Relative sensitivity factors

Pragmatically, the relative concentrations of elements are determined from the measured ion beam ratios by the application of relative sensitivity factors, which are determined experimentally from standard samples ... 

In principle, therefore, the surface concentration of an element can be calculated from the intensity of a particular photoelectron emission, according to Eq. (2.6). In practice, the method of relative sensitivity factors is in common use. If spectra were recorded from reference samples of pure elements A and B on the same spectrometer and the corresponding line intensities are and respectively, Eq. (2.6) can be written as... 

Thus, as for XP S, the average surface concentration Na can, in principle, be calculated by measurement of the Auger current, according to Eq. (2.13). Again, as in XPS, relative sensitivity factors are generally used. The Auger current for the same transition XYZ in a standard of pure A is measured under the same experimental conditions as in the analysis of A in M, whereupon the ratio of the atomic concentrations is... 

Sa is the relative sensitivity factor. Normally, values of Sa are derived empirically or semi-empirically. Tables of such sensitivity factors have been published by Payling... 

Because measuring A can be problematic, quantification is normally performed by relative sensitivity factor (RSF) methods. If a species A on the surface is detected by the ion X , the ratio of the detected ion current /a(X ) to the primary ion current 7pi and the surface density 9 (A) is called the practical sensitivity factor Sp (X (A)) ... 

The most accurate - and most popular - method of quantifying matrix effects is to analyze the unknown sample with a similar sample of known composition. The relationship between measured intensity and the content of each sample is, usually, defined by the relative sensitivity factor (RSF) ... 

Values of Y (X (A)) for elements typically range from 10 to 10 in NR-laser-SNMS, and from 10 to 10 in R-laser-SNMS. If the experimental conditions are not well known, the concentration of A can also be quantified by using the relative sensitivity factor (RSF) method (Eqs (3.8) and (3.9) in Sect. 3.1.3). 

Tab. 3.1. NR-laser-SNMS Relative sensitivity factors S (Me, ESi) and detection limits DL for metals on Si wafer surfaces.

If relative sensitivity factors are used, reference measurement of standard samples is not necessary. The ratio of two different elemental concentrations in one sample is given by ... 

The ratio Db/Da is a so-called relative sensitivity factor D. This ratio is mostly determined by one element, e. g. the element for insulating samples, silicon, which is one of the main components of glasses. By use of the equation that the sum of the concentrations of all elements is equal to unity, the bulk concentrations can be determined directly from the measured intensities and the known D-factors, if all components of the sample are known. The linearity of the detected intensity and the flux of the sputtered neutrals in IBSCA and SNMS has been demonstrated for silicate glasses [4.253]. For SNMS the lower matrix dependence has been shown for a variety of samples [4.263]. Comparison of normalized SNMS and IBSCA signals for Na and Pb as prominent components of optical glasses shows that a fairly good linear dependence exists (Fig. 4.49). 

Quantified using standards and relative sensitivity factors (RSFs). 

Relative sensitivity factors (RSFs) may be more readily applied if the intensities are not referred to an internal reference standard, when the concentration is given by... 

The indexed relative sensitivity factor approach obviates the necessity of measuring the relative sensitivity factors from all possible matrices, by transferring relative sensitivity factors for elements between different matrices by using the matrix-dependence of characteristic intensity ratios in the spectra. Calibration curves are constructed relating RSFs for an element in a matrix to the matrix ion species ratio (e.g. M2+/M+ for element M) generated from a single standard. 

Secondary ion mass spectrometry (SIMS) was used to characterise the coatings for their Ti, Ru and O stoichiometry on the surface and as a function of depth into the coating. A PHI 6650 Quadrupole mass spectrometer, with Cs+ as the ion source was used in these studies. The conversion of the measured secondary ion counts to concentration was performed using relative sensitivity factors, which were first determined with a standard sample containing known amounts of RuC>2 and TiC>2. All of the SIMS profiles were repeated several times, to determine the measurement precision, which was typically +10%. 

For a given detector and a given pair of elements the last two factors give a single constant (Icab) that can be treated as a relative sensitivity factor. Both that factor and the method obtained their names after the two people who introduced them, Cliff and Lorimer (1975). The simplicity originates from the fact that the Uab factor does not depend either on the rest of elements also present in the sample or on the other parameters of the sample (thickness, density), as far as the thin film criterion is fulfilled. The Cliff-Lorimer factors can either be calculated using the known parameters of the detector or can be measured if a well-characterized thin film sample (standard) is available. In the first case the method is standardless. In the second case the known weight fractions and the measured intensity ratio provides the Cliff-Lorimer factor for the pair of elements. 

Figure 2.31 Relative sensitivity factors of several elements for silicon analysis a) positive secondary ions were measured using Oj primary ion beam (R. C. Wilson, F. A. Stevie and C. W. Magee, Secondary Ion Mass Spectrometry 1989. Reproduced by permission of John Wiley Sons, Inc.)...

MS—solid bronze electrodes made from drillings. Tin content from slurry technique (above) used for internal standard. Rel. sensitivity factors obtained from NBS standards C1115, C1116, C1117, 124, and CA4. 

The energy of the primary electron beam was 2.5 keV and the energy of the Ar+ used in the depth profiles was 2 keV. The relative sensitivity factors were determined using pure standard samples and are presented in Table 14.1. Some samples were studied using an Auger spectrometer equipped with a retarding field analyser. The thickness of the films was... 

Typically, a broad energy (0-1000 eV) survey spectrum was acquired from each sample for elemental detection and then high resolution data for each element were collected to determine the surface chemistry and compositions of different samples. The elemental compositions of different samples were determined from the integrated area intensities of respective photoelectron peaks after normalizing for their relative sensitivity factors [20]. 

U. Relative sensitivity factors for atomic ions out of a given matrix vary by only one, in extreme cases by about two, orders of magnitude at least for the high irradiance techniques (J ). This holds even for ions f only+one polarity, i.e. under suitable conditions ions such as F or Cl are detected in positive ion spectra. This contrasts favourably with e.g. SIMS where sensitivity factors vary by several orders of magnitude. 

Empirical Methods. The empirical methods use calibration standards to derive sensitivity factors that can be used to determine the unknown concentration of given elements in similar matrices [3. The sensitivity factors are derived from calibration curves that plot measured secondary ion intensities versus the known concentration of standards. Three types of sensitivity factors have been used the absolute sensitivity factor, the relative sensitivity factor, and the indexed relative elemental sensitivity factor. 

Newbury [70] conducted comparative SIMS studies of selected glasses ana- steels with laboratories in the U.S., Japan, and Europe. He had each laboratory calculate relative sensitivity factors for several elements under a variety of experimental conditions. These results were astonishing and showed that a given relative sensitivity factor varied from 5 to 60. He also compared the measured concentrations with predicted values from physical models. 

---