Secondary Ion Formation

Low primary ion current densities on the order of 1 nA/cm are necessary to eject intact molecules without the sample damage observed for dynamic SIMS which employs current densities 1 iiA/cm. The nature and preparation of the support and solution are also important. For instance. Fig. 23 illustrates the Influence of substrate material upon (M+H)+ and (M-H) emission from glycine. In the sub-monolayer range, no (M+H) ions are ejected on the Cu support presumably due to (M-H) -Cu complex formation. On an inert substrate such as Au, dimer formation occurs between two adsorbed glycine molecules giving rise to (M+H)+-(M-H) surface complexes and hence a similar trend is observed in (M+H)+ and (M-H) emission. Acid-etching the metal substrate. 

Copyright 6 1989 William Andrew Publishing/Noyes Retrieved from www.knovel.com 

Since its introduction into the modem world of chemical analysis methods 1 K. Siegbahn, et al (1), photoelectron spectroscopy has become an increasingly important method for studying semiconductor surfaces. Not only is it widely emplc ed as a surface analytic method but also it finds wide application in chemically characterizing layered structures and interfaces which are important to semiconductor device manufacture. In this tutorial paper, a brief outline of the photoemission experiment will be presented. Modern instrumentation employed in semiconductor characterization will be surveyed and examples will be discussed which demonstrate the power of photoelectron spectroscopy in characterizing semiconductors and semiconductor device structures. 

Photoemission from a solid is a surface event. In figure 2, a photon absorption event is shown. In solid materials, x-rays can be absorbed over a considerable depth (several microns) into the solid surface depending on the photon energy (3). When photoemission occurs, as shown, the electron is freed from an atom in the crystal lattice and moves freely in the conduction band of the material. In most materials, electron-electron interactions readily occur resulting in a loss of electron kinetic 

---

Because of the complex situation on the surface, satisfactory theoretical description of the ionization process leading to secondary ion formation has not yet been possible. Different ionization mechanisms have been proposed ... 

The limitations of SIMS - some inherent in secondary ion formation, some because of the physics of ion beams, and some because of the nature of sputtering - have been mentioned in Sect. 3.1. Sputtering produces predominantly neutral atoms for most of the elements in the periodic table the typical secondary ion yield is between 10 and 10 . This leads to a serious sensitivity limitation when extremely small volumes must be probed, or when high lateral and depth resolution analyses are needed. Another problem arises because the secondary ion yield can vary by many orders of magnitude as a function of surface contamination and matrix composition this hampers quantification. Quantification can also be hampered by interferences from molecules, molecular fragments, and isotopes of other elements with the same mass as the analyte. Very high mass-resolution can reject such interferences but only at the expense of detection sensitivity. 

The advantages of SIMS are its high sensitivity (detection limit of ppms for certain elements), its ability to detect hydrogen and the emission of molecular fragments that often bear tractable relationships with the parent structure on the surface. Disadvantages are that secondary ion formation is a poorly understood phenomenon and that quantification is often difficult. A major drawback is the matrix effect secondary ion yields of one element can vary tremendously with chemical environment. This matrix effect and the elemental sensitivity variation of five orders of magmtude across the periodic table make quantitative interpretation of SIMS spectra oftechmcal catalysts extremely difficult. 

The results presented in this paper focus on a systematic investigation of secondary ion formation from polymeric films prepared from TEOS on hydrophilic silicon wafer surfaces. 

The formation of secondary ions is the most difficult feature in SIMS. Where sputtering is relatively well understood, the process of sputter ionization is not, and a theory that describes the process of secondary ion formation satisfactorily does not yet exist. A number of trends can be rationalized, though. 

As said before, the ionization probability, which accompanies sputtering, is at best qualitatively understood. There have been several attempts to develop models for secondary ion formation. The interested reader may consult the literature for reviews [2,4]. Here we will briefly describe one model that accounts quantitatively for a number of observations on metals, namely the perturbation model of Nprskov and Lundquist [11]. This assumes that the formation of a secondary ion occurs just above the surface, immediately after emission. Then ... 

Clearly an unambiguous examination of the chemical nature of sorbed complexes using SIMS in these measurements is complicated by the presence of manganese and manganese-cobalt containing ions in the spectra. The greater relative ion intensities and the intensity distribution differences for the pH 10 sample compared to the pH 7 material, may arise due to the presence of different surface amine species. Alternatively, the difference may be related to different secondary ion formation processes for sorbed species. 

Sample Preparation. The methods of sample preparation affect the chemical and physical properties of the sample molecules and hence can profoundly influence the secondary ion formation/ emission process. In earlier molecular SIMS studies the samples were prepared by placing a dilute solution of the compound onto an acid-etched Ag foil [87, 88]. The acid etched surface provides, a substrate onto which thin layers of the compound can be deposited from solutions with extended concentration ranges. If on the other hand, the substrate was not etched and the concentration of the solution was too high, the adsorbed molecular film would grow too thick and consequently quench the secondary ion emission. 

The most common species used with SIMS sources are Ar+, 02+, 0 , and N2+. These ions and other permanent gas ions are formed easily with high brightness and stability with the hollow cathode duoplasmatron. Ar+ does not enhance the formation of secondary ions but is popular in static SIMS, in which analysis of the undisturbed surface is the goal and no enhancement is necessary. 02+ and 0 both enhance positive secondary ion count rates by formation of surface oxides that serve to increase and control the work function of the surface. 02+ forms a more intense beam than 0 and thus is used preferentially, except in the case of analyzing insulators (see Chapter 11). In some cases the sample surface is flooded with 02 gas for surface control and secondary ion enhancement. An N2+ beam enhances secondary ion formation, but not as well as 02+. It is very useful for profiling and analysis of oxide films on metals, however. It also is less damaging to duoplasmatron hollow cathodes and extends their life by a factor of 5 or more compared to oxygen. 

Although several ideas have been proposed, the mechanism of secondary ion formation is not at all well established. 

Figure 11 Kinetic emission mechanism of secondary ion formation (from Ref. 10).

Benninghoven, A., Lange, W., Jirikowsky, M., Holtkamp, D. (1982) Investigations on the mechanism of secondary ion formation from organic compounds amino acids. 

Wandass, J.H., Schmitt, R.L., Gardella, J.A. Jr. (1989) Secondary ion formation from Langmuir-Blodgett films studies of positive molecular ions. Appl. Surf Sci, 40, 85-96. 

The surface mass spectrum characterises the surface chemical structure. The spectral intensities can be used to determine the relative surface concentrations of the different surface species. Both positive and negative ion detection modes are possible in SIMS, as in all mass spectrometry techniques. A comparison of the positive and negative ion spectra can often substantially improve the analysis of the results. In SIMS, the charged fraction of the secondary particle flux is very small (10 ). Moreover, the number of sputtered ions per incident primary ion (i.e. the secondary-ion yield) is matrix dependent. With such yield variations direct quantification of surface species based on the number of desorbed secondary ions (i.e. from the SIMS data) is generally impossible [123]. Wucher et al. [143] have recently described a method to determine the secondary ion formation probability, i.e. the ionisation probability of sputtered particles in a direct and quantitative manner. 

Secondary ion formation by impact of primary ions and neutrals... 

That being said, there remains debate surrounding the details of secondary ion formation/survival, which in itself, more than justifies the ongoing research being carried out in this area. 

Figure 3.1 Highly simplified illustration of a two-step process in which atomic secondary ion emission proceeds via knock-on sputtering resulting from keV primary ion impact (left box) followed by secondary ion formation/survival (right box) from the sputtered population. M in the right box can represent any element with the superscript referring to its associated charge and e" referring to an electron. The ejection of molecular ions appears to follow similar albeit more complicated routes.

Reasons for this stem from the fact that the charge transfer process responsible for secondary ion formation/survival, which ever process it may be (see Section 3.3.1 for possibilities), is highly sensitive to the chemistry of the outermost surface of the substrate at the instant or shortly after the sputtering event. Indeed, the information needed to accurately predict the ionization/neutralization processes active would preclude the necessity of analyzing the respective solid. 

Analysis-induced mass fractionation is encapsulated in what is referred to as the Instrument Mass Fractionation (IMF). Because this is matrix dependent, this can only be assessed through comparison of known isotope compositions of some standard reference material with the isotopic composition derived via SIMS analysis of the above-mentioned reference. The fact that a matrix-dependent IMF is observed leads to the realization that examination into the IMF can provide the much-needed information on the fundamentals of the secondary ion formation/survival process. 

The influence of core holes on secondary ion formation/survival is, however, not limited to that described by the Kinetic Enaission model. Indeed, other processes can include, but are not limited to, the formation of anti-bonding states, which can then lead to the stimulated desorption of various atoms, ions, and/or molecules from various surfaces, and excitation of valence states, which can then de-excite back into their original levels or other levels dependent on the substrate in question. 

---