**1. Introduction**

With an ancient history starting in the Preclassic period (1200–400 B.C.) with consumption among the Olmec and other pre-Colombian populations of the Americas [1], cocoa is now a ubiquitous food. Cocoa appears today in many different forms, mainly in chocolate, with consumption averaging around 8 kg per person per annum in many European countries [2]. The top four countries, which account for nearly 65% of the total world chocolate production, are the USA, Germany, Switzerland, and Belgium, which in terms of retail sales reach respectively USD 20, USD 10, USD 14, and USD 12 billion per year [3]. The consumption of cocoa and cocoa-based products is of great interest both for the highly appreciated sensorial profile and for the possible beneficial health effects which are being studied in current times [4,5].

Cocoa beans represent the essential raw material for chocolate, and they are obtained from the *Theobroma cacao* L. tree, which is almost exclusively cultivated in tropical/developing countries, where it represents a source of export earnings both at the level

**Citation:** Forte, M.; Currò, S.; Van de Walle, D.; Dewettinck, K.; Mirisola, M.; Fasolato, L.; Carletti, P. Quality Evaluation of Fair-Trade Cocoa Beans from Different Origins Using Portable Near-Infrared Spectroscopy (NIRS). *Foods* **2023**, *12*, 4. https:// doi.org/10.3390/foods12010004

Academic Editors: Mourad Kharbach and Samuli Urpelainen

Received: 23 November 2022 Revised: 15 December 2022 Accepted: 16 December 2022 Published: 20 December 2022

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

of families, communities, and nations [6]. Africa covers more than 75% of the world's total cocoa beans production and Côte d'Ivoire is the largest cocoa bean exporter in the world, with more than USD four billion exported in 2020, followed by Ghana and Ecuador.

Once the cacao pods have been harvested, they must undergo post-harvest operations on farms and plantations before becoming the so-called cocoa beans, which are then traded in the international market and processed into final industrial products. The post-harvest processing comprise pod opening and removal of beans from the pod, bean-pulp mass fermentation and bean drying. In this sequence, the fermentation constitutes an essential critical step for the development of desired flavor attributes of the commercial cocoa beans. In the further processing, cocoa beans are roasted, cracked and ground to give a powdery mass from which fat is expressed [7] and the release of fat ultimately leads to a liquid-like ingredient, namely cocoa liquor. Additionally, many processes are implemented in the chocolate industry that originate several products with different forms and functionalities.

Fermentation and drying constitute key farm(er)-based unit operations with strong influences in the final quality of cocoa beans and subsequent products [8].

Since 2010, approximately 4 million tons of cocoa beans have been produced annually around the world [9], and the three biggest importers are the Netherlands, Germany, and USA, with product worth 2,375,923, 1,209,366 and 1,026,931 USD imported in 2020, respectively (Sources: ITC calculations based on UN COMTRADE and ITC statistics).

The socio-economic importance and international interest towards cocoa beans are intelligible as an estimated five million farming households depend on cacao as a cash crop, and 70 per cent of cocoa is produced by smallholders living on less than USD 2 per day and relying on cocoa for 60 to 90 per cent of their income [9].

Cocoa beans are divided worldwide between "fine or flavor" (mainly Criollo and Trinitario) and "bulk" (mainly Forastero) varieties, with a common belief that fine and flavor cacao varieties receive significant price premiums in international markets. However, recent studies show how post-harvest processing has a central role in causing heterogeneity in cocoa prices, independently of the variety grown [10]. This is because although the primary factors influencing the quality attributes of cocoa beans are the cocoa tree cultivar and genotype, it is well-established that the agronomic and environmental conditions together with the harvest and post-harvest steps are crucial elements in the determination of the final quality of commercial cocoa beans [11–15].

Partially fermented or unfermented beans are prone to bitterness and astringency with poor chocolate flavor and aroma [2]. Moreover, an appropriate drying process will reduce the beans' water content of 55% to around 7% [16], preventing fermentation from continuing uncontrolled, slowing the development of molds that could give rise to unwanted, unpleasant flavors and equilibrating the beans acidity that would otherwise be excessive in the final products [17,18].

Although fermentation is considered as the "core stage" of the cocoa transformation process from seeds to chocolate, it is currently performed mostly by small third-world producers in an empirical way, with little or no technification, without control in processing conditions, originating cocoa batches of low and heterogeneous quality [13].

Chocolate and chocolate-based products sell in a very competitive market, where quality is crucial, and value is enormous. It should be clear that if the quality of cocoa beans is poor, final products will suffer this deficiency as well, and the whole industry sector will be affected [19].

Most of the existing commercial standards for cocoa beans base their quality requirements on the results of the cut test, or on sensory estimation by trained panels (Aculey et al., 2010). The cut test consists of cutting cocoa beans lengthwise, observing the number of defective beans. The ISO 1114 states that both halves of each bean shall be visually examined, and the result for each kind of defect shall be expressed as a percentage of the 300 beans examined. The ISO also defines nine categories of defects: those related to poor fermentation (slaty and violet/purple beans) and those being indicators of high FFA levels, poor flavor and/or other contaminants (bean clusters, broken beans, smoky beans, moldy beans, germinated beans, flat beans, insect-damaged/infested beans) [20].

On the one hand, the cut test is limited by its subjective nature and does not represent a sufficiently reliable methodology for a comprehensive description of the main quality contributors [21]. On the other hand, laboratory methodologies are often demanding in terms of time and cost effectiveness, which can be critical factors, and they are mainly based on destructive determinations. Moreover, in cocoa producing countries the availability of laboratory infrastructures is poor [22].

In this context, the application of near infrared (NIR) spectroscopy, particularly with the recent developments of portable NIR spectrometers, may represent a valuable solution for providing a cocoa beans' quality profile, in a rapid, non-destructive, and reliable way. This analytical technique could be useful to both cocoa bean producers, mostly in the developing countries, and processors, mostly in the developed countries, alike.

Many researchers have already investigated quality parameters of cocoa beans through NIR spectroscopy quantifying, for instance, fat, sugars, proteins, moisture, pH and titratable acidity, polyphenols, and other volatile and non-volatile compounds [23]. Even the assessment of the authenticity of cocoa powder has been studied by identifying the country of origin of raw materials, varietal purity, or the presence of adulterants [24].

It must be noted, however, that most of the studies that successfully predicted cocoa beans' quality parameters through NIR spectroscopy mainly analyzed samples which had been purposely subjected to different degrees of fermentation, e.g., analyzing the beans at different days during the fermentation process. Doing so, the samples are not in the status in which importers/exporters normally trade them in the international market, fully fermented and dried, therefore this might not be a representative "working condition" for these actors in the cocoa supply chain. Moreover, only cocoa bean samples coming from one or few countries are most often utilized in the previous studies, hence limiting the variability that can be included in the NIR prediction model. Finally, most of the studies present in literature utilize benchtop NIR instruments, which have some notable practical disadvantages if compared with the more recent portable NIR spectrometers, which are recently being considered in the literature [25,26].

On this basis, our study aimed to predict some quality parameters of commercial cocoa bean samples using portable NIR spectrometers, also in comparison with a benchtop spectrometer, on both whole and ground samples. Cocoa beans were provided by an Italian fair-trade importer and all the samples have been produced to meet internationally accepted merchantable quality standards: well-fermented and dry. The ultimate purpose of the study was, therefore, to evaluate the possibility of using portable NIR spectrometers in commercial contexts, while also assessing the practicability of using NIR spectroscopy on whole fermented coca beans to rapidly predict main quality parameters.

#### **2. Materials and Methods**

#### *2.1. Samples*

Fifty-six samples from commercially available cocoa beans have been provided by Altromercato Impresa Sociale Soc. Coop. (Via 9. Crispi, Bolzano, Italy), a major importer of fair-trade products in Italy in 2021. Thirty-three of them came from Africa while the remaining twenty-three came from South America. All the samples have been produced to meet internationally accepted merchantable quality standards i.e., well-fermented and dry, free from smoky beans and abnormal or foreign odors, free from evidence of adulteration, reasonably free from living insects, virtually free from broken beans, pieces of shell, and foreign matter. African countries included Togo, Uganda, Madagascar, and Sierra Leone, while American beans came from Honduras, Ecuador, Perú, Dominican Republic, Nicaragua, and Venezuela (Supplementary Table S1). Although it was not possible to obtain accurate information on the beans' variety for each sample, documents reported that only Trinitario and Forastero varieties were utilized.

The beans were shipped in 25-ton containers which roughly corresponds to 360 bags (70 kg of cocoa beans/bag). Around 4 kg of cocoa beans from each container were sampled as representative for the 33% of the total bags' number. Of these, 200 g of cocoa beans were randomly taken for analysis and stored at −20 ◦C in plastic bags until assay in a LGPv 8420 MediLine refrigerator (Liebherr, Kirchdorf an der Iller, Germany).

### *2.2. Beans Peeling and Grinding*

Three random aliquots of 20 g of dry and nitrogen frozen cocoa beans from each sample were de-husked by hand. The peels and nibs were carefully collected and weighted to determine average shell percentage on a 4 digits balance (Adventurer model ARRV70, OHAUS, Parsippany, NJ, USA). About 100 g of nibs (de-husked cocoa beans) were ground in a multi-purpose grinder for 45 s (3 intervals of 15 s with 10 s pause), then sifted on a 0.5 mm sieve. Before grinding, the beans were frozen with liquid nitrogen to make them brittle and avoid becoming a mash. The obtained cocoa powder for each sample was stored at −20 ◦C in the dark prior to the following analyses.
