Favoritar este trabalho
Como citar esse trabalho?
Resumo

Aiming to add objectivity to the quality control of roasted and ground coffee, done conventionally by sensorial analysis, we investigated the potential of using the near-infrared spectroscopy (NIR) combined with chemometrics. As a good quality profile, coffees awarded at sensory contest were analyzed, being 41 of raw and 52 of roasted coffee. In addition to these samples, we got some adulterants as wheat, soybeans, rye, corn, and we select black beans as defective grains. All samples were roasted, milled, sieved to 22 mesh and their moisture determined by halogen lamp thermocouple at 105 °C for 7 min. On average the samples presented 6% of moisture. After these, trigonelline and caffeine content were determined by primary technique using liquid chromatography with diode array detection (HPLC-DAD). To create a database for each type of sample all NIR spectra were acquired from 400 to 2400 nm using 2.5 g of sample directly in a spinning quartz cell. Thus, multivariate calibration was performed for moisture, trigonelline and caffeine, with NIR spectra processed using PLS, PCR and PCA. Experimental results for moisture and caffeine differed in less than 5% (p <0.05) considering predictive models compared to the values measured by primary methods, while the error was less than 10% (p <0.1) for trigonelline. In conclusion, NIR spectra were able to distinguish the raw and roasted coffee of the adulterants and black beans studied, splitting them into groups, endorsing the assessments made by the sensory analysis.

Compartilhe suas ideias ou dúvidas com os autores!

Sabia que o maior estímulo no desenvolvimento científico e cultural é a curiosidade? Deixe seus questionamentos ou sugestões para o autor!

Faça login para interagir

Tem uma dúvida ou sugestão? Compartilhe seu feedback com os autores!

Instituições
  • 1 UEL
Eixo Temático
  • ANA - Química Analítica
Palavras-chave
Sensorial analysis
NIR
Coffee quality control
adulterants
Defects