Concentration of polycyclic aromatic hydrocarbons in roasted coffee beans: a systematic review

Coffee stands as one of the globe's most beloved beverages, with its consumption constantly on the rise. The development of coffee-based products can be attributed to its delightful taste, invigorating effects, and the numerous health advantages it offers individuals^[1]. Polyphenols, especially chlorogenic acids in coffee, show properties such as antioxidant, anti-inflammatory, antibacterial, and anti-atherosclerotic effects^[2,3]. Coffee beverages are made from roasted coffee beans^[4]. Roasting plays a crucial role in coffee production as it fosters the enhancement of flavor, aroma, and color. The time and temperature of roasting significantly impacts the formation of flavor elements^[5].

Coffee beans come from the coffee plant. Two types of Coffea arabica, commonly known as Arabica, and C. canephora, commonly known as Robusca, are cultivated for this purpose^[4].

Polycyclic aromatic hydrocarbons (PAHs) are produced by the incomplete combustion or decomposition of organic matter^[4]. PAHs are a collection of over 200 hydrophobic compounds composed of two or more fused aromatic rings^[4]. Based on the number of aromatic rings, PAHs can be categorized as either light compounds (2–3 rings) or heavy compounds (4–6 rings)^[6]. The United States Environmental Protection Agency (USEPA) has categorized 16 PAHs based on their prevalence and toxicity. These include benzo[a]anthracene (BaA), benzo[b]fluoranthene (BbF), indeno[1,2,3-cd]pyrene (IND), benzo[k]fluoranthene (BkF), dibenzo[a,h]anthracene (DBA), acenaphthene (Acp), acenaphthylene (AcPy), fluorene (Flu), phenanthrene (PA), anthracene (Ant), fluoranthene (FL), pyrene (Pyr), naphthalene (Nap), chrysene (CHR), benzo[a]pyrene (BaP), and benzo[g,h,i]perylene (BghiP)^[7].

Many PAHs are recognized as carcinogens, mutagens, and teratogens, presenting significant risks to human health and well-being^[8,9]. These compounds lead to colon, breast, and lung cancers^[4]. These compounds are also endocrine disruptors and increase the risk of cardiovascular disease^[8,10].

The most important human exposure to PAHs is through food^[11]. They are formed in foods during several processes such as roasting, frying, and grilling. The coffee roasting process can lead to the formation of PAHs^[12]. The procedure commonly occurs within the temperature range of 120 to 230 °C^[3]. Roasting time is usually 15 to 18 min^[13]. During coffee roasting, various processes occur including Maillard reactions, sugar decompositions, lipid oxidations, and pyrolysis. These reactions are responsible for the creation of the aroma, flavor, and color of coffee^[14]. During the roasting process of coffee beans, existing compounds such as fat, carbohydrates and amino acids are pyrolyzed and PAHs are formed^[4]. In addition, the coffee plant also receives PAH compounds through air, water, and soil^[15] (Fig. 1).

Figure 1. 

Pathways to PAH entry into coffee.

Various coffee sample preparation and analytical methods have been studied for PAH determination. These methods should demonstrate high specificity, repeatability, and reproducibility^[16]. Typically, traditional sample preparation methods involve extracting different solvents and then performing a suitable cleanup process^[3]. Some extraction methods are QuEChERS (quick, easy, cheap, effective, rugged, and safe), sonication, alkaline hydrolysis, supramolecular solvent microextraction (SUPRAS), Soxhlet extraction together with saponification, liquid–liquid extraction, accelerated solvent extraction, and ultrasound-assisted extraction. For the detection and quantification of PAH in coffee, commonly used methods include gas chromatography (GC), high-performance liquid chromatography (HPLC)^[3,17−20].

This systematic review aims to comprehensively examine the presence and levels of polycyclic aromatic hydrocarbons (PAHs) in coffee, factors affecting PAH levels, review analytical methods, sample preparation, compare PAH levels in coffee with other similar products, and summarize the risk assessments conducted.

According to the inclusion and exclusion criteria, two authors performed all the steps based on the PRISMA checklist to prevent bias.

Search strategy

On July 22, 2024, English articles from PubMed databases, Science Direct, Web of Science, and Scopus were received. This study was performed without a time limit. Five hundred and fourty four articles were received based on the search keywords including (PAHs or 'polycyclic aromatic hydrocarbon') and (coffee). The title and abstract were checked in the first stage, and according to the criteria, articles unrelated to the main topic were removed. Three authors, including AA, BM, and PS reviewed the studies to extract valuable information.

Inclusion and exclusion criteria

During the manuscript screening stage, the title and abstract were carefully read, and those that did not meet the inclusion criteria were excluded from this study. Exclusion criteria included review articles, book chapters, non-English articles, clinical studies, animal studies, and reviews of other contaminants. Inclusion criteria include all original articles that investigated the amount of PAHs in coffee. Furthermore, special attention was also paid to analytical methods and sample preparation to determine the PAH levels in different coffee samples. This information was extracted from the manuscripts and entered into this systematic study and discussed.

Figure 2 shows a diagram of the study. Five hundred and fourty four articles were obtained from the review of four databases. After removing the similar ones, 348 articles remained for title and abstract review. Two hundred and ninety four articles were excluded due to the use of animal studies, clinical studies, book chapters, review articles, and other contaminants. The full text of 54 articles was received, and in the final stage, 32 studies were selected for data extraction and inclusion in the systematic article.

Figure 2. 

PRISMA diagram.

Descriptive results of screened manuscripts

Authors Amirhossein Abedini and Bahare Mohamadi extracted important data shown in Table 1, and if a conflict arose, it was corrected by the third author (PS). Country, type of PAHs, PAH measurement method and main findings in Table 1 were extracted from 32 articles. The analytical method for determining the PAH levels in coffees was based on both liquid chromatography and gas chromatography.

Figure 3 shows the geographical distribution of studies. The most studies were conducted in Asia Pacific, followed by Europe and South America. Among countries, the most studies were conducted in Brazil. Brazil is one of the exporting countries and one of the largest consumers of coffee^[46,47].

Figure 3. 

The geographical distribution of studies (deep blue: 4 studies, medium blue: 2 or 3 studies, and light blue: 1 study).

This systematic review provides a comprehensive overview of current research on the prevalence and levels of polycyclic aromatic hydrocarbons (PAHs) in coffee. The analysis of polycyclic aromatic hydrocarbons (PAHs) in coffee from various countries shows significant variation in the types and concentrations of these compounds. These differences are due to several factors, especially the roasting conditions of the coffee beans. Roasting is done at different temperatures and for different lengths of time. As a rule, the higher the temperature and length of roasting, the greater the amount of PAHs produced^[36]. The health implications of these findings are substantial, especially considering the universal popularity of coffee. The detection of carcinogenic PAHs raises serious concerns regarding potential health risks, highlighting the need for careful monitoring and regulatory interventions to safeguard consumers.

Factors affecting the concentration of PAHs in coffee

Some studies have tracked the levels of benzo[a]pyrene in coffee. A notable finding is that benzo[a]pyrene, a carcinogenic PAH, was detected in varying amounts in the samples. For instance, its concentration in coffee grounds from Italy was measured at 9.0 ng/g^[15]. While brewed coffee from Brazil showed levels ranging from 0.0 to 46 ± 0.6 μg/L^[26]. In another study, the level of benzo[a]pyrene in instant coffee samples was reported to be around 0.1–0.4 μg/kg^[35].

Most studies considered total PAHs. In a study, the total PAH levels in roasted coffees was detected in the range of 13.75 ± 2.90 μg/kg^[27]. In another study, total PAH levels ranging from 3.5 to 16.4 μg/kg were detected in roasted coffee^[36]. The PAH levels in foods depend greatly on cooking conditions and temperature^[31]. A study revealed that higher roasting temperatures were correlated with increased PAH concentrations^[18]. Similarly, research conducted by Nguyen et al., from Vietnam indicated that longer roasting times and higher temperatures resulted in elevated levels of various PAHs^[24]. The authors of this study considered a temperature of 200 °C for 50 min to be appropriate^[24]. In a previous study, increasing the roasting temperature of coffee to 260 °C resulted in the formation of two PAHs, pyrene and chrysene^[38].

These findings highlight the importance of carefully controlling the roasting processes to reduce PAH formation. Therefore, manufacturers should adopt standardized roasting protocols to alleviate health risks. It has been observed that coffee beans roasted at lower temperatures and lighter colors have lower PAHs than darker coffee beans roasted at higher temperatures^[38]. Similar to the results of this study, research conducted in the United States found that darker roasted coffees had higher levels of PAHs than their lighter counterparts^[40]. Another study also observed that as the roasting temperature increased, heavier molecular-weight PAHs were formed^[39]. Research undertaken in Brazil has shown that PAH profiles can differ significantly depending on the degree of roasting^[44]. The authors of a study stated that if mild roasting is done, the PAH levels will not increase. They also emphasize the use of electric heating systems^[32]. Also, to reduce the amount of PAHs during roasting, some researchers have suggested roasting in hot steam. Under these conditions, only five PAHs were detected, including acenaphthene, fluorene, phenanthrene, and anthracene acenaphthylene^[29].

The region where the coffee is produced also affects the PAH levels. Plants obtain PAHs from their environment, which includes water, soil, and air, so environmental conditions also affect the PAH levels of the environment^[18]. Research conducted in Iran showed that PAH levels differ depending on the coffee's country of origin, indicating that geographical sourcing plays a significant role in PAH contamination^[25,27]. It was also observed that the PAH levels of green Arabica coffee samples from different countries were different^[32].

Regarding the effect of species on PAHs, a study measured both roasted coffee species, Arabica and Robusta. B[a]P levels were reported in both ND^[33]. However, this topic is suggested for future research. In particular, one research study found that the Robusca species had higher levels of caffeine, suggesting the formation of a caffeine-PAH complex that may facilitate the transfer of these compounds to the brew^[44]. Therefore, there are some ambiguities in this regard.

The brewing conditions also affect the PAH levels. In the study, the total PAH levels in brewed coffee samples was detected in the range of 0.52 to 1.8 μg/L^[45]. Analyses of the extracted data show that different brewing methods affect PAH levels^{[3,4,9,37,38,42,43]}. It has been observed that if a coffee maker is used to brew coffee, greater amounts of PAHs are released from the coffee than when the coffee beans are immersed in boiling water^[18,20].

Some studies have been conducted on instant coffee. In a study by Grover et al., the PAH levels of four popular brands of instant coffee were measured. Significant differences were observed between the PAHs levels of different brands. The values ranged from 831.7–1,589.7 μg/kg. The authors attribute the observed difference to the manufacturing process^[7]. In a study, roasted coffee and instant coffee were compared. The sum of 25 PAHs was higher in instant coffee^[34]. This difference is due to the instant coffee production process, which increases the concentration of PAHs. However, PAHs increase with lower molecular weight, which are less dangerous^[34].

Health risk assessment

A few studies have conducted a risk assessment. Most studies did not identify any risk to the consumer by calculating carcinogenic and non-carcinogenic risks^[24]. In another similar study, the carcinogenic risk was calculated after measuring the PAHs of coffees and no risk was observed for coffee consumers^[25]. In a study by Roudbari et al., after determining the amount of PAHs in roasted coffee samples, the carcinogenic risk was calculated, and the values ​​were negligible^[27]. In one study, the carcinogenic risk for four carcinogenic PAHs was calculated for roasted coffee and no results of concern were found^[34]. In addition, in the study by Wan et al., a risk assessment was also conducted for brewed coffee samples, which found a low level of concern^[20].

The PAH levels in roasted coffee is lower than in some grilled and smoked foods. In a study that looked at the PAH levels in a variety of foods, the total PAHs in roasted coffee was lower than in charcoal-grilled sausages and smoked fish^[31]. In a previous study, the PAH levels of three beverages based on coffee, tea, and cocoa were measured. Coffee had lower PAH levels than tea and cocoa^[39].

Analytical method and sample preparation for determining PAHs

Given that an acceptable limit for determining the PAH values in coffees has not been established, a valid measurement method must be used to measure these compounds^[15]. The choice of analytical method significantly influences the reported concentrations of PAHs and is probably one of the reasons for the heterogeneity between the results of different studies. The analytical method in most studies was either GC-MS or HPLC-based (Table 1). Regarding the analytical method, a relatively high recovery of the present compounds was observed with the gas chromatography coupled with mass spectrometry (GC-MS) in studies. The recovery percentage in one study using this method was reported to be around 94% to 106%^[34]. In two studies, GC with a FID detector was used. The recovery percentages were relatively good but lower than with GC-MS^[22,39]. Both UV and FLD detectors have been used in HPLC. However, the use of FLD detectors is more common^[28,35]. In two studies, the recovery rate using HPLC was about 87%^[33,35]. This method has been shown to have good sensitivity in measuring PAHs in coffee beans and brewed coffee^[18]. A study has shown that HPLC with an FLD detector performs analysis quicker and with greater sensitivity than GC-MS^[31].

Coffee has a complex matrix, so the sample preparation for the analytical method plays an important role in determining the PAH values^[4]. Regarding sample preparation, SPE cartridges were used in some studies^[35,45]. In some studies, the liquid–liquid extraction method was used and C-18 columns were used for the clean-up step^[30,32]. Some studies also used the QuEChERS method^[20,31]. This method is both accurate and reproducible^[31]. In some cases, the saponification process with alkalis was carried out to remove fatty acids^[36,37]. Coffee contains 15% fat. Therefore, saponification is an effective method for removing fat^[42]. In a study by Guatemala-Morales et al., it was observed that the percentage recovery increased after saponification^[36].

In more recent studies, synthetic adsorbents were used in the extraction step. In one study, Poly(o-phenylenediamine)-Zn composite was used for this purpose^[22]. Furthermore, the magnetic composite NiFe₂O₄@Ti₃C₂TX was used to prepare the samples. Under these conditions, the recovery percentage was about 84.5% to 112.6%^[21]. Also, in another study, a (Fe₃O₄@COF (Tp-NDA), which is a magnetic covalent organic framework nanocomposite was used for the preparation. The percentage recovery range was recorded as 74.6%–101.8%^[43].

In this systematic review, the detected levels of PAHs in coffee varieties were investigated. In almost all studies, it was concluded that, with increasing roasting temperature and duration, the amount of PAHs with higher molecular weight was higher. Information and education on the need for gentle roasting are essential. Continuous monitoring and improvements in brewing techniques may also help reduce the potential health risks associated with PAHs in coffee. The method of brewing coffee will also affect the amount of PAHs in the cup. Brewing with a coffee machine is usually higher. The location of the coffee is also an influence. This is due to environmental contamination of unprocessed green coffee. The measurement method was mostly based on GC-MS, but HPLC with a fluorescence detector is also sensitive. Risk assessments conducted in some studies indicated that the risk was negligible. For future studies, the difference between the PAH levels in the two types of coffee, Arabica, and Robusta, under the same roasting conditions is recommended.