Steady state sputtering

Steady-state molecular beam studies of the reaction of methylacetylene on reduced Ti02 (001) surfaces were undertaken to determine whether this reaction could be performed catalytically under UHV conditions. A representative experiment is presented in Figure 1. Prior to each experiment, the surface was sputtered and annealed to a temperature between 400 K and 550 K surfaces prepared in this manner have the highest fraction of Ti(+2) sites (ca. 30% of all surface cations) of any surface we have been able to create by initial sputtering [3]. Thus these are the surfaces most active for cyclotrimerization in TPD experiments [1]. Steady-state production of trimethylbenzene (as indicated by the m/e 105 signal detected by the mass spectrometer) was characterized by behavior typical of more traditional catalysts a jump in activity upon initial exposure of the crystal to the molecular beam, followed by a decay to a lower, constant level of activity over a longer time scale. Experiments of up to 6 hours in duration showed... 

Figure 10.4 Area-normalized CL spectra of Pt4/7/2 for the pure Pt (dotted Une), Pt5gCo42 (solid line), and PtgoRu4o (dashed line) alloys with respect to p (a) as-prepared (h) after electrochemical stabilization. The samples were thin film pure Pt or Pt-based alloys (diameter 8 mm and thickness 80 nm) prepared on Au disks by DC sputtering. Electrochemical stabilization of Pt58 C042 was performed by repeated potential cycling between 0.075 and 1.00 V at a sweep rate of 0.10 V s in 0.1 M HCIO4 under ultrapure N2 (99.9999%) until CV showed a steady state. PtgoRu4o was stabilized by several potential cycling between 0.075 and 0.80 V at 0.10 V s in 0.05 M H2SO4 under ultrapure N2. (From Wakisaka et al. [2006], reproduced by permission of the American Chemical Society.)...

Molecular dynamics simulations have yielded a great deal of information about the sputtering process. First, they have demonstrated that for primary ion energies of a few keV or less, the dynamics which lead to ejection occur on a very short timescale on the order of a few hundred femtoseconds. This timescale means that the ejection process is best described as a small number of direct collisions, and rules out models which rely on many collisions, atomic vibrations and other processes to reach any type of steady state . Within this same short-timescale picture, simulations have shown that ejected substrate atoms come from very near the surface, and not from subsurface regions. 

X 10 — steady state sputtering condition has been reached... 

When a steady state condition has been achieved. Equation 21 implies that the relative surface concentrations are only functions of the bulk concentrations and the sputtering coefficients. This point cannot be overemphasized. Many authors have misinterpreted their data because they did not understand the consequences of this result. Once the sputtering coefficients are known, then thermodynamic properties, such as a tendency towards surface segregation, do not affect the surface concentration. However, the sputtering yields themselves are partially determined by binding energies and the type of compounds which are present in the surface region. These parameters are, of course, influenced by thermodynamic considerations. 

The glow-discharge is particularly suitable for atomic absorption (AA) techniques as the overwhelming majority of sputtered species are neutral atoms. This ability to produce a steady-state population directly from a solid matrix has obvious advantages that led Russell and Walsh to suggest the use of GD in the early development of AA [204]. 

Fig. 32 Effect of chemical shift of Pt 4d5/2 binding energy in Pt skin layer, prepared on a bare PtsoFeso alloy surface by sputtering deposition, onto (top) the kinetically controlled H2 oxidation current (/ <) and (bottom) the steady-state CO coverage (6co) at room temperature in 0.1 M HCIO4 solution saturated with 100 ppm CO/H2 balance [96, 98]. (Reprinted with permission from Ref. 96, Copyright 2003 by John Wiley Sons Ltd).

Breakdown of Major Sputtered Species as Average Yield per Ion During Steady-State Etching for Ion Incident Energies of 25.50.100, and 200 eV ... 

The steady-state surface concentrations are relatively easy to obtain. Consider the implantation of ion species A into the host material B, where NA and VB are the concentrations (per unit volume) of A atoms and B atoms, respectively, at the surface of the sample. Therefore NA/NB represents the surface composition. Let JA and JB be the flux of sputtered atomic species A and B, respectively. Then... 

This is the steady-state surface composition, which is roughly inversely proportional to the total sputtering yield Y, but multiplied by the preferential sputtering factor r. For r = 1, (12.10) reduces to NA/(NB + NA) = MY. 

Since ion implantation is also a bombardment of energetic ions, there is always some sputtering, especially when heavy ions and high doses are used. Sputtering makes the sample surface recede therefore it affects the implantation profile and also removes atoms that have been implanted. This eventually leads to a steady-state condition in which there is no further increase in the amount of implanted species retained in the material. 

For example, if there is preferential sputtering where r > 1, the sputtering yield of A is greater than that of B, and the surface will be enriched in B. This enrichment produces an increase in the sputtering yield of B (more B atoms) and a decrease in the sputtering yield of A (fewer A atoms). As the process continues with macroscopic amounts (greater than 10 nm) of material removed, the increased concentration of B balances out the preferential sputtering of A. Therefore, at steady state the surface concentration ratio will differ from that of the bulk ratio when r 1. 

---