Graphite detection limits

Atomic absorption spectroscopy is more suited to samples where the number of metals is small, because it is essentially a single-element technique. The conventional air—acetylene flame is used for most metals however, elements that form refractory compounds, eg, Al, Si, V, etc, require the hotter nitrous oxide—acetylene flame. The use of a graphite furnace provides detection limits much lower than either of the flames. A cold-vapor-generation technique combined with atomic absorption is considered the most suitable method for mercury analysis (34). 

Atomic absorption spectroscopy is an alternative to the colorimetric method. Arsine is stiU generated but is purged into a heated open-end tube furnace or an argon—hydrogen flame for atomi2ation of the arsenic and measurement. Arsenic can also be measured by direct sample injection into the graphite furnace. The detection limit with the air—acetylene flame is too high to be useful for most water analysis. 

Electrothermal vaporization can be used for 5-100 )iL sample solution volumes or for small amounts of some solids. A graphite furnace similar to those used for graphite-furnace atomic absorption spectrometry can be used to vaporize the sample. Other devices including boats, ribbons, rods, and filaments, also can be used. The chosen device is heated in a series of steps to temperatures as high as 3000 K to produce a dry vapor and an aerosol, which are transported into the center of the plasma. A transient signal is produced due to matrix and element-dependent volatilization, so the detection system must be capable of time resolution better than 0.25 s. Concentration detection limits are typically 1-2 orders of magnitude better than those obtained via nebulization. Mass detection limits are typically in the range of tens of pg to ng, with a precision of 10% to 15%. 

The presence of redox catalysts in the electrode coatings is not essential in the c s cited alx)ve because the entrapped redox species are of sufficient quantity to provide redox conductivity. However, the presence of an additional redox catalyst may be useful to support redox conductivity or when specific chemical redox catalysis is used. An excellent example of the latter is an analytical electrode for the low level detection of alkylating agents using a vitamin 8,2 epoxy polymer on basal plane pyrolytic graphite The preconcentration step involves irreversible oxidative addition of R-X to the Co complex (see Scheme 8, Sect. 4.4). The detection by reductive voltammetry, in a two electron step, releases R that can be protonated in the medium. Simultaneously the original Co complex is restored and the electrode can be re-used. Reproducible relations between preconcentration times as well as R-X concentrations in the test solutions and voltammetric peak currents were established. The detection limit for methyl iodide is in the submicromolar range. 

SSMS can be classified among the milliprobe techniques (Figure 8.3), i.e. it is a unique link between microprobe techniques and macroanalytical methods that are characterised by poor lateral and in-depth resolutions (as in OES), or that have no lateral resolution whatsoever (as in NAA). Also, the achievable precision and accuracy are poor, because of the irreproducible behaviour of the r.f. spark. Whereas analysis of metals, semiconductors and minerals is relatively simple and the procedures have become standardised, the analysis of nonconducting materials is more complex and generally requires addition of a conducting powder (e.g. graphite) to the sample [359]. Detection limits are affected by the dilution, and trace contamination from the added components is possible. These problems can be overcome by the use of lasers [360]. Coupled with isotope dilution, a precision of 5% can be attained for SSMS. 

Wakabayashi et al. [51] determined penicillamine in serum by HPLC. Serum (0.1 mL) was vortex-mixed for 30 s with 50 pL of 0.1% EDTA and 0.2 mL of 10% TCA. The solution was centrifuged at 1500 x g and filtered. A 5 pL portion was analyzed on a Shodex C18 column (15 cm x 4.6 mm i.d.), using a mobile phase of 19 1 methanolic 0.05 M phosphate buffer (pH 2.8) containing 1 mM sodium octylsulfate and 10 pM EDTA. Liver or kidney samples were similarly extracted, and the extracts were cleaned up on a Bond-Elut cartridge prior to HPLC analysis. Detection was effected with an Eicom WE-3G graphite electrode maintained at +0.9 V versus Ag/AgCl. The calibration graph was linear up to 500 ng, and the detection limits were 20 pg. For 1 pg of penicillamine added to serum, liver, or kidney, the respective relative standard deviations (n = 5) were 3.6, 5.1, and 4.4%. 

Sturgeon et al. [59] have described a hydride generation atomic absorption spectrometry method for the determination of antimony in seawater. The method uses formation of stibene using sodium borohydride. Stibine gas was trapped on the surface of a pyrolytic graphite coated tube at 250 °C and antimony determined by atomic absorption spectrometry. An absolute detection limit of 0.2 ng was obtained and a concentration detection limit of 0.04 pg/1 obtained for 5 ml sample volumes. 

Bishop [75] determined barium in seawater by direct injection Zeeman-modulated graphite furnace atomic absorption spectrometry. The V203/Si modifier added to undiluted seawater samples promotes injection, sample drying, graphite tube life, and the elimination of most seawater components in a slow char at 1150-1200 °C. Atomisation is at 2600 °C. Detection is at 553.6 nm and calibration is by peak area. Sensitivity is 0.8 absorbance s/ng (Mo = 5.6 pg 0.0044 absorbance s) at an internal argon flow of 60 ml/min. The detection limit is 2.5 pg barium in a 25 ml sample or 0.5 pg using a 135 ml sample. Precision is 1.2% and accuracy is 23% for natural seawater (5.6-28 xg/l). The method works well in organic-rich seawater matrices and sediment porewaters. 

Guevremont et al. [117] studied the use of various matrix modifiers in the graphite furnace gas method of determination of cadmium in seawater. These included citric acid, lactic acid, aspartic acid, histidine, and EDTA. The addition of less than 1 mg of any of the compounds to 1 ml seawater significantly decreased matrix interference. Citric acid achieved the highest sensitivity and reduction of interference, with a detection limit of 0.01 pg cadmium per litre. 

Pruszkowska et al. [135] described a simple and direct method for the determination of cadmium in coastal water utilizing a platform graphite furnace and Zeeman background correction. The furnace conditions are summarised in Table 5.1. These workers obtained a detection limit of 0.013 pg/1 in 12 pi samples, or about 0.16 pg cadmium in the coastal seawater sample. The characteristic integrated amount was 0.35 pg cadmium per 0.0044 A s. A matrix modifier containing di-ammonium hydrogen phosphate and nitric acid was used. Concentrations of cadmium in coastal seawater were calculated directly from a calibration curve. Standards contained sodium chloride and the same matrix modifier as the samples. No interference from the matrix was observed. 

The chemiluminescence technique has been used to determine trivalent chromium in seawater. Chang et al. [187] showed Luminol techniques for determination of chromium (III) were hampered by a salt interference, mainly due to magnesium ions. Elimination of this interference is achieved by seawater dilution and utilising bromide ion chemiluminescence signal enhancement (Fig. 5.7). The chemiluminescence results were comparable with those obtained by a graphite furnace flameless atomic absorption analysis for the total chromium present in samples. The detection limit is 3.3 x 10 9 mol/1 (0.2 ppb) for seawater with a salinity of 35%, with 0.5 M bromide enhancement. 

A Cis column loaded with sodium diethyldithiocarbamate has been used to extract copper and cadmium from seawater. Detection limits for analysis by graphite furnace atomic absorption spectrometry were 0.024 pg/1 and 0.004 xg/l, respectively [283]. 

Bolshov et al. [358] used this technique to determine low lead concentrations. A detection limit of 0.05 pg ml 1 was achieved in studies with aqueous solutions as reference using a graphite atomiser. 

In contrast, the coupling of electrochemical and spectroscopic techniques, e.g., electrodeposition of a metal followed by detection by atomic absorption spectrometry, has received limited attention. Wire filaments, graphite rods, pyrolytic graphite tubes, and hanging drop mercury electrodes have been tested [383-394] for electrochemical preconcentration of the analyte to be determined by atomic absorption spectroscopy. However, these ex situ preconcentration methods are often characterised by unavoidable irreproducibility, contaminations arising from handling of the support, and detection limits unsuitable for lead detection at sub-ppb levels. 

In this method, inorganic lead in seawater samples are converted to tetra-ethylead using sodium tetraethylboron (NaB(C2H5)4) which is then trapped in a graphite furnace at 400 °C. Quantitation is achieved by using a simple calibration graph prepared from aqueous standards. An absolute detection limit of (3relative standard deviation. 

Graphite furnace atomic absorption spectrometry has also been used to determine zinc [610,611] in seawater with a detection limit of 2 ig [611]. Guevre-mont [610] has discussed the use of organic matrix modifiers for the direct determination of zinc. 

Cadmium, copper, and silver have been determined by an ammonium pyrrolidine dithiocarbamate chelation, followed by a methyl isobutyl ketone extraction of the metal chelate from the aqueous phase [677], and finally followed by graphite furnace atomic absorption spectrometry. The detection limits of this technique for 1% absorption were 0.03 pmol/1 (copper), 2 nmol/1 (cadmium), and 2 nmol/1 (silver). 

Molybdenum is an element for which platform atomisation does not offer an advantage. Just the opposite is the case sensitivity is very poor and memory effects are very strong. The Zeeman detection limit for wall atomisation in a pyrocoated graphite tube using 100 til of reference solution is 0.03 xl (for both peak height and peak area evaluation) [709]. 

---