No abstract

Recebido em 10/12/04; aceito em 20/5/05; publicado na web em 8/12/05 CHEMOMETRICS II: SPREADSHEETS FOR EXPERIMENTAL DESIGN CALCULATIONS, A TUTORIAL. This work describes, through examples, a simple way to carry out experimental design calculations applying an spreadsheets. The aim of this tutorial is to introduce an alternative to sophisticated commercial programs that normally are too complex in data input and output. An overview of the principal methods is also briefly presented. The spreadsheets are suitable to handle different types of computations such as screening procedures applying factorial design and the optimization procedure based on response surface methodology. Furthermore, the spreadsheets are sufficiently versatile to be adapted to specific experimental designs.Keywords: spreadsheets; experimental design; chemometrics. INTRODUÇÃOAtualmente, com o advento acelerado dos meios computacionais, cálculos muitas vezes simples podem ficar mascarados quando realizados por pacotes computacionais complexos. Os planejamentos fatoriais [1][2][3] , por ex., exigem após sua execução, cálculos simples, mas muitas vezes tediosos devido à grande quantidade de dados a serem processados. A maioria dos estudantes e profissionais aprende a utilizar softwares que realizam tais cálculos, entretanto, em muitos casos, sem o interesse e estímulo necessários para entender como os mesmos são processados.A importância do entendimento dos procedimentos realizados pelos softwares é fundamental para avaliação dos resultados obtidos, bem como para o questionamento da maneira pelo qual tais softwares os realizam.Há no mercado diversos tipos de softwares que realizam cálcu-los de planejamentos experimentais, tais como Statistica 4 , Unscrambler 5 , Statgraphics 6 , Design Expert 7 , Modde 8 , Minitab 9 dentre outros. Entre os programas gratuitos destaca-se o pacote de algoritmos executáveis disponível no web site Chemkeys 10 , sendo este até o momento, o único gratuito em português. Nota-se, portanto, que a maioria destes programas não são gratuitos e muitas vezes processam resultados complicados de serem entendidos por um pesquisador inexperiente. Ao contrário, as planilhas eletrôni-cas, já bastante difundidas e conhecidas, são práticas no sentido da entrada de dados e equações, além de proporcionarem excelente visualização dos resultados, fácil transferência de dados, gráficos e tabelas, sendo ferramentas poderosas para implementar e realizar diferentes tipos de cálculos, como os de planejamentos experimentais. É válido destacar que existem versões gratuitas disponí-veis na web tanto para sistemas operacionais Windows quanto para Linux, como a encontrada no pacote Openoffice 11 . O objetivo deste trabalho foi desenvolver, de maneira simples e didática, planilhas eletrônicas utilizando-se o software Excel ® da Microsoft 12 para realizar cálculos de planejamentos fatoriais e otimização empregando Metodologias de Superfície de Resposta (RSM). Desta maneira, o trabalho descreve, de modo amplo e objetivo, os métodos implementados empre...

A new procedure with high ability to enhance prediction of multivariate calibration models with a small number of interpretable variables is presented. The core of this methodology is to sort the variables from an informative vector, followed by a systematic investigation of PLS regression models with the aim of finding the most relevant set of variables by comparing the cross-validation parameters of the models obtained. In this work, seven main informative vectors i.e. regression vector, correlation vector, residual vector, variable influence on projection (VIP), net analyte signal (NAS), covariance procedures vector (CovProc), signal-to-noise ratios vector (StN) and their combinations were automated and tested with the main purpose of feature selection. Six data sets from different sources were employed to validate this methodology. They originated from: near-Infrared (NIR) spectroscopy, Raman spectroscopy, gas chromatography (GC), fluorescence spectroscopy, quantitative structure-activity relationships (QSAR) and computer simulation. The results indicate that all vectors and their combinations were able to enhance prediction capability with respect to the full data sets. However, regression and NAS informative vectors from partial least squares (PLS) regression, both built using more latent variables than when building the model presented in most of tested data sets, were the best informative vectors for variable selection. In all the applications, the selected variables were quite effective and useful for interpretation.

No abstract

No abstract

Pectin was extracted from passion fruit peel using three different acids (citric, hydrochloric or nitric) at different temperatures (40-90°C), pH (1.2-2.6) and extraction times (10-90 min), with and without skins using a 2 4 factorial design. Temperature, pH and extraction time had highly significant effects on the pectin yield. A central composite design with face centring was used to optimise the extraction process conditions for citric acid without skins. Pectin yields varied from 10% to 70%. The optimal conditions for maximisation of pectin yield were the use of citric acid at 80°C and pH 1 with an extraction time of 10 min considering model extrapolation.

No abstract

a b s t r a c tThis study examined variables that influence the nitric acid-extraction of pectins from cacao pod husks. An initial screening study tested the main parameters influencing yield and uronic acid content by a factorial fractional 3 3−1 design. Additional studies used response surface methodology using a central composite design. Yield was optimized by increasing temperature. Uronic acid content was higher when pH is decreased and temperature is increased, or when pH is increased and temperature is decreased. To optimize both parameters, extraction conditions of pH 1.5 at 100 • C for 30 min provided a predicted yield and uronic acid content of approximately 9.5% and 80%, respectively. The obtained experimental value for the yield was 9.0 ± 0.4%, the pectin presenting 66.0% uronic acid. Chemical and spectroscopic analyses showed that this pectin is mainly a homogalacturonan highly esterified and acetylated (DE 56.6%; DA 17.1%) with some rhamnogalacturonan insertions.