed that the indium concentration coefficient \({x}\) was inversely dependent on temperature– but furthermore, that In\(_{x}\)Ga\(_{1-x}\)N nanopillars could not be deposited at temperature below 575 degrees Celsius. Other nanopillar properties–the density and diameter– likewise bore relationships with substrate temperature, indicting that the control granted by this synthesis was somewhat constrained. But while the temperature cutoff may have limited the In\(_{x}\)Ga\(_{1-x}\)N nanopillars short of the full spectrum possible, Vajpeyi and associates did achieve both range (from 450 to 625 nm) while maintaining limits on emission wavelength.
More recently, Guo et al. attempted a comprehensive evaluation of MBE-produced InGaN nanowires, using plasma-assisted MBE (PA-MBE) to vary the relative In\(_{x}\)Ga\(_{1-x}\)N composition not through substrate temperature, but rather through changing the In flux during MBE nanowire production\cite{Guo_2010}. As a result, Guo’s group managed to not only create nanowires of varying In compositions previously not seen in MBE synthesis, but also created wires in which the In\(_{x}\)Ga\(_{1-x}\)N ratio varied over the length of the wire. The photoluminescence spectra illustrated a manner of InGaN synthesis which not only has high precision for different ratios of In\(_{x}\)Ga\(_{1-x}\)N, but also could control (to a degree) the broadness of emission, allowing more or less focused wavelengths of light, and showing the true potential of MBE.