388: Fourier Transform Infrared Spectroscopy Combined With Chemometrics for Rapid Estimation of Brewers’ Spent Grain Proximate Analysis

Introduction

Brewer’s spent grain (BSG) is a major byproduct of breweries. Although currently used as an animal feed, dried BSG is an excellent source of fibers (~60%) and proteins (~20%). There is a growing interest in exploring the application of BSG in food formulations due to its nutritional value. However, determination of its chemical composition by traditional wet chemistry techniques is time consuming, labor intensive, and expensive. In this study, we investigated the application of Fourier Transform Infrared (FTIR) spectroscopy in combination with chemometrics for quantifying protein, carbohydrates, and lipids in BSG.

Methods

A total of 55 BSG samples received from local breweries around Columbus, OH. The amount of protein, carbohydrates, and lipids were determined using wet chemistry methods. The samples were scanned in a wavenumber range of 4000 to 400 cm-1 using a portable FTIR spectrometer (model 4500 series, Agilent, Santa Clara, USA). The absorbance spectra from the scans were examined using a supervised classification technique (soft independent modeling of class analogy, SIMCA). The calibration models were developed using partial least squares regression (PLSR).

Results

Prediction models were developed for protein, carbohydrates, and lipids. The values obtained from the calibration models were comparable to those obtained with the wet chemistry methods. The developed models had a very low standard error of cross validation (SECV) between 1% and 5%. Similarly, the coefficient of determination (R2) varied from 0.90 to 0.95 and this explained most of the variance in predicated values. These findings suggest that the FTIR spectroscopy in combination with chemometrics can provide reliable and fast results (within a few seconds) to estimate BSG composition.

Significance

The studied FTIR spectroscopy is simple to use for the rapid determination of BSG macronutrients. This method eliminates the need for any special sample preparation in addition to being simple to use. The quality of prediction model needs to be periodically validated to check the accuracy of values. Demand for upcycled food products is growing and BSG is starting to be used in formulations of food. The FTIR technique will complement the food industry or third-party laboratories involved in a successful incorporation of BSG in food products.

Authors: Veeramani Karuppuchamy; Shreya Nuguri; Luis Rodriguez-Saona; Osvaldo Campanella