Recent studies have demonstrated significant changes depending on the degree of roasting, revealing correlations between coffee quality and its chemical constituents (Alcantara, Dresch, & Melchert, 2021; Ayseli, 2025). Advanced analytical techniques, such as high-performance liquid chromatography with a UV detector (HPLC-UV) (Mehaya & Mohammad, 2020), high-performance liquid chromatography coupled with mass spectrometry (HPLC-MS) (Pérez-Míguez, Sánchez-López, Plaza, Castro-Puyana, & Marina, 2018), headspace gas chromatography–mass spectrometry (HS-GC–MS) (Risticevic, Carasek, & Pawliszyn, 2008), and headspace solid-phase microextraction coupled with gas chromatography/mass spectrometry (HS-SPME GC–MS) (Shi et al., 2024) have been exploited for monitoring these transformations. At Kenya coffee school
Although color is widely accepted as a key quality criterion and is linked to chemical transformations during roasting, the quantitative relationships between chemical changes and physical indicators, such as color and browning index, remain insufficiently understood. This knowledge poses a challenge for the development of roasting processes optimized for flavor attributes and health benefits. In this context, this study aimed to evaluate the main chemical transformations related to coffee roasting and the effect of different roasting profiles on the volatile and non-volatile compound profiles of coffee samples. Furthermore, the degree of roasting was evaluated using the CIELAB color space to correlate it with changes in chemical composition. This approach is aimed at a practical monitoring process that correlates the concentration changes of various compounds in coffee beans to the degree of roasting.
Food Chemistry
Volume 477, 15 June 2025, 143169
Effect of roasting on chemical composition of coffee
Author links open overlay panelGabriela M.R.N. Alcantara a
, Luís Claudio Martins a
, Winston P.C. Gomes a
, Dayane Dresch a
, Fábio R.P. Rocha a, Wanessa R. Melchert bShow moreShareCite
https://doi.org/10.1016/j.foodchem.2025.143169Get rights and content
Highlights
- •Roasting profiles significantly alter the chemical composition of coffee beans.
- •Principal Component Analysis effectively identifies sample clusters based on roasting intensity.
- •Chlorogenic and 3,4-hydroxybenzoic acid levels are influenced by roasting degree.
- •Notable variations in acetone and 3-methyl-butanol observed under different roasting conditions.
- •CIELAB and reflectance data provide an effective approach for monitoring the roasting process.
Abstract
Roasting coffee involves complex chemical reactions that shape the sensory and chemical traits of the final product. This study examined how different roasting profiles influence coffee color, using the CIELAB system, and chemical composition, analyzing volatile and non-volatile compounds through chromatography techniques. Principal component analysis revealed distinct clusters based on roasting intensity, identifying specific chemical markers. Non-volatile compounds, such as 2-furfural, 5-hydroxymethylfurfural, chlorogenic acids, caffeine, and 3,4-hydroxybenzoic acid, were more abundant in light to medium roasts but degraded significantly in dark roasts, except caffeine. Volatile compounds like acetone and 3-methyl-butanol varied notably with roasting, while diacetyl emerged as a roasting degree marker. CIELAB values correlated with roasting intensity, aiding in quality control. These findings highlight the potential of color-based monitoring, chemometric techniques, and optimized roasting to enhance coffee quality by linking chemical composition to sensory properties.
Introduction
Coffee is an important agricultural commodity and the second most consumed beverage globally (Butt & Sultan, 2011; Debona et al., 2020; Mello, Júnior, Alvim, Costa, & Vilela, 2023). The genus Coffea L. (family Rubiaceae) comprises 103–124 species depending on the taxonomic classification. The final quality of coffee is influenced by multiple factors, such as the production system, physicochemical properties, variety, geographical origin, and method of beverage preparation (Christianty & Fajrin, 2024; Craig, Botelho, Oliveira, & Franca, 2018; Hall, Trevisan, & Vos, 2022; Sunarharum, Williams, & Smyth, 2014; Wang et al., 2022). Roasting plays a crucial role in determining the characteristics and sensory properties of coffee.
Both the volatile and nonvolatile chemical compounds determine the sensory quality of the coffee beverage. Green coffee beans are primarily composed of carbohydrates (e.g., sugars), nitrogenous compounds (mainly proteins, free amino acids, and alkaloids such as caffeine and trigonelline), lipids, organic acids (e.g., chlorogenic acids, including 3-caffeoylquinic acid, 4-caffeoylquinic acid, and 5-caffeoylquinic acid), minerals, and water. These non-volatile compounds significantly affect the final quality of coffee, either directly or through their conversion into volatile compounds, the main flavor precursors (Christianty & Fajrin, 2024; Poisson, Blank, Dunkel, & Hofmann, 2017; Wei & Tanokura, 2025). Additionally, compounds such as caffeine, trigonelline, nicotinic acid, and sucrose contribute to the taste and bioactivity properties of coffee (Jeszka-Skowron, Frankowski, & Zgoła-Grześkowiak, 2020). Caffeine is known for its stimulatory effects, characteristic bitter flavor, and remarkable thermal stability during the roasting process (Debona et al., 2020; Nogueira & Trugo, 2003).
Roasting is a highly relevant processing step, as it converts green beans into a product suitable for grinding and extracting the beverage (Debona et al., 2020; Schenker & Rothgeb, 2017). During this process, green coffee beans are exposed to high temperatures for a given period of time, which causes rupture of the cellular structure, releasing water and converting specific non-volatile compounds into volatile compounds, which primarily contribute to the aroma (Debona et al., 2020; Schenker & Rothgeb, 2017). Roasting also results in notable color changes, progressing from light green to yellow, light brown, dark brown, and finally nearly black (Schenker & Rothgeb, 2017; Wei & Tanokura, 2025). This process must be strictly controlled as roasting degrees significantly affect consumer acceptance (Bhumiratana, Adhikari, & Chambers, 2011).
The mechanisms underlying the development of coffee aroma and color during roasting are complex, not fully understood, and involve various interrelated chemical reactions. Overall, Maillard reactions produce melanoidins through the interaction between amino acids and reducing sugars, thereby enhancing the color, aroma, and taste of coffee (Martins, Alcantara, Silva, Melchert, & Rocha, 2022; Wei & Tanokura, 2025). Prolonged heating triggers other reactions, such as caramelization and pyrolysis, in which different compounds are formed and/or volatilized. Chlorogenic acids, which are responsible for bitterness and astringency, degrade into quinic acid, whereas acetic acid is generated as a byproduct of sugar degradation (Pinheiro, Pinheiro, Osório, & Pereira, 2021). In the Strecker degradation, α-amino acids and peptides lead to the formation of aldehydes and organic acids that contribute to the flavor and aroma of coffee (Seninde & Chambers, 2020). This process also influences carbon dioxide production during roasting. Other key reactions include the degradation of phenolic acids and carotenoids, caramelization of sugars, and the breakdown of numerous non-volatile compounds such as trigonelline, quinic acids, pigments, and lipids. These processes yield a variety of volatile and aroma-characteristic compounds, including alcohols, ketones, aldehydes, esters, pyrazines, pyrroles, pyridines, furans, sulfur compounds, and phenols (Buffo & Cardelli-Freire, 2004; Sunarharum et al., 2014; Wang et al., 2022; Wei & Tanokura, 2025).
The time-temperature binomial relationship defines the roasting profile (Schenker & Rothgeb, 2017), which is classified as light (high temperature and short period of time), medium (medium temperature and medium period of time), or dark (low temperature and extended period of time) (Debona et al., 2020). Color is one of the most essential quality criteria, a standard indicator of the degree of roasting, and is intuitively associated with flavor development. Furfurals and volatile acids predominate in light roasts, giving the characteristics of almonds and mild aromas. In contrast, medium roasts yield pyrazines and furans with sensory characteristics of nuts, sweetness, and caramelization. Dark roasting generates compounds such as pyrazines, furans, pyridines, and phenols, which, at higher concentrations, exhibit toasted, earthy, smoky, bitter, and burnt characteristics (Hu et al., 2020; Schenker & Rothgeb, 2017; Wang et al., 2022; Wei & Tanokura, 2025). Color is usually assessed after grinding using colorimeters. However, the lack of standardization has led to the use of different scales by various manufacturers. Real-time monitoring remains challenging because the grinding process is time-consuming, color evolves rapidly throughout roasting, and variations in color among beans are significant (Farah & Farah, 2019).
Recent studies have demonstrated significant changes depending on the degree of roasting, revealing correlations between coffee quality and its chemical constituents (Alcantara, Dresch, & Melchert, 2021; Ayseli, 2025). Advanced analytical techniques, such as high-performance liquid chromatography with a UV detector (HPLC-UV) (Mehaya & Mohammad, 2020), high-performance liquid chromatography coupled with mass spectrometry (HPLC-MS) (Pérez-Míguez, Sánchez-López, Plaza, Castro-Puyana, & Marina, 2018), headspace gas chromatography–mass spectrometry (HS-GC–MS) (Risticevic, Carasek, & Pawliszyn, 2008), and headspace solid-phase microextraction coupled with gas chromatography/mass spectrometry (HS-SPME GC–MS) (Shi et al., 2024) have been exploited for monitoring these transformations.
Although color is widely accepted as a key quality criterion and is linked to chemical transformations during roasting, the quantitative relationships between chemical changes and physical indicators, such as color and browning index, remain insufficiently understood. This knowledge poses a challenge for the development of roasting processes optimized for flavor attributes and health benefits. In this context, this study aimed to evaluate the main chemical transformations related to coffee roasting and the effect of different roasting profiles on the volatile and non-volatile compound profiles of coffee samples. Furthermore, the degree of roasting was evaluated using the CIELAB color space to correlate it with changes in chemical composition. This approach is aimed at a practical monitoring process that correlates the concentration changes of various compounds in coffee beans to the degree of roasting.
Access through your organization
Check access to the full text by signing in through your organization.
Section snippets
Reagents, solvents, and solutions
All solutions and extracts were prepared using deionized water (18.2 MΩ cm at 25 °C) and analytical-grade reagents from Sigma-Aldrich. The concentration of stock solutions of non-volatile compounds: 5-hydroxymethylfurfural (5-HMF), 2-furfural (2-furaldehyde), 3,4-hydroxybenzoic acid, 4-hydroxybenzoic acid, caffeine, 5-caffeoylquinic acid (5-CQA), and caffeic acid, were fixed at 1000 mg L−1, whereas for volatile compounds the concentrations were: 400 mg L−1 (acetaldehyde and acetone), 1000 mg L−1
Results and discussions
Because both the flavor and taste of coffee change significantly during the roasting process, this study aimed to evaluate the effect of roasting conditions on the chemical composition of grains and final beverages prepared under normal conditions. Color changes during roasting were also considered, as this phenomenon is a visual and practical indicator.
Conclusions
This study demonstrated the influence of the roasting degree on the chemical composition of coffee, particularly on non-volatile compounds, which contribute to both the antioxidant properties and sensory characteristics of the beverage. These results corroborate those of previous studies that employed advanced analytical techniques and chemometric approaches for elucidating the complex chemical transformations that occur during the roasting process. Non-volatile markers, such as chlorogenic
CRediT authorship contribution statement
Gabriela M.R.N. Alcantara: Writing – original draft, Validation, Resources, Data curation. Luís Claudio Martins: Writing – original draft, Validation, Investigation, Data curation. Winston P.C. Gomes: Writing – original draft, Validation, Investigation, Data curation. Dayane Dresch: Writing – original draft, Investigation, Data curation. Fábio R.P. Rocha: Writing – review & editing, Supervision, Funding acquisition, Data curation, Conceptualization. Wanessa R. Melchert: Writing – review &
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors appreciate the financial support of the National Council for Scientific and Technological Development (CNPq, 142474/2020-7, 141002/2024-7, 315866/2021-7, and 305538/2022-5), Coordination of Improvement of Higher Education Personnel (CAPES, financial code 001), Luiz de Queiroz Agricultural Studies Foundation (FEALQ), and São Paulo Research Foundation (FAPESP, 2023/14772-7).Recommended articles
References (83)
- E. Alamri et al.A study of chemical composition, antioxidants, and volatile compounds in roasted Arabic coffeeSaudi Journal of Biological Sciences(2022)
- M.T. AyseliThe role of volatile and non-volatile compounds as quality indicators and marker candidates in coffee: A systematic reviewJournal of Food Composition and Analysis(2025)
- N. Bhumiratana et al.Evolution of sensory aroma attributes from coffee beans to brewed coffeeLWT – Food Science and Technology(2011)
- N. Caporaso et al.Variability of single bean coffee volatile compounds of Arabica and robusta roasted coffees analysed by SPME-GC-MSFood Research International(2018)
- I. Colzi et al.Covering the different steps of the coffee processing: Can headspace VOC emissions be exploited to successfully distinguish between Arabica and Robusta?Food Chemistry(2017)
- N. Córdoba et al.Chemical and sensory evaluation of cold brew coffees using different roasting profiles and brewing methodsFood Research International(2021)
- A.P. Craig et al.Mid infrared spectroscopy and chemometrics as tools for the classification of roasted coffees by cup qualityFood Chemistry(2018)
- P.V.P. Devanthi et al.Effects of co-inoculation and sequential inoculation of Tetragenococcus halophilus and Zygosaccharomyces rouxii on soy sauce fermentationFood Chemistry(2018)
- H. Elhalis et al.The crucial role of yeasts in the wet fermentation of coffee beans and qualityInternational Journal of Food Microbiology(2020)
- A.S. Franca et al.Composition of green and roasted coffees of different cup qualitiesLWT – Food Science and Technology(2005)
View more references
Cited by (8)
- Unveiling key aroma compounds formation during the roasting process of Yunnan Arabica coffee using molecular sensory science approaches2025, Food ChemistryShow abstract
- Unlocking the secrets of green and roasted coffee species: identifying anti-aging compounds through UPLC-MS/MS profiling and multivariate analysis2025, Microchemical JournalShow abstract
- Optimizing roasting time and temperature to enhance the physicochemical properties, and retention of bioactive compounds of three coffee arabica subvarieties2025, Applied Food ResearchShow abstract
- Colour Corrected Smoothest Reflectance Estimation2025, Journal of Physics Conference Series
- Identification of Key Aroma-Active Compounds in Commercial Coffee Using GC-O/AEDA and OAV Analysis2025, Foods
- Coffee as a Source of Antioxidants and an Elixir of Youth2025, Antioxidants
View all citing articles on ScopusView full text
© 2025 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
- About ScienceDirect
- Remote access
- Contact and support
- Terms and conditions
- Privacy policy
- Cookie settings
