RESUMEN

The chemical composition of complex inorganic materials, such as copper concentrate, may influence the economics of their further processing because most smelters, and particularly the producers of high-purity electrolyte copper, have strict limitations for the permissible concentration of impurities. These components might be harmful to the quality of the products, impair the production process and be hazardous to the environment. The goal of the present work is the development of a method for the determination of fluorine in copper concentrate using high-resolution graphite furnace molecular absorption spectrometry and direct solid sample analysis. The molecular absorption of the diatomic molecule CaF was measured at 606.440nm. The molecule CaF was generated by the addition of 200µg Ca as the molecule-forming reagent; the optimized pyrolysis and vaporization temperatures were 900°C and 2400°C, respectively. The characteristic mass and limit of detection were 0.5ng and 3ng, respectively. Calibration curves were established using aqueous standard solutions containing the major components Cu, Fe, S and the minor component Ag in optimized concentrations. The accuracy of the method was verified using certified reference materials. Fourteen copper concentrate samples from Chile and Australia were analyzed to confirm the applicability of the method to real samples; the concentration of fluorine ranged from 34 to 5676mgkg-1. The samples were also analyzed independently at Analytik Jena by different operators, using the same equipment, but different target molecules, InF and GaF, and different operating conditions; but with a few exceptions, the results agreed quite well. The results obtained at Analytik Jena using the GaF molecule and our results obtained with CaF, with one exception, were also in agreement with the values informed by the supplier of the samples, which were obtained using ion selective electrode potentiometry after alkaline fusion. A comparison will also be made for the three target molecules and the three independently developed methods for the determination of fluorine, although all three methods used direct solid sample analysis.

RESUMEN

A method has been developed for the determination of lead in biomass, bio-oil, pyrolysis aqueous phase, and biomass ashes by high-resolution continuum source graphite furnace atomic absorption spectrometry (HR-CS GF AAS) and direct solid or liquid sample analysis. All measurements were performed without chemical modifier and calibration could be carried out using aqueous standard solutions. A pyrolysis temperature of 800°C and an atomization temperature of 2200°C were applied. The limits of detection and quantification were, respectively, 0.5 µg kg(-1) and 2 µg kg(-1) using the analytical line at 217.001 nm and 6 µg kg(-1) and 19 µg kg(-1) at 283.306 nm. The precision, expressed as relative standard deviation, was between 3% and 10%, which is suitable for direct analysis. The lead concentrations found for the solid samples varied between 0.28 and 1.4 mg kg(-1) for biomass and between 0.25 and 2.3 mg kg(-1) for ashes, these values were much higher than those found for bio-oil (2.2-16.8 µg kg(-1)) and pyrolysis aqueous phase (3.2-18.5 µg kg(-1)). After the determination of lead in the samples, it was possible to estimate the relative distribution of this element in the fractions of the pyrolysis products, and it was observed that most of the lead present in the biomass was eliminated to the environment during the pyrolysis process, with a significant portion retained in the ashes.

Asunto(s)

RESUMEN

High-resolution continuum source graphite furnace atomic absorption spectrometry, because of the use of only one radiation source for all elements, offers the possibility of sequential determination of two or more elements from the same sample aliquot if their volatilities are significantly different. Cd and Cr were determined sequentially in samples of biomass and biomass ashes employing direct solid sample analysis. The use of a chemical modifier was found to be not necessary, and calibration could be carried out using aqueous standard solutions. A pyrolysis temperature of 400°C and an atomization temperature of 1500°C were used for the determination of Cd; no losses of Cr were observed at this temperature. After the atomization of Cd the wavelength was changed and Cr atomized at 2600°C. The limits of detection (LOD) and quantification (LOQ) were 1.1 µg kg(-1) and 3.7 µg kg(-1), respectively, for Cd and 21 µg kg(-1) and 70 µg kg(-1), respectively, for Cr using the most sensitive line at 357.869 nm, or 90 µg kg(-1) and 300 µg kg(-1), respectively, using the less sensitive line at 428.972 nm. The precision, expressed as relative standard deviation was around 10%, which is typical for direct solid sample analysis. The values found for Cd in biomass samples were between <1.1 µg kg(-1) and 789 µg kg(-1), whereas those for Cr were between 7.9 mg kg(-1) and 89 mg kg(-1); the values found in the ashes were significantly lower for Cd, between <1.1 µg kg(-1) and 6.3 µg kg(-1), whereas the trend was not so clear for Cr, where the values were between 3.4 mg kg(-1) and 28 mg kg(-1).

Asunto(s)

RESUMEN

A method has been developed to determine 10 elements in Brazilian red wines using high-resolution continuum source flame atomic absorption spectrometry, a technique that allows the fast sequential determination of an essentially unlimited number of elements per sample, each one under previously optimized conditions. All measurements were made without sample preparation, using aqueous standard solutions for calibration. The results were in agreement within 99% of confidence (t-test) with those obtained by inductively coupled plasma optical emission spectrometry. The same grape, Cabernet sauvignon, was used in all experiments, and the wines from each region were prepared especially for this investigation in order to avoid any confusion due to grapes from other regions, which are often used in commercial wines. The elements K, Mn, Rb and Sr were found to be the best indicators for the origin of the wines, based on a Principal Component Analysis.

Asunto(s)

RESUMEN

A simple method has been developed to determine antimony and lead in pewter alloy cups produced in Brazil, using fast sequential determination by high-resolution continuum source flame atomic absorption spectrometry. The samples were dissolved in HCl and H(2)O(2), employing a cold finger system in order to avoid analyte losses. The main resonance line of lead at 217.001 nm and a secondary line of antimony at 212.739 nm were used. The limits of detection for lead and antimony were 0.02 and 5.7 mg L(-1), respectively. The trueness of the method was established by recovery tests and comparing the results obtained by the proposed method with those obtained by inductively coupled plasma optical emission spectrometry. The results were compared using a student's t-test and there was no significant difference at a 95% confidence interval. With the developed methods, it was possible to determine accurately antimony and lead in pewter samples. The lead concentration found in the analysed samples was around 1 mg g(-1), which means that they are not lead free; however, the content was below the maximum allowed level of 5 mg g(-1). The antimony content, which was found to be between 40 and 46 mg g(-1), is actually of greater concern, as antimony is known to be potentially toxic already at very low concentrations, although there is no legislation yet for this element.

Asunto(s)

RESUMEN

Three different procedures for sample preparation have been compared for the determination of Cu, Mo and Sb in airborne particulate matter (APM) collected on glass fiber filters using high-resolution continuum source graphite furnace atomic absorption spectrometry (HR-CS GF AAS). Direct solid sample analysis of the ground filters was compared with microwave-assisted acid leaching with aqua regia and ultrasound-assisted extraction also using aqua regia. The main absorption line at 324.754 nm or the secondary line at 216.509 nm was used for the determination of Cu, depending on the analyte content in the samples. The primary absorption line at 313.259 nm was used for Mo and the secondary line at 212.739 nm for Sb determination. The limits of detection (LOD, 3σ) found for the direct solid sampling method, based on ten atomizations of an unused filter were 15 µg g(-1) for all three analytes, corresponding to 40 ng m(-3) for a typical air volume of 1,440 m(3) collected over a period of 24h. The LOD for the other two methods were less than a factor of two inferior, but the total time required for an analysis was significantly longer. The repeatability of the measurements was between 3 and 9% (n=5), and the results obtained with the three methods did not show any significant difference. The ratio between the three analytes on the filters from areas of intense traffic was found to be around Cu:Mo:Sb≈4:1:1.4, which suggests that the source of all three elements is brake linings, i.e., related to automobile traffic. When the ratio deviated significantly from the above values, the source of contamination was assumed to be of different origin.

Asunto(s)

RESUMEN

Antimony is one of the constituents of pewter, an alloy composed of a minimum of 90% tin with the balance being made up with copper, antimony and perhaps some bismuth. A method has been developed to determine Sb in acetic acid leachates from pewter cups. The employed instrumentation, an atomic absorption spectrometer, equipped with a quartz trap-and-atomizer device, is simple and relatively inexpensive with low running costs. Interferences due to the presence of tin and ways to control them were investigated in detail. The applied approach made possible to overcome potentially serious interference of Sn leached from the cup material (which was shown to take place in the atomizer), by a combination of (i) high concentration of HCl, which decreases the efficiency of stannane generation and (ii) in-atomizer collection. The resulting Sn tolerance limit was between 10 and 20 mg L(-1). The advantages of the in-atomizer collection are a lower tin interference in the atomizer, and a much better limit of detection (LOD), which makes possible reducing the atomization interference further by working with more diluted sample solutions. Besides the Sn interference, an interference of an unknown volatile compound transported to the atomizer together with stibine was identified in the measured sample solutions. This interference could be controlled using the analyte addition technique. The applicability of the method was tested on solutions containing a wide range of interferents leached from the pewter cups, obtained at leaching times between 1 and 24h. The LOD in the sample solutions was found to be 0.03 µg L(-1) Sb.

Asunto(s)

RESUMEN

Two simple methods have been developed to determine cadmium and lead in different kinds of beverages and vinegar leached from pewter cups produced in Brazil. Leaching experiments have been carried out with different solutions: beer, sugar cane spirit, red and white wine, vinegar and a 3% acetic acid solution. The solutions were kept in cups with and without solder for 24h. Lead and cadmium have been determined using graphite furnace atomic absorption spectrometry with deuterium background correction. The limits of detection were 0.05 and 1.4 µg L(-1), and the characteristic mass was 1.0 pg and 19 pg for Cd and Pb, respectively. With the developed methods it was possible to determine accurately cadmium and lead by direct analysis in these liquids and to evaluate the leaching of these metals from pewter cups. The results presented in this work show that pewter cups are not cadmium- and lead-free; this point goes against the manufacturers' declaration that their products are lead-free.

Asunto(s)

RESUMEN

A fast routine screening method for the simultaneous determination of cadmium and iron in bean and soil samples is proposed, using high-resolution continuum source graphite furnace atomic absorption spectrometry and direct solid sampling. The primary absorption line at 228.802 nm has been used for the determination of cadmium, and an adjacent secondary line, at 228.726 nm, for iron. Fourteen bean samples and 10 soil samples from nine states all over Brazil have been analyzed. The limits of detection (3 sigma, n = 10) were 2.0 microg kg(-1) for Cd and 4.5 mg kg(-1) for Fe. The relative standard deviation ranged from 4 to 7% for Cd and from 5 to 28% for Fe, which is usually acceptable for a screening method. The accuracy of the method has been confirmed by the analysis of two certified reference materials; the results were in agreement with the certified values at a 95% confidence interval.

Asunto(s)

RESUMEN

Cadmium and iron are antagonistic elements in the sense that they produce different effects in the human body. Both elements have to be determined routinely in grain products, cadmium because of its toxicity, and iron because all grain products, according to Brazilian law, have to contain a minimum of 42 mg kg(-1) Fe to combat anemia. A routine screening method has been developed for the quasi simultaneous determination of cadmium and iron using high-resolution continuum source electrothermal atomic absorption spectrometry and direct solid sampling. The primary absorption line at 228.802 nm has been used for Cd, and an adjacent secondary line at 228.726 nm for the determination of Fe. Various chemical modifiers have been investigated, and a mixture of tungsten and iridium, applied as a permanent modifier, showed the best performance; it stabilized Cd up to a pyrolysis temperature of 700 degrees C and did not over-stabilize Fe. Two atomization temperatures were used sequentially, 1700 degrees C for Cd and 2600 degrees C for Fe, because of their significantly different volatilities. The characteristic masses obtained were 0.9 pg for Cd and 1.2 ng for Fe. The limits of detection (3 sigma, n=10) were 0.6 microg kg(-1) for Cd and 0.5 mg kg(-1) for Fe. The relative standard deviation ranged from 3 to 7% for Cd and from 4 to 13% for Fe, which is satisfactory for the purpose. The accuracy of the method was confirmed by the analysis of three certified reference materials; the results were in agreement with the certified values at a 95% confidence interval. The Cd content in the investigated grain products was between 0.9 and 10.5 microg kg(-1), but most of them did not contain the required minimum amount of iron.

Asunto(s)

RESUMEN

Two analytical methods for the determination of cadmium in wheat flour by electrothermal atomic absorption spectrometry without prior sample digestion have been compared: direct solid sampling analysis (SS) and slurry sampling (SlS). Besides the conventional modifier mixture of palladium and magnesium nitrates (10 microg Pd+3 microg Mg), 0.05% (v/v) Triton X-100 has been added to improve the penetration of the modifier solution into the solid sample, and 0.1% H(2)O(2) in order to promote an in situ digestion for SS. For SlS, 30 microg Pd, 12 microg Mg and 0.05% (v/v) Triton X-100 have been used as the modifier mixture. Under these conditions, and using a pyrolysis temperature of 800 degrees C, essentially no background absorption was observed with an atomization temperature of 1600 degrees C. About 2 mg of sample have been typically used for SS, although as much as 3-5 mg could have been introduced. In the case of SlS multiple injections had to be used to achieve the sensitivity required for this determination. Calibration against aqueous standards was feasible for both methods. The characteristic mass obtained with SS was 0.6 pg, and that with SlS was 1.0 pg. The limits of detection were 0.4 and 0.7 ng g(-1), the limits of quantification were 1.3 and 2.3 ng g(-1) and the relative standard deviation (n=5) was 6-16% and 9-23% for SS and SlS, respectively. The accuracy was confirmed by the analysis of certified reference materials. The two methods were applied for the determination of cadmium in six wheat flour samples acquired in supermarkets of different Brazilian cities. The cadmium content varied between 8.9+/-0.5 and 13+/-2 ng g(-1) (n=5). Direct SS gave results similar to those obtained with SlS using multi-injections; the values of both techniques showed no statistically significant difference at the 95% confidence level. Direct SS was finally adopted as the method of choice, due to its greater simplicity, the faster speed of analysis and the better figures of merit.

Asunto(s)

RESUMEN

A procedure is proposed for the direct determination of manganese and iron in wine employing fast sequential flame atomic absorption spectrometry and the reference element technique to correct for matrix effects. Cobalt, silver, nickel and indium have been tested as reference elements. The results demonstrated that cobalt and indium at a concentration of 2 and 10mgL(-1) were efficient for quantification of manganese and iron, respectively. Under these conditions, manganese and iron could be determined with quantification limits of 27 and 40microg L(-1), respectively. The proposed method was applied to the determination of manganese and iron in 16 wine samples. The content of manganese varied from 0.78 to 2.89mgL(-1) and that of iron from 0.88 to 9.22mgL(-1). The analytical results were compared with those obtained by inductively coupled plasma optical emission spectrometry after complete mineralization using acid digestion. The statistical comparison by a t-test (95% confidence level) showed no significant difference between the results.

Asunto(s)

RESUMEN

A method has been developed for the determination of lead in wine by electrothermal atomic absorption spectrometry without any sample preparation and calibration against aqueous standards, using 7.5 microg Pd as a chemical modifier. The results obtained for seven wines using the proposed method and an acid digestion procedure did not show any significant difference using a Student's t-test. Atomization in a transversally heated filter atomizer (THFA) was compared with atomization in a conventional transversally heated platform furnace. The former provided a 2.6-fold higher sensitivity, improving the characteristic mass from 34 to 12 pg and a 1.6-fold better limit of detection (0.3 microg L(-1) compared to 0.5 microg L(-1)) for aqueous solutions using the same injection volume of 20 microL. However, the average precision, expressed as the relative standard deviation for the determination of lead in wine under routine conditions was improved from 4.6% with platform atomization to 0.6% in the THFA. The lead content found in seven arbitrarily chosen white and red wines, five from Brazil, one from Chile and one from Spain, ranged from 6 to 60 microg L(-1) Pb with an average content of 11.4 microg L(-1) Pb for the wines from South America.

Asunto(s)

RESUMEN

A new method for F AAS determination of sodium and potassium in biodiesel using water-in-oil microemulsion as sample preparation is proposed. The method was investigated for biodiesel produced from different sources, as soybean, castor and sunflower oil and animal fat and was also applied for vegetable oils. The optimized condition for microemulsion formation was 57.6% (w/w) of n-pentanol, 20% (w/w) of biodiesel or vegetable oil, 14.4% (w/w) of Triton X-100 and 8% (w/w) of water (aqueous standard of KCl or NaCl in/or diluted HNO(3)). The optimized instrumental parameters were: aspiration rate of 2 mL min(-1) and the flame composition of 0.131 of C(2)H(2)/air ratio. For comparison purpose, the determination of sodium and potassium were also carried out according to European norms (EN 14108 and EN 14109, respectively). These norms are applied for determination of sodium and potassium in fatty acid methylic ester samples and consist in the sample dilution using organic solvent and determination by F AAS. The stability of microemulsified aqueous standards and samples was investigated and it was found to be stable for at least 3 days while the organic standard diluted with xylene showed a decrease around of 15% in the analytical signal in 1h. The limits of detection were 0.1 microg g(-1) and 0.06 microg g(-1) and the obtained characteristic concentrations were 25 microg L(-1) and 28 microg L(-1) for sodium and potassium, respectively. The proposed method presented two times better limits of detection and better precision (0.4-1.0%) when compared with the dilution technique (1.5-4.5%). The accuracy of the method was evaluated through recovery tests and comparison with the results obtained by dilution technique. The recoveries ranged from 95% to 115% for biodiesel and 90% to 115% for vegetable oil samples. Comparison between the results obtained for biodiesel by both methods showed no significant differences at the 95% confidence level according to a Student's t-test. This study shows that the proposed method based on microemulsion as sample preparation can be applied as an efficient alternative for sodium and potassium determination in biodiesel samples.

Asunto(s)

RESUMEN

In recent work, it has been shown that electrothermal atomic absorption spectrometry (ET AAS) can be used to differentiate between volatile and non-volatile nickel and vanadium compounds in crude oil. In the present work, the distribution of these two groups of compounds over different fractions of crude oil was investigated. For this purpose two crude oil samples were separated in two steps: firstly, the asphaltenes were precipitated with n-heptane, and secondly, the maltenes were loaded on a silica column and eluted with solvents of increasing polarity. The four fractions of maltenes eluted from silica column were: F1, saturated and light aromatics; F2, polyaromatics; F3, resins; and F4, polar compounds. Fractions F1 and F2 were further investigated using gas chromatography, and all fractions were characterized by CHN analysis, confirming the increase of aromatics in the fractions 2, 3, 4 and asphaltenes. For the determination of Ni and V by ET AAS, oil-in-water emulsions were prepared. The speciation analysis was carried out measuring without chemical modifier (stable compounds) and with 20 microg palladium (total Ni and V) and the volatile fraction was calculated by difference. The limits of detection were 0.02 microg g(-1) and 0.06 microg g(-1), for Ni and V, respectively, based on an emulsion of 2g of oil in 10 mL. The volatile species of Ni and V were associated with fractions F3 and F4, while only thermally stable Ni and V was precipitated in part together with the asphaltenes.

Asunto(s)

RESUMEN

The literature about direct solid sample analysis of the past 10-15 years using electrothermal atomic absorption spectrometry has been reviewed. It was found that in the vast majority of publications aqueous standards were reported as having been used for calibration after careful program optimization. This means the frequently expressed claim that certified reference materials with a matrix composition and analyte content close to that of the sample have to be used for calibration in solid sample analysis is not confirmed in the more recent literature. There are obviously limitations, and there are examples in the literature where even calibration with certified reference materials did not lead to accurate results. In these cases the problem is typically associated with spectral interferences that cannot be corrected properly by the systems available for conventional line source atomic absorption spectrometry, including Zeeman-effect background correction. Using high-resolution continuum source atomic absorption spectrometry, spectral interferences become visible owing to the display of the spectral environment at both sides of the analytical line at high resolution, which makes program optimization straightforward. Any spectrally continuous background absorption is eliminated automatically, and even rapidly changing background absorption does not cause any artifacts, as measurement and correction of background absorption are truly simultaneous. Any kind of fine-structured background can be eliminated by "subtracting" reference spectra using a least-squares algorithm. Aqueous standards are used for calibration in all published applications of high-resolution continuum source atomic absorption spectrometry to direct solid sample analysis.

RESUMEN

A method has been developed for the determination of cobalt, copper and manganese in green coffee using direct solid sampling electrothermal atomic absorption spectrometry (SS-ET AAS). The motivation for the study was that only a few elements might be suitable to determine the origin of green coffee so that the multi-element techniques usually applied for this purpose might not be necessary. The three elements have been chosen as test elements as they were found to be significant in previous investigations. A number of botanical certified reference materials (CRM) and pre-analyzed samples of green coffee have been used for method validation, and inductively coupled plasma optical emission spectrometry (ICP OES) after microwave-assisted acid digestion of the samples as reference method. Calibration against aqueous standards could be used for the determination of Mn and Co by SS-ET AAS, but calibration against solid CRM was necessary for the determination of Cu. No significant difference was found between the results obtained with the proposed method and certified or independently determined values. The limits of detection for Mn, Cu and Co were 0.012, 0.006 and 0.004mugg(-1) using SS-ET AAS and 0.015, 0.13 and 0.10mugg(-1) using ICP OES. Seven samples of Brazilian green coffee have been analyzed, and there was no significant difference between the values obtained with SS-ET AAS and ICP OES for Mn and Cu. ICP OES could not be used as a reference method for Co, as essentially all values were below the limit of quantification of this technique.

RESUMEN

A method for the direct determination of volatile and non-volatile nickel and vanadium compounds in crude oil without previous treatment using direct solid sampling graphite furnace atomic absorption spectrometry is proposed. The crude oil samples were weighed directly onto solid sampling platforms using a microbalance and introduced into a transversely heated solid sampling graphite tube. In previous work of our group losses of volatile nickel and vanadium compounds have been detected, whereas other nickel and vanadium compounds were thermally stable up to 1300 and 1600 degrees C, respectively. In order to avoid this problem different chemical modifiers (conventional and permanent) have been investigated. With 400microg of iridium as permanent modifier, the signal started to drop already after two atomization cycles, possibly because of an interaction of nickel (which is a catalyst poison) with iridium. Twenty micrograms of palladium applied in each determination was found to be optimum for both elements. The palladium was deposited on the platform and submitted to a drying step at 150 degrees C for 75s. After that the sample was added onto the platform and submitted to the furnace program. The influence of sample mass on the linearity of the response and on potential measurement errors was also investigated using four samples with different nickel content. For the sample with the lowest nickel concentration the relationship between mass and integrated absorbance was found to be non-linear when a high sample mass was introduced. It was suspected that the modifier had not covered the entire platform surface, which resulted in analyte losses. This problem could be avoided by using 40microL of 0.5g L(-1) Pd with 0.05% Triton X-100. Calibration curves were established with and without modifier, with aqueous standards, oil-in-water emulsions and the certified reference material NIST SRM 1634c (trace metals in residual fuel oil). The sensitivity for aqueous standards and emulsions was close to that for SRM 1634c, making possible the use of aqueous standards for calibration. The limits of detection and quantification obtained for nickel and vanadium under this condition were found to be 0.02 and 0.06microg g(-1), respectively, for both elements, based on 10mg of sample. Nickel and vanadium were determined in the samples with (total Ni and V) and without the use of Pd (thermally stable compounds), and the concentration of volatile compounds was calculated by difference. The results were compared with those obtained by high-resolution continuum source graphite furnace atomic absorption spectrometry by emulsion technique; no significant differences were found for total Ni and V at the 95% confidence level according to a Student's t-test.

RESUMEN

Vanadium is recognized worldwide as the most abundant metallic constituent in petroleum. It is causing undesired side effects in the refining process, and corrosion in oil-fired power plants. Consequently, it is the most widely determined metal in petroleum and its derivatives. This paper offers a critical review of analytical methods based on atomic spectrometric techniques, particularly flame atomic absorption spectrometry (FAAS), electrothermal atomic absorption spectrometry (ET AAS), inductively coupled plasma optical emission spectrometry (ICP OES), inductively coupled plasma mass spectrometry (ICP-MS). In addition an overview is provided of the sample pretreatment and preparation procedures for vanadium determination in petroleum and petroleum products. Also included are the most recent studies about speciation and fractionation analysis using atomic spectrometric techniques.

RESUMEN

The amount of sample that is available for analysis in laboratory plant cultivation experiments is usually very limited. Highly sensitive analytical techniques are therefore required, even for elements that are present in the plants at mg g(-1) concentrations, and graphite furnace atomic absorption spectrometry (GFAAS) was chosen in this work because of its micro-sampling capability, and its relatively simple operation. Four micro-methods were investigated for the determination of iron in roots and leaves of rice plants: i) a micro-digestion with nitric and hydrochloric acids, ii) a slurry procedure using tetramethylammonium hydroxide (TMAH) tissue solubilizer, iii) a slurry prepared in 1.4 mol L(-1) nitric acid, and treated in an ultrasonic bath, and iv) the direct analysis of solid samples. The micro-digestion was suffering from high blank values and contamination problems, so that it could not be recommended for routine purposes. The TMAH method exhibited poor precision and occasional low recoveries, particularly for real samples. Direct solid sampling analysis gave results similar to those obtained with the slurry technique with ultrasonic agitation for the determination of iron in certified reference materials with iron content up to about 100 microg g(-1), but was too sensitive for the investigated rice plants, which had an iron content up to several mg g(-1). The slurry technique with ultrasonic treatment of the samples, suspended in dilute nitric acid, was finally adopted as the method of choice. The method was then applied for the determination of iron in the leaves and in different compartments of the roots of two rice cultivars, one sensitive to iron toxicity, an important nutritional disorder, and the other one resistant to iron toxicity. The results suggest that the higher resistance to iron toxicity of the second cultivar is due to a smaller uptake of iron from the soil, resulting in lower iron levels in all compartments of the plant.