Green Coffee Beans Research Articles

Non-targeted profiling of aldehydes and ketones in coffee beans via isotope labelling-assisted UHPLC-Q-Orbitrap-MS and effects of roasting

This study aimed to investigate the aldehydes and ketones (AKs) composition in coffee beans, particularly assess the differences between green and roasted beans. Using a novel isotope-labelling-assisted liquid chromatography-high-resolution mass spectrometry strategy, 142 potential AKs were identified in coffee beans extracts. Notably, the presence of tri-carbonyl in roasted coffee beans is revealed for the first time. Compared to green beans (GB), roasted beans showed a much higher abundance of AKs, highlighting the crucial role of roasting in the formation of these compounds. The total content indicated that unsaturated aliphatic AKs were abundant in GB, whereas other groups of AKs, such as diketones, furfurals, and tri-carbonyls, were present at higher levels in roasted beans. Roasting significantly increased the contents of 2-methylbutyraldehyde, acetoin, acetol, pyruvic acid, 5-hydroxymethylfurfural, methylglyoxal, 2,3-butanedione, 2,3-pentanedione, and 2,3-hexanedione, which were 4–83 times more than those in GB. Some of them are intermediates for subsequent reactions, since their content decreased with increasing roasting degrees. Moreover, 9 AKs were selected as potential markers for distinguishing between green and roasted coffee beans and could be used to monitor the degree of roasting. Overall, this study offers valuable insights for understanding the composition and changes in AKs during coffee roasting.

Implementing a deep learning model for defect classification in Thai Arabica green coffee beans

Arabica coffee is a significant economic driver in Northern Thailand and has substantial opportunities for market growth. However, the Thai coffee business must ensure consistent quality standards and is currently heavily dependent on manual labor, to first identify, and then remove substandard unroasted coffee beans. This research developed a classification model based on a Convolutional Neural Network to detect 17 types of defects in green coffee beans. The image augmentation phase was enhanced by rotating images at 45, 90, 135, 180, 225, and 270° and expanding the dataset from 979 original images across 17 defect types to a robust 6,853 images. Several architectures including MobileNetV2, MobileNetV3, EfficientNetV2, InceptionV2, and ResNetV2 were assessed. Following extensive evaluations, MobileNetV3 emerged as the best-performing model and underwent further fine-tuning, achieving significant accuracy improvements through hyperparameter optimization. The model's robustness and generalizability were validated via 5-fold cross-validation, with accuracy ranging from 98.78 % to 99.84 % across all defect types. When tested with unseen data, the model achieved an accuracy of 88.63 %. A web application prototype was also developed for real-time coffee bean defect classification and its usability was tested. Seven farmers reported high satisfaction with the ease of use and effectiveness of the application in classifying coffee bean defects, with 71.4 % expressing a strong likelihood of recommending the application to others. These promising results demonstrate the practical utility of the model in enhancing quality sorting processes in the coffee industry.

Occurrence of ochratoxins in coffee and risk assessment of ochratoxin a in a Costa Rican urban population

Costa Rica is a coffee producer and consumer country, but this product is prone to ochratoxin contamination; therefore, this study aims evaluates the human health risk associated with ochratoxin exposure among coffee consumers in the Costa Rica. Ochratoxin A (OTA) is a nephrotoxic compound classified as a Group 2B carcinogen, produced by the fungi Aspergillus section Circumdati, Aspergillus section Nigri and Penicillium spp. The presence of OTA and ochratoxin B (OTB) in Costa Rican coffee products (n = 175) was determined by HPLC with fluorescence detection. OTA was detected in 58.2% of the green coffee beans (1.01 ± 0.85 ng g −1), in 36.8% of the pure roasted coffee (2.59 ± 4.41 ng g −1), in 23.1% of the sugar-added roasted coffee (1.59 ± 0.33 ng g−1) and 75% of the instant coffee samples (0.69 ± 0.58 ng g−1). The contamination with OTB was 45.5% (1.28 ± 0.83 ng g −1), 31.6% (1.60 ± 2.04 ng g −1), 30.8% (1.42 ± 0.86 ng g −1), and 41.7% (2.64 ± 2.07 ng g −1), respectively. The dietary exposure to OTA of the Costa Rican population was assessed by a probabilistic approach. The mean estimated daily intake (EDI) of OTA from coffee was: 0.184 (90% IC: 0.179–0.189) ng kg−1 bw day−1 for the total population (0.189 [90% IC: 0.184–0.194] ng kg−1 bw day−1 for males and 0.181 [90% IC: 0.176–0.186] ng kg−1 bw day−1 for females). The EDIs were lower than the tolerable human intake benchmarks for OTA set by international food safety authorities (even though more than 80% of OTA is extracted during coffee preparation). The results evidence a low risk (related to ochratoxin) for coffee consumers in Costa Rica.

Comparative analysis of YOLO models for green coffee bean detection and defect classification

The quality and uniformity of green coffee beans significantly influence the overall flavor and value of the product. In the coffee industry, automated flaws and bean-type identification offer numerous advantages. This study examines the effectiveness of multiple YOLO (You Only Look Once) models for identifying and classifying green coffee beans. Various YOLO variants, including YOLOv3, YOLOv4, YOLOv5, YOLOv7, YOLOv8, and custom models, are compared with a focus on computational efficiency, accuracy, and speed. Utilizing a dataset of 4,032 training and 506 testing images encompassing diverse bean types, defects, and lighting conditions, we assessed the performance. The bounding boxes generated by our models accurately encompass coffee beans, with the background typically uniform and set to black. Our analysis reveals the superior performance of the custom-YOLOv8n model, which we meticulously customized for green coffee bean detection. This model achieved high precision, recall, f1-score, and mAP, demonstrating its potential for real-world implementation in coffee bean quality control systems. The customization process involved fine-tuning the model to focus on significant features relevant to green coffee bean detection and employing specific labeling strategies. Customization allows you to fine-tune the model to focus on important features relevant to green coffee bean detection. This sensitivity ensures that the model can effectively distinguish between different bean types and detect even subtle defects. This paper clarifies our primary objective of evaluating YOLO models’ performance in identifying and categorizing green coffee beans, with potential implications for enhancing efficiency and consistency in the coffee industry. A succinct key sentence underscores the benefits of our research for readers seeking efficient YOLO model selection and implementation in agricultural systems.

Development of an analytical solution to diffusion equation for multidimensional solids of arbitrary shape

Mass transfer processes between solids and fluids are ubiquitous in the food engineering field and using simple analytical models for their simulation or estimation of mass diffusivities remains widespread. Unfortunately, these solutions are available only for a limited number of simple solid geometries which might not be applicable to all food shapes; thus, this study introduces a simple method to obtain the volume average concentration in multidimensional solids of arbitrary shape (SAS). The proposed method approximates the (analytical or numerical) solution of the SAS from the analytical solution of a properly sized box by minimizing a weighted similitude index between the original and box shapes. Besides, the method was generalized to consider the isotropic shrinkage of foods (that is, size reduction while maintaining the aspect ratio). The applicability of the equivalent box approach was exemplified by solving inverse problems for the estimation of caffeine diffusivity during aqueous extraction of green coffee beans and water diffusivity during lentil drying using data available from literature. The results were compared with those obtained by the finite element solution of the 3D mass transfer model using the real product shape. Mass diffusivities for caffeine in green coffee and for water in lentils estimated with the equivalent box approach were not statistically different (p>0.05) to those estimated by numerically solving the 3D mass transfer model under the same assumptions. This method represents a simple and reliable way to model mass transfer in complex-shaped foods.

Edible Alginate-Lecithin Films Enriched with Different Coffee Bean Extracts: Formulation, Non-Cytotoxic, Anti-Inflammatory and Antimicrobial Properties.

The aim of this study was to analyze the functional properties of newly obtained films based on sodium alginate and lecithin with the addition of antioxidant-rich coffee extracts and to verify their potential as safe edible food packaging materials. In our study, we developed alginate-lecithin films enriched with green or roasted coffee bean extracts. The roasting process of coffee beans had a significant impact on the total phenolic content (TPC) in the studied extracts. The highest value of TPC (2697.2 mg GAE/dm3), as well as antioxidant activity (AA) (17.6 mM T/dm3), was observed for the extract of light-roasted coffee beans. Films with the addition of medium-roasted coffee extracts and baseline films had the highest tensile strength (21.21 ± 0.73 N). The addition of coffee extract improved the barrier properties of the films against UV light with a decrease in the transmittance values (200-400 nm), regardless of the type of extract added. Studies on Caco-2, HepG2 and BJ cells showed that digestated films were non-cytotoxic materials (100-0.1 μg/cm3) and had no negative effect on cell viability; an increase was noted for all cell lines, the highest after 48 h in a dose of 1 μg/cm3 for a film with medium-roasted coffee (194.43 ± 38.30) for Caco-2. The tested films at 20% digestate concentrations demonstrated the ability to reduce nitric oxide (NO) production in the RAW264.7 cell line by 25 to 60% compared to the control. Each of the tested films with coffee extracts had growth inhibitory properties towards selected species of bacteria.

Evaluating the antiviral efficacy and specificity of chlorogenic acid and related herbal extracts against SARS-CoV-2 variants via spike protein binding intervention

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and influenza A virus infect the respiratory tract through surface proteins, causing similar symptoms. Since the onset of the COVID-19 pandemic, both viruses have posed significant ongoing global health threats. Like the influenza virus, SARS-CoV-2 evolves into variants that can reduce vaccine efficacy. Thus, herbal medicines are being explored as supplementary options to enhance protection against these infections. This study aimed to investigate the therapeutic potential of chlorogenic acid (3-CQA) and related extracts from green coffee beans and Echinacea purpurea against SARS-CoV-2 variants and H1N1 infection. The methods employed included an ELISA-based trimeric spike protein binding assay, viral infection assays, plaque assays, and molecular docking studies. Results showed that 3-CQA blocked spike protein/angiotensin converting enzyme 2 (ACE2) binding for most SARS-CoV-2 variants of concern, except the Delta variant. Extracts from Green coffee bean and E. purpurea effectively blocked all variants tested. Additionally, antibodies blocked spike protein binding up to Omicron BA.2. Molecular docking suggested that 3-CQA binding to Omicron BA.1, BA.2, and BA.4, though not to the Delta spike protein, may lead to steric hindrance, preventing receptor-binding domain interactions with ACE2. Finally, both 3-CQA and E. purpurea extract showed preventive effects against H1N1 viral infection, though with lower potency compared to SARS-CoV-2. In conclusion, 3-CQA has potential as a phytoconstituent marker for herbs with bioactive properties against SARS-CoV-2 and H1N1 viral infections.

Microstructural Modification and Sorption Capacity of Green Coffee Beans.

To enhance the pore structure of green coffee beans (GCB) and detect the sorption capacity and extraction characteristics of flavor compounds before roasting, this study employed several methods: hot air drying (HD), freeze-drying (FD), 3-levels short-time heating with puffing (SH1P, SH2P, and SH3P), and 3-levels microwave with puffing (MW45P, MW60P, and MW75P). These methods were applied to GCBs pre-soaked in water for different times. The effects of these treatments on color change, porosity, microstructure, citric acid sorption capacity, and caffeine and chlorogenic acid extraction yield were investigated. Results indicated that, except for GCBs treated with SH1P, SH2P, SH3P, and MW75P, all other modified GCBs showed minimal color change. GCBs treated with MW60P exhibited favorable pore structures. MW60P treatments significantly improved the extraction yield of caffeine and chlorogenic acid. Furthermore, the increased porosity and improved pore size distribution of GCB after MW60P resulted in a significant increase in the sorption of citric acid onto modified GCB. The rate of the sorption reaction followed the pseudo-first-order kinetics. In conclusion, MW60P are effective treatments for enhancing the microstructure of GCB, improving sorption capacity, and improving the extraction yield of flavor compounds.

Alternative Method for Glyphosate Determination in Unroasted Green Coffee Beans by Liquid Chromatography Tandem Mass Spectrometry (LC-MS/MS).

This research proposes an alternative method to detect and quantify glyphosate residues in unroasted green coffee beans by LC-MS/MS. The sample preparation was conducted without derivatization steps, with integrated cleanup, which improves the analytical method's frequency. Validation results were consistent with the requirements of the regulatory guidelines employed. Specificity, linearity (r2 = 0.9991), precision (RSD ≤ 9%), and recovery (92-112%) were ensured, with a satisfactory limit of quantification (LOQ = 0.48 mg kg-1). These data demonstrate that the method is suitable for monitoring glyphosate residues in unroasted coffee beans while offering simplicity and speed in sample preparation. The method was applied to analyze authentic unroasted coffee bean samples, in which two of them were contaminated with glyphosate (<LOQ). These results exhibit the importance of glyphosate monitoring in coffee samples and enhance that the method can be successfully implemented as a tool to guarantee food quality and safety.

Geographic origin characterization of Brazilian green coffee beans via untargeted metabolomics

Coffee is a widely popular beverage worldwide, known for its distinct sensory properties which are greatly affected by geographical origin. Herein, we performed an untargeted metabolomic evaluation of green coffee beans (n = 40) from four different regions in Brazil: Cerrado Mineiro, Sul de Minas, Caparaó, and Mogiana Paulista; by using UHPLC-HRMS (ultra-high performance liquid chromatography coupled with high-resolution mass spectrometry). The most significant metabolites responsible for coffee characterization were theobromine, zeatin, phenylacetaldehyde, 2-acetyl-1-pyrroline, chlorogenic acids, ferulic acid, p-coumaric acid, abscisic acid, and jasmonic acid. Our findings demonstrate that the green coffee cultivated in Cerrado Mineiro, the most valuable among the four samples evaluated, exhibits a unique and typical metabolite profile, setting it apart from the coffee beans grown in other regions. Finally, the findings reported may be relevant for coffee producers in the Cerrado Mineiro area, as they contribute to establishing a certificate of origin for their high-quality product.

Green Coffee Bean Quality Inspection Using Computer Vision Systems

Green Coffee Bean Quality Inspection Using Computer Vision Systems

Evaluation of the Effect of Different Doses of X-Ray Irradiation on the Physicochemical and Microbiological Profiles of Liberica Green Coffee Beans

Liberica coffee has significant potential for cultivation and trade in Indonesia. The coffee trade requires effective post-harvest processing to maintain commodity quality. X-ray irradiation offers several advantages as a post-harvest processing method. However, research on X-ray irradiation remains limited. This study was designed to determine the effect of different doses of X-ray irradiation on the microbiological and physicochemical characteristics of Liberica green coffee beans and to identify the optimal dose for treatment. The research employed a randomized block design (RBD) method with five different doses of X-ray irradiation: 0 kGy, 2.7 kGy, 5.4 kGy, 7.2 kGy, and 10.8 kGy. The results indicated that varying doses of X-ray irradiation had no discernible effect on the color, water, protein, and fat content of Liberica green coffee beans. However, differences in the X-ray irradiation dose significantly affected the degree of acidity (pH), caffeine content, total sugar, antioxidants, and phenol levels in the beans. Additionally, there was a decrease in the Total Plate Count (TPC) with increasing doses of X-ray irradiation. The best treatment achieved was at an irradiation dose of 7.2 kGy.

Co-Encapsulation of Coffee and Coffee By-Product Extracts with Probiotic Kluyveromyces lactis.

Coffee and coffee by-products contain several chemical compounds of great relevance, such as chlorogenic acid (CGA), trigonelline, and caffeine. Furthermore, yeasts have been the target of studies for their use as probiotics because of their interesting biochemical characteristics. The combined administration of probiotic microorganisms with components that provide health benefits mediated by alginate encapsulation is an alternative that ensures the stability of cells and chemical compounds. In this context, the aim of this work was to co-encapsulate the probiotic yeast Kluyveromyces lactis B10 and extracts of green coffee beans, coffee silverskin, and PVA (black, green or immature, and sour coffee beans). The bioactive composition, antioxidant and antimicrobial activities of the extracts, microcapsule morphological characteristics and encapsulation efficiency, ability of the encapsulation to protect the yeast cells subjected to gastrointestinal conditions, and antioxidant activity of the microcapsules were evaluated. All the evaluated extracts showed antioxidant activity, of which PVA showed 75.7% and 77.0%, green coffee bean showed 66.4% and 45.7%, and coffee silverskin showed 67.7% and 37.4% inhibition of DPPH and ABTS•+ radicals, respectively, and antimicrobial activity against the pathogenic bacteria E. coli, Salmonella, and S. aureus, with high activity for the PVA extract. The microcapsules presented diameters of between 1451.46 and 1581.12 μm. The encapsulation efficiencies referring to the yeast retention in the microcapsules were 98.05%, 96.51%, and 96.32% for green coffee bean, coffee silverskin, and PVA, respectively. Scanning electron microscopy (SEM) showed that the microcapsules of the three extracts presented small deformations and irregularities on the surface. The K. lactis cells encapsulated in all treatments with the extracts showed viability higher than 8.59 log CFU/mL, as recommended for probiotic food products. The addition of green coffee bean, coffee silverskin, and PVA extracts did not reduce the encapsulation efficiency of the alginate microcapsules, enabling a safe interaction between the extracts and the K. lactis cells.

Coffee and mineral oil hydrocarbons: potential dietary intake

Levels of mineral oil hydrocarbons were measured in a large range of green and roasted coffee beans or ground powder. To better understand the consumer exposure to mineral oil hydrocarbons, the transfer to the brewed coffee was assessed under three different preparations. As a result, less than 5% of mineral oil hydrocarbons were transferred to the cup. With this low transfer rate, the coffee contribution to the mineral oils daily intake can be assessed to be very low, below 0.8% of the total exposure.

Potential Risk of Caffeoylquinic Acids, the Main Polyphenol Components in Coffee, on the Health of Piglets.

As the main coffee polyphenols, caffeoylquinic acids (CQAs) are abundant in coffee-derived products and have the potential to act as novel feed additives for animals. However, research on the side effects of dietary CQAs supplementation is scarce, especially in young animals. Here, we explore the safety of CQAs derived from green coffee beans. Results showed that ingesting 50, 125, 250, and 500 mg/kg of dietary CQAs for 55 days is associated with greater final body weight, average daily gain, and feed efficiency in piglets compared with the control group (P < 0.05). CQAs also increased the apparent digestibility of dry matter, crude protein, and gross energy at a dose over 50 mg/kg (P < 0.05). Interestingly, CQAs supplementation with 500 mg/kg increased the white blood cell count (P < 0.05). Moreover, CQAs supplementation at a dose over 50 mg/kg decreased the serum total cholesterol concentration but increased the immunoglobulin M level in serum (P < 0.05). Importantly, CQAs supplementation had no side effects on organ histopathology and organ weight (P > 0.05). These results suggest that CQAs could serve as a secure and effective additive to improve growth performance without negatively affecting the organs of piglets.

Non-linear machine learning coupled near infrared spectroscopy enhanced model performance and insights for coffee origin traceability

Over the past decade, there has been overwhelming interest in rapid and routine origin tracing and authentication methods, such as near infrared (NIR) spectroscopy. In a systematic and comprehensive approach, this study coupled NIR with advanced machine learning models to explore the origin classification of coffee at various scales (continental to regional level). Speciality green coffee beans were sourced from three continents, eight countries, and 22 regions. The dispersive bulk NIR spectra were used for spectral registration in the reflectance mode, and the obtained spectra were preprocessed with extended multiplicative scatter correction and mean centering. The classical linear partial least squares-discriminant analysis adequately predicted origin at the continental and country level, and showed promise at the regional level. Non-linear machine learning models improved predictions further, with the best accuracy found using random forest with accuracies up to 0.99. Discriminating wavelength regions and constituents were identified at each origin scale, with more minor wavelength regions selected by random forest. This proof of concept work demonstrated the potential of NIR spectroscopy coupled with machine learning for rapid origin classification of coffee from the continental to the regional level.

Evaluating the Impact of Green Coffee Bean Powder on the Quality of Whole Wheat Bread: A Comprehensive Analysis.

The current investigation focuses on the effect of different concentrations of green coffee bean powder (GCBp) on the physicochemical, microbiological, and sensory characteristics of whole wheat bread (WWB). C1 bread formulation (containing 1% GCBp) exhibited the highest loaf volume, suggesting optimal fermentation. Moisture analysis revealed minor alterations in the moisture retention attributes of the bread formulations. Impedance analysis suggested that C1 exhibited the highest impedance with a high degree of material homogeneity. Swelling studies suggested similar swelling properties, except C5 (containing 5% GCBp), which showed the lowest swelling percentage. Furthermore, color and microcolor analysis revealed the highest L* and WI in C1. Conversely, higher concentrations of GCBp reduced the color attributes in other GCBp-containing formulations. FTIR study demonstrated an improved intermolecular interaction in C1 and C2 (containing 2% GCBp) among all. No significant variation in the overall textural parameters was observed in GCBp-introduced formulations, except C2, which showed an improved gumminess. Moreover, the TPC (total phenolic content) and microbial analysis revealed enhanced antioxidant and antimicrobial properties in GCBp-incorporated formulations compared to Control (C0, without GCBp). The sensory evaluation showed an enhanced appearance and aroma in C1 compared to others. In short, C1 showed better physicochemical, biological, and sensory properties than the other formulations.

Production, proximate, antioxidants and sensory properties of bread incorporated with Green Coffee (&lt;i&gt;Arabica&lt;/i&gt; spp) Beans powder

Production, proximate, antioxidants and sensory properties of bread incorporated with Green Coffee (&lt;i&gt;Arabica&lt;/i&gt; spp) Beans powder

Peningkatan cita rasa kopi robusta terdekafeinasi melalui fermentasi ulang menggunakan mucilage tiruan

Some consumers prefer coffee's flavor but not caffeine's adverse effects. Therefore, a decaffeination method is used to reduce the amount of caffeine in coffee. However, decaffeination can decrease the flavor of the coffee. Accordingly, this study aimed to examine the influence of re-fermentation utilizing imitative mucilage from sweet potato and passion fruit pulp on the taste of decaffeinated coffee. The study's Robusta coffee green bean samples were treated: non- decaffeination and re-fermentation; non-decaffeination and re-fermententation; decaffeination and fermentation; and fermentation and decaffeination. Each sample was roasted and evaluated for flavor using the cupping test, and tests to green beans were also conducted for total acid, total dissolved solids, pH, phenolic, and volatile compounds. The results of this study indicate that decaffeination can reduce the taste of robusta coffee, as seen from the decreased cupping test results. However, decaffeinated coffee beans re-fermented with imitative mucilage exhibited increased cupping test values, total phenols, and volatile compounds. The decaffeination and further fermentation treatments decreased the total acid and dissolved solids. In contrast, the re-fermentation treatment impacted the increase in pH, total phenolic, and volatile compounds

From beverage to anticancer agent: The repurposing of green coffee bean extract loaded in solid lipid nanoparticles

Green coffee bean extract (GCBE) captivated the world with its various health benefits. However, its topical use was limited by the hampered penetration of hydrophilic GCBE through the skin. So, this study aimed to incorporate GCBE-loaded solid lipid nanoparticles (SLNs) into water-in-oil (W/O) cream, to explore its repurposed application as an anticancer drug. The anticancer activity was assessed in vitro using MTT assay against human breast cancer cell lines (MCF-7) and in vivo using the Ehrlich ascites carcinoma (EAC) model in mice compared to free GCBE followed by immunohistochemical analysis of caspase-3 and estrogen receptor. The optimized GCBE-SLNs results of the MTT assay exhibited a concentration-dependent reduction in the viability of MCF-7 with approximately 36 times more toxic effect against MCF-7 compared to free GCBE. The results of the in vivo Ehrlich ascites carcinoma model revealed that GCBE-SLNs cream exhibited statistically significant higher expression of caspase-3, lower expression of estrogen receptor, and increased necrotic index compared to the control group. Conclusively, GCBE-SLNs cream seems to have reassuring promises as anticancer remediation in both in vivo and in vitro studies. The present work is considered the first pre-clinical evidence for the potential merits of GCBE-SLNs in cancer management.