Graphite filter atomizers (GFA) for electrothermal atomic absorption spectrometry (ETAAS) show substantial advantages over commonly employed electrothermal vaporizers and atomizers, tube and platform furnaces, for direct determination of high and medium volatility elements in matrices associated with strong spectral and chemical interferences. Two factors provide lower limits of detection and shorter determination cycles with the GFA: the vaporization area in the GFA is separated from the absorption volume by a porous graphite partition; the sample is distributed over a large surface of a collector in the vaporization area. These factors convert the GFA into an efficient chemical reactor. The research concerning the GFA concept, technique and analytical methodology, carried out mainly in the author's laboratory in Russia and South Africa, is reviewed. Examples of analytical applications of the GFA in AAS for analysis of organic liquids and slurries, bio-samples and food products are given. Future prospects for the GFA are discussed in connection with analyses by fast multi-element AAS.

The main aim of this thesis has been to contribute to the ongoing work with development of new diode-laser-based spectroscopic techniques and metho­do­lo­gies for sensitive detection of molecules in gas phase. The techniques under scrutiny are tunable diode laser absorption spectrometry (TDLAS) and Faraday modulation spectrometry (FAMOS). Conventional distributed-feedback (DFB) tele­communication diode lasers working in the near-infrared (NIR) region have been used for detection of carbon monoxide (CO) and temperature in hot humid media whereas a unique frequency-quadrupled external-cavity diode laser producing mW powers of continuous-wave (cw) light in the ultra violet (UV) region have been used for detection of nitric oxide (NO). A methodology for assessment of CO in hot humid media by DFB-TDLAS has been developed. By addressing a particular transition in its 2nd overtone band, and by use of a dual-fitting methodology with a single reference water spectrum for background correction, % concentrations of CO can be detected in media with tens of percent of H2O (≤40%) at T≤1000 °C with an accuracy of a few %. Moreover, using an ordinary DFB laser working in the C-band, a technique for assessment of the temperature in hot humid gases (T≤1000 °C) to within a fraction of a percent has been developed. The technique addresses two groups of lines in H2O that have a favorable temperature dependence and are easily accessed in a single scan, which makes it sturdy and useful for industrial applications. A technique for detection of NO on its strong electronic transitions by direct absorption spectrometry (DAS) using cw UV diode laser light has been deve­loped. Since the electronic transitions are ca. two or several orders of magnitude stronger than of those at various rotational-vibra­tional bands, the system is capable of detecting NO down to low ppb∙m concentrations solely using DAS. Also the FAMOS technique has been further developed. A new theoretical description expressed in terms of both the integrated line strength of the transition and 1st Fourier coefficients of a magnetic-field-modulated dispersive lineshape functions is presented. The description has been applied to both ro-vib Q-transitions and electronic transitions in NO. Simulations under different pressures and magnetic field conditions have been made that provide the optimum conditions for both cases. A first demonstration and characteri­zation of FAMOS of NO addressing its electronic transitions in the UV-region has been made, resulting in a detection limit of 10 ppb∙m. The characterization indicates that the technique can be significantly improved if optimum conditions can be obtained, which demonstrates the high potential of the UV-FAMOS technique.

Cleaning validation, in which cleaned surfaces are analysed for residual material, is an important process in pharmaceutical manufacturing and research facilities. Current procedures usually consist of either swab or rinse-water sampling followed by analysis of the samples. The analysis step is typically either rapid but unselective (conductivity, pH, total organic carbon, etc.), or selective but time-consuming (HPLC). This thesis describes the development of an in situ surface-spectroscopic analysis that removes the need for swab sampling and is both rapid and selective. This method has the potential to complement existing analyses to increase the efficiency of cleaning-validation protocols. The spectrometric system consists of a Fourier-transform infrared (FTIR) spectrometer coupled to a fibre-optic grazing-angle reflectance probe, and allows the measurement of infrared reflection-absorbance spectra (IRRAS) from flat surfaces in ~10 s. Multivariate chemometric methods, such as partial least squares (PLS) regression, are used to exploit the high information content of infrared spectra to obtain selective analyses without physical separation of the analyte or analytes from whatever interfering species may be present. Multivariate chemometric models require considerably more effort for calibration and validation than do traditional univariate techniques. This thesis details suitable methods for preparing calibration standards by aerosol deposition, optimising and validating the model by cross- and test-set validation, and estimating the uncertainty by resampling and formula-based approaches. Successful calibration models were demonstrated for residues of acetaminophen, a model active pharmaceutical ingredient (API), on glass surfaces. The root-mean-square error of prediction (RMSEP) was ~0.07 µg cm⁻². Simultaneous calibration for acetaminophen and aspirin, another API, gave a similar RMSEP of 0.06 µg cm⁻² for both compounds, demonstrating the selectivity of the method. These values correspond to detection limits of ~0.2 µg cm⁻², well below the accepted visual detection limit of ~1-4 µg cm⁻². The sensitivity of the method with a stainless steel substrate was found to depend strongly on the surface finish, with highly polished surfaces giving more intense IRRAS. RMSEP values of 0.04- 0.05 µg cm⁻² were obtained for acetaminophen on stainless steel with three different finishes. For this system, severe nonlinearity was encountered for loadings 1.0 µg cm⁻². From the results presented in this thesis, it is clear that IRRAS has potential utility in cleaning validation as a complement to traditional techniques.

O objetivo desse trabalho foi o desenvolvimento de métodos analíticos para a determinação simultânea de Cd/Pb em cimento e Cr/Mn em alumina por espectrometria de absorção atômica com atomização eletrotérmica (SIMAAS) e amostragem de suspensão. Considerando os erros de amostragem de suspensão e as condições compromissadas impostas para a determinação simultânea, foi também objetivo do trabalho estudar a aplicação de padrão interno para corrigir erros sistemático e aleatório. A viabilidade de se usar o Tl como padrão interno foi explorada na determinação de Cd e Pb em cimento. O uso do padrão interno se baseou na hipótese de que o elemento Tl deveria adsorver quantitativamente sobre as partículas do material em suspensão. Foi feito o estudo térmico dos elementos Cd, Pb e Tl em soluções aquosas e suspensões, na presença e ausência de diferentes modificadores químicos. O melhor modificador químico (5 µg Pd + 3 µg Mg) foi escolhido com base nos gráficos de correlação. As melhores temperaturas de pirólise e atomização para a determinação simultânea de Cd e Pb foram 600°C e 2000°C, respectivamente. A preparação das amostras foi feita com suspensões contendo 30 µg L-1 de Tl em 0,1 % (v v-1) de HNO3 + 0,025 % (m v-1) de Triton X-100. A avaliação da exatidão foi feita com a determinação de Cd e Pb em cimento de construção civil e dentário com análise de solução e análise de materiais de referência de sedimento marinho (MESS1 e BCSS 1) e de rio (Buffalo River, IEAEA 356). Os resultados foram concordantes em um nível de confiança de 95% (teste t-student). Os limites de detecção (LD) obtidos foram 0,14 µg L-1 (1,4pg) para o Cd e de 1,2 µg L-1 (12 pg) para Pb. As massas características foram 1,2 pg e 25 pg para Cd e Pb, respectivamente. Na determinação simultânea de Cr e Mn em alumina não foi utilizado padrão interno. O estudo envolveu a obtenção de curvas de pirólise e atomização em presença de diferentes modificadores. Estudos de interferência mostraram que 100 mg L-1 de Al provocaram aumento nos sinais analíticos dos elementos que foi eliminado com o modificador químico NaOH (0,2 mol L-1) + HF (0,4 mol L-1) combinado com 350 µg de Nb. Nessa condição as temperaturas de pirólise e atomização foram 1400°C e 2400°C, respectivamente. As suspensões de alumina foram preparadas em 2 % (v v-1) HNO3. A avaliação da exatidão foi feita com a determinação de Cr e Mn em material de referência de alumina (Alumina Reduction Grade - 699) e os resultados foram concordantes em um nível de confiança de 95% (teste t-student). Os limites de detecção (LD), estimados a partir da variabilidade de 20 medidas consecutivas de uma solução contendo 2 % (v v-1) HNO3 foram 0,11 µg L-1 (2,2 pg) para Cr e 0,17 µg L-1 (3,5 pg) para o Mn. As massas características foram 10 pg e 1,3 pg para Cr e Mn, respectivamente. / The aim of this work was the development of analytical methods for the simultaneous determination of Cd/Pb in cement and Cr/Mn in alumina by electrothermal atomic absorption spectrometry (SIMAAS) and slurry sampling. Considering the errors of slurry sampling and the imposed compromissed conditions for the simultaneous determination, it was also the aiming of this work to study the application of internal standard for the sistematic and random errors correction. The viability of using Tl as internal standard was investigated for Cd and Pb determination in cement. The internal standard aplicability was based on the hypothesis of the element Tl would have to adsorver quantitatively on the particles of the material in suspension. The thermal behaviour of Cd, Pb and Tl in aqueous solutions and suspensions was obtained in presence and absence of different chemical modifiers. The best chemical modifier (5 mg Pd + 3 mg Mg) was chosen based on correlation graphs. The pyrolysis and atomization temperatures for the simultaneous determination of Pb were 600°C and 2000°C, respectively. The sample the preparation was made with 30 mg L-1 of Tl in 0,1% (v v-1) of HNO3 + 0.025% (m v-1 of Triton X-100. The evaluation of the accuracy was made with the determination of Cd and Pb in cement, microwave extraction in acid media and analysis of marine sediment reference materials (MESS1 and BCSS 1) and river sediment (Buffalo River, IEAEA 356). The results are concordant at 95% of confidence level using Studen\'s t-test. The detention limits (LD) was 0.14 µg L-1 (1.4 pg) for Cd and 1.2 µg L-1(12 pg) for Pb. The characteristic masses were 1.2 pg and 25 pg for Cd and Pb, respectively. The simultaneous determination of Cr and Mn in alumina was done without internal standard. Pyrolysis and atomization temperaturas were obtained in presence of different modifiers. Studies of interference shown that 100 mg L-1 of Al provoked increase in the analytical signals of Cr and Mn that was eliminated with the chemical modifier NaOH (0,2 mol L-1) + HF (0,4 mol L-1) combined with 350 mg of Nb as permanent chemical modifier. In this condition pyrolysis and atomization temperatures were 1400 °C and 2400 °C, respectively. The alumina suspensions were prepared in 2% (v v-1) HNO3. The evaluation of the accuracy was made with the determination of Cr and Mn in alumina reference material (Reduction Alumina Grating - 699) and the results were concordant at 95 % of the confidence level using Student\'s t-test. The detention limits (LD) were 0.11 µg L-1 (2.2 pg) for Cr and 0.17 µg L-1 (3.5 pg) for Mn. The characteristic masses were 10 and 1.3 pg for Cr and Mn, respectively.

Cleaning validation, in which cleaned surfaces are analysed for residual material, is an important process in pharmaceutical manufacturing and research facilities. Current procedures usually consist of either swab or rinse-water sampling followed by analysis of the samples. The analysis step is typically either rapid but unselective (conductivity, pH, total organic carbon, etc.), or selective but time-consuming (HPLC). This thesis describes the development of an in situ surface-spectroscopic analysis that removes the need for swab sampling and is both rapid and selective. This method has the potential to complement existing analyses to increase the efficiency of cleaning-validation protocols. The spectrometric system consists of a Fourier-transform infrared (FTIR) spectrometer coupled to a fibre-optic grazing-angle reflectance probe, and allows the measurement of infrared reflection-absorbance spectra (IRRAS) from flat surfaces in ~10 s. Multivariate chemometric methods, such as partial least squares (PLS) regression, are used to exploit the high information content of infrared spectra to obtain selective analyses without physical separation of the analyte or analytes from whatever interfering species may be present. Multivariate chemometric models require considerably more effort for calibration and validation than do traditional univariate techniques. This thesis details suitable methods for preparing calibration standards by aerosol deposition, optimising and validating the model by cross- and test-set validation, and estimating the uncertainty by resampling and formula-based approaches. Successful calibration models were demonstrated for residues of acetaminophen, a model active pharmaceutical ingredient (API), on glass surfaces. The root-mean-square error of prediction (RMSEP) was ~0.07 µg cm⁻². Simultaneous calibration for acetaminophen and aspirin, another API, gave a similar RMSEP of 0.06 µg cm⁻² for both compounds, demonstrating the selectivity of the method. These values correspond to detection limits of ~0.2 µg cm⁻², well below the accepted visual detection limit of ~1-4 µg cm⁻². The sensitivity of the method with a stainless steel substrate was found to depend strongly on the surface finish, with highly polished surfaces giving more intense IRRAS. RMSEP values of 0.04- 0.05 µg cm⁻² were obtained for acetaminophen on stainless steel with three different finishes. For this system, severe nonlinearity was encountered for loadings 1.0 µg cm⁻². From the results presented in this thesis, it is clear that IRRAS has potential utility in cleaning validation as a complement to traditional techniques.

Diss. (sammanfattning) Uppsala : Sveriges lantbruksuniversitet, 2007. / Härtill 4 uppsatser.

O objetivo desse trabalho foi o desenvolvimento de métodos analíticos para a determinação simultânea de Cd/Pb em cimento e Cr/Mn em alumina por espectrometria de absorção atômica com atomização eletrotérmica (SIMAAS) e amostragem de suspensão. Considerando os erros de amostragem de suspensão e as condições compromissadas impostas para a determinação simultânea, foi também objetivo do trabalho estudar a aplicação de padrão interno para corrigir erros sistemático e aleatório. A viabilidade de se usar o Tl como padrão interno foi explorada na determinação de Cd e Pb em cimento. O uso do padrão interno se baseou na hipótese de que o elemento Tl deveria adsorver quantitativamente sobre as partículas do material em suspensão. Foi feito o estudo térmico dos elementos Cd, Pb e Tl em soluções aquosas e suspensões, na presença e ausência de diferentes modificadores químicos. O melhor modificador químico (5 µg Pd + 3 µg Mg) foi escolhido com base nos gráficos de correlação. As melhores temperaturas de pirólise e atomização para a determinação simultânea de Cd e Pb foram 600°C e 2000°C, respectivamente. A preparação das amostras foi feita com suspensões contendo 30 µg L-1 de Tl em 0,1 % (v v-1) de HNO3 + 0,025 % (m v-1) de Triton X-100. A avaliação da exatidão foi feita com a determinação de Cd e Pb em cimento de construção civil e dentário com análise de solução e análise de materiais de referência de sedimento marinho (MESS1 e BCSS 1) e de rio (Buffalo River, IEAEA 356). Os resultados foram concordantes em um nível de confiança de 95% (teste t-student). Os limites de detecção (LD) obtidos foram 0,14 µg L-1 (1,4pg) para o Cd e de 1,2 µg L-1 (12 pg) para Pb. As massas características foram 1,2 pg e 25 pg para Cd e Pb, respectivamente. Na determinação simultânea de Cr e Mn em alumina não foi utilizado padrão interno. O estudo envolveu a obtenção de curvas de pirólise e atomização em presença de diferentes modificadores. Estudos de interferência mostraram que 100 mg L-1 de Al provocaram aumento nos sinais analíticos dos elementos que foi eliminado com o modificador químico NaOH (0,2 mol L-1) + HF (0,4 mol L-1) combinado com 350 µg de Nb. Nessa condição as temperaturas de pirólise e atomização foram 1400°C e 2400°C, respectivamente. As suspensões de alumina foram preparadas em 2 % (v v-1) HNO3. A avaliação da exatidão foi feita com a determinação de Cr e Mn em material de referência de alumina (Alumina Reduction Grade - 699) e os resultados foram concordantes em um nível de confiança de 95% (teste t-student). Os limites de detecção (LD), estimados a partir da variabilidade de 20 medidas consecutivas de uma solução contendo 2 % (v v-1) HNO3 foram 0,11 µg L-1 (2,2 pg) para Cr e 0,17 µg L-1 (3,5 pg) para o Mn. As massas características foram 10 pg e 1,3 pg para Cr e Mn, respectivamente. / The aim of this work was the development of analytical methods for the simultaneous determination of Cd/Pb in cement and Cr/Mn in alumina by electrothermal atomic absorption spectrometry (SIMAAS) and slurry sampling. Considering the errors of slurry sampling and the imposed compromissed conditions for the simultaneous determination, it was also the aiming of this work to study the application of internal standard for the sistematic and random errors correction. The viability of using Tl as internal standard was investigated for Cd and Pb determination in cement. The internal standard aplicability was based on the hypothesis of the element Tl would have to adsorver quantitatively on the particles of the material in suspension. The thermal behaviour of Cd, Pb and Tl in aqueous solutions and suspensions was obtained in presence and absence of different chemical modifiers. The best chemical modifier (5 mg Pd + 3 mg Mg) was chosen based on correlation graphs. The pyrolysis and atomization temperatures for the simultaneous determination of Pb were 600°C and 2000°C, respectively. The sample the preparation was made with 30 mg L-1 of Tl in 0,1% (v v-1) of HNO3 + 0.025% (m v-1 of Triton X-100. The evaluation of the accuracy was made with the determination of Cd and Pb in cement, microwave extraction in acid media and analysis of marine sediment reference materials (MESS1 and BCSS 1) and river sediment (Buffalo River, IEAEA 356). The results are concordant at 95% of confidence level using Studen\'s t-test. The detention limits (LD) was 0.14 µg L-1 (1.4 pg) for Cd and 1.2 µg L-1(12 pg) for Pb. The characteristic masses were 1.2 pg and 25 pg for Cd and Pb, respectively. The simultaneous determination of Cr and Mn in alumina was done without internal standard. Pyrolysis and atomization temperaturas were obtained in presence of different modifiers. Studies of interference shown that 100 mg L-1 of Al provoked increase in the analytical signals of Cr and Mn that was eliminated with the chemical modifier NaOH (0,2 mol L-1) + HF (0,4 mol L-1) combined with 350 mg of Nb as permanent chemical modifier. In this condition pyrolysis and atomization temperatures were 1400 °C and 2400 °C, respectively. The alumina suspensions were prepared in 2% (v v-1) HNO3. The evaluation of the accuracy was made with the determination of Cr and Mn in alumina reference material (Reduction Alumina Grating - 699) and the results were concordant at 95 % of the confidence level using Student\'s t-test. The detention limits (LD) were 0.11 µg L-1 (2.2 pg) for Cr and 0.17 µg L-1 (3.5 pg) for Mn. The characteristic masses were 10 and 1.3 pg for Cr and Mn, respectively.

A new spectrometric technique for measuring diffusion coefficients of Ordovician Queenston Formation shale from the Michigan Basin of southwest Ontario, Canada is presented; in this case pore diffusion coefficients (Dp) were determined for the conservative (iodide, I^-) and reactive (cesium, Cs^+) tracers in porous media. Furthermore, diffusion-reaction parameters such as Cation exchange capacity (CEC) were obtained for the reactive tracer, Cs^+ by the reactive-transport modeling. The principle of X-ray absorption spectrometry (XAS) is based on the attenuation of high atomic number of diffusive species (e.g. Cs^+and I^-) in porous media, allowing for the quantification of the resulting spatial changes. The technique employs a 1.1 mm collimated X-ray beam to resolve the tracer presence in a slice; small region of a rotating sample for a fixed time. The x-ray beam is incident on a sample of porous rock while the transmitted beam is detected on the opposite side of the sample, allowing acquisition of a transmitted X-ray energy spectrum vs intensity (counts). Additional analyses such as mineralogical studies by scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDS) were conducted to improve the understanding of solute transport. The comparison of the data obtained by the XAS method indicating a good agreement with established radiography and other conventional diffusion methods provide an alternate approach for quantifying the diffusion coefficient of porous media.

A água de coco é um isotônico natural, uma bebida pouco calórica e muito nutritiva, representando 1,4 % do mercado de refrigerantes e de isotônicos artificiais. Aliado ao crescente mercado interno e externo, surge os processos de conservação com o intuito de preservação da água de coco fora do fruto, facilitando a comercialização e aumentando o tempo de vida de prateleira. Considerando o pressuposto, esse trabalho teve como objetivo principal avaliar a influência dos processos de conservação, ultrafiltração, congelamento e pasteurização, na concentração de Al, Ca, Cu, Fe, K, Mg, Mn, Na, Se e Zn em água de coco. As determinações foram feitas por espectrometria de emissão óptica com plasma indutivamente acoplado e espectrometria de absorção atômica com atomização eletrotérmica. Investigação no sentido de determinar espécies livres de Se4+ e Se6+, em água de coco, por geração de hidretos e pré-concentração na superfície de um tubo de grafite modificada com Ir também foi objetivo desse trabalho. Os estudos foram realizados por meio das determinações das concentrações totais dos elementos e em diferentes frações (filtrado e resíduo), obtidas após ultrafiltração com membrana de celulose (0,100 e 0,025 µm). Em todos os processos de conservação estudados, exceto na ultrafiltração, as concentrações totais dos elementos de interesse foram preservadas. A ultrafiltração com filtro de 0,025 μm promoveu a retenção de uma quantidade significativa de Fe (30%) e Zn (50 %). A influência dos processos de conservação, com base na distribuição dos elementos de interesse, mostrou uma tendência de variação na distribuição de Cu, Fe e Zn na água de coco pasteurizada e em todos os elementos na congelada, sobretudo para aquelas com tempo de congelamento superior a 30 dias. A especiação de Se6+ e Se4+ foi feita após redução do Se6+ com 2,0 mol L-1 HCl + 1 mol L-1 KBr e aquecimento por microondas focalizadas em microfrascos, apresentando cerca de 95 % de eficiência na redução na ausência da amostra de água de coco. / The coconut water is a natural isotonic, nutritive and low caloric drink, corresponding to 1.4 % of the soft drink and artificial isotonic market store. The conservation processes of the coconut water appear in association to the expansion of the internal and external market, increasing the shelf-life of the product and its comercialization. The objective of this work was evaluate the influence of the conservation processes, such as, ultrafiltration, freezing and pasteurization based on the concentrations of Al, Ca, Cu, Fe, K, Mg, Mn, Na, Se and Zn in the coconut water. The determinations were done by inductively coupled plasma-optical emission spectrometry and electrothermal atomic absorption spectrometry. The investigation of free-species of Se4+ and Se6+, in coconut water, by hydride generation and pre-concentration on the modified graphite tube surface with Ir was also the aim of this project. The studies were carried out through the determinations of the total concentration of the interest elements and in different fractions (filtrate and residue), obtained after ultrafiltration with cellulose membrane filters (0.100 and 0.025 µm). In all processes of conservation studied, except the ultrafiltration, the total concentrations were preserved. The ultrafiltration with filters of 0.025 µm promoted the retention of significant concentration of Fe (30%) e Zn (50%). Considering the distribuituion of the interest elements, the influence of the conservation processes showed a tendency of the variation in the Cu, Fe and Zn distribuition for the pasteurized coconut water, and for all elements in the frozen, mainly for frozen samples storaged for more than 30 days. The speciation of Se4+ and Se6+ was carried out after reduction of Se6+ with 2 mol L-1 HCl + 1 mol L-1 KBr and using microvials heated in a focused microwave. The reduction efficiency of the Se6+ to Se4+ was 95 %, in absence of coconut water.

O objetivo desse trabalho foi o desenvolvimento de métodos analíticos para a determinação de Cd, Pb, Cu, Ni, Co, Cr e Mn em cimentos Portland por espectrometria de absorção atômica com atomização em forno de grafite e amostragem direta de sólidos. Não foi necessário nenhum tipo de pré-tratamento das amostras, pois apresentaram tamanhos médios de partículas (~12 µm) adequados para aplicação em análise direta de sólidos. Os programas de aquecimento foram otimizados a partir da avaliação térmica de cada um dos elementos (curvas de pirólise e atomização), do uso de modificadores químicos e de matriz e da comparação de curvas de calibração em solução aquosa e na presença da amostra sólida. Todas as determinações dos elementos foram feitas usando curvas de calibração obtidas em meio de solução aquosa. Devido a alta concentração nas amostras, no caso do Cu, Ni, Co, Cr e Mn foi necessário o uso da correção de fundo por efeito Zeeman com 3 campos, aumentando a representatividade. Nessa condição, foi possível a pesagem de alíquotas da amostra de até 200 µg para análise. As melhores condições para as determinações de Cd e Pb foi com o uso do modificador químico 5 µg de Pd em 0,1 % m v-1 de Triton X-100. Nessa condição, as melhores temperaturas de pirólise e atomização foram 600°C e 2000°C para Cd e 900°C e 2100°C para Pb, respectivamente. As determinações de Cu e Ni foram possíveis sem o uso de modificador químico ou de matriz. As temperaturas de pirólise e atomização foram 1200°C e 2600°C, respectivamente, para os dois elementos. As determinações de Co, Cr e Mn foram realizadas com o uso de modificador de matriz (Na2CO3 + ZnO) que promoveu uma fusão in situ durante o programa de aquecimento. Nessas condições, as temperaturas de pirólise e atomização foram 1400 °C e 2600 °C, respectivamente. Os limites de detecção (3σ) para as determinações diretas em cimento Portland foram 0,011 µg g-1 para Cd, 0,28 µg g-1 para Pb, 0,25 µg g-1 para Cu, 0,24 µg g-1 para Ni, 0,11 µg g-1 para Co, 1,1 µg g-1 para Cr e 1,97 µg g-1 para Mn. As concentrações características foram 0,0098 µg g-1 para Cd, 0,21 µg g-1 para Pb, 0,17 µg g-1 para Cu, 0,24 µg g-1 para Ni, 0,068 µg g-1 para Co, 0,62 µg g-1 para Cr e 0,14 µg g-1 para Mn. As exatidões dos métodos foram avaliadas a partir da análise de materiais de referência de cimento Protland (NIST 1886a e 1889a) e de sedimentos marinhos (NBCC MESS-1 e NBCC BCSS-1). Todos os resultados concordaram a um nível de 95% de confiança, quando aplicado o teste t-Student. / The objective of this work was the development of analytical methods for the determination of Cd, Pb, Cu, Ni, Co, Cr e Mn in Portland cement by solid sampling graphite furnace atomic absorption spectrometry. It was not necessary any pretreatment of the samples, because the particles size (12 µm) were apropriated for direct solid analysis. The heating programs were optimized by thermal evaluation of each element (pyrolysis and atomization curves), using chemical and matrix modifiers and comparison of the calibration curves obtained in presence of aqueous solutions and solid samples. All determinations of the elements were performed against aqueous calibration curves. Due to the high concentration in the samples, for Cu, Ni, Co, Cr e Mn determinations were necessary 3-field Zeeman background correction, improving the representativity. Using this condition, it was possible weight aliquots of samples up to 200 µg. The best condition for Cd and Pb determinations was 5 µg of Pd in 0.1% m v-1 de Triton X-100 as chemical modifier. The pyrolysis and atomization temperatures were 600°C and 2000°C for Cd and 900°C and 2100°C for Pb, respectively. The determinations of Cu e Ni were possible without chemical modifier. The pyrolysis and atomization temperatures were 1200°C and 2600°C, respectively, for both elements. For the determinations of Co, Cr e Mn were used a matrix modifier (Na2CO3 + ZnO) for the in situ fusion, during the heating program. In this condition, the pyrolysis and atomization temperatures were 1400°C and 2600°C, respectively, for all elements. The detection limits (3σ) for the direct determination of Portland cement were 0.011 µg g-1 for Cd, 0.28 µg g-1 for Pb, 0.25 µg g-1 for Cu, 0.24 µg g-1 for Ni, 0.11 µg g-1 for Co, 1.1 µg g-1 for Cr and 1.97 µg g-1 for Mn. The characteristic concentrations were 0.0098 µg g-1 for Cd, 0.21 µg g-1 for Pb, 0.17 µg g-1 for Cu, 0.24 µg g-1 for Ni, 0.068 µg g-1 for Co, 0.62 µg g-1 for Cr and 0,14 µg g-1 for Mn. The accuracy of the methods was evaluated by reference materials of Portland cement (NIST 1886a and 1889a) and marine sediments (NBCC MESS-1 and NBCC BCSS-1) analysis. All results are in good agreement with the recommended concentration at 95 % (t-Student test) as confidence level.