Trends Sci. 2026; 23(9): 13444

The Dynamic Changes of Chlorogenic Acids and Alkaloids in Coffee Processing and Brewing: A Systematic Literature Review

Nazwa Nurhaliza^1,2, Dian Herawati^1,2 and Nuri Andarwulan^1,2,*

¹Division of Food Science and Technology, Faculty of Engineering and Technology, IPB University,

IPB Dramaga Campus, Bogor 16680, Indonesia

²South-East Asia Food & Agricultural Science and Technology (SEAFAST) Center, IPB University,

IPB Dramaga Campus, Bogor 16680, Indonesia

(^*Corresponding author’s e-mail: [email protected])

Received: 25 January 2026, Revised: 26 March 2026, Accepted: 2 April 2026, Published: 15 April 2026

Abstract

Coffee is one of the most consumed beverages worldwide, and its quality and health-related properties are strongly influenced by bioactive compounds such as chlorogenic acids (CQAs) and alkaloids (caffeine, trigonelline, theobromine). Their concentrations vary considerably depending on processing steps, including postharvest, roasting, and brewing. This systematic review, conducted following PRISMA 2020 guidelines, evaluated the stability and changes of CQAs and alkaloids in Arabica (Coffea arabica) and Robusta (Coffea canephora) across green beans, roasted beans, and brewed coffee. Articles were retrieved from multiple databases using specific keywords, and quantitative data were extracted and analyzed using Microsoft Excel and R Studio. Results showed that Arabica green beans contained 12.50 - 160.10 mg/g CQAs, which de-creased by up to 99% after roasting and brewing, while Robusta initially contained 32.10 - 185.60 mg/g with reductions of 56% - 92%. Caffeine was more stable, averaging 22.15 ± 15.22 mg/g in Arabica and 45.58 ± 22.14 mg/g in Robusta green beans, with moderate reductions through processing. Trigonelline consistently decreased by 80% - 94% in Ara-bica and 68% - 77% in Robusta, while theobromine remained at low levels with further reductions. Overall, roasting and brewing significantly degrade CQAs and trigonelline, while caffeine shows relative stability, providing insight into processing strategies that may optimize coffee’s bioactive profile.

Keywords: Caffeine, Chlorogenic acids, Coffea arabica, Coffea canephora, Trigonelline, Theobromine, Roasting, Brewing

Abbreviations

a.s.l Above sea level

CQAs Total chlorogenic acids

CQA Chlorogenic acid

Cf Caffeine

Concentration initial

Concentration final

DAD Diode Array Detector

diCQA Dicaffeoylquinic acids

db Dry basis

HPLC High-Performance Liquid Chromatography

Med Median

Min Minimal

PDA Photodiode Array

PRISMA Preferred Reporting Items for Systematic Reviews and Meta-Analyses

SD Standard Deviation

Tr Trigonelline

Th Theobromine

UV Ultra Violet

Introduction

Coffee is one of the most widely consumed and traded food commodities in the world, ranking second only to oil in terms of global trade [1]. Beyond its economic importance, coffee has attracted increasing scientific attention due to its complex chemical composition and its potential implications for human health and sensory quality. In recent years, growing consumer demand for specialty coffee, functional beverages, and evidence-based health claims has intensified the need to understand how bioactive compounds dynamically change throughout the coffee processing chain. In the 2023/2024 period, Arabica coffee production reached 102.2 million 60 kg bags, while Robusta accounted for 75.8 million 60 kg bags [2]. These 2 species differ significantly in cultivation conditions and chemical characteristics. Arabica is typically cultivated at higher altitudes (1,000 - 2,000 m a.s.l.) and is appreciated for its smooth and complex flavor, while Robusta is more tolerant to lower altitudes (400 - 700 m a.s.l.), produces a more bitter taste, and contains higher levels of caffeine [1,3].

The functional and sensory qualities of coffee are strongly influenced by its bioactive compounds, primarily chlorogenic acids (CQAs) and alkaloids such as caffeine, trigonelline, and theobromine. Chlorogenic acids are esters of hydroxycinnamic acids (caffeic, ferulic, or p-coumaric acid) and quinic acid, accounting for approximately 4% - 12% of dry weight depending on species and origin [4]. Importantly, these compounds are chemically unstable and do not remain static; their concentrations continuously change in response to environmental exposure, fermentation, thermal treatment, and extraction conditions. CQAs are susceptible to degradation through isomerization, pyrolysis, hydrolysis into quinic acid, and further transformation into phenolic derivatives during roasting [5]. They contribute to antioxidant capacity, acidity, and bitterness, while also being associated with potential protective effects against metabolic disorders [6].

Caffeine, a purine alkaloid derived from xanthine, ranges from 0.9% - 1.3% in Arabica and 1.5% - 2.5% in Robusta [4]. Unlike CQAs, caffeine is relatively thermally stable, although slight reductions under severe roasting have been reported [7,8]. Trigonelline, a niacin-related alkaloid, occurs at higher levels in Arabica (0.6% - 2.0%) compared to Robusta (0.6% - 0.7%) [4]. During roasting, trigonelline undergoes degradation to form volatile compounds such as pyridines and nicotinic acid, contributing to aroma formation [7]. Theobromine is present in smaller amounts and exhibits inconsistent behavior during roasting, with some studies reporting increases due to demethylation reactions and others reporting degradation [9,10]. These transformations demonstrate that alkaloids and CQAs follow distinct kinetic pathways across processing stages, leading to complex compositional shifts rather than simple linear degradation trends.

Post-harvest processing (dry, semi-washed, or full-washed methods) significantly influences the bioactive profile of green coffee beans. Extended fermentation and soaking steps may promote leaching, enzymatic hydrolysis, and oxidation of CQAs and alkaloids [10-12]. Environmental variables such as altitude, rainfall, temperature, and soil composition further modulate precursor accumulation [13]. Thus, the chemical composition of coffee reflects a continuous transformation process that begins with cultivation conditions and extends through post-harvest, roasting, and ultimately brewing.

Roasting represents the most critical transformation stage in shaping the chemical and sensory characteristics of coffee. During roasting, complex thermochemical reactions occur, including Maillard reactions, caramelization, Strecker degradation, and pyrolysis, which collectively reshape the chemical matrix of coffee beans [14]. Chlorogenic acids may decrease by more than 80% - 90% at dark roast levels [8,15], partly due to conversion into chlorogenic acid lactones and other volatile compounds. The degradation of trigonelline contributes to the formation of volatile compounds associated with aroma complexity [9], whereas caffeine remains comparatively stable under thermal conditions [8,15,16]. These chemical transformations directly influence sensory attributes such as bitterness, aroma complexity, and body, and may also affect physiological activity, as degradation products can exhibit different bioavailability or bioactivity compared with their precursor compounds.

Brewing further determines the final concentration of bioactives in the cup. Extraction efficiency depends on physicochemical parameters such as water temperature, pressure, grind size, brewing time, and coffee-to-water ratio [17]. High-temperature percolation methods may accelerate extraction kinetics of CQAs and alkaloids, whereas cold brew methods favor selective extraction of less polar compounds. Consequently, variations in brewing conditions generate substantial variability in reported compound concentrations.

Despite numerous studies, reported values ​​for chlorogenic acid and alkaloids vary widely across studies. This variation is not only biological but also methodological. Differences in roast classification, analytical approach, reporting units (mg/g, %, mg/mL), moisture correction, and extraction basis often hinder direct comparisons. Some studies report concentrations per dry weight of coffee beans, others per brew volume, and some studies do not include green coffee bean composition as a reference at all. As a result, data are sometimes contradictory, particularly regarding the optimal roasting or brewing conditions for maximizing retention of bioactive compounds.

Although several narrative reviews have discussed coffee bioactives, there is currently no systematic review that quantitatively harmonizes concentration data across all major processing stages from post-harvest treatment to roasting and brewing while simultaneously comparing Arabica and Robusta. Existing reviews generally focus on health effects roasting chemistry or individual compounds rather than integrating the full transformation pathway [14,18]. In addition previous reviews rarely address data standardization challenges or evaluate methodological heterogeneity among studies which limits their ability to draw robust comparative conclusions. This gap highlights the need for a structured quantitative synthesis approach.

The present systematic review therefore aims to systematically collect and harmonize reported concentrations of chlorogenic acids and major alkaloids across processing stages analyze compositional shifts along the transit pathway from green bean to brewed coffee and identify key processing parameters influencing compound stability. Unlike narrative summaries this review applies unit standardization and cross stage comparison to improve interpretability of heterogeneous datasets.

The inclusion of studies within a defined recent timeframe was intended to ensure that the data reflect contemporary analytical standards in compound quantification, where improvements in separation efficiency, detection sensitivity, and validation rigor have enhanced measurement accuracy and reproducibility. By synthesizing data generated under these more consistent conditions and harmonizing concentrations across species and processing stages, this study provides a quantitative integration that moves beyond descriptive reporting toward a more coherent understanding of bioactive transformation in coffee processing.

Materials and methods

Literature search

This review followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [19] with precisely specified keywords, inclusion criteria, and exclusion criteria. The review protocol was not registered in PROSPERO because the study does not evaluate clinical outcomes; however, the methodology followed PRISMA 2020 guidelines with predefined eligibility criteria and extraction strategy. The systematic screening process resulted in the inclusion of 50 eligible articles, as illustrated in the PRISMA flowchart (Figure 1).

Figure 1 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram illustrating the systematic literature selection process for this review. 50 Articles included in the review presents 242 data points in the content of total chlorogenic acids (3-CQA, 4-CQA, 5-CQA, 3,4-diCQA, 3,5-diCQA, 4,5-diCQA) and alkaloids (caffeine, trigonelline, theobromine) of Arabica and Robusta Coffee. The list of articles included in the review is presented in supplementary materials.

Further data is processed and analyzed using Microsoft Excel 2021 and R Studio. All studies including data present data about chlorogenic acids (3-CQA, 4-CQA, 5-CQA, 3,4-diCQA, 3,5-diCQA, 4,5-diCQA) and alkaloids (caffeine, trigonelline, theobromine) in Arabica and Robusta Coffee. The unit used for data processed is mg/g dry basis, with some studies requiring unit conversion. Publication searches were conducted using several scientific databases and search engines, including Google Scholar, ScienceDirect, Springer, MDPI, and PubMed, using specific keywords. ResearchGate was used only as a supplementary platform to access full-text articles that had already been identified through these databases. The keyword list is presented in Table 1. Inclusion and Exclusion Criteria are presented in Table 2.

Table 1 Specific keywords and search terms used in the database search

Table 2 Inclusion and exclusion criteria.

Data processing

Data on the concentrations of total chlorogenic acids (CQAs) and their isomers, caffeine (Cf), trigonelline (Tr), and theobromine (Th) were grouped according to coffee processing stages: green beans, roasted beans, and brewed coffee. Reported concentrations of CQAs, Cf, Tr, and Th across the reviewed studies were expressed in various units, including μg/g, mg/g, mg/mL, mg/100 mL, mg/100 g, mg/L, mg/kg, g/100 g, and g/kg. For consistency, all values were converted to mg/g dry basis for green beans, roasted beans, and brewed coffee.

In addition to concentration data extraction, methodological transparency regarding analytical performance was evaluated. Although no predefined minimum inclusion criteria were established for analytical validation parameters (e.g., limit of detection (LOD), limit of quantification (LOQ), recovery, and linearity), the reporting of these parameters was documented when available. Variability in analytical sensitivity and validation reporting across studies was considered as a potential source of inter-study heterogeneity. Furthermore, the risk of bias of each included study was assessed across 5 domains: Selection bias, performance bias, detection bias, attrition bias, and reporting bias. Each domain was evaluated based on the clarity of methodological descriptions and reporting completeness.

For green and roasted beans, concentrations were converted to mg/g dry basis (db) using reported moisture content. When moisture values were unavailable, reference moisture ranges were applied (green beans ≈ 12% [13]; roasted beans ≈ 1.34% - 3.41%) [13]. Conversion followed:

where represents the concentration on a dry basis (mg/g), represents the concentration on wet basis, and MC presents moisture content (%).

For brewed coffee, concentrations originally reported on a volumetric basis (mg/mL, mg/100 mL or mg/L) were first standardized to mg/mL. To enable cross-stage comparison on a uniform dry mass basis, brewed coffee concentrations were subsequently expressed relative to the dry mass of coffee powder used during preparation. When brewing parameters were reported, the actual coffee-to-water ratio and serving volume were applied. The dry basis conversion for brewed coffee was calculated as:

where represents the compound concentration in the beverage (mg/mL), V presents beverage volume (mL), and represents the dry mass coffee powder used (g).

The standardized dataset was processed using Microsoft Excel 2021, and box plots for CQAs, Cf, Tr, and Th were generated using R Studio. Prior to comparative statistical analysis, normality and homogeneity of variance were assessed using the Shapiro-Wilk test and Levene’s test, respectively. Because unequal variances were observed between groups, differences in chlorogenic acids and alkaloid levels between Arabica and Robusta coffee were subsequently evaluated using Welch’s t-test implemented in R Studio. The percentage of compound degradation was calculated from the relative difference between concentrations at each processing stage using the following equation:

where represents the concentration in green beans and represents the concentration in roasted beans or brewed coffee.

Daily intake was estimated based on beverage concentration, serving volume, and assumed consumption frequency. Intake was calculated using the following equation:

Results and discussion

Number of publication coffee origin

The literature search was limited to publications from 2015 to July 2025. Eligible studies that met the inclusion criteria comprised references within this time frame. Furthermore, during the screening process, publications reporting non-quantitative data on chlorogenic acids and alkaloids were excluded. The search and screening process resulted in a total of 50 articles being included. The number of studies published on Arabica and Robusta coffee, as well as the countries of origin of the respective samples, are presented in Figures 2(a) and 2(b).

Figure 2 Distribution of included studies based on coffee species (a) and geographical origin of coffee samples by country (b) for publications published between 2015 and 2025.

Figure 2(a) shows the number of publications on Arabica and Robusta coffee included in this review during the period 2015 - 2025, indicating an increasing scientific interest in these species and their bioactive compounds. In the last 10 years, research has been dominated by Arabica coffee (60%) compared to Robusta coffee (40%). This predominance is noteworthy because the 2 species differ substantially in biochemical composition, genetic background, and ecological determinants, all of which shape their adaptive responses and baseline bioactive profiles. Robusta generally contains higher levels of caffeine and total chlorogenic acids than Arabica, which may be linked to ecological adaptation to greater pest pressure and more extreme environmental conditions, where caffeine functions as a natural defense compound [20]. In contrast, Arabica typically exhibits lower caffeine levels and distinct chlorogenic acid profiles influenced by cultivation at higher altitudes. In this review, Arabica and Robusta data were analyzed separately to maintain comparative accuracy, ensuring that differences in study proportion do not bias interspecies interpretation.

Figure 2(b) illustrates the origin of the coffee samples analyzed for chlorogenic acids and alkaloids (caffeine, trigonelline, and theobromine). The majority of the coffee samples included in this review originate from 3 major regions: Asia, Latin America, and Africa. Within Asia, Indonesia is the most frequently reported country of origin, followed by Laos, Vietnam, Thailand, China, India, Korea, and Jordan. In Latin America, the samples predominantly originate from Brazil and Colombia, along with Guatemala, Honduras, Peru, and Mexico. In Africa, the represented countries include Ethiopia, Uganda, Kenya, Rwanda, and the Democratic Republic of Congo. Geographical origin plays a critical role in determining the baseline composition of bioactive compounds prior to processing. Terroir-related factors such as soil composition, altitude, temperature regime, and rainfall patterns significantly influence the biosynthesis and accumulation of chlorogenic acids and alkaloids in green coffee beans [13]. Therefore, regional grouping of sample origins helps explain part of the inter-study variability observed in reported compound concentrations.

Altitude is one of the most consistently reported environmental determinants affecting coffee chemistry. Several studies have shown that coffee cultivated at higher elevations tends to accumulate higher concentrations of certain chlorogenic acids, potentially due to slower maturation rates and increased environmental stress responses [21,22]. Temperature gradients associated with altitude may also modulate enzymatic pathways involved in phenolic metabolism.

In addition to altitude, other pre-harvest factors such as rainfall distribution and nitrogen fertilization practices also play a central role in shaping alkaloid biosynthesis and nitrogen metabolism. Increased nitrogen availability has been associated with enhanced caffeine synthesis, whereas water stress may alter phenolic accumulation patterns. Under nitrogen-deficient conditions, caffeine may partially function as a recyclable nitrogen pool through limited degradation into allantoin, while this pathway appears less active when nitrogen is sufficient [23]. These findings highlight that pre-processing compositional differences are often established at the cultivation stage rather than during roasting or brewing alone.

Chlorogenic acids and alkaloid in green bean

The total chlorogenic acid content in green coffee beans analyzed from the literature included in this review is presented in Figure 3. The analysis showed that the mean total chlorogenic acids in Arabica green beans ranged from 45.39 to 71.97 mg/g db, while in Robusta green beans it ranged from 50.40 to 142.88 mg/g db.

Figure 3 Total chlorogenic acids content in green beans at various post-harvest processes in Arabica and Robusta coffee; *data of Robusta coffee.

These findings show that the CQAs content in Robusta coffee was higher than in Arabica coffee, consistent with the results reported by Jeszka-Skowron et al. [13]. This difference was particularly evident in natural-processed Robusta (142.88 mg/g db), which contained nearly twice the amount of CQAs compared to Arabica (68.23 mg/g db). In Arabica coffee, the honey and natural processes showed similar ranges and similar mean CQA levels, whereas the wine, semi-washed, and full-washed processes exhibited comparable but slightly lower concentrations. Differences in CQAs content among post-harvest procedures are associated with the loss of compounds during fermentation and their solubilization in water during the washing process, as well as the degradation of components caused by sun-drying [10].

Figure 4 showed that the mean caffeine content in Arabica coffee ranged from 11.29 to 25.43 mg/g db, while in Robusta it ranged from 23.90 to 56.40 mg/g db. Trigonelline content in Arabica coffee was between 7.20 and 12.95 mg/g db, whereas in Robusta it ranged from 5.00 to 9.74 mg/g db. Theobromine content in Arabica coffee was relatively low, ranging from 0.12 to 0.18 mg/g db, while in Robusta it was only detected in the natural process at 0.03 mg/g db.

Figure 4 Alkaloid contents (caffeine, trigonelline, theobromine) in green beans at various post-harvest processes in Arabica and Robusta coffee; *data of Robusta coffee; **data of Arabica coffee.

The highest caffeine content in Arabica coffee was found in the natural process (25.43 mg/g db), while in Robusta it was observed in the wine process (56.40 mg/g db). According to the findings of Halagarda and Obrok [24], coffees processed using natural and anaerobic fermentation methods showed higher caffeine levels compared to those processed with full-wash methods. These results indicate that natural and wine processing methods are more effective in preserving caffeine content in coffee beans.

Overall, the variability in chlorogenic acid and alkaloid concentrations observed in green coffee beans across studies reflects the combined influence of geographic origin, agronomic conditions, and post-harvest practices. Beyond cultivation factors, differences in post-harvest processing methods (dry, semi-washed, or full washed) can further modify the basic composition of green coffee beans through microbial fermentation, leaching, and enzymatic transformations. Consequently, the chemical profile established at the green bean stage represents the baseline from which subsequent transformations occur, and therefore plays an important role in shaping the compositional changes observed during roasting and brewing.

Chlorogenic acids and alkaloid in roasted bean

Figure 5 shows the changes in total chlorogenic acid (CQAs) content in Arabica and Robusta coffee at different roasting levels. At light roasting, Robusta showed higher CQAs content (56.3 mg/g db) compared to Arabica (average 32.69 mg/g db). At medium roasting, Arabica showed higher levels (36.86 mg/g db) than Robusta (12.26 mg/g db). A similar pattern was observed at the dark level, where Arabica (11.7 mg/g db) showed higher CQAs than Robusta (8.25 mg/g db), before sharply decreasing at very dark roasting (11.17 mg/g db).

Figure 5 Total chlorogenic acids content in roasted beans at various roasting levels (light, medium, dark) of Arabica and Robusta Coffee.

These findings indicate that Robusta consistently showed a reduction in CQAs content as roasting progressed, while Arabica showed a slight increase at medium roasting before dropping sharply at very dark roasting. These findings indicate that Robusta consistently shows a decrease in CQAs content with roasting, while Arabica shows a slight increase at medium roasts before declining sharply at very dark roasts. A similar trend was also observed in Arabica coffee from Ethiopia showed 21% increase in chlorogenic acid content from light to medium roasting [25]. These trends reflect the thermal instability of chlorogenic acids during roasting. Increasing roasting temperature and duration promote several reactions, including isomerization, hydrolysis, and thermal degradation of CQAs. Among the various isomers, 5-CQA is considered the most thermally susceptible, leading to a progressive reduction in total CQAs as roasting intensity increases [18]. During roasting, chlorogenic acids can undergo ester bond cleavage, producing caffeic acid and quinic acid, which subsequently participate in further reactions such as oxidation, decarboxylation, and condensation reactions. In addition, CQAs can also undergo lactonization, forming chlorogenic acid lactones that contribute to the bitterness and complexity of roasted coffee flavor. Further thermal degradation may generate volatile phenolic compounds such as guaiacol and 4-vinylguaiacol, which play important roles in the aroma profile of roasted coffee [16].

Figure 6 illustrates the variation in alkaloid content of Arabica and Robusta coffees across different roasting levels. The average caffeine content in Arabica ranged from 11.29 to 25.43 mg/g db, compared to 23.90 to 56.40 mg/g db in Robusta, indicating consistently higher levels in the latter. Trigonelline concentrations in Arabica varied be-tween 7.20 and 12.95 mg/g db, with the highest levels observed at medium roast, com-pared to 5.00 to 9.74 mg/g db in Robusta. Theobromine content was relatively low in both species, ranging from 0.53 to 0.02 mg/g db in Arabica and from 0.03 to 0.01 mg/g db in Robusta.

Figure 6 Alkaloid contents (caffeine, trigonelline, theobromine) in roasted beans at various roasting levels of Arabica and Robusta Coffee.

Caffeine content decreased slightly with increasing roasting level, a trend consistent with previous findings [26]. As an odorless and bitter compound, caffeine is considered highly stable during roasting, with only minor losses attributable to sublimation or thermal degradation. Trigonelline in Arabica coffee exhibited a pattern similar to chlorogenic acids, increasing from light to medium roast before declining sharply at dark roast. In contrast, Robusta coffee showed a consistent decrease in trigonelline content across all roasting levels.

The coffee roasting process induces several chemical reactions that significantly affect the stability of bioactive components, particularly chlorogenic acids and trigonelline. The thermal degradation pathways of these compounds are illustrated in Figures 7 and 8.

Figure 7 Proposed thermal degradation pathway of chlorogenic acid during coffee roasting with formation of non-volatile compounds (modified from Gigl et al. [27]). Drawn using Chem Draw Ultra 12.0.

Chlorogenic acid, an ester between caffeic acid and quinic acid, which undergoes thermal degradation during roasting. According to Gigl et al. [27], heating causes the cleavage of the ester bond, resulting in caffeic acid and quinic acid as the primary products. Caffeic acid subsequently undergoes further reactions such as decarboxylation, oxidation, and alkylation, producing 3,4-dihydroxybenzene, hydroxyhydroquinone, and 4-ethylcatechol, respectively. These transformations contribute to the development of brown colored compounds and the characteristic flavor profile of roasted bean coffee.

Meanwhile, trigonelline also undergoes significant thermal degradation. According to Stadler et al. [28], the main reactions involved include N-demethylation and de-carboxylation. The N-demethylation pathway produces nicotinic acid. Meanwhile, the decarboxylation pathway produces 1-methylpyridinium, which can undergo further reactions through nucleophilic reactions, methyl rearrangements, and methyl dis-placement, resulting in the formation of various pyridine derivatives such as 2-hydroxy-1-methyl-1,2-dihydropyridine, 2- and 4-methylpyridine, and 1,2-, 1,3-, and 1,4-dimethylpyridinium. These products are known to contribute to the formation of volatile compounds in roasted bean coffee.

Figure 8 Proposed thermal degradation pathway of trigonelline during coffee roasting with formation of non-volatile compounds (modified from Stadler et al. [28]). Drawn with Chem Draw Ultra 12.0.

Chlorogenic acids and alkaloid in brewed coffee

Figure 9 presents the total chlorogenic acid and alkaloid levels in Arabica and Robusta coffees prepared using different brewing methods. In Arabica coffee, espresso yielded the highest total chlorogenic acid (4.20 mg/g db) and caffeine content (1.64 mg/g db). In contrast, Robusta coffee brewed as cold brew showed the highest levels of chlorogenic acid (4.04 mg/g db), caffeine (1.94 mg/g db), and trigonelline (0.58 mg/g db). For Arabica, both trigonelline (0.76 mg/g db) and theobromine (0.26 mg/g db) were also highest in cold brew. Robusta, however, exhibited detectable theobromine only in boiled, at 0.01 mg/g db.

Figure 9 Total chlorogenic acids (CQA’s) and alkaloid contents (caffeine, trigonelline, theobromine) in brewed coffee prepared using various brewing methods of Arabica and Robusta coffee; *data of Arabica coffee.

Brewed coffee retains suspended compounds such as chlorogenic acids and alkaloids, with their concentrations being affected by factors including filtration, pressing, and the grind size of the coffee grounds [29]. The type of extraction method and the length of extraction are the main drivers of variation across different brewing techniques. As a result, adjustments in brewing parameters lead to differences in the levels of these bioactive compounds present in the final cup of coffee.

Changes in Chlorogenic acids and alkaloid from coffee bean to brew

Figure 10 illustrates the variation in total chlorogenic acids (CQAs) and their major isomers (3-CQA, 4-CQA, and 5-CQA) in Arabica and Robusta coffees, from green beans to roasted beans and brewed coffee. In green beans, Robusta exhibited substantially higher CQAs content (107.99 mg/g db) compared to Arabica (64.41 mg/g db). Following roasting, CQAs levels decreased markedly in both species, with Arabica retaining 30.43 mg/g db, while Robusta showed a sharp decline to 15.26 mg/g db. In brewed coffee, only trace amounts of CQAs were detected, averaging 0.87 mg/g db in Arabica and 0.32 mg/g db in Robusta. These results demonstrate that although Robusta initially contains more CQAs, Arabica tends to retain a greater proportion after roasting and brewing.

Figure 10 Total chlorogenic acids (CQA’s) and mono-caffeoylquinic acids (3-CQA, 4-CQA, 5-CQA) content across green beans, roasted beans, and brewed coffee of Arabica and Robusta coffee.

Figure 11 and Table 3 summarize the average concentration of dicaffeoylquinic acid (diCQA) isomers (3,4-diCQA, 4,5-diCQA, and 3,5-diCQA) in green beans, roasted beans, and brewed Arabica coffee.

Figure 11 Dicaffeoylquinic acids (di-CQAs) content in green beans, roasted beans, and brewed coffee of Arabica samples (n = 27) *mean value.

Table 3 Change dicaffeoylquinic acids of Arabica coffee

* Change after roasting process (green bean – roasted bean); ** change after brewing (green bean - brewed).

In green beans of Arabica coffee, diCQAs were consistently present at relatively high levels, with 3,5-diCQA reported as the dominant isomer (3.62 mg/g db), followed by 4,5-diCQA (1.37 mg/g db) and 3,4-diCQA (0.96 mg/g db). Across the literature, roasting was associated with pronounced reductions in diCQA content, reaching 78.82%, 79.81%, and 95.58% losses for 3,4-diCQA, 4,5-diCQA, and 3,5-diCQA, respectively, indicating that 3,5-diCQA is the most heat-sensitive isomer. Further decreases were consistently observed in brewed coffee, where only trace amounts of 3,4-diCQA (0.02 mg/g db, 97.49% reduction) and 3,5-diCQA (0.06 mg/g db, 98.26% reduction) remained, while 4,5-diCQA was no longer detectable.

Figure 12 shows the variation in caffeine, trigonelline, and theobromine content in Arabica and Robusta coffees from green beans to roasted beans and brewed beverages.

Figure 12 Alkaloid contents (caffeine, trigonelline, theobromine) across green beans, roasted beans, and brewed coffee of Arabica and Robusta coffee.

Based on the analyzed literature included in the review, in green beans, Robusta contained a markedly higher caffeine concentration (mean 45.58 mg/g db) compared to Arabica (22.15 mg/g db). After roasting, caffeine levels declined significantly in both species, with Arabica retaining 13.28 mg/g db and Robusta 18.06 mg/g db. In brewed coffee, only trace levels were detected, averaging 1.19 mg/g db in Arabica and 1.78 mg/g db in Robusta.

Trigonelline followed a similar trend, with Arabica green beans containing 10.68 mg/g db compared to 8.31 mg/g db in Robusta. Roasting reduced these values to 6.31 mg/g db in Arabica and 2.99 mg/g db in Robusta, while brewed coffee contained less than 1 mg/g db in both species 0.66 mg/g db in Arabia and 0.43 mg/g db in Robusta. Theobromine levels were low in both coffees, with Arabica showing slightly higher concentrations. Green beans of Arabica averaged 0.19 mg/g db, which decreased to 0.15 mg/g db in roasted beans and 0.06 mg/g db in brewed coffee. Robusta, by contrast, exhibited negligible levels across all stages, ranging from 0.03 to 0.02 mg/g db.

Table 4 presents the changes in total chlorogenic acids and alkaloids in Arabica and Robusta coffee during roasting and after brewing. The contents of total chlorogenic acids and trigonelline showed significant differences between Arabica and Robusta in green beans, roasted bean, and brewed coffee (p < 0.05), except for theobromine, which showed no significant variation (p > 0.05). Caffeine also exhibited significantly differences between the 2 species in green and roasted beans (p < 0.05), although no significant difference was observed in brewed coffee (p > 0.05). The average decrease in chlorogenic acids and alkaloids across processing stages is illustrated in Figure 13.

Table 4 Concentration of total chlorogenic acid and alkaloid of Arabica and Robusta coffee.

* Change after roasting process (green bean – roasted bean); ** Change after brewing (green bean – brewed); Value with different superscript in the same row are significantly different (p > 0.05) (t-Test).

Figure 13 Changes in total chlorogenic acids (a), caffeine (b), trigonelline (c), and theobromine (d) of Arabica and Robusta across processing stages. Values are the mean and standard deviation.

As presented in Table 4, the percentage reduction of total chlorogenic acids from green beans to roasted beans ranged from 31.98% - 79.76% in Arabica and 69.66% - 76.63% in Robusta. Caffeine decreased by 62.33% - 78.18% in Arabica and 44.43% - 82.90% in Robusta. Trigonelline showed reductions of 1.78% - 54.41% in both Arabica and Robusta, while theobromine decreased by 85.83% in Arabica and 22.57% - 40.17% in Robusta. In brewed coffee, total chlorogenic acids in Arabica were reduced by 99.00% - 99.16%, caffeine by 53.76% - 92.54%, trigonelline by 80.58% - 94.51%, and theobromine by 6.86% - 91.03%. Mean-while, in Robusta brews, total chlorogenic acids decreased by 56.13% - 92.69%, caffeine by 88.03% - 97.06%, trigonelline by 68.00% - 77.55%, and theobromine by 62.25% -78.17%. These values highlight that although both species undergo substantial losses of bioactive compounds during roasting and brewing, Robusta generally shows greater reductions compared to Arabica.

Processing conditions also play a crucial role. Roasting intensity represents the main driver of chlorogenic acid degradation, whereas brewing parameters such as coffee-to-water ratio, extraction time, and temperature influence the final concentration of bioactive compounds in the beverage. Analytical variability across studies, including differences in extraction procedures and chromatographic detection methods, may also contribute to the dispersion of reported values.

In addition, methodological quality across the included studies was evaluated through a risk of bias assessment covering selection, performance, detection, attrition, and reporting domains. Most studies showed a low risk of bias in the selection and attrition domains, reflecting adequate reporting of sample sources and replication procedures. However, detection bias was frequently categorized as unclear due to limited reporting of analytical validation parameters, particularly the absence of limit of detection (LOD) and limit of quantification (LOQ) values in several studies. Detailed assessments are provided in Supplementary Table S2.

Despite these sources of heterogeneity, the compiled evidence indicates that the transformation of coffee bioactive compounds during processing follows predictable compound-specific stability patterns. Chlorogenic acids are highly sensitive to thermal degradation, trigonelline acts as a precursor of roasting derived aroma compounds, and caffeine remains chemically stable. These patterns suggest that processing intensity rather than extraction alone is the primary determinant of phenolic losses in coffee.

Contribution of coffee brew to the daily intake of chlorogenic acid and caffeine

The contents total of chlorogenic acids (CQAs) and caffeine in coffee are influenced by both the coffee species and the brewing method used. These variations affect the estimated daily intake required to obtain potential health benefits. Table 5 presents the range of CQAs and caffeine compositions in various types of Arabica and Robusta brews, as well as the estimated daily intake values.

Table 5 Concentration of total chlorogenic acid and alkaloid of Arabica and Robusta coffee.

^aCoffee brewed with a serving volume of 100 mL/cup; ^bCoffee brewed with a serving volume of 30 mL/cup; Values represents the minimum-maximum range.

Based on the data in Table 5, Arabica brews coffee shows total chlorogenic acids levels ranging from 13.00 - 337.70 mg/cup and Robusta brews coffee has CQAs levels between 329.00 - 411.00 mg/cup. Accordingly, the estimated intake of chlorogenic acids and alkaloids from brewed coffee remains within ranges commonly associated with potential health benefits when coffee is consumed in moderate amounts. Previous studies have reported that the intake of CQAs in the range of approximately 13.5 - 1,200 mg/day, which can be achieved through moderate coffee consumption, has been associated with several beneficial physiological effects. These include reduce fasting blood glucose levels, improved glucose tolerance, weight control, and improved blood pressure in hypertensive patients [30]. The metabolic benefits of coffee are largely attributed to the biological activity of chlorogenic acids, which are known to influence glucose metabolism and insulin sensitivity [31]. Therefore, moderate coffee consumption may serve as a practical dietary source of bioactive compounds, particularly CQAs, that may contribute to metabolic health while remaining within the recommended daily safety limits for caffeine intake [32].

Conclusions

This review indicates that variations in chlorogenic acid, caffeine, trigonelline, and theobromine concentrations between Arabica and Robusta coffees are influenced by several factors, particularly roasting and brewing processes. On average, the total concentrations of chlorogenic acids and caffeine in green Robusta beans are higher than those in Arabica. The total chlorogenic acids decreased by 52% - 99% in Arabica and 86% - 97% in Robusta during roasting and brewing. Trigonelline levels declined by 41% - 94% in Arabica and 64% - 95% in Robusta, while caffeine decreased by 40% - 95% and 60% - 96% in Arabica and Robusta, respectively. Theobromine exhibited smaller changes, with re-ductions of 21% - 68% in Arabica and 33% - 67% in Robusta. Selecting an appropriate brewing method can greatly influence the levels of bioactive compounds and, consequently, the overall quality of coffee. By optimizing the brewing process, coffee can be made to better align with consumer preferences and health considerations.

This systematic review synthesized quantitative evidence on the dynamic changes of chlorogenic acids and major alkaloids (caffeine, trigonelline, and theobromine) in Arabica and Robusta coffee across key processing stages, including green beans, roasting, and brewing. The compiled data confirm that species differences strongly influence the baseline composition of coffee beans, with Robusta generally exhibiting higher concentrations of caffeine and total chlorogenic acids than Arabica. Among the processing stages, roasting represents the most influential transformation step, leading to substantial degradation of several bioactive compounds. Total chlorogenic acids decreased by approximately 52% - 99% in Arabica and 86% - 97% in Robusta, while trigonelline declined by 41% - 94% and 64% - 95%, respectively. Caffeine showed comparatively greater stability, with reductions of 40% - 95% in Arabica and 60% - 96% in Robusta, whereas theobromine exhibited smaller changes, ranging from 21% - 68% in Arabica and 33% - 67% in Robusta. These results highlight that roasting intensity and brewing conditions play critical roles in determining the final concentration of bioactive compounds in brewed coffee.

Despite this, considerable variability was observed across studies. Differences in geographical origin, agronomic conditions, post-harvest processing, roasting levels, brewing parameters, and analytical reporting practices contribute to substantial heterogeneity in the collected dataset. Although concentration values were aligned to a common dry-basis unit, synthesis remains limited by inconsistent experimental designs and the absence of formal heterogeneity modeling across studies. Therefore, future research should prioritize standardized reporting of processing parameters and analytical methodologies, while integrating advanced molecular approaches such as targeted metabolomics and isotopic tracing to further elucidate the biochemical transformation pathways of coffee bioactive compounds throughout the processing chain.

Acknowledgements

This research was financial support provided by the Ministry of Higher Education, Science, and Technology of the Republic of Indonesia with the Grant No. 006/C3/DT.05.00/PL/2025 is grate-fully acknowledged.

Declaration of Generative AI in Scientific Writing

The authors acknowledge the assistance of generative AI tools for language refinement and grammatical editing only. No part of the content, data analysis, or interpretation was generated by AI. The authors remain fully responsible for the originality, accuracy, and conclusions presented in this manuscript.

CRediT Author Statement

Nazwa Nurhaliza: Conceptualization; Methodology; Software; Validation; Formal analysis; Investigation; Resources; Data curation; Writing - Original draft preparation; Visualization; Supervision; Project administration; Funding acquisition. Dian Herawati: Conceptualization; Methodology; Software; Validation; Formal analysis; Investigation; Resources; Data curation; Writing - Review and editing; Visualization; Supervision; Project administration; Funding acquisition. Nuri Andarwulan: Conceptualization; Methodology; Software; Validation; Formal analysis; Investigation; Resources; Data curation; Writing - Review and editing; Visualization; Supervision; Project administration; Funding acquisition.

References

[1] N Farhaty and Muchtaridi. Tinjauan kimia dan aspek farmakologi senyawa asam klorogenat pada biji kopi: Review (in Indonesian). Jurnal Farmaka 2016; 14(1), 214-227.

[2] International Coffee Organization. Coffee report and outlook. International Coffee Organization, London, 2023.

[3] HL Wachamo. Review on health benefit and risk of coffee consumption. Medicinal & Aromatic Plants 2017; 6(4), 1-12.

[4] A Farah. Coffee: Emerging health effect and disease prevention. 1^st ed. John Wiley and Sons, New Jersey, 2012.

[5] LT Anh-Dao, T Chi-Thien, N Nhut-Truong, T Tu-Chi, D Minh-Huy, N Thanh-Nho and N Cong-Hau. Changes in the total phenolic contents, chlorogenic acid, and caffeine of coffee cups regarding different brewing methods. Food Research 2024; 8(6), 71-79.

[6] S Mangiwa and YR Yabansabra. Kadar trigonelin dalam biji kopi Arabika (Coffea arabica) asal Wamena, Kabupaten Jayawijaya, Papua. SAINS: Jurnal MIPA Dan Pengajarannya 2016; 16(1), 29-34.

[7] PC Ayu. 2017, Pengembangan model penentuan kandungan kimia utama pembentuk flavour biji kopi java preanger menggunakan FT-NIRS (in Indonesian). Graduate’s Thesis. Institut Pertanian Bogor, Bogor, Indonesia.

[8] CF Tsai and IPJ Jioe. The analysis of chlorogenic acid and caffeine content and its correlation with coffee bean color under different roasting degree and sources of coffee (Coffea arabica typica). Processes 2021; 9(11), 2040.

[9] M Jeszka-Skowron, R Frankowski and A Zgola-Grzeskowiak. Comparison of methylxantines, trigonelline, nicotinic acid and nicotinamide contents in brews of green and processed Arabica and Robusta coffee beans-Influence of steaming, decaffeination and roasting processes on coffee beans. LWT 2020; 125, 109344.

[10] Y Yulianti, N Andarwulan, DR Adawiyah, D Herawati and D Indrasti. Physicochemical characteristics and bioactive compound profiles of Arabica Kalosi Enrekang with different postharvest processing. Food Science and Technology 2022; 42, 67622.

[11] KM Santosa, S Supriyadi and Y Rahmadian. Sensory analysis, caffeine, chlorogenic acid and non-volatile taste compounds of arabica coffee (Coffea arabica) fermented with sugar addition for brew taste. Indonesian Food and Nutrition Progress 2020; 17(2), 37-44.

[12] M Reza, D Herawati and F Kusnandar. Perbandingan asam fenolat dan total fenolik kopi Arabika Bogor pengolahan pascapanen dan tingkat sangrai berbeda (in Indonesian). Jurnal Mutu Pangan 2023; 10(2), 93-99.

[13] M Jeszka-Skowron, A Sentkowska, K Pryzynska and MPD Pena. Chlorogenic acids, caffeine content and antioxidant properties of green coffee extracts: Influence of green coffee bean preparation. European Food Research and Technology 2016; 242(8), 1403-1409.

[14] DS da Costa, TG Albuquerque, HS Costa and APA Bragotto. Thermal contaminants in coffee induced by roasting: A review. International Journal of Environmental Research and Public Health 2023; 20(8), 5586

[15] FBR Yeison, NG Guzman and JG Hernandez. Effect of the postharvest processing method on the biochemical composition and sensory analysis of arabica coffee. Engenharia Agrícola 2020; 40(2), 177-183.

[16] R Rusinek, B Dobrzanski, M Gawrysiak-Witulska, A Siger, A Zytek, H Karami, A Umar, T Lipa and M Gancarz. Effect of the roasting level on the content of bioactive and aromatic compounds in Arabica coffee beans. International Agrophysics 2024; 38(1), 31-42.

[17] D Herawati, MS Armawan, N Nurhaliza, FA Mu’Arij, D Hunaefi and S Noviasari. Impact of bean origin and brewing methods on bioactive compounds, bioactivities, nutrition, and sensory perception in coffee brews: An Indonesian coffee gastronomy study. International Journal of Gastronomy and Food Science 2024; 35, 100892.

[18] K Socala, A Szopa, A Serefko, E Poleszak and P Wlaz. Neuroprotective effects of coffee bioactive compounds: A review. International Journal of Molecular Sciences 2020; 22(1), 107.

[19] PRISMA 2020. PRISMA statement, Available at: http://www.Prisma-Statement.Org/, accessed March 2025

[20] P Orpong, W Sriwiang, S Chookaew, J Eamsiri, Saengkorakot, K Kehemthong, K Kamdee, R Photinam and R Picha. Comprehensive analysis of elemental profiles, caffeine and chlorogenic acid contents, antioxidant activities, and isotope ratios in Thai single-origin green coffee beans. Measurement: Food 2025; 20, 100265.

[21] RCE Dias, SEW Opitz and C Yeretzian. Bioactive compounds in blends of coffee defects originating from the harvesting. Coffee Science 2022; 17, e172027.

[22] RCE Dias, SEW Opitz and C Yeretzian. Bioactive compounds in blends of coffee defects originating from the harvesting. Coffee Science 2022; 17, e172027.

[23] MF Pompelli, GM Pompelli, AF de Oliveira and WC Antunes. The effect of light and nitrogen availability on the caffeine, theophylline and allantoin contents in the leaves of Coffea arabica L. AIMS Environmental Science 2013; 1(1), 1-11.

[24] M Halagarda and P Obrok. Influence of post-harvest processing on functional properties of coffee (Coffea arabica L.). Molecules 2023; 28(21), 7386.

[25] M Bolka and S Emire. Effects of coffee roasting technologies on cup quality and bioactive compounds of specialty coffee beans. Food Science & Nutrition 2020; 8(11), 5765-6297.

[26] M Gorecki and E Hallmann. The antioxidant content of coffee and its in vitro activity as an effect of its production method and roasting and brewing time. Antioxidants 2020; 9(4), 308

[27] M Gigl, O Frank, L Irmer and T Hofmann. Identification and quantitation of reaction products from chlorogenic acid, caffeic acid, and their thermal degradation products with odor-active thiols in coffee beverages. Journal of Agricultural and Food Chemistry 2022; 70(17), 5427-5437.

[28] RH Stadler, N Varga, J Hau, FA Vera and DH Welti. Alkylpyridinium. 1. Formation in model system via thermal degradation of trigonelline. Journal of Agricultural and Food Chemistry 2002; 50, 1192-1199.

[29] A Santanatoglia, E Schievano, I Menegazzo, L Fioretti, G Caprioli, S Vittori, G Sagratini and L Alessandroni. Cafestol and Kahweol content in different specialty coffee brews: Exploration by NMR analysis and evaluation of brewing parameters. Journal of Food Composition and Analysis 2025; 137, 106929.

[30] Y Yu, Z Zhang and C Chang. Chlorogenic acid intake guidance: Sources, health benefits, and safety. Asia Pacific Journal of Clinical Nutrition 2022; 31(4), 602-610.

[31] H Lu, Z Tian, Y Cui, Z Liu and X Ma. Chlorogenic acid: A comprehensive review of the dietary sources, processing effects, bioavailability, beneficial properties, mechanisms of action, and future directions. Comprehensive Reviews in Food Science and Food Safety 2020; 19(6), 3130-3158.

[32] European Food Safety Authority. Scientific opinion on the safety of caffeine. Efsa Journal 2015; 13(5), 4102.

[33] G Angeloni, L Guerrini, P Masella, M Innocenti, M Bellumori and A Parenti. Characterization and comparison of cold brew and cold drip coffee extraction methods. Journal of the Science of Food and Agriculture 2019; 99(1), 391-399.

[34] D Bauer, J Abreu, N Jordao, JSD Rosa, O Freitas-Silva and A Teodoro. Effect of roasting levels and drying process of Coffea canephora on the quality of bioactive compounds and cytotoxicity. International Journal of Molecular Sciences 2018; 19(11), 3407.

[35] M Worku, B De Meulenaer, L Duchateau and P Boeckx. Effect of altitude on biochemical composition and quality of green arabica coffee beans can be affected by shade and postharvest processing method. Food Research International 2018; 105, 278-285.

[36] Z Asfew and A Dekebo. Determination of caffeine content in Wollega Zones, Ethiopian coffee bean, pulp and leaves by high performance liquid chromatography. Trend in Phytochemical Research 2019; 3(4), 261-274.

[37] CA Pinheiro, LL Pereira, DB Fioresi, DD Oliveira, VM Osorio, JA da Silva, UA Pereira, MAG Ferrao, EM Riva-Souza, AFA da Fonseca and PF Pinheiro. Physico-chemical properties and sensory profile of Coffea canephora genotypes in high-altitudes. Australian Journal of Crop Science 2019; 13(12), 2046-2052.

[38] P Wongsa, N Khampa, S Horadee, J Chaiwarith and N Rattanapanone. Quality and bioactive compounds of blends of Arabica and Robusta spray-dried coffee. Food Chemistry 2019; 283, 579-587.

[39] N Chindapan, S Soydok and S Devahastin. Roasting kinetics and chemical composition changes of Robusta coffee beans during hot air and superheated steam roasting. Journal of Food Science 2019; 84(2), 292-302.

[40] IW Budiastra, M Sutrisno, S Widyotomo and PC Ayu. Determination of trigonelline and chlorogenic acid (CGA) concentration in intact coffee beans by NIR spectroscopy. AgricEbgInt: CIGR Journal 2020; 22(1), 162-168

[41] S Jung, S Gu, SH Lee and Y Jeong. Effect of roasting degree on the antioxidant properties of espresso and drip coffee extracted from Coffea arabica cv. Java. Applied Sciences 2021; 11(15), 7025.

[42] S Awwad, R Issa, L Alnsour, D Albals and I Al-Momani. Quantification of caffeine and chlorogenic acid in green and roasted coffee samples using HPLC-DAD and evaluation of the effect of degree of roasting on their levels. Molecules 2021; 26(24), 7502.

[43] J Montenegro, LS dos Santos, RG de Souza, LGB Lima, DS Mattos, BPPB Viana, ASCF Bastos, L Muzzi, CA Conte-Junior, EPR Gimba, O Freitas-Silva and AJ Teodoro. Bioactive compounds, antioxidant activity and antiproliferative effects in prostate cancer cells of green and roasted coffee extracts obtained by microwave-assisted extraction (MAE). Food Research International 2021; 140, 110014.

[44] MC Espindula, FJE Botelho, ADS Clemente, AL Teixeira, GQF Alves and RME Oliveira. Quality evaluation of Coffea canephora ‘Apoata’ seeds for rootstock production. Coffee Science 2018; 13(1), 1.

[45] A Duangjai, S Saokaew, BH Goh and P Phisalprapa. Shifting of physicochemical and biological characteristics of coffee roasting under ultrasound-assisted extraction. Frontiers in Nutrition 2021; 8, 724591.

[46] A Muzykiewicz-Szymanska, A Nowak, D Wira and A Klimowicz. The effect of brewing process parameters on antioxidant activity and caffeine content in infusions of roasted and unroasted Arabica coffee beans originated from different countries. Molecules 2021; 26(12), 3681.

[47] N Cordoba, FL Moreno, C Osorio, S Velasquez and Y Ruiz. Chemical and sensory evaluation of cold brew coffees using different roasting profiles and brewing methods. Food Research International 2021; 141, 110141.

[48] A Analianasari, EW Kenali, D Berliana, M Yulia and S Shintawati. Evaluasi pasca panen, cacat mutu dan atribut kimia (kafein, asam klorogenat) kopi robusta Lampung Barat (in Indonesian). Jurnal Teknologi & Industri Hasil Pertanian 2022; 27(1), 42-52.

[49] CDS Portela, IF de Almeida, TADD Reis, BRB Hickmann and MDT Benassi. Effects of brewing conditions and coffee species on the physicochemical characteristics, preference and dynamics of sensory attributes perception in cold brews. Food Research International 2022; 151, 110860.

[50] MC Espindula, FJE Botelho, ADS Clemente, AL Teixeira, GQF Alves and RME Oliveira. Quality evaluation of Coffea canephora ‘Apoata’ seeds for rootstock production. Coffee Science 2018; 13(1), 1.

[51] A Prasetya, MA Moulina, HM Arif and R Nofriansyah. Chemical quality, preference, and business analysis of Kepahiang Robusta coffee by roasting equipment type. SINTA Journal (Science, Technology, and Agricultural) 2022; 3(1), 33-40.

[52] LE Yulianti, DP Putri, IF Azizah, S Witman, MA Karim and A Rahayuningtyas. Physicochemical properties of ‘Cisalak’ Robusta coffee with hot air based roasting method. BIO Web of Conferences 2023; 69, 03014.

[53] A Santanatoglia, S Angeloni, D Bartolucci, L Fioretti, G Sagratini, S Vittori and G Caprioli. Effect of brewing methods on acrylamide content and antioxidant activity: Studying eight different filter coffee preparations. Antioxidants 2023; 12(10), 1888.

[54] AF Ramadhani and FC Nisa. Differences in chemical characteristics due to different roasting of robusta coffee beans. Coffee Science 2023; 18, e182116.

[55] R Dong, M Zhu, Y Long, Q Yu, C Li, J Xie, Y Huang and Y Chen. Exploring correlations between green coffee bean components and thermal contaminants in roasted coffee beans. Food Research International 2023; 167, 112700.

[56] AA Oksari, N Hasanah and RTM Sutamihardja. Characteristics of ground coffee quality on variations in temperature and roasting time for Robusta coffee (Coffea canephora Pierre Ex A.Froehner) green bean. Indonesian Journal of Applied Research 2023; 4(1), 11-16.

[57] J Gallardo-Ignacio, A Santibanez, O Oropeza-Mariano, R Salazar, RM Montiel-Ruiz, S Cabrera-Hilerio, M Gonzales-Cortazar, F Cruz-Sosa and P Nicasio-Torres. Chemical and biological characterization of green and processed coffee beans from Coffea arabica varieties. Molecules 2023; 28(12), 4685.

[58] AP Arum, SB Patricia, OF Alit, B Kusmanadhi, Setiyono and DA Savitri. Study of organic acid and caffeine content of Arabica coffee beans (Coffea arabica) processing of the Java Ijen Sukosari Bondowoso Farmers Group. IOP Conference Series: Earth and Environmental Science 2025; 1497(1), 012023.

[59] M Linda, F Bastian, M Mahendradatta, AB Tawali and A Laga. The effect of several postharvest processing on the quality of Robusta Coffee (Coffea canephora). IOP Conference Series: Earth and Environmental Science 2023; 1200, 012011.

[60] M Misto, S Indarti, Y Safitri, M Mutmainnah, T Mulyono and DD Alawiyah. Identification of chlorogenic acid, caffeine, melanoidin, sucrose, and protein content of local Indonesia Arabica coffee base on its cupping and variety variation. BIO Web of Conferences 2024; 101, 01003.

[61] Y Abubakar, MR Alya, A Karim and YD Fazlina. Effect of variety and farm altitude on chemical content of Arabica coffee (Coffea arabica) grown in Gayo Highland, Indonesia. IOP Conference Series: Earth and Environmental Science 2024; 1356, 012030.

[62] RA Firnabillah, E Mustaro and E Marwani. Biochemical composition and volatile profile analysis of three varieties of Coffea arabica and their correlation with the microclimate of Mount Tangkuban Perahu, West Java, Indonesia. Journal of Biological Diversity 2024; 25(2), 3264-3276.

[63] LB Acre, T Viencz, JS Francisco, RB Rocha, EA Alves and MT Benassi. Composition of Coffea canephora varieties from the Western Amazon. Journal of the Brazilian Chemical Society 2024; 35(8), e-20240031.

[64] LF Echeverri-Giraldo, VO Perez, CT Arboleda, LJV Gutierrez and LCI Quinchua. Content of acidic compounds in the bean of Coffea arabica L., produced in the Department of Cesar (Colombia), and its relationship with the sensorial attribute of acidity. Separations 2024; 11(2), 52.

[65] I Rahmawati and SH Asrori. Comparative analysis of caffeine content in cold and hot brewed robusta coffee using high-performance liquid chromatography (HPLC). JKPK 2024; 9(1), 15.

[66] YA Jeon, P Natraj, SC Kim, JK Moon and YJ Lee. Comparative analysis of phytochemical and functional profiles of Arabica coffee leaves and green beans across different cultivars. Foods 2024; 13(23), 3744.

[67] N Fitri, N Nurhaliza, ES Anggraeni, D Herawati, D Hunaefi and S Noviasari. Chemical composition, antioxidant activity, and sensory profile of espresso-based Arabica coffee from different bean origins. Beverage Plant Research 2024; 5(1), e008.

[68] BW Duke, MA Mokonon, ND Mekonin and KD Kuffi. Effects of roasting degree and grinding size on caffeine content and sensorial quality of coffee. Journal of Food Quality 2025; 2025, 2405668.

[69] LC Souza, MAG Ferrao, AFAD Fonseca, PS Volpi, M Comerio, LL Pereira, JAM Carvalho, SH Saraiva, PF Pinheiro, ALD Silva Junior and TCB Soares. An interdisciplinary approach for evaluating beverage quality in Coffea canephora. Anais da Academia Brasileira de Ciencias 2025; 97(2), e20240084.

[70] RS Katarzyna, AP Anna, B Maciej, SM Lidia, S Tomasz and SS Kinga. Roasting temperature as a factor modifying the caffeine and phenolic content of Ethiopian coffee. Processes 2025; 13(7), 2037.

Supplementary Materials

Table S1 Articles included in the review.

Table S2 Risk of Bias Assessment of included in the review.