Calibration atomic emission spectroscopy

In the inductively coupled plasma atomic emission spectroscopy (ICPAES) method (ASTM DD 5600), a sample of petroleum coke is ashed at 700°C (1292°F) and the ash is fused with lithium borate. The melt is dissolved in dilute hydrochloric acid, and the resulting solution is analyzed by inductively coupled plasma atomic emission spectroscopy using aqueous calibration standards. Because of the need to fuse the ash with lithium borate or other suitable salt, the fusibility of ash may need attention (ASTM D1857). 

M. Glick, K. R. Brushwyler and G. M. Hieftje, Multivariate calibration of a photodiode array spectrometer for atomic emission spectroscopy, Appl. Spectrosc., 45, 1991, 328-333. 

M. L. Griffiths, D. Svozil, P. J. Worsfold, S. Denham and E. H. Evans, Comparison of traditional and multivariate calibration techniques applied to complex matrices using inductively coupled plasma atomic emission spectroscopy, J. Anal. At. Spectrom., 15, 2000, 967-972. 

The analytical application of atomic-absorption or atomic-emission spectroscopy generally involves obtaining the sample in an appropriate solution for measurement and calibrating the instrument properly. Commonly used methods for different materials are described below. Frequently, a releasing agent will have to be added, or a solvent extraction will be required to concentrate the element and increase the sensitivity. Standards should be treated in a similar manner. 

Today, atomic emission spectroscopy always makes use of relative quantitation, i. e. unknown samples are quantitatively analysed after calibration with samples of known composition. The most common approach to calibration is internal standardisation. The underlying assumption, introduced by Gerlach in 1925, is that the ratio of the analyte mass to the mass of the internal standard, matching the analyte in its chemical properties, emission wavelength, energy of the line, and ionisation... 

In atomic emission spectroscopy flames, sparks, and MIPs will have their niche for dedicated apphcations, however the ICP stays the most versatile plasma for multi-element determination. The advances in instrumentation and the analytical methodology make quantitative analysis with ICP-AES rather straightforward once the matrix is understood and background correction and spectral overlap correction protocols are implemented. Modern spectrometer software automatically provides aids to overcome spectral and chemical interference as well as multivariate calibration methods. In this way, ICP-AES has matured in robustness and automation to the point where high throughput analysis can be performed on a routine basis. 

The second approach named laser-induced breakdown spectrometry (LIBS) is based on atomic emission spectroscopy. In this method, a laser is focused on a solid sample and forms a microplasma that emits light characteristics of the elemental composition of the sample. The emitted light is collected, spectrally resolved, and detected to monitor concentrations of elements via their unique spectral signatures. When calibrated, the method can also provide quantitative measurements. 

While a deviation from a straight line calibration is often predictable in principle from physical theory, a quantitative account is usually lacking there is no known reason why a true calibration graph should be a quadratic function or higher order polynomial. Indeed they are often of a somewhat different shape. This leads to a degree of lack of fit between the true function and the fitted function. Figure 2.23 shows an example, where a quadratic function has been fitted to closely spaced data of the slightly different shape typical of inductively coupled plasma atomic emission spectroscopy (ICP-AES) calibrations. 

In Inductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES), a gaseous, solid (as fine particles), or liquid (as an aerosol) sample is directed into the center of a gaseous plasma. The sample is vaporized, atomized, and partially ionized in the plasma. Atoms and ions are excited and emit light at characteristic wavelengths in the ultraviolet or visible region of the spectrum. The emission line intensities are proportional to the concentration of each element in the sample. A grating spectrometer is used for either simultaneous or sequential multielement analysis. The concentration of each element is determined from measured intensities via calibration with standards. 

In emission spectroscopy the molecule or atom itself serves as the somce of light with discrete frequencies to be analyzed. In some cases, such as Exp. 39, which deals with the emission spectrum of molecular iodine vapor, excitation by a monochromatic or nearly monochromatic laser or mercury lamp is utilized. For other cases, such as the emission from N2 molecules, electron excitation of nitrogen in a discharge tube provides an intense somce whose spectrum is analyzed to extract information about the electronic and vibrational levels. Such low-pressure (p < 10 Torr) line somces are available with many elements, and lamps containing Hg, Ne, Ar, Kr, and Xe are often used for calibration purposes. The Pen-Ray pencil-type lamp is especially convenient for the visible and... 

The method of standard additions is widely used in atomic spectroscopy (e.g. determination of Ca2+ ions in serum by atomic emission spectrophotometry) and, since several aliquots of sample are analysed to produce the calibration graph, should increase the accuracy and precision of the assay... 

Because the atomic fluorescence is measured at a right angle to the source, spectral interferences are minimal and a simple cutoff filter may often be used to isolate the emission line. The intensity of the fluorescence is directly proportional to the analyte concentration. As the analyte concentration within the flame becomes large, self-absorption of resonance fluorescence becomes significant, as it does in flame emission spectroscopy. Under these conditions, the linearity of the instrumental response breaks down and a calibration curve must be used or the analyte solutions diluted accordingly. 

The predominant method for the analysis of aluminum-base alloys is spark source emission spectroscopy. Solid metal samples are sparked directly, simultaneously eroding the metal surface, vaporizing the metal, and exciting the atomic vapor to emit light in proportion to the amount of material present. Standard spark emission analytical techniques are described in ASTM ElOl, E607, E1251 and E716 (36). A wide variety of well-characterized solid reference materials are available from major aluminum producers for instrument calibration. 

FIGURE 9-12. Flame emission calibration curves for magnesium (2852A) (A) 0-10, (B) 0-50, (C) 0-100 /.ig/vn. [From W. G. Schrenk, in Flame Emission and Atomic Absorption Spectroscopy, Vol. 2, Edited by J. A. Dean and T. C. Rains, Marcel Dekker, New York, (1971), Chapter 12. 

Reference should be made to Section 7 in Chapter 9, since many of the techniques described there also are applicable to atomic absorption spectroscopy. As in flame emission, a variety of read-out devices are used for atomic absorption. These include visual meters, chart recorders, digital read-out, devices that convert data to absorbance units, and even digital devices that can be calibrated to present analytical data directly in concentration units. The method used to process the read-out data depends on the nature of the read-out signal. In the case of direct read-out in concentration units the data require no further processing. Such data may be fed to the input of a computer for a direct printing of analytical results. 

In flame emission spectroscopy, the concentration of electronically excited atoms in the cooler, outer part of the flame is lower than in the warmer, central part of the flame. Emission from the central region is absorbed in the outer region. This selfabsorption increases with increasing concentration of analyte and gives nonlinear calibration curves. In a plasma, the temperature is more uniform, and self-absorption is not nearly so important. Table 20-4 states that plasma emission calibration curves are linear over five orders of magnitude compared with just two orders of magnitude for flames and furnaces. 

---