In this study, the effect of synthesis conditions and annealing process on the sensitivity and stability of gas sensors made of flame-synthesized Zn-doped γ-Fe2O3 particles was investigated. Zn-doped γ-Fe2O3 particles were synthesized by flame spray pyrolysis using either H2/Air or H2/O2 coflow diffusion flames. The particles were then annealed at 325~350˚C in a tube furnace under air atmosphere. Both as-synthesized and annealed particles were used as gas sensing materials to construct gas sensors. Transmission electron microscopy (TEM), X-ray diffraction (XRD), Brunauer-Emmett-Teller surface area measurement (BET), Williamson and Hall (WH) method were employed to characterize the particles. Gas sensors were fabricated by applying the as-synthesized and annealed particles on interdigitated electrodes. The response of the gas sensor to acetone vapor, H2 in dry synthetic air was measured before and after three days of aging. High-temperature flame (H2/O2) generated nanometer-sized particles; lower temperature flame (H2/Air) generated micrometer-sized particles. Fe2O3 particles doped with 15% Zn showed the highest sensitivity. The sensors made from as-synthesized particles showed a gas sensing sensitivity that was 20 times higher than the literature value. The sensors made of microparticles lost their sensing ability after three days of aging, but sensors made of nanoparticles did not show significant change after aging. Sensors made of annealed particles (either micro or nano) did not have significant gas sensing ability, but annealing process improved the stability of gas sensors. Analysis using the WH method showed that the microstrains decreased significantly in both H2/O2 and H2/Air flame particles after annealing. The results showed that sensors made of nanoparticles have higher gas sensing signal, and more resistant toward aging than sensors made of microparticles. In addition, annealing process affected on the stability favorably due to reduction of structural defects.

Made available in DSpace on 2014-06-11T19:35:05Z (GMT). No. of bitstreams: 0 Previous issue date: 2003Bitstream added on 2014-06-13T18:46:11Z : No. of bitstreams: 1 andrade_avc_dr_araiq.pdf: 5138584 bytes, checksum: bbedecd07acf7ed59aafdb0934ad7d94 (MD5) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / Métodos de determinação de tamanho de cristalito e microdeformação da rede cristalina utilizando difratometria de raios X por pó foram discutidos e aplicados em três sistemas cerâmicos de interesse tecnológico: LiCoO2 preparado por combustão ou pelo método de Pechini; SnO2 não dopado e dopado com 5% de cério ou 5% de ítrio ou 5% de lantânio, tratados termicamente a 550oC, 720oC, 820oC, 920oC, 1000oC e 1100oC; e hidroxiapatita preparada pelo método da emulsão. Foram discutidas (i) as análises qualitativas de tamanho de cristalito-deformação (td) utilizando gráficos da largura-total-a-meia-altura (fwhm) em função do ângulo de difração 2θ; (ii) a análise isotrópica pelo método de Rietveld utilizando a função Thompson-Cox-Hastings, modificada por Young e Desai (TCHZ); (iii) o gráfico de Williamson-Hall (iv) o método “double-Voigt” e (v) o método fenomenológico de Peter Stephens aplicado ao método de Rietveld. Foram desenvolvidos dois programas escritos para o programa MathCad® 7.0 que calculam o tamanho de cristalito e a microdeformação utilizando o gráfico de Williamson-Hall e o método “double-Voigt”, respectivamente. O método da máxima entropia foi utilizado neste trabalho como complementar ao método de Rietveld para a obtenção dos mapas de densidades eletrônicas da hidroxiapatita. A análise qualitativa de tamanho de cristalito e microdeformação da rede cristalina foi aplicada ao sistema LiCoO2. Este estudo revelou que a fase LiCoO2 trigonal predomina, acompanhada de uma pequena quantidade da fase espinélio Co3O4 cúbica. Observou-se que tanto os cristalitos quanto as microdeformações são maiores quando se utiliza o método de Pechini. Nas análises utilizando o método de Rietveld e a TCHZ nas amostras de SnO2, observou-se que enquanto os cristalitos aumentam com o aumento da temperatura de tratamento, as microdeformações diminuem... / Methods for crystallite size and microstrain determination using x-ray powder diffractometry are discussed and applied to three interesting technological ceramics systems: LiCoO2 prepared by combustion or Pechini methods, SnO2 non doped and doped with 5% of cerium or 5% of yttrium or 5% of lanthanum, heat-treated at 550oC, 720oC, 820oC, 920oC, 1000oC and 1100oC, and hydroxylapatite prepared by emulsion method. Are discussed (i) the crystallite size-deformation (td) qualitative analyses using fwhm as 2θ function; (ii) the isotropic analysis by Rietveld method using the modified Young and Desai Thompson-Cox-Hastings function (TCHZ); (iii) the Williamson-Hall plot, (iv) the double-Voigt method and (v) Peter Stephens' phenomenological method applied to the Rietveld method. Two programs to calculate the crystallite size and the microstrains are developed to be used in the MathCad® 7.0 program, using the Williamson-Hall plot and double-Voigt method, respectively. The maximum entropy method was used in this work as complemental to the Rietveld method for obtaining the electron densities maps of hydroxyapatite. The size-strain qualitative analysis was applied to the LiCoO2 system. This study revealed that LiCoO2 trigonal phase prevails, followed by of a small amount of spinel Co3O4 cubic phase. It is observed that crystallites and microstrains are larger when the Pechini method is used. Analysis using the Rietveld method and TCHZ function in SnO2 samples, allow to observe that, while crystallite size increases when the temperature treatment increase, the microstrains decreases. The non doped samples show greater sizes and smaller strains. The hydroxyapatite size and strain analysis showed differences among the used methods. The Rietveld method using the Peter Stephens' model showed cristallite sizes between 918 Å and 869 Å, in the parallel and perpendicular directions of broadening axis...(Complete abstract click electronic access below)

Resumo: Métodos de determinação de tamanho de cristalito e microdeformação da rede cristalina utilizando difratometria de raios X por pó foram discutidos e aplicados em três sistemas cerâmicos de interesse tecnológico: LiCoO2 preparado por combustão ou pelo método de Pechini; SnO2 não dopado e dopado com 5% de cério ou 5% de ítrio ou 5% de lantânio, tratados termicamente a 550oC, 720oC, 820oC, 920oC, 1000oC e 1100oC; e hidroxiapatita preparada pelo método da emulsão. Foram discutidas (i) as análises qualitativas de tamanho de cristalito-deformação (td) utilizando gráficos da largura-total-a-meia-altura (fwhm) em função do ângulo de difração 2θ; (ii) a análise isotrópica pelo método de Rietveld utilizando a função Thompson-Cox-Hastings, modificada por Young e Desai (TCHZ); (iii) o gráfico de Williamson-Hall (iv) o método "double-Voigt" e (v) o método fenomenológico de Peter Stephens aplicado ao método de Rietveld. Foram desenvolvidos dois programas escritos para o programa MathCad® 7.0 que calculam o tamanho de cristalito e a microdeformação utilizando o gráfico de Williamson-Hall e o método "double-Voigt", respectivamente. O método da máxima entropia foi utilizado neste trabalho como complementar ao método de Rietveld para a obtenção dos mapas de densidades eletrônicas da hidroxiapatita. A análise qualitativa de tamanho de cristalito e microdeformação da rede cristalina foi aplicada ao sistema LiCoO2. Este estudo revelou que a fase LiCoO2 trigonal predomina, acompanhada de uma pequena quantidade da fase espinélio Co3O4 cúbica. Observou-se que tanto os cristalitos quanto as microdeformações são maiores quando se utiliza o método de Pechini. Nas análises utilizando o método de Rietveld e a TCHZ nas amostras de SnO2, observou-se que enquanto os cristalitos aumentam com o aumento da temperatura de tratamento, as microdeformações diminuem...(Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Methods for crystallite size and microstrain determination using x-ray powder diffractometry are discussed and applied to three interesting technological ceramics systems: LiCoO2 prepared by combustion or Pechini methods, SnO2 non doped and doped with 5% of cerium or 5% of yttrium or 5% of lanthanum, heat-treated at 550oC, 720oC, 820oC, 920oC, 1000oC and 1100oC, and hydroxylapatite prepared by emulsion method. Are discussed (i) the crystallite size-deformation (td) qualitative analyses using fwhm as 2θ function; (ii) the isotropic analysis by Rietveld method using the modified Young and Desai Thompson-Cox-Hastings function (TCHZ); (iii) the Williamson-Hall plot, (iv) the "double-Voigt" method and (v) Peter Stephens' phenomenological method applied to the Rietveld method. Two programs to calculate the crystallite size and the microstrains are developed to be used in the MathCad® 7.0 program, using the Williamson-Hall plot and "double-Voigt" method, respectively. The maximum entropy method was used in this work as complemental to the Rietveld method for obtaining the electron densities maps of hydroxyapatite. The size-strain qualitative analysis was applied to the LiCoO2 system. This study revealed that LiCoO2 trigonal phase prevails, followed by of a small amount of spinel Co3O4 cubic phase. It is observed that crystallites and microstrains are larger when the Pechini method is used. Analysis using the Rietveld method and TCHZ function in SnO2 samples, allow to observe that, while crystallite size increases when the temperature treatment increase, the microstrains decreases. The non doped samples show greater sizes and smaller strains. The hydroxyapatite size and strain analysis showed differences among the used methods. The Rietveld method using the Peter Stephens' model showed cristallite sizes between 918 Å and 869 Å, in the parallel and perpendicular directions of broadening axis...(Complete abstract click electronic access below) / Orientador: Carlos de Oliveira Paiva Santos / Coorientador: José Caetano Zurita da Silva / Doutor