Roasted Coffee Beans Research Articles

Stability and degradation of chlorogenic acids in green and roasted coffee beans during long-term storage.

The study estimated the shelf life of specialty coffee packaged in six types of packaging (Tocuyo bag (TB), Double-bilaminate foil and aluminuim bag (DFAB), Ecotac vacuum bag (EV), Pressed cardboard box (PCB), Double-laminated bag without valve, with opening and zipper (DBOZ), Double-laminated bag with degassing valve and zipper (DBDVZ) and Triple-laminated bag with degassing valve and zipper (TBDVZ)). The estimation of shelf life was conducted by means of cup scores provided by six coffee tasters for coffee stored at 40, 50, and 60 °C. The Arrhenius equation was employed to obtain accelerated models for predicting shelf life. It was determined that green coffee beans are most effectively preserved in DBOZ, maintaining their freshness for a period of up to 55.13 days. The second-best option was EV, which has a shelf life of up to 35.21 days. The sole packaging alternative that was subjected to testing for roasted coffee beans was found to allow for their preservation for a period of up to 32 days. However, for roasted and ground coffee, of the four alternatives evaluated, the TBDVZ proved to be the optimal alternative, at 12.18 days. However, the other alternatives (DBOZ and DBDVZ) allow for very similar storage times, at 11.99 and 11.48 days, respectively. PCB does not appear to be a viable packaging alternative for roasted and ground coffee (7.85 days). Finally, we found that coffee stored in DFAB and aluminum bags at 20 °C has been shown to retain its quality for up to 250 days. Furthermore, if the temperature is reduced to 10 °C, the coffee’s shelf life is extended to more than 600 days. The insights derived from this research are of significant value to industry stakeholders, consumers, and developers of specialty coffee packaging.

From green bean to brewed coffee: A Lipidomic perspective on coffee lipid composition.

The Creator Coffee Roaster is a speciality coffee processing business based in Yogyakarta and has been established since 2012. The company focuses on producing roasted coffee beans of various types, including Robusta, Arabica, and house blends that are uniquely crafted by the company. The raw coffee beans are sourced from multiple coffee-producing regions in Indonesia, giving their products a diverse richness of flavour. As consumer interest in speciality coffee continues to grow, the demand for The Creator Coffee Roaster’s products has also increased steadily. This study evaluates the efficiency of the supply chain using the Data Envelopment Analysis (DEA) method, with input variables such as cash-to-cash cycle time and order fulfilment cycle time, and output variables such as delivery performance and order fulfilment. The analysis results show that farmers 2 and 4 achieved 100% efficiency, while farmers 1 and 3 were still inefficient, with scores of 57.1% and 66.7%, respectively. The novelty of this research lies in utilising DEA, which adopts input and output variables from the SCOR 12.0 method to provide a data-driven evaluation, helping speciality coffee roastery SMEs optimise their supply chain strategy. Based on these findings, it is recommended that specialty coffee SMEs establish partnerships with efficient farmers, provide improvement feedback to those who are less efficient, and adopt data-driven supply chain management to enhance competitiveness and business sustainability.

Coffee plants and beans are prone to fungal contamination that pose health risks to consumers by producing mycotoxins like ochratoxin A (OTA). Thus, the present study aimed to analyze the mycobiota of Costa Rican coffee beans, focusing on potentially ochratoxigenic species and their in vitro susceptibility patterns to antifungal agents. Fungal isolates were obtained from cherry, green, and roasted coffee beans from Costa Rica; they were identified by morphology, MALDI-TOF technology, and sequencing. The isolation frequency (FR) was 33.10 % for all the samples analyzed, 49.51 % for the cherry fruits, 37.67 % for the green coffee beans, and 17.33 % for the roasted beans. The cherry beans were mainly contaminated with Geotrichum klebahni (46.34 % FR and 90.91 % relative density [RD]), while the green and roasted coffee beans were mainly infected with Aspergillus spp. (22.00 % FR and 55.23 % RD and 13.83 % FR and 77.57 % RD, respectively). A total of 46.67 % of A. westerdijkiae and 20.00 % of A. ochraceus produced fluorescence in YES broth related with ochratoxin production. The isolates of the Aspergillus section Circumdati were susceptible to the azole antifungals. Costa Rican coffee beans could be contaminated with mycotoxigenic fungi during storage.

<p><em>The classification of coffee bean roasting levels is an important aspect of ensuring coffee product quality. This study compares the performance of two deep learning architectures, ResNet50 and VGG16, in classifying coffee bean images into three roasting levels: light, medium, and dark. The dataset consists of 1,800 images with a resolution of 224×224 pixels, divided into training, validation, and testing sets. Both models were trained with identical configurations using transfer learning and partial fine-tuning. The evaluation results show a very small accuracy difference of only 0.01 point, with ResNet50 slightly outperforming VGG16. This indicates that both models are equally reliable for roast level classification. However, ResNet50 is more time-efficient, requiring only about 10 minutes of training compared to over 25 minutes for VGG16. This difference is suspected to be related to the complexity of VGG16’s architecture. The study concludes that ResNet50 offers high efficiency with competitive accuracy. Further research is recommended to optimize VGG16’s architecture to improve computational efficiency without compromising accuracy.</em><em></em></p><p><strong><em>Keywords: </em></strong><em>Machine learning; Restnet50; VGG16; Coffee bean roasting</em><em></em></p><p align="center"><strong><em> </em></strong></p><p align="center"><strong>Abstrak</strong></p><p>Klasifikasi tingkat roasting biji kopi merupakan aspek penting dalam penjaminan mutu produk kopi. Penelitian ini membandingkan performa dua arsitektur deep learning, ResNet50 dan VGG16, dalam mengklasifikasikan citra biji kopi pada tiga tingkat roasting: light, medium, dan dark. Dataset berisi 1.800 citra beresolusi 224×224 piksel, dibagi menjadi data latih, validasi, dan uji. Kedua model dilatih dengan konfigurasi identik menggunakan transfer learning dan fine-tuning parsial. Hasil evaluasi menunjukkan selisih akurasi sangat tipis, hanya 0,01 poin, dengan ResNet50 sedikit unggul. Hal ini menunjukkan kedua model sama-sama andal untuk klasifikasi tingkat roasting. Namun, ResNet50 lebih efisien secara waktu, hanya memerlukan sekitar 10 menit pelatihan dibandingkan VGG16 yang lebih dari 25 menit. Perbedaan ini diduga terkait kompleksitas arsitektur VGG16. Disimpulkan bahwa ResNet50 menawarkan efisiensi tinggi dengan akurasi kompetitif. Penelitian lanjutan disarankan mengevaluasi optimasi arsitektur VGG16 untuk meningkatkan efisiensi komputasi tanpa mengorbankan akurasi.</p><p> </p>

The paper presents an analysis of the content of caffeine, tocopherols, and phenolic compounds as well as the volatile compound profile and volatile compound emission intensity in relation to the cultivation parameters of the Typica variety of Arabica coffee from Peru, Costa Rica, Guatemala, and Ethiopia. The study provides a detailed description of the cultivation and post-harvest parameters of the coffee types selected for the analyses. Special emphasis was placed on the analysis of the plantation altitude effect on instrumentally determined aromatic parameters and bioactive properties. The analyses were performed with the use of high-performance liquid chromatography, gas chromatography, and an electronic nose. The investigation results indicate a significant effect of the altitude of coffee cultivation on the most important biological and chemical properties of coffee beans, e.g. caffeine content, phenolic content, intensity of volatile compound emission, and coffee aroma. Other factors, such as shading and post-harvest processing, were also found to interact with the cultivation altitude and influence the content of these coffee attributes.

Optimization of gradient separation conditions using liquid chromatography with tandem mass spectrometric detection for analysis of phenolic profile during coffee roasting.

Coffee beans are rich in phenolic compounds such as phenolic acids, polyphenols and flavonoids which contain antioxidant activity. Robusta coffee (Coffea canephora) contains much higher levels of caffeine and chlorogenic acid than arabica coffee. The two compounds above have high antioxidant and UV filter activity so they can be formulated as sunscreen preparations. The aim of this study was to compare the antioxidant activity and sun protection factor of samples using green and roasted robusta coffee formulated into sunscreen hydrogel preparations from the methanol fraction of robusta coffee beans (Coffea canephora). Antioxidant activity testing using the DPPH (1,1-diphenyl-2- picrylhydrazyl) method at a wavelength of 515nm, showed the highest activity shown by the hydrogel preparation of the methanol fraction of green robusta coffee beans (Coffea canephora) with a free radical inhibition percentage value of 17.12ppm. Meanwhile, the percentage value of free radical inhibition from the hydrogel preparation from the methanol fraction of roasted robusta coffee beans (Coffea canephora) was 13.30ppm. The sun protection factor values for the methanol fraction hydrogel preparation of green and roasted robusta coffee beans (Coffea canephora) are 27.57 (ultra protection) and 10.66 (maximum protection).

Application of High-Performance Liquid Chromatography-Tandem Triple Quadrupole Mass Spectrometry to the Assessment of Acrylamide Contamination of Roasted Coffee Beans

This study examines the effect of temperature, roasting time, and coffee bean weight on the maturity level of roasted coffee beans, utilizing fuzzy logic to model and predict optimal roast outcomes. Coffee bean roasting level, a critical factor in determining flavor, is assessed using Mamdani-type fuzzy logic, which takes three input parameters—temperature, time, and weight—divided into ranges for specific roast outcomes. Case study parameters, including a coffee bean weight of 450 grams, roasting time of 39 minutes, and temperature of 130°C, were analyzed to classify the roast maturity. The fuzzy logic output indicates that this specific case falls within the medium roast level by the similarity of centroid between moment calculation and matlab calculation is 10,7, demonstrating the model's capability to provide precise roast classifications. This structured approach to evaluating coffee roasting parameters contributes to enhancing consistency and quality in the coffee industry, highlighting the potential of fuzzy logic for robust decision support in coffee quality control.

Indonesia is the 3rd largest coffee producing country in the world in 2022-2023 with coffee production reaching 11.85 million bags per 60 kg of coffee. One of the important processes in coffee production is roasting because the roasting level of coffee beans can affect the taste and aroma of coffee. The problem faced is that the process of assessing the level of coffee roasting is traditionally carried out through visual observation by an expert (roaster). This method produces a subjective level of assessment and requires high skills and experience, making the assessment of the level of coffee roasting less efficient and prone to human error. Therefore, in this study the author aims to develop a Convolutional Neural Network (CNN) model for the classification of the level of coffee bean roasting that can achieve better and faster accuracy. In this study, the author compared two CNN architecture approaches for the classification of the level of coffee bean roasting. The first approach is full learning with an architecture consisting of three convolution layers. The second approach is transfer learning based on the VGG-16 model. From the results of the analysis, it is known that the full learning model has a better level of accuracy and a faster running time than the VGG-16 transfer learning. The CNN full learning model for coffee bean roasting level classification is able to classify the coffee bean roasting level, with an accuracy of 98.75% and a running time of 856 ms per step. The application of CNN for coffee roasting level classification can provide benefits such as improving quality control and reducing the level of subjectivity of a roaster in assessing the roasting level of coffee beans.

Acrylamide, a potential occupational carcinogen, is a natural by-product formed during the thermal processing of starchy foods and roasted coffee beans. Recent studies have also reported high levels of acrylamide in various thermally treated fast foods in Saudi Arabia. This study aims to meet the critical need for effective antioxidant therapies to mitigate acrylamide-induced liver damage. By comparing the protective effects of selenium and quercetin nanoparticles, it seeks to identify the more potent nano-antioxidant, thereby contributing to the advancement of safer and more efficient strategies for preventing chemically induced hepatotoxicity. Twenty adult male Albino Wistar rats were randomly divided into four groups: control, acrylamide-treated, acrylamide + SeNP, and acrylamide + QNP. Acrylamide exposure (Acrylamide exposure (50 mg/kg/day, orally for 21 days)) significantly elevated serum levels of cholesterol (CHO), triglycerides (TG), low-density lipoprotein (LDL), alanine transaminase (ALT), aspartate transaminase (AST), creatinine, and urea, cholesterol (233.33 ± 7.50 mg/dL), (238.33 ± 4.93 mg/dL), (67.33 ± 2.51 mg/dL), (80.33 ± 3.51 U/L), and AST (80.00 ± 3.00 U/L) respectively, compared to the control group (CHO: 155.33 ± 8.02, TG: 150.00 ± 7.93, LDL: 39.33 ± 4.16, ALT: 16.66 ± 3.78, AST: 22.66 ± 2.08). while significantly reducing glutathione (GSH) and superoxide dismutase (SOD) levels in liver tissues compared to the control group. Treatment with SeNPs and QNPs led to a marked reduction in these altered biochemical parameters and improved liver histopathology. In conclusion, selenium and quercetin nanoparticles exhibited a protective effect against acrylamide-induced hepatotoxicity in male Albino Wistar rats, suggesting their potential use in mitigating liver damage caused by environmental toxins.

Coffee by-products are promising reservoirs of antioxidants and fermentable sugars. Coffee silverskin (CSS) is a significant by-product of coffee bean roasting. In this study, glucose release from CSS was optimized to investigate the pretreatment parameters using the Taguchi method. For this purpose, four different acids (HCl, H2SO4, HNO3, and CH3COOH) and two alkalis (NaOH and KOH) were tested. The factors were acid/alkali type and amount (1, 2, 3, and 4%), CSS amount (2.5, 7.5, 12.5, and 17.5%), and treatment time (15, 30, 45, and 60 min). Sixteen experimental runs were performed using the L16 orthogonal array. The conditions under which maximum glucose release was obtained and the effects of these conditions on the process were examined. The maximum glucose yield (Yg= 0.037 gglucose/gbiomass) was obtained by autoclaving a 4% (v/v) HCl solution containing 17.5% CSS (w/v) for 60 min. This study evaluated the effectiveness of pre-enzymatic CSS treatments to produce fermentable sugars that are valuable for various industrial applications.

Comprehensive evaluation of volatile compounds and sensory profiles of coffee throughout the roasting process.

Large dataset on Fourier transform near infrared (FT-NIR) spectroscopy of green and roasted specialty coffee: Preprocessed infrared spectra and sensory scores for machine learning-based quality monitoring.

Immature green coffee beans, which originate from green coffee cherries, are defective beans that are difficult to distinguish from mature green beans. Roasting these immature green beans produces quaker beans, which can reduce coffee quality. Acid treatment has been studied to improve the aroma quality of coffee. Acetic acid, an organic acid in coffee produced during fermentation and commonly used as vinegar, requires further study to determine its effects on immature coffee beans. This study investigated the effect of acetic acid treatment on the sensory, physicochemical and antioxidant properties of immature Robusta coffee beans. Immature green beans were soaked in acetic acid solutions of varying concentrations (0, 1, 2, and 3 %) and durations (30, 60, and 90 min) at 35 °C. Immature coffee beans had lighter brown color after roasted, and have higher chlorogenic acid (65.39 mg/g), caffeine (26.08 mg/g) and total soluble phenolic contents (65.70 mgGAE/g) than mature beans (p < 0.05). The roasted coffee of CI3-90 was the best acetic acid treatment sample with the highest cupping score (82.9) and the most intensely brown color, which resulted in decreased brightness (38.64 to 30.54) and yellowness (29.91 to 25.79). The acetic acid treatment significantly reduced total soluble phenolic content (61.58 to 45.26 mgGAE/g), chlorogenic acid content (20.10 to 14.64 mg/g) and trigonelline content (7.40 to 6.07 mg/g) of immature roasted coffee beans. Nevertheless, acetic acid treatment did not significantly influence caffeine content or antioxidant activity in either the green or roasted immature coffee beans (p > 0.05). HIGHLIGHTS Acetic acid treatment can improve the sensory and color quality of roasted immature coffee beans. Treating immature coffee beans with 3 % acetic acid for 90 min is the best treatment, resulting in the highest cupping score and roasted bean colors that are close to the roasted color of mature coffee beans. The chlorogenic acid and trigonelline concentrations in green and roasted immature coffee exhibited a tendency to decrease with increasing acetic acid concentrations and treatment durations. The antioxidant activities of immature green and roasted coffee beans remained relatively stable after acetic acid treatment. GRAPHICAL ABSTRACT

The coffee brewing that is most often found in the community is manual or brewing coffee. However, this method is considered to still have weaknesses and is less practical because it leaves coffee grounds in the drink. One effort to overcome this problem is coffee in pouch packaging, and the addition of spices to diversify products. This research aims to study the effect of the degree of roasting of Arabica coffee beans and the percentage of added lemongrass powder on the chemical and sensory properties of coffee bag products. This research used a factorial completely randomized design (RAL). Experimental factors used in this study are the degree of roasting of Arabica coffee beans (D) which consists of 3 levels including 1900C (light) (D1), 2100C (medium) (D2) and 2400C (dark)(D3). The second factor is the concentration of kitchen lemongrass powder (S) which consists of 2 levels, namely 8% (S1) and 10% (S2). The research data obtained is then analyzed using Analysis of Variance (ANOVA) and followed with Duncan's Multiple Range Test (DMRT) if there are differences between treatments. The results of various types of roasting degree treatments had an effect on reducing water content, ash content, and total dissolved solids. The value of water content and ash content decreases as the degree of roasting increases, the value of total dissolved solids increases as the degree of roasting increases. The percentage of adding kitchen lemongrass powder also has an effect on reducing the ash content. The ash content value increased as the percentage of lemongrass powder was added. The interaction between the degree of roasting and the addition of lemongrass powder affects pH and hedonics (aroma, color and taste). The pH value increases as the degree of roasting increases and the addition of lemongrass powder. The average value of the aroma hedonic test was 3.55 (neutral), color hedonic 3.62 (liked) and taste hedonic 3.35 (neutral). The best treatment was obtained at a light roast level with an additional percentage of 8% kitchen lemongrass powder (D1S1) followed by antioxidant activity testing with a result of 52%.

West Lampung Robusta coffee has a higher caffeine content than other types of coffee. The variations in extraction time in ultrasound-assisted extraction (UAE) produced different caffeine contents than conventional extraction methods. This study aims to determine extraction time variation (15, 30, 45) influences caffeine contents in roasted coffee beans. The organoleptic test and alkaloid test evaluated Robusta coffee qualitatively. The extract obtained was measured by measuring the % MC, yield value, and wavelength scanning. Quantitative analysis using UV-Vis Spectrophotometry determined the caffeine content with different extraction times, and then each group was assessed using SPSS software. The coffee bean powder was dark brown, had a distinctive coffee aroma, bitter taste, and smooth texture, and contained alkaloids with the Mayer, Wagner, and Dragendorff tests. The yield and % water content of extract at 30 (36.938%; 6.39%) were higher than at 15 (24.111%; 6.07%) and 45 minutes (22.170%; 6.37%). Alkaloids of the extract were investigated at a wavelength of 273 nm. Caffeine contents were 0.977 ± 0.018 mgCE/mg at the 15, 0.792 ± 0.054 mgCE/mg at the 30, and 0.979 ± 0.026 mgCE/mg at the 45 minutes in significant (P<0.05). In conclusion, the difference in extraction time impacted the caffeine content of the robusta coffee bean.

The unique aroma of arabica coffee beans will come out after passing through the roasting step, therefore the quality control is important. Quality control in the company is very important to produce products that have a slick quality value so consumers had their own satisfaction when buying products. This study aims to analyze (1) The level of product defects of arabica roasted coffee beans (2) The level of process stability and processing capacity of arabica roasted coffee beans products in achieving zero defects (3) Quality improvement efforts to minimize product defects of arabica roasted coffee beans. The sample used in this study was 100 samples of arabica roasted coffee beans products using non-probability sampling techniques due to the limited time and cost of the study in drawing random probability samples. This research was conducted at Drama Coffee Roastery, Pasar Tunjungan, first floor Unit 08, Jalan Tunjungan 88, Surabaya City which had wide marketing of arabica roasted coffee beans products, covering Surabaya area or outside the city, and has a product license and legality. The method used in this research was six sigma method with DMAIC (Define, Measure, Analyze, Improvement, and Control) approach. The results showed that Drama Coffee Roastery had 4 types of defects product, was overdeveloped, underdeveloped, cracked, and striped. Drama Coffee Roastery had a sigma level value of 3.35 which explains that UMKM Drama Coffee Roastery had the ability to process product quality within the scope of the average industry in Indonesia. This indicates that the product requires quality control to achieve zero defects in order to entry international coffee market. The study found that six sigma method was effective in solved problems that occur at Drama Coffee Roastery by reviewing green beans, green beans roasting time, sorting improvements, and green beans storage improvements that will have an impact on the supply chain efficiency of arabica roasted coffee beans products.