Chemical vapor deposition (CVD) is a process used to grow a film of solid material by transporting a chemical precursor in its vapor phase and decomposing it through heat. Typically, a material in its solid form is vaporized and diluted with an organic reactant, which will assist the material’s surface mobility. The chemical combination is transported to the heated surface where it decomposes, leaving the material on the heated surface to migrate to the growing film. This technique is widely used in IC microfabrication as a deposition method for the layers comprising a semiconductor wafer, specifically the oxide and nitride layers. It significantly helped transistor shrinking by establishing very thin film deposits in the substrate.
CVD is also used for fabrications of novel powder, fiber, components of dense structural parts, preforms of ceramic composites, coatings for corrosion and wear resistance, and synthetic diamonds. Basically, CVD is used in deposition of film in surfaces that other methods cannot achieve. Over other deposition methods, CVD is very versatile, and can form pure and dense films at low temperature and low atmospheric pressure. It can deposit a wide range of materials in a number of orientations such as monocrystalline, polycrystalline, epitaxial, uniaxial, and amorphous. These materials can be a metallic element like tungsten and aluminum, or non-metallic like carbon, or even a compound such as fluorocarbon, nitride, carbide, oxide, etc.
In a typical CVD process, the actual deposition process takes place in a reactant chamber. Figure 1 details the phases of reactions in CVD, which starts at mass transport of the reactants to the chamber (a). This produces film precursors and undesirable by-products during the gas phase reaction (b). Then the film precursors move towards the surface (c) where it is adsorbed (d). The selectively occurring adatoms reacts with the surface and migrates the film forming material to the growth sites (e). Once, at the growth site, the material will be incorporated to the growing film (f) whereas both the by-products and volatile reactants are desorbed (g) and removed from the chamber through vacuum and exhaust systems (h). Some by-products are treated appropriately before releasing it to the atmosphere to eliminate any danger it can introduce to the environment, especially the toxic, flammable, or corrosive gases.
Since the materials involved in CVD is in gaseous form, this process requires volatile and stable precursors. The precursors usable for CVD can be categorized as halides, hydrides, metal organic compounds, alkyls, alkoxides, carbonyls, dialylimides, and diketonates, to name a few. In general, only a single element from the precursor compound is deposited during the CVD process, while the rest of it is volatized. For instances that the precursor provides two or more elements, the transport mechanism is faster and simpler since it requires lesser amount of reactants to deposit the material.
Another important aspect of every CVD system is the energy source (thermal, electron or laser) needed to ignite the chemical reaction of the film precursors with the target surface. Among these energy sources, the most widely used is the thermal energy. In a CVD process with a thermal energy source, the chamber is first heated to a certain temperature. Once the diffused precursors reached the heated surface, equilibrium shifts towards the material to be deposited, which has lower temperature.
The rate of deposition or the flux of the material is greatly affected by gas diffusion and surface processes. Flux defines the amount of substance that will flow through an area at a given time. As postulated in Fick’s first law, flux will always flow from the area with higher concentration to the area with lower concentration. It relates the flux of the material to the diffusivity of the gas, and the concentration of the layer separating the surface target surface and the bulk flow. Mathematically, the flux of the depositing material is given by:
Fl = D(dc/dx) Equation 1
Fl is the diffusion flux of the material in amount of substance per unit area per time (mol/m2.s);
D is the diffusion coefficient in square of length over time (m2/s);
dc is the concentration of the layer in amount of substance per unit volume (mol/m3); and
dx is the boundary layer thickness (m).
The boundary layer thickness, which is a function of the distance x along the substrate and can computed from
δ(x)=√(ηx/ρU) Equation 2
Moreover, the average boundary layer in a surface is modeled by Prandtl as:
δ = 1/L ∫0L δ(x)dx= 2/3 [L√(η/ρUL)] Equation 3
η is gas viscosity in pressure and time (Pa·s);
ρ is gas density in units of mass per volume (kg / m3);
U is gas stream velocity in unit of length over time (m / s); and
L is the length of the surface being deposited on (m);
is a reverse of dimensionless gas constant called Reynold’s number Re. In fluid motion, Reynold’s number is the ratio of the inertial forces to the viscous effects for a given flow motion. It is defined as:
Re = ρUL/η Equation 4
Substituting equation 4 in equation 3 yields to:
δ = 2L/(3√( Re)) Equation 5
Lastly, substituting equation 5 in equation 1 will simplify Fick’s Law equation to:
Fl = D Δc/2L 3√( Re) Equation 6
This demonstrates that aside from diffusion coefficient and concentration gradient, the deposition rate of a material also depends on the square root of the gas velocity, gas viscosity, and gas density.
If you are interested in deposition services, please visit our deposition service page.
Subscribe to our newsletter to receive our new articles directly in your mail box.
If you liked this article, please give it a quick review in StumbleUpon, Facebook or Pinterest.