Loading...
Please wait, while we are loading the content...
Similar Documents
Identificação e quantificação de adulterantes em whey protein concentrado empregando espectroscopia no infravermelho próximo e resolução multivariada de curvas
| Content Provider | Semantic Scholar |
|---|---|
| Author | Ravazzi, Caroline Gomes |
| Copyright Year | 2019 |
| Abstract | Whey protein is one of the most popular dietary supplements worldwide. It corresponds to a complex mixture of globular proteins extracted from raw milk during cheese manufacturing. In this dissertation, Near Infrared Spectroscopy (NIRS) and Near Infrared Chemical Imaging (NIR-CI) techniques were evaluated as strategies to identify and to quantify adulterants in whey protein concentrate (WPC) samples. Empirical models for quantification were constructed using Multivariate Curve Resolution (MCR-ALS) method. Adulterations were carried out through the individual addition of starch, wheat, soybean, corn, rice and urea to WPC in the concentration range of 1 to 30% (w/w). Reflectance spectra of adulterated samples were obtained using two instruments: a benchtop spectrometer (Spectrum 100N FT-NIR Perkin Elmer) and a miniaturized portable instrument (DLP NIRscan Nano Texas Instruments). WPC samples were also adulterated trough the addition of caffeine in the concentration range of 0.1 to 10% (w/w) and analyzed using an imaging system (Spotlight 400N FT-NIR Imaging System, Perkin-Elmer). MCR-ALS models for quantification presented absolute errors lower than 6% for all adulterants evaluated and recovered spectra showed high correlation with reference spectra. Randomization tests were applied to compare the performance between MCR-NIRS-Benchtop and MCR-NIRSPortable models. Despite of the reduced resolution and narrower spectral range, the data from portable instrument generated models with predictive capacity equivalent to the benchtop instrument models. In addition, NIR-CI proved to be more efficient than NIRS for analyzing samples at lower concentration ranges (1-10% w/w), making it possible to identify the presence of caffeine at concentration levels of 0.5%. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.repositorio.unicamp.br/bitstream/REPOSIP/334286/1/Ravazzi_CarolineGomes_M.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |