Database Needs for Modeling and Simulation of Plasma Processing (1996)
Chapter: SURFACE REACTION DATABASE AND DIAGNOSTICS
applications can be extrapolated from flame studies, but apart from the data of Piper and co-workers31 on nitrogen-silane reactions, the database for quenching excited species in other etching and deposition environments is sparse.
TABLE 3.2 Candidate Species for Application of TALIF
|
|
Excitation |
Fluorescence |
|||||
|
Species |
λ (nm) |
Transition |
λ (nm) |
Transition |
Titration Reaction |
||
|
H |
205 |
1s2S |
-3d2D |
656 |
3d2D |
-2p2P |
|
|
C |
280 |
2p23p |
-2p3p3P |
910 |
3p3P |
-3p3P0 |
|
|
N |
207 |
2 p3 4S0 |
-2p23p S0 |
747 |
3p4S0 |
-3p4p |
|
|
O |
225 |
2p43p |
-2p33p3P |
844 |
3p3P |
-3s3S |
|
|
Cl |
210 |
3p52p |
-3p44p2F |
904 |
4p2F |
-4s2D |
|
|
Cl |
233 |
3p2p0 |
-4p4S0 |
725-775 |
4p4S0 |
-4s4P |
|
|
SOURCE: Data from A.D. Tserepi, J.R. Dunlop, B.L. Preppernau, and T.A. Miller, J. Appl. Phys. 72:2638 (1992); Alden et al., Optics Communications 71(5):263 (1989); Das et al., J. Chem. Phys. 79:725 (1983). |
|||||||
Another technique reported in a number of very encouraging recent studies32 is degenerate four-wave mixing (DFWM). This technique has several advantages: first, only two ports are needed, and they can be in line, as opposed to one large-aperture port at right angles to the entrance probe beam port; second, because of the fully resonant nature of DFWM, it is significantly more sensitive than other four-wave mixing techniques such as coherent anti-Stokes Raman scattering (CARS). Examples of the sensitivity include OH in a flame, about 4 × 1011 cm-3; and NO, about 8 × 1011 cm-3. The molecules C3 and SiC2 have been measured using DFWM and stimulated emission pumping (SEP) at sensitivities of 1012 cm-3 per rotational state. The main advantage of DFWM in the present context, however, is an insensitivity to quenching, allowing measurements with laser-induced fluorescence (LIF) levels of sensitivity without the corrections for molecular interactions required for LIF quantification.
Surface Reaction Database and Diagnostics
Surface reactions are often paramount in controlling the concentrations of atoms and other reactive radicals produced in etching and deposition plasmas. Many measurements have been made of surface reactions, generally using ex situ techniques. Some progress is being made with in situ sensors and ellipsometry (including infrared ellipsometry). Other recent studies have either used measurements of the steady-state (time-averaged) species concentration profiles adjacent to surfaces to extract reaction rates with surfaces, or made time-gated measurements of species decay after switch-off. The first method is more applicable at higher pressures in the reactor. The dynamic range of this approach depends on the accuracy and spatial resolution of the probing method. Generally the limited depth resolution will make extraction of data from finely structured surfaces quite difficult. Recent measurements on defined large areas of different materials adjacent to each other have revealed that surface reactivity is very sensitive to very small concentrations of sputtered material. An alternative diagnostic method is to make time-resolved measurements of the concentration decay, after the radio-frequency supply has been turned off or reduced to a significantly lower power level. Modulated plasma experiments can be used also to determine the gas phase kinetics, the negative ion formation rates, and the formation of clusters. At lower pressures, surface recombination and reaction often dominate the losses, and coefficients can be extracted from the species decay constants and time-dependent profiles adjacent to the surface. There are many indications that the surface reactivity often depends not only on the material and its history, but also on the fluxes and synergisms of other species (ions and neutrals) arriving at the surface. Except for a few cases, relatively little is known about these synergisms.
