Aplicação das técnicas de espectroscopia de infravermelho com transformada de Fourier (FTIR) nos intervalos médio (MIR) e próximo (NIR), para avaliação do conteúdo em nalcanos e álcoois de cadeia longa de amostras vegetais e fecais
Data
2019-02-20
Autores
Título da revista
ISSN da revista
Título do Volume
Editora
Resumo
O conhecimento dos fatores que influenciam a relação dinâmica entre o animal e a pastagem, em especial o comportamento alimentar dos animais revela-se fundamental para o desenvolvimento de estratégias adequadas de gestão dos herbívoros em pastoreio. Os nalcanos e os álcoois de cadeia longa (LCOH) têm sido utilizados como marcadores fecais para estimar diversos parâmetros nutricionais. No entanto, as suas análises químicas são morosas e dispendiosas, existindo, desta forma, a necessidade de se desenvolverem metodologias mais expeditas para avaliar as concentrações destes marcadores. Este trabalho teve como objetivo avaliar a utilização da tecnologia de Espectroscopia de Infravermelho com Transformada de Fourier (FTIR), no intervalo do infravermelho próximo (NIR) e no intervalo do infravermelho médio (MIR), para a determinação da concentração em n-alcanos e LCOH em amostras de distintas espécies vegetais e amostras fecais de herbívoros domésticos. Nesse sentido, foram registados, através destas técnicas, espectros de 33 amostras de alimentos (Trifolium repens, Erica spp., Lolium perenne, Ulex gallii, herbáceas, herbáceas do monte e gramíneas) e 181 amostras de fezes (bovinos e equinos). De modo a desenvolver calibrações para a previsão de concentrações n-alcanos e LCOH recorreu-se à análise estatística multivariada. Relativamente aos modelos desenvolvidos para as amostras vegetais, os melhores resultados foram observados para as calibrações com recurso ao NIR. Os melhores coeficientes de determinação de validação externa (R 2 v) obtidos foram de 0,90 e 0,79 para LCOH e nalcanos, respetivamente. Para as amostras fecais, no intervalo NIR, os resultados indicam previsões na validação externa (R 2 v) similares para as duas espécies (0,64). Pelo contrário, no intervalo MIR foram observadas diferenças de R 2 v entre amostras fecais de bovinos (0,70) e equinos (0,57). Em relação aos modelos criados para as fezes de ambas as espécies animais, observámos uma tendência para os LCOH e os n-alcanos presentes em maiores concentrações, serem os marcadores com melhores estimativas. Os resultados obtidos neste estudo sugerem que a seleção da técnica a utilizar poderá depender do tipo de matriz, sendo a homogeneidade das matrizes um dos fatores mais importantes para o seu sucesso. No sentido de melhorar a precisão e robustez dos modelos criados para as estimativas das concentrações destes marcadores utilizando estas metodologias, a base de dados (maior variabilidade) utilizada para as calibrações destes modelos terá que ser obrigatoriamente aumentada.
Understanding the factors that influence the dynamic relationship between animal and pasture, especially the grazing behaviour of the animals, is fundamental for the development of adequate management strategies of grazing herbivores. N-alkanes and long chain alcohols (LCOH) have been used as faecal markers to estimate various nutritional parameters. However, their chemical analyses are time-consuming and expensive, making it necessary to develop more expeditious methodologies to evaluate the concentrations of these markers. The objective of this work was to evaluate the use of Fourier Transform Infrared Spectroscopy (FTIR) technology in the near infrared (NIR) and mid-infrared (MIR) intervals, for the determination of the concentration of n-alkanes and LCOH, in samples of different plant species and faecal samples of domestic herbivores. In this sense, spectra of 33 food samples (Trifolium repens, Erica spp., Lolium perenne, Ulex gallii, herbaceous, grass herbaceous and grass) and 181 faecal samples (bovine and equine) were recorded using these techniques. In order to develop calibrations for the prediction of n-alkane concentrations and LCOH, a multivariate statistical analysis was used. Regarding the models developed for the plant samples, the best results were observed for the calibrations using the NIR. The best external validation determination coefficients (R 2 v) obtained were 0,90 and 0,79 for LCOH and nalkanes, respectively. For faecal samples, in the NIR range, the results indicate similar external validation predictions (R 2 v) for both species (0,64). However, in the MIR interval, differences of R 2 v were observed between faecal samples of cattle (0,70) and equines (0,57). Relative to the models developed for faeces of both animal species, it was observed that LCOH and n-alkanes presented in higher concentrations, were the markers with the best estimates. The results obtained in this study suggest that the selection of the used technique may depend on the type of matrix, and the homogeneity of the matrices is one of the most important factors for its success. In order to improve the accuracy and robustness of the created models, the database (greater variability) used for the calibrations of these models will have to be increased for the assessment of the concentrations of these markers using these methodologies.
Understanding the factors that influence the dynamic relationship between animal and pasture, especially the grazing behaviour of the animals, is fundamental for the development of adequate management strategies of grazing herbivores. N-alkanes and long chain alcohols (LCOH) have been used as faecal markers to estimate various nutritional parameters. However, their chemical analyses are time-consuming and expensive, making it necessary to develop more expeditious methodologies to evaluate the concentrations of these markers. The objective of this work was to evaluate the use of Fourier Transform Infrared Spectroscopy (FTIR) technology in the near infrared (NIR) and mid-infrared (MIR) intervals, for the determination of the concentration of n-alkanes and LCOH, in samples of different plant species and faecal samples of domestic herbivores. In this sense, spectra of 33 food samples (Trifolium repens, Erica spp., Lolium perenne, Ulex gallii, herbaceous, grass herbaceous and grass) and 181 faecal samples (bovine and equine) were recorded using these techniques. In order to develop calibrations for the prediction of n-alkane concentrations and LCOH, a multivariate statistical analysis was used. Regarding the models developed for the plant samples, the best results were observed for the calibrations using the NIR. The best external validation determination coefficients (R 2 v) obtained were 0,90 and 0,79 for LCOH and nalkanes, respectively. For faecal samples, in the NIR range, the results indicate similar external validation predictions (R 2 v) for both species (0,64). However, in the MIR interval, differences of R 2 v were observed between faecal samples of cattle (0,70) and equines (0,57). Relative to the models developed for faeces of both animal species, it was observed that LCOH and n-alkanes presented in higher concentrations, were the markers with the best estimates. The results obtained in this study suggest that the selection of the used technique may depend on the type of matrix, and the homogeneity of the matrices is one of the most important factors for its success. In order to improve the accuracy and robustness of the created models, the database (greater variability) used for the calibrations of these models will have to be increased for the assessment of the concentrations of these markers using these methodologies.
Descrição
Dissertação de Mestrado de Engenharia Zootécnica
Palavras-chave
Herbívoros domésticos , álcoois de cadeia longa , n-alcanos , espectroscopia de infravermelho , calibrações