Authentication of roasted and ground coffee samples containing multiple adulterants using NMR and a chemometric approach

Highlights

• A fast, eco-friendly, and easy method for determination of coffee adulterants.

• Quantification of six different adulterants in roasted coffee using ¹H NMR.

• Chemometric tools assist in identifying adulterated coffee samples.

• PCA differentiation between pure and adulterated roasted coffee samples.

Abstract

Brazil is still the world's largest producer and exporter of coffee. In order to maximize profits, some producers may add lower cost materials (such as corn, barley, or even coffee husks) to commercial coffee. In view of the growing market for coffee products and the importance of coffee for the Brazilian economy, it is necessary to have a rapid, simple, and reliable methodology to identify and quantify coffee adulterants. NMR has proved to be a versatile and robust tool for the identification of adulterants in foods and beverages. Here, we explore the versatility of ¹H NMR assisted with chemometric tools, avoiding laborious data analysis, for the quantification of coffee adulteration. Six different adulterants were considered: barley, corn, coffee husks, soybean, rice, and wheat. The NMR-based methodology described here provided satisfactory LOD values (0.31–0.86%) for adulterants in medium and dark roast coffees. The statistical techniques PCA and SIMCA were employed for pattern recognition and the identification of pure and adulterated samples. Use of the SIMCA model enabled 100% correct classification for both training and prediction sets, ensuring the accuracy, traceability, and reliability of the results.

Introduction

Coffee is one of the most widely consumed beverages worldwide, with social aspects including the provision of hospitality and welcoming environments. The moderate consumption of coffee has been reported to provide several benefits to human health, such as the prevention of Alzheimer's and Parkinson's diseases, decreased risk of developing tumors, increased capacity for concentration, and decrease of fatigue (ABIC, 2009a; Alves, Casal, & Oliveira, 2009; George, Ramalakshmi, & Rao, 2008). Coffee consumption has also been suggested to reduce the risk of developing type 2 diabetes (Carlstro & Larsson, 2018).

World trade in coffee beans is worth billions of dollars, with Brazil being the largest producer of coffee and the greatest exporter of the raw beans (International Coffee Organization, 2019). In the year 2018, Brazilian producers exported almost 2.13 million tons, generating revenue of US$5.14 billion. Brazilians drink the equivalent of 5.1 kg per capita of roasted coffee annually, with consumption having grown over the years (Matos, 2018). Along with the growth of the coffee market, the consumer demand for higher quality coffee has also increased.

Due to its high commercial value, unfortunately, coffee is often a target of food fraud, which is an issue that receives constant attention by the global media. By definition, food fraud and adulteration are the deliberate substitution, addition, adulteration, or misrepresentation of food or food ingredients, for economic gain (FAO & WTO, 2017). There are innumerable examples of fraud involving food products (Tibola, da Silva, Dossa, & Patrício, 2018) with different matrices (Callao & Ruisánchez, 2018; Hong et al., 2017; Moore, Spink, & Lipp, 2012; Riedl, Esslinger, & Fauhl-Hassek, 2015; Ropodi, Panagou, & Nychas, 2016). Coffee is one such product that deserves special attention.

The value of the coffee market, in Brazil and elsewhere, means that it is necessary to impose strict regulation of coffee products. In Brazil, the National Health Surveillance Agency (ANVISA) has established a maximum permissible limit of 1% for the content of foreign substances in coffee (ANVISA, 1999). Several techniques are available for the detection of contaminants in coffee samples (Toci, Farah, Pezza, & Pezza, 2016), although few have sufficient versatility and robustness to be able to precisely identify and quantify different adulterants employed in coffee fraud. A private agency in Brazil, called the Brazilian Association of Coffee Industries (ABIC) has, since 1989, provided the “ABIC purity stamp”, which is awarded to products that contain only coffee in their composition (ABIC, 2009b). Although the “ABIC purity stamp” is not mandatory for coffee marketed in Brazil, its presence on a coffee label should be a sign of good production practices. However, the technique used by ABIC to determine coffee adulteration is outdated and subject to operator error (Amboni, de, de Francisco, & Teixeira, 1999; de Menezes Jr. & Bicudo, 1951, pp. 13–47). The authentication of coffee analysis methodologies should be performed, in order to ensure the reliability, traceability, and comparability of results (López, Callao, & Ruisánchez, 2015).

Nuclear magnetic resonance spectroscopy (NMR) is a promising analytical tool for the identification of different types of food and beverage adulteration (Consonni & Cagliani, 2010; Hachem et al., 2016; Hatzakis, 2019; Hu et al., 2015). It remains underutilized for this purpose, but has already been used for the authentication of oils, cereals, grains, alcoholic beverages, and fruit juices (Hong et al., 2017). This technique provides information about the structure and chemical composition of the major constituents of the sample (Tavares & Ferreira, 2006). In the case of coffee, NMR has been used to identify the origin of the coffee (Consonni, Cagliani, & Cogliati, 2012), to differentiate between Coffea canephora and Coffea arabica (Gunning et al., 2018; Monakhova et al., 2015; Schievano, Finotello, De Angelis, Mammi, & Navarini, 2014), and to elaborate a coffee fingerprint (Toci et al., 2018). Major advantages of NMR are that it only requires a small amount of sample, with very simple pretreatment. It avoids the production of toxic waste and is a nondestructive technique (de Moura Ribeiro, Boralle, Redigolo Pezza, Pezza, & Toci, 2017). Compared to other methods available, ¹H NMR offers simplicity and rapid analysis.

Chemometrics have been successfully used together with NMR for the analysis of foods, for example in the discrimination of beers (da Silva et al., 2019), determination of the fat content in powered milk (Nascimento et al., 2017), and detection of peanut oil adulteration (Zhu, Wang, & Chen, 2017), demonstrating the benefits that can be achieved by combining these tools. With the assistance of chemometric treatment, it is possible to analyze large amounts of analytical data in less time.

The study described in this paper was undertaken to develop a methodology based on NMR combined with chemometric tools to enable the identification of adulterated samples and the quantification of six different coffee adulterants. The chemometric tools selected to improve the reliability of the results were principal component analysis (PCA) and soft independent modeling of class analogies (SIMCA). The proposed ¹H NMR methodology was successfully applied to several commercial ground coffee blends from different origins and with different degrees of roast, without the need for any previous separation procedures, providing a robust method that can be easily employed on a daily basis in routine analysis, with reliable and accurate results.