Submitted by Leticia Schettini (leticia@ufrrj.br) on 2016-09-19T15:16:21Z No. of bitstreams: 1 2010 - Juarez Vicente.pdf: 1012625 bytes, checksum: 30ce03687174fc4cf4226e455adde13a (MD5) / Made available in DSpace on 2016-09-19T15:16:21Z (GMT). No. of bitstreams: 1 2010 - Juarez Vicente.pdf: 1012625 bytes, checksum: 30ce03687174fc4cf4226e455adde13a (MD5) Previous issue date: 2010-05-07 / From the scientific viewpoint, the word ?uncertainty? expresses doubt about an analytic result. The implementation of the concept of uncertainty of measurement is a crucial step that the Brazilian laboratories must take in the process for obtaining the certification of the ability to perform essays in accordance with the Brazilian technical norm ABNT NBR ISO/IEC 17025. Regardless the technical area of application, the determination of the uncertainty of measurement associated with the result of an analysis is extremely important, especially in the food industry. The expanded uncertainty is related with the degree of reliability of an analytical result, and by means of this information it is possible to evaluate if the result of an analysis is consistent, i.e., if the uncertainty associated to the measurement remains inside an acceptable range. When the percentage of uncertainty exceeds a threshold, it is necessary to find the process variables that more intensely contribute to increasing the uncertainty and, in the sequence, to take corrective actions in order to minimize the impact caused for those variables over the uncertainty of measurement. The aim of this study was to detect the uncertainty of measurement associated with the analysis of total reducing sugars (TRS) by the Lane-Eynon method. The equipments used were analytical and semi analytical balances, burettes with 10mL and 25mL, and the food matrix studied was wild honey. To perform the calculations of the standard uncertainty, u, information about the uncertainty of the laboratory materials used in the analysis ? obtained from the certificate of calibration. The combined uncertainty, uc, was obtained by deriving the influence factors and considering, if necessary, the effective degree of freedom, ?eff, as stated in the Welch-Satterthwaite equation. The expanded uncertainty, U, was obtained by multiplying uc by the appropriate coverage factor, k = 2 (at level of confidence of 95,45%). The results showed that the expanded uncertainty obtained for the analysis of TRS was less than 2,0% when the analytical balance was used. The process variable having the greatest impact was the volume (close to 80,0%), followed by the title (close to 20,0%). The analysis of the expanded uncertainty obtained for the analysis of TRS when the semi analytical balance, and 10mL as well as 25mL burettes were used provided reliable results. However, care must be taken in such a combination of equipments, especially when using the 25mL burettes, since the percentage of uncertainty was great than 3,5%, too close to the limit established by the Codex Alimentarius, 4,0%. It was observed that, in this case, the variable having the greatest impact on the uncertainty was the title, with 79,82% (10mL burettes), and 87,10% (25mL burettes). / Do ponto de vista t?cnico-cient?fico, a palavra ?incerteza? expressa a d?vida em rela??o a um resultado anal?tico. A aplica??o do conceito de incerteza de medi??o ? fundamental para os laborat?rios brasileiros que almejam obter o reconhecimento da capacidade de realiza??o de ensaios em conformidade com a norma ABNT NBR ISO/IEC 17025. Independentemente da ?rea de atua??o, ? de extrema import?ncia o estudo e o conhecimento da incerteza de medi??o associada a um resultado de an?lise, em particular quando aplicado ? ind?stria de alimentos. A incerteza expandida informa o grau de confiabilidade de um resultado anal?tico e, atrav?s dela, ? poss?vel avaliar, a um n?vel de abrang?ncia estabelecido, se o resultado de uma an?lise transmite credibilidade, ou seja, se o valor da incerteza est? em uma faixa de concentra??o aceit?vel ou n?o. Quando o percentual de incerteza ultrapassa o limite aceit?vel, ? necess?rio conhecer quais as vari?veis do m?todo que mais contribuem para a incerteza e executar as medidas corretivas a fim de minimizar este impacto e, consequentemente, reduzir a incerteza total. O objetivo geral deste trabalho foi determinar a incerteza expandida associada ? an?lise titulom?trica de a??cares redutores totais (ART) pelo m?todo de Lane-Eynon, tendo como matriz uma amostra de mel, utilizando balan?a anal?tica e semi anal?tica e buretas de 10mL e 25mL. Para o c?lculo da incerteza padr?o (u), utilizaram-se os dados obtidos nos certificados de calibra??o de equipamentos e vidrarias empregados na an?lise. A incerteza combinada ( ) foi obtida derivando os fatores de influ?ncia e considerando, quando necess?rio, o grau de liberdade efetivo (?eff), dado pela Equa??o de Welch-Satterthwaite. Obteve-se a incerteza expandida (U) multiplicando-se por 2 (dois) o valor de para um n?vel de confian?a de 95,45%. Os resultados obtidos mostraram que a incerteza expandida de ART utilizando balan?a anal?tica n?o foi significativa (<2%) e que a vari?vel que mais influenciou foi o volume de amostra gasto (~80%), seguido do T?tulo (~20%). A an?lise da incerteza expandida de ART obtida com balan?a semi anal?tica e buretas de 10mL e 25mL gerou resultados confi?veis. No entanto, deve haver cautela neste caso, especialmente ao se utilizar bureta de 25mL, visto que o valor percentual de incerteza superou 3,5%. Observa-se que a vari?vel que mais contribui para a incerteza foi o T?tulo com 79,82% (bureta de 10mL) e 87,10% (bureta de 25mL).