What are Vapor-Phase Epitaxy (VPE) and Metal-Organic Vapor-Phase Epitaxy (MOVPE)?
This is a continuation from the previous tutorial - what is liquid-phase epitaxy (LPE)?
Vapor-Phase Epitaxy
In vapor-phase epitaxy (VPE), the source chemicals from which the epitaxial layers are gown are gaseous. The technique has been widely used for the growth of several III-V compound semiconductors.
VPE is often classified as one of two different methods, the chloride and the hydride techniques.
In the chloride method, \(\text{AsCl}_3\) or \(\text{PCl}_3\) is passed over elmental \(\text{Ga}\) or \(\text{In}\) to form metal chlorides. These metal chlorides then react with \(\text{AsH}_3\) or \(\text{PH}_3\) near the \(\text{InP}\) substrate to form epitaxial layers of \(\text{InGaAsP}\) on \(\text{InP}\).
The metal chlorides can also be formed by using source pieces of \(\text{GaAs}\) or \(\text{InP}\) instead of elemental \(\text{Ga}\) or \(\text{In}\).
In the hydride method, metal chlorides are formed by passing \(\text{HCl}\) gas over hot \(\text{In}\) or \(\text{Ga}\) metal.
VPE of \(\text{InGaAsP}\) material for laser devices has often used the hydride method. The chloride method is more useful for fabrication of the field-effect transistors because it produces layers with lower background doping levels.
Figure 4-9 shows a schematic of the growth reactor for the growth of \(\text{InGaAsP}\).

The deposition is started by passing \(\text{HCl}\) gas over hot (\(850-900^\circ\text{C}\)) \(\text{In}\) or \(\text{Ga}\) metal, which forms metal chlorides. Arsine and phosphine are mixed with the metal chlorides in the mixing zone.
The \(\text{InP}\) substrate is held in a chamber at room temperature that is flushed with \(\text{H}_2\). Prior to growth, the substrate is heated to a temperature of \(~700^\circ\text{C}\) in an atmosphere of \(\text{AsH}_3\) and \(\text{PH}_3\) to prevent surface decomposition.
The substrate is then inserted into the growth chamber for a desired period of time. Here the growth takes place. The growth rate generally lies in the range of 0.1-1 μm/min. All reactant flows are controlled by precision mass-flow controllers.
P-type doping (with \(\text{Zn}\)) can be accomplished by flowing \(\text{H}_2\) over hot \(\text{Zn}\) metal, which carries \(\text{Zn}\) vapor to the mixing chamber. Similarly, n-type doping can be accomplished using 100 ppm \(\text{H}_2\text{S}\) along with \(\text{AsH}_3\) and \(\text{PH}_3\) in the mixing chamber.
The composition of the layers to be grown in a VPE reactor is changed by altering the flow rates of the reactants.
It takes a certain amount of time to establish gaseous equilibrium in the growth chamber after the flow rates are changed. Because of this time delay, it is difficult to grow successive epitaxial layers with different compositions and abrupt interfaces in a one-chamber reactor of the form shown in Figure 4-9.
For the growth of high-quality layers of different compositions, multi-chamber growth reactors have been used. Figure 4-10 shows a schematic of a dual-chamber growth reactor.

In each chamber the gas-flow ratio, which is specific. to a certain composition, is first established.
Then the substrate is introduced into one chamber for the first growth to take place. For the second growth, the substrate is transferred to the other chamber. The transfer time is ~ 2 seconds.
Heterojunction-interface widths of about 5-6 nm have been obtained using this technique.
Metal-Organic Vapor-Phase Epitaxy
Metal-organic vapor-phase epitaxy (MOVPE), also known as metal-organic chemical vapor deposition (MOCVD), is a variant of the VPE technique that uses metal alkyls as sources from which the epitaxial layers form.
MOVPE has been extensively studies by Dupuis, Daps, and coworkers using the \(\text{AlGaAs}\) material system.
The low-pressure MOVPE technique, wherein the reaction takes place at a gas pressure of ~ 0.1 atm. has been used for the growth of \(\text{InGaAsP}\) by Hertz, Razeghi, and coworkers.
Several other researchers have also reported on the growth of III-V compounds by MOVPE.
Figure 4-11 shows a schematic of a low-pressure MOVPE system.

Group III alkyls [\(\text{Ga}(\text{C}_2\text{H}_5)_3\) and \(\text{In}(\text{C}_2\text{H}_5)_3\)] and group V hydrides [\(\text{AsH}_3\) and \(\text{PH}_3\)] are introduced into a quartz reaction chamber that contains a substrate placed on a radio-frequency (RF) heated (\(\sim500^\circ\text{C}\)) carbon susceptor.
The gas flow near the substrate is laminar, with velocities in the range of 1-15 cm/s for a working pressure between 0.1 and 0.5 atm.
It is thought that a stagnant boundary layer is formed near the hot susceptor and gas molecules diffuse to the hot surface of the substrate.
At the hot surface the metal alkyls and the hydrides decompose, producing elemental \(\text{In}\), \(\text{Ga}\), \(\text{P}\), and \(\text{As}\). The elemental species deposit on the substrate, forming an epitaxial layer.
The gas-flow rates are controlled by mass-flow controllers. \(\text{Zn}(\text{C}_2\text{H}_5)_3\) and \(\text{H}_2\text{S}\) are used as sources for p-type and n-type doping respectively.
In the low-pressure MOVPE technique, the velocities of the reactants are higher than those for hydride VPE. This allows quick changes in composition at heterojunction interfaces since new gas compositions are rapidly established. The growth rates are typically in the range of 2-4 μm/h.
Dupuis et al. in 1985 were able to grow high-quality \(\text{InGaAsP}\) double-heterostructures using an atmospheric pressure MOVPE system.
The next tutorial introduces what is molecular-beam epitaxy (MBE).