=

Figure 11. (upper right) Redistribution of boron after two step oxidation (wet-dry). Curve A: calculated distribution after first oxidation (1100°C, wet, xo 1 μm) (initial condition for second oxidation). Curve B: calculated redistribution after second oxidation (1200°C, dry, x = 500 Å). Curve calculated redistribution after second oxidation (1200°C, dry, n = 700 A). Curve D: calculated redistribution after second oxidation (1200°C, dry x 1000 Å). measured +++ measured distribution corresponding to curve A, redistribution corresponding to Curve D (Margalit et al., Ref. 10).

C:

...

=

ETCHING RATE (A/sec)

Figure 14. Relationship among silicon nitride film composition, etch rate, and electrical conductivity as influenced by NH3/SiH4 ratio during silicon nitride deposition (Gyulai et al., Ref. 11).

(a)

F

Figure 16.

800 850 900 950 1000 1050 1100 1150 1200

DEPOSITION TEMPERATURE (°C)

Relative amounts of (111), (110)

and (100) texture in 5 μm thick polycrystalline silicon films (Kamins and Cass, Ref.13).

(c)

(d)

Figure 15. Scanning electron micrographs of aluminum alloy deposits with (a) 12% Cu, (b) 8% Cu, (c) 4% Cu, (d) 0% (a)-(c) 1.7% Si, (d) 2.5% Si (Learn, Ref. 12).

Cu;

Figure 17. Relative corrosion intensity versus weight percent phosphorous after 24 hours at 95°C/100% RH/20 volts (Paulson and Kirk, Ref. 14).

INTRODUCTION

Although low energy ion scattering spectrometry (ISS) has been described by several authors as a surface and in-depth analytical tool1,2,3, characteristics of the method will be illustrated in this paper by using analytical examples relating to the topic of this workshop-Si, Si02, and related materials. After consideration of the technique in general, some fundamental studies of Si and the Si/SiO2 interface will be discussed in which ISS is particularly suited.

METHOD

Figure 1 is a schematic diagram of the scattering process in which rare gas ions of mass M1 and initial energy Eo are scattered by surface atoms of mass M2 in elastic, binary collisions. Although this ion-surface interaction results in the neutralization of about 99.9 percent of the incoming ions, a significant, measurable number survive as scattered ions, and the energy loss suffered by these ions can be related to the mass of the scattering center by the equation given at the bottom of the figure. At a lab scattering angle of 90° this equation reduces to the simple relation

=

E1/E0 (M2 M1)/M2 + M1)

where El is the energy of the scattered ion and the other terms have been defined above. Since M1 and Eo are known quantities and one measures El in the experiment, it is a simple matter to deduce M2. The rare gas ions used are generally in the 1-3 keV energy range and will thus penetrate most solids on the order of 10-60 Å. Survival of these projectiles as scattered ions from a single collision, however, can be restricted almost entirely to the first average atomic layer and thus defines ISS as a true surface technique. Of course, bombarding a solid with such energetic ions also causes sputtering, and for this reason the technique naturally provides information as a function of depth.

To better put this technique in perspective with related methods, Figure 2 contrasts the principles just described with those of high energy backscattering and secondary ion mass spectroscopy (SIMS). High energy ISS uses

MeV projectiles which penetrate thousands to tens of thousands of A into a solid, and the energy loss suffered by the scattered particle (this particle does not have to remain ionized to be energy analyzed at these energies) is due not only to the mass of the scattering center, but its depth as well. It should be noted that this technique does not depend on the sputtering process for obtaining information in depth, and can thus be used very effectively to calibrate difficult, layered samples. SIMS, on the other hand, detects not the scattered ions, but those sputtered particles which escape the solid as ions. The sputtered ions are then extracted into a mass spectrometer and analyzed according to their mass/charge ratio. A point sometimes forgotten when comparing surface techniques is that SIMS data come from the sputtered species, which in the steady state are representative of the bulk sample composition. Low energy ISS and other techniques such as Auger Electron Spectroscopy analyze the surface that remains after sputtering, so that differential sputtering effects must be considered in the interpretation of data.

The instrumentation which in practice detects the low energy scattering event is commercially available from the 3M Company, 5 and is schematically shown in Figure 3. Rare gas ions are formed in a filament ion source and accelerated toward the sample surface at energies between 500 and 3000 eV with a very small energy spread. The ions are focussed to a spot which is variable from about 1-3 mm in diameter (recent enhancements of the ion gun have reduced the beam size to 100 μm). This beam can be kept stationary for highest sensitivity with relatively poor depth resolution or rastered in combination with a small area detection gate for optimum resolution in depth. Ions that scatter from the sample surface at 90° enter a 127° electric sector for energy analysis and are detected with a channel electron multiplier operating in a pulse counting mode. A feature of great importance in this instrumentation is a charge neutralization filament located near the sample surface. Insulators, such as glass or thermal layers of Si02, which can be a problem for techniques employing high density ion or electron beams, are