RETRACTED ARTICLE: Use of electrochemical methods to determine the effect of brewing techniques (Espresso, Turkish and Filter coffee) and roasting levels on the antioxidant capacity of coffee beverage

Abstract

Coffee is a complex mixture of chemicals, which provide biologically active compounds with various health benefits. The some biologically active compounds arising from both its natural structure and formed after processing were determined as an antioxidant capacity of coffee beverages. In this study, we aimed to determine how roasting levels of Arabica coffee seed (light, medium, dark) and three brewing techniques—decoction methods (Turkish coffee), infusion method (filter coffee) and pressure methods (Espresso)—affect total antioxidant capacity in a cup of coffee beverage by electrochemical methods such as square wave stripping voltammetry (SWSV), differential pulse stripping voltammetry (DPSV) and cyclic voltammetry (CV). Antioxidant capacities of the coffee samples in terms of the equivalent amounts were determined according to standard oxidation peaks of rutin and caffeic acid. The highest antioxidant capacity was found in espresso coffee prepared at light roasting coffee seeds as equivalent the routine and caffeic at 9.4 ± 0.2 g/L and 19.7 ± 0.7 g/L, respectively with SWSV on a carbon paste electrode. As a result, SWSV, DPSV and CV voltammetric methods, fast, reliable, fully validated and without any pretreatment are alternative to conventional analytical methods to evaluation antioxidant values in any food samples.

Introduction

The coffee is an important beverage and accounting for 75% of the regular soft drink consumption. Around 500 billion cups of coffee is consumed per year in the world. The coffee are rich in compounds which have various kinds of health benefits with radical scavenging capability (Ciaramelli et al. 2019; Cano-Marquina et al. 2013). Rich source of biologically active compounds with determined antioxidant capacity of coffee and coffee beverages arising from compounds in both its natural structure and formed after processing (Kitzberger et al. 2014). Diversity of phenolic components provide potential health benefits and phenols play an important role in the formation of coffee flavor. Of the more than 800 volatile compounds found in roasted coffee flavors, the only 42 have been identified as phenols (Cordoba et al. 2019). Due to high potential health benefits, it is considered as a functional beverage (Farah and Lima 2019).

Polyphenols are mainly classified into phenolic acids and flavonoids. Hydroxycinnamic acids are non-flavonoid subclass of phenolic compounds in plants and found mostly as caffeic acid, a type of trans-cinnamic acid, and its derivative, chlorogenic acids (CGA) which is a major polyphenol of coffee (Cano-Marquina et al. 2013; Farah and Lima 2019). It is an ester formed between caffeic acid and ferulic acid, and quinic acid (Cano-Marquina et al. 2013) and hydrolyzed by intestinal microflora into various aromatic acid metabolites including caffeic acid and quinic acid (Itagaki, et al. 2011; Flament 2001). CGA takes part in formation of color, flavor and aroma during roasting. It’s amount in the coffee beverage depends on degree of roasting, the grinding, the ratio of coffee to water, the brewing method, the water temperature and contact time of coffee with water (Kitzberger et al. 2014; López et al. 2016). CGAs is responsible from astringent taste of coffee which can be easily recognized in a coffee brews (Flament 2001). In the in vitro study of Itagaki et al. (2011) showed that, antioxidant activity of caffeic acid is stronger than that of CGA and uptake of the CGA by Caco-2 cells was much less than that of caffeic acid. Caffeic acid (3,4-dihydroxycinnamic) is one of the hydroxycinnamate and phenylpropanoid metabolites in plant foods such as coffee drinks, blueberries, apples and cider (Magnani et al. 2014).

The pleasant aroma, taste and color of coffee beverages prepared from roasted coffee beans, are complex system, and physical–chemical properties are highly influenced by process conditions during roasting (Kitzberger et al. 2014). The roasting degree is controlled by roasting time and temperature and categorized as a light, medium or dark roast coffee bean (Vignoli et al. 2014). During roasting a number of different chemical reactions including Strecker degradation, Maillard reactions, oxidation, carbohydrate caramelization, degradation of polyphenols and formation of complex mixture of aroma compounds take place (Wongsa et al. 2019). Antioxidative capacity of roasted coffee are associated with the degradation of chlorogenic, malic and citric acid of green coffee and the formation of melanoidins and quinic acid (Król et al. 2020). Phenols produced and their components influenced by time and temperature of roasting (Wongsa et al. 2019). Studies showed that antioxidant activity of coffee decreases as the roasting degree increases due to the degradation of CGA (Vignoli et al. 2014). The biologically active substances such as pyrogallic acid, nicotinic acid, quinolinic acid, trigonelline, tannic acid, pyrogallic acid, caffeine, hydroxycinnamic acids, and Maillard reaction products, such as melanoidins are potential antioxidants of roasted coffee seed and beverages (Vignoli et al. 2011; Farah and Lima 2019).

Health benefits of coffee are mainly attributed to its antioxidant capacity of biological compounds such as phenolics which exhibit a wide range of physiological properties, such as anti-allergenic, anti-inflammatory, anti-artherogenic, antioxidant, antimicrobial, anti-thrombotic, vasodilatory effects, cardioprotective, aging, and antiproliferative properties (Lin, et al. 2016; Ciaramelli et al. 2019; Farah and Lima 2019). Flavonoids modulate a number of biological functions such as anti-inflammatory and anti-microbial activities with their ability to terminate free radicals, chelate metal ions, scavenge free oxygen (Lin et al. 2016). According to studies, drinking coffee may prevent cardiovascular diseases (CVD), Parkinson’s disease, reduce risk of developing type 2 diabetes, stroke, prevent the formation of gallstones, Alzheimer’s disease, gallbladder diseases and gout by reducing the level of uric acid in blood (Lim et al. 2020). In the systematic review and meta-analysis of Ding et al. (2019), they concluded that drinking coffee at 3–5 cups per day associated with reduction in CVD risk and heavy coffee consumption wasn’t associated with increase in CVD risk (Ding et al. 2014). Coffee which is an important source of the caffeine (purine alkaloid; 1,3,7-trimethylxanthine) has very important physiological effects such as increase awaking periods, reduce fatigue (Ludwig, et al. 2014), enhances alertness, concentration, and mental and physical performance (George et al. 2008). Another important potential of caffeine was found as suppressing body weight gain by stimulating thermogenesis, extending sympathetic stimulation, suppressing food intake and reducing adipose tissue mass (Ballis 2019).

The antioxidant capacity of coffee can be influenced by several factors, such as the variety and origin of coffee, the type and degree of roasting, the type of brewing techniques and properties of machines used in brewing (Fig. 1) (López et al. 2016). During the brewing process, there are several parameters that is effective on extraction mechanism of coffee soluble solids. Volatile and non-volatile flavour compounds are extracted from the ground coffee and form the final quality of coffee beverage. The size reduction of the roasted beans, contact of water with roasted coffee grounds is essential for the antioxidant composition and health properties of a coffee brew, and it is crucial step for extraction of coffee compounds (Fig. 1) (Ludwig, et al. 2012; Cordoba et al. 2019, 2020; López et al. 2016; Caprioli et al. 2015).

Fig. 1.

Illustration of main effective parameters on extraction mechanism for quality of coffee brews [11,47,48]

Coffee beverages can be prepared with different methods (Caprioli et al. 2015). Mostly three methods were used: decoction methods (boiled coffee, Turkish coffee, percolator coffee, and vacuum coffee), infusion methods (filter coffee and Napoletana), and pressure methods (Plugger, Moka, and espresso) (Cordoba et al. 2020). The filter, espresso and Turkish coffee are the widely known coffee brewing methods and differences between their extraction techniques with the other popular brewing methods are illustrated in Fig. 2. Although the Turkish coffee was the oldest type, there are few scientific papers about chemical attributes and flavor profiles compared to the filtered coffee and espresso brewing techniques (Cordoba et al. 2020).

Fig. 2.

Extraction techniques of coffee solubles in brewing methods: decoction methods (Turkish coffee, boiled coffee, percolator coffee, and vacuum coffee), infusion methods (filter coffee), and pressure methods (Plugger, Moka, and espresso) [48]

The Turkish coffee is the most popular brewing techniques prepared by the decantation methods. It is different than espresso and filter coffee beverage where no sediment is present and particle size of grinded coffee seed is very small like flour. In order to prepare Turkish coffee finely ground roasted coffee seeds are mixed with water at room temperature and heated until allowing to form foams on the surface, when foams are started to increase at nearly 92–95 °C, than poured carefully in a special coffee cup to prevent foam disruption (Fig. 1). Strong coffee with foam layer on the top and settled coffee particles as a sediment (not for drinking) at the bottom is known as Turkish coffee. It is prepared optimally with in the 3 min and not boiled (Özdestan 2014; Caprioli et al. 2015; Cordoba et al. 2020).

For all these reasons, the aims of this study were to determine the total antioxidant capacity, caffeic acid and rutin amount of the three different coffee brewing techniques (Turkish coffee, filter coffee and espresso) at three roasting degrees (light, medium, dark) by using CV, SWS and DPV techniques, and to investigate effect of roasting degree and brewing techniques on the responsible groups of antioxidants (caffeic acid and rutin content) and the total antioxidant capacity and also to compare results with other methods used before. This paper is the first study for comparison of antioxidant level of Turkish coffee with filter and espresso brewing techniques at different roasting temperatures and aimed to optimize a novel, fast, simple method for determination of antioxidant levels of coffee with CV, SWS and DPV techniques.

Material and method

Apparatus

The voltammetric measurements were carried by using a Vertex® One (Ivium) electrochemical analyzer with a solid electrode cell stand to which the three-electrode system was connected. The solid electrochemical cell connected with a carbon paste electrode (CPE, BASi MF-2010) as the working electrode, an Ag/AgCl (BASi, MF-2052) as the reference electrode and a platinum wire as the auxiliary electrode (BASi, MW-1032). In order to obtain electrochemical signals, cyclic voltammetry (CV), differential pulse stripping voltammetry (DPSV) and square wave stripping voltammetry (SWSV) were used under the optimized conditions. For the SWSV, accumulation potential, accumulation time, pulse amplitude, step potential, and frequency were found 0 mV, 30 s, 50 mV, 5 mV and 100 Hz, respectively. Furthermore, accumulation potential and accumulation time deduced 0 mV and 30 s for the DPSV method at pulse time of 10 ms and pulse amplitude of 50 mV on CPE in pH 4.0 Brittion-Robinson (B-R) buffer solutions. The pH measurements were carried out by using Mettler Toledo pH meter with an accuracy of ± 0.05 and electrochemical processes were performed at room temperature.

Reagents

Standard phenolic compounds of caffeic acid and rutin were purchased from Aldrich-Sigma at analytical standard. The all standard phenolic solutions were prepared as a concentration of 500 mg/L by their solvents. Britton Robinson (B–R), pH 4.0 buffer solution used as a support electrolyte, was equipped by 0.04 mol/L of acetic acid, ortho phosphoric and boric acid. The 2.0 M NaOH or 2.0 M HCl solutions were used to adjust supporting electrolyte at B-R buffer at pH 4.0. Distilled water was used in all solution preparation, washing and electrode cleaning processes. Also, all prepared stock solutions were stored in a refrigerator at 4 °C in a dark environment when not in use.

Process preparation of coffee samples

Most preferred handcrafted roasted Arabica coffee seed blends from Colombia, Costarica and Guatemala were used in the experiments were obtained from 3rd wave coffee producer company (Gastro Coffee Roaster). Equal amount of seeds were mixed and blend were prepared. Company roasted the coffee beans with same standards in the roaster (A roasting machine with PLC control system and a 15 kg capacity destoner connection By HasGaranti) at three different temperatures as light (105 Agtron Gourmet/180 °C), medium (85 Agtron Gourmet/205 °C) and dark (50 Agtron Gourmet/210 °C) and then grounded in the Industrial Mill by Has Garanti mill as fine powder for Turkish coffee samples, coarse size for filter coffee and bean size for espresso was prepared before brewing (The espresso machine grinds the core itself).

Turkish coffee was prepared using the automatic Turkish coffee maker (Arzum Okka). 7.0 gr grounded coffee and 80 ml distilled water at room temperature were added, mixed and heated until foamed twice (92–95 °C). Espresso coffee was prepared with 7.0 gr grounded coffee and nearly 35 mL distilled water by using the Jura Impressa XS9 Classic espresso machine at pressure of 9 bar (nearly 9 atm), and 90–95 °C. Filter coffee samples were prepared with the 7.0 gr grounded coffee with 125 mL purified water. All coffee brews were cooled to room temperature and then centrifuged for 5 min. The samples were stored at − 20 °C for further use. Before the entire brewing process, the coffee machines was thoroughly washed with ultra-pure water.

Results and discussion

Cyclic voltammetry measurements

In this study, we obtained cyclic voltammograms of rutin and caffeic acid (1 mg/L) to identification of standard peak current at a scan rate of 100 mV/s on the CPE between − 200 and + 1300 mV potentials (Fig. 3). It is also known that the peak position of phenolic compounds varies depending on the pH values. Therefore, a single support electrolyte as a pH 4.0 B-R buffer solution was used for each substance. Among these substances, only the caffeic acid exhibited a reversible electrode reaction, while other has an irreversible reaction as an oxidation according to the CV data’s. Furthermore, the caffeic acid exhibited an oxidation peak at 0.38 V by CV on CPE. But also, rutin (0.44 V and 1.03 V) has two oxidation peaks (Fig. 3). The common feature of both substances is that they have oxidation peaks at about 0.4 V. Therefore, qualitative analyzes of coffee samples for the antioxidant capacity evolution were determined based on this peak potential for standard phenolic compounds. In addition, in order to reckon quantitate analysis for total antioxidant capacity (TAC) of coffee samples by CV, concentration of rutin and caffeic acid was used as 1 mg/L. The main reason for this is that the peaks of caffeic acid and rutin are very sensitive and very intense on CPE at pH 4.0 B-R buffer solution. As can be seen in Fig. 3, the peak current values of the caffeic acid and routine at approximately 0.4 V were found to be 0.801 ± 0.025 µA and 0.331 ± 0.009 µA, respectively by CV at 100 mV/s scan rate in pH 4.0 B-R buffer solutions on CPE.

Fig. 3.

Peak currents of samples; F1:The CV for the standard phenolic compounds at 100 mV/s scan rate in pH 4.0 B-R buffer solutions. F2: The CV of Espresso coffee samples prepared with light, medium and dark roasted coffee beans at 100 mV/s scan rate in pH 4.0 B-R buffer solutions. F3: The SWSV for the standard phenolic compounds under the optimum condition in pH 4.0 B-R buffer solutions. F4: The SWSV for the filtered coffee samples brewing from various coffee beans (E_acc = 0 mV, t_acc = 30 s, ΔE = 50 mV, Es = 5 mV, f = 100 Hz in pH 4.0 B-R buffer solutions). F5: The DPSV for the standard phenolic compounds under the optimum condition in pH 4.0 B-R buffer solutions. F6: DPSV for the Espresso coffee samples brewing from various coffee beans ((E_acc = 0 mV, t_acc = 30 s, ΔE = 50 mV and ΔE_t = 10 ms in pH 4.0 B-R buffer solutions)

The cyclic voltammograms were taken to evolution total antioxidant capacity of the light, medium and dark roasted coffee beans by its brewing samples of Turkish coffee, Espresso and Filter coffee at 100 mV/s on the CPE in pH 4.0 B-R buffer solution. Referring to Fig. 3, the CV obtained at three roasted degree of Turkish coffee samples were seen, and the slightly roasted Turkish coffee sample has a reversible oxidation peaks at about 0.4 V. This obtained peaks represent exactly caffeic acid standard due to its similar peak potential of caffeic acid has one reversible peak at nearly 0.4 V in pH 4.0. Furthermore, total antioxidant capacities of coffee beverages were calculated in terms of equivalent routine and caffeic acid (Table 1).

Table 1.

Equivalent amount of rutin and caffeic acid of the Turkish coffee, filter coffee and espresso brewing at three roasting temperature determined by CV, SWSV, DPSV

Coffee Sample/ Method	Equivalent rutin (g/L)			Equivalent caffeic acid (g/L)
	Brewing technique			Brewing technique
	Turkish coffee	Espresso	Filter coffee	Turkish coffee	Espresso	Filter coffee
CV
LR	8.2 ± 0.3	225.5 ± 8.2	6.1 ± 0.3	47.0 ± 1.9	542.8 ± 9.9	14.7 ± 0.8
MR	0.5 ± 0.0	41.44 ± 1.1	4.3 ± 0.8	99.8 ± 2.7	99.8 ± 2.7	10.4 ± 1.8
DR	ND	ND	ND	ND	ND	ND
SWSV
LR	2.3 ± 0.1	9.4 ± 0.2	5.2 ± 0.2	5.0 ± 0.1	19.7 ± 0.7	11.7 ± 0.3
MR	0.4 ± 0.0	3.3 ± 0.1	2.4 ± 0.1	0.9 ± 0.0	7.4 ± 0.2	5. 3 ± 0.2
DR	0.1 ± 0.0	2.4 ± 0.1	0.8 ± 0.0	0.19 ± 0.0	4.9 ± 0.1	1. 8 ± 0.1
DPSV
LR	3.0 ± 0.1	13.4 ± 0.1	1.5 ± 0.1	5.7 ± 0.2	25.7 ± 1.0	2.8 ± 0.5
MR	0.2 ± 0.0	5.0 ± 0.1	1.0 ± 0.0	0.3 ± 0.0	9.6 ± 0.2	1.8 ± 0.0
DR	ND	3.6 ± 0.1	0.6 ± 0.0	ND	7.0 ± 0.1	1.1 ± 0.1

CV: Cyclic voltammetry

SWSV: Square wave stripping voltammetry

DPSV: Differential pulse stripping voltammetry

According to the results obtained by CV for the three roasted temperature and three brewing techniques, the maximum amount of antioxidant capacity was found in espresso brewing technique with light roasted coffee seed. However, no peak was observed in the dark roasted bean samples. It can be said that roasting removes phenolic compounds especially caffeic acid in the coffee samples by the effect of temperature. As a result, we can conclude that roasting level and brewing techniques directly affect antioxidant capacity of coffee beverage. The maximum antioxidant capacity was found in light roasted Espresso as equivalent the routine and caffeic by 542.750 ± 19.857 g/L and 225.475 ± 8.249 g/L, respectively.

Square wave stripping voltammetry measurements

We researched electrochemical behavior of standard rutin (0.5 mg/L) and caffeic acid (0.5 mg/L) for identification of references peak current and potential for the phenolic compounds under the optimum conditions in pH 4.0 on CPE. Furthermore, the caffeic acid exhibited an oxidation peak at nearly 0.38 V but also rutin has two oxidation peaks at 0.45 V and 1.03 V (Fig. 3). While the peak potential values obtained by SWSV were almost the similar for both items according to the CV data, a large increase in peak currents on SWS voltammograms was observed for the both agents. This also proves that SWSV is very sensitive than CV method, even at 4 times less concentration. The basic common feature of both substances is that they have an oxidation peak at about 0.4 V. Therefore, for the qualitative determination of the antioxidant capacity of coffee beverages, the peak potential and peak intensity values at approximately 0.4 V of the standard phenolic compounds by SWSV were referenced. Furthermore, in order to reckon quantitate analysis of total antioxidant capacity (TAC) for the coffee beverages by SWSV, concentration of rutin and caffeic acid was used as 0.5 mg/L under the optimum condition in pH 4.0 B-R buffer solution. The main reason for this is that the peaks of caffeic acid and routine in the SWSV are very sensitive than CV. As can be seen in Fig. 3, the peak current values of the caffeic acid and routine at approximately 0.4 V were found to be 48.212 ± 1.743 µA and 102.112 ± 3.682 µA, respectively by SWSV under the optimum condition on CPE.

SWS measurements were taken to determine the total antioxidant in different brewed coffee samples. Each measurement was repeated three times under the optimum conditions in pH 4.0 B-R buffer solutions on CPE. As can be seen in Fig. 3, in the SWS voltammograms for the Turkish coffee brews at three roasting temperature (light, medium and dark) have an oxidation peak at nearly 0.4 V (Fig. 3). The obtained oxidation peak at 0.4 V by SWSV for coffee samples may represent the peak as an antioxidant reference peak, since it generally has the same potential of phenolic compounds. In addition, the total antioxidant capacities of coffee samples were calculated in terms of routine and caffeic acid based on peak intensities of the standards (Table 1).

According to the results obtained by SWSV for coffee brews, the maximum antioxidant capacity was found at light roasted coffee bean brewed by espresso. Moreover, while antioxidant capacity of dark roasted coffee beans was not determined by CV, it was seen that quantitative total antioxidant determination was possible by SWSV method. The results obtained by SWSV are consistent with the CV results. Results showed that different roasting levels and brewing techniques have a direct effect on the phenolic compound contents and thus have a significant effect on their antioxidant capacity. The antioxidant capacity of espresso at light roasting degree was the maximum when compared to the Turkish and filter coffee brewing as equivalent the routine and caffeic by 9.358 ± 0.166 g/L and 19.700 ± 0.746 g/L, respectively. Increasing temperature of roasting resulted in the decrease in antioxidant capacity of coffee brews.

Differential pulse stripping voltammetry measurements

The moderated condition for the DPSV such as pulse time of 10 ms, pulse amplitude of 50 mV, step potential of 5 mV, accumulation time of 30 s and accumulation potential of 0 mV were optimized in pH 4.0 B-R buffer solutions. Afterwards, electrochemical measurements were carried out by DPSV in the presence of standard routine and caffeic acid at 0.5 mg/L, in order to deduce the total antioxidant amount in the coffee samples (Fig. 3). In the Fig. 3, there is a single oxidation peak for the caffeic acid at approximately 0.38 V. But also, two oxidation peaks at 0.45 V and 1.03 V were obtained for the routine by DPSV. Potential values of peaks designated by DPSV for both substances are almost identical to CV and SWSV. It therefore proves that total antioxidant capacities of coffee samples can be identified using the peak potential of 0.4 V for three voltammetric methods of standard substances. Peak height values ofstandard substances were determined depending on the concentration under moderated conditions. According to the Fig. 3, the peak current values of the caffeic acid and routine at approximately 0.4 V were found as 10.471 ± 0.450 µA and 20.101 ± 0.162 µA, respectively by DPSV on CPE under the optimum condition in pH 4.0 BR buffer solutions. According to these results, DPSV method is very sensitive compared to CV and sensitivity is low compared to SWSV.

Antioxidant capacity of three coffee beverages varied as espresso <filter coffee> Turkish coffee at three roasted levels by DPSV measurements (Table 1). No peak was obtained for dark roasted beans of Turkish coffee by the DPSV method in pH 4.0 B-R buffer solution (Fig. 3). The obtained values by DPSV are consistent with CV and SWSV results. The highest antioxidant capacity of coffee brews between three roasting levels as equivalent routine and caffeic acid was found in espresso at light roasted degree 13.378 ± 0.536 g/L and 25.682 ± 1.029 g/L, respectively.

Effect of roasting and brewing

Antioxidant capacities of Arabica coffee at three different roasting temperatures brewed with espresso, Turkish and filter coffee techniques were determined by electrochemical methods such as SWSV, DPSV and CV in terms of the equivalent amounts according to the standard rutin and caffeic acid which exhibited peaks at approximately 0.4 V in pH 4.0.

Results showed that the antioxidant capacity of coffee beverages changed significantly with both brewing techniques and roasting temperatures. Antioxidant activities was highest in espresso, moderate in filter coffee and lowest in Turkish coffee. As a result, SWSV, DPSV and CV voltammetric methods, fast, reliable, fully validated and without any pretreatment are alternative to conventional analytical methods to evaluation antioxidant values in any food samples. The amount of rutin and caffeic acid decreased as roasting temperature increased. Roasting intensity, brewing type with changing time of exposure, type of filtering, coffee grinding size, amount of water used may effected the chemical composition of coffee beverage (Caprioli et al. 2015; López et al. 2016; Cordoba et al. 2020).

In order to observe the effect of roasting degree and brewing techniques on the antioxidant activity, Eq. (1) is used. The effect was followed by using the antioxidant value at light roasting (LR) as an initial value. Percent loss were calculated by dividing antioxidant level of medium and dark roasting to their initial LR value. The values of rutin and caffeic acid obtained from the Table 1. The results were represented on Fig. 4.

$$\text{Loss}\%=\frac{\text{Antioxidant}\text{Value}}{\text{Antioxidant}\text{Value}\text{at}\text{LR}}\times 100$$ 1

Fig. 4.

Percent loss in caffeic acid and rutin at medium roasting (MR) and dark roasting (DR) of coffee beverages prepared by three different brewing technique (TC: Turkish coffee, Espr: Espresso, FC: Filter coffee)

First variable that result in the large differences between rutin and caffeic acid content in three brewing techniques was the roasting degree where amount of metabolites is decreased by the reaction occurring during the roasting. The temperatures were 180 °C for light, 205 °C for medium and 210 °C for dark roasting level. As seen from the temperatures, very narrow temperature increase needed to reach dark roasting level from medium roasting, nearly 5–10 °C heating. Coffee seeds reached dark level in a short time, showed that after a point, as temperature increase reaction rate increase rapidly, coffee seeds becomes more thermolabile. The rutin and cafeic acid has the similar temperature sensitivity where decrease rate of each brews was nearly similar at same roasting degree (Fig. 4.). For example, reduction in both caffeic acid and rutin content of filter coffee was 55% for SWSV, 29% for CV and 34% for DPSV method at medium roasting. Compared to brewing techniques, espresso has highest rutin (225.5 g/L) and caffeic content (542,8 g/L) then filter coffee and Turkish coffee (CV method) at light roasting degree. There were no available data studied the rutin and caffeic acid content of coffee beverages in the literature to compare.

Determination for effects of geographic regions, roasting degree profiles of CGA isomers on coffee beverages brewed by 10 min. contact of coffee with boiling water (100 ⁰C) at a ratio of 1:20 (w/v) showed highest decrease which was higher than 85% at dark roasting compared to light and medium roasting degree. Also antioxidant activity of coffee samples obtained from different regions was changing at same conditions (Liang et al. 2016).

During roasting process decrease in carbohydrates, proteins, CGAs, free amino acids, crude lipids, minerals and aliphatic acid level of green coffee beans were observed. Caffeine is stable against heating (Caprioli et al. 2015). CGAs are thermolabile compounds. Significant reduction in CGAs together with incorporation of a part of hydroxycinammic acids into the melanoidins were observed during roasting and brewing methods (Ciaramelli et al. 2019; Farah and Lima 2019; Derossi et al. 2018; Liang et al. 2016). Ozdestan (2014) concluded that most of the free amines and total amine content in green coffee decreased during roasting and also total bioactive amine amount of brewed coffees were significantly different from each other (Özdestan 2014).

Total phenol content of espresso, Turkish and Americano coffee was 22.01 mg GAE/mL, 9.87 mg GAE/mL and 5.05 mg GAE/mL, total antioxidant activity was 10.91 mg Trolox/mL for 5.45 mg Trolox/mL and 3.47 mg Trolox/mL respectively in the research of Derossi et al. (2018). They found that grinding levels had slight effect on the caffeine and phenol activity of coffee prepared by espresso, Turkish and Americano brewing. Controversial to the Dersossi et al. (2018), Severini et al. (2015) concluded that among the variables of the quality of espresso coffee, grinding was one of the most important factor at the constant volume, 9 bar pressure and 92 °C. Coarsest coffee ground provided more water flow and thus highest percolation rate and high caffeine amount (Derossi et al. 2018).

The result of a study showed that in espresso brews, CGA was higher than dripped/filtered or other brewing methods like Turkish coffee. According to results, CGA and caffeine content of espresso 3–6 times higher than moka and filtered coffees (Cordoba et al. 2020). The reasons were explained by higher ratio of grounded roasted coffee per water and high pressure techniques used for extraction in the espresso machine (mostly 9 bar). Lower CGA in cold brewing method was attributed to low water temperature used in extraction with short infusion periods. The studies concluded that heating up to 95 °C provides higher extraction of CGA but holding coffee brews at elevated temperatures may result in reduction in antioxidant values like in Turkish coffee brewing method (Farah and Lima 2019).

Studies of Sridevi et al. (2011) and Demir et al. (2020a, b) showed that the method of coffee brewing and roasting temperature had significant effect on total solid content and chemical composition in coffee brew (Demir et al. 2020b; Sridevi et al. 2011). Coffee prepared by espresso had highest solid content than Turkish coffee and filter coffee was the least one. As the roasting temperature increased decrease in the concentrations of free form of cafestol and kahweol were determined (Sridevi et al. 2011). Zhang et al. (2012) explained that cafestol extraction yield of coffee beverage depends on brewing type and degree of roasting. Coffee beverage prepared by boiled and French press at light roasted coffee seed had the highest cafestol level than Turkish coffee and Mocha and lowest level was found at dark roasted brewed with Turkish coffee and Mocha (Zhang et al. 2012).

Garjonovic et al. (2017) determined antioxidant capacity of coffee for instant, espresso, filter and Turkish brews by DC polarographic assay method. They found that the antioxidant capacity of espresso, filter and Turkish coffee samples were similar and instant coffee has highest value. There were no any information about the origin and roasting degree of coffee in the study (Gorjanovic´ et al. 2017). Result of Garjonovic et al. (2017) different than Demir et al. (2020a; b). Demir et al. (2020a, b) determined antioxidant capacity of light (180 °C), medium (205 °C) and dark roasted (210 °C) Arabica seed brewed with Turkish coffee and filer coffee as equivalent gallic acid and quercetin by electrochemical methods. Results showed that light roasted coffee beverage has the highest antioxidant capacity and filter coffee brew antioxidant capacity was higher than Turkish coffee at both three roasting level (Demir et al. 2020b) as in this study.

According to the most of studies, decrease in the CGA content of coffee beverages were observed with the increasing roasting temperature and extended time as in our study. Loss in the amount of CGA in the browning reactions such as Maillard reactions doesn’t mean a decrease in antioxidant capacity because there are also browning products formation (Liang et al. 2016). Maillard reactions are noncovalent reactions of CGA isomers and melanoidins that result in products with increasing antioxidant activity (Liang et al. 2016; Kocadağlı and Gökmen 2016).

Effect of extraction time

Extraction time is explained as another important factor that determine the coffee beverage quality (Cordoba et al. 2020; López et al. 2016; Caprioli et al. 2015). It is contact time of water with grounded coffee seed in the brewing methods. Soluble compounds are dissolved based on the extraction techniques and washed away with water. When extraction time is completed, coffee beverage with extracted soluble solids is ready for drinking (Fig. 1 and Fig. 2). Extraction time was 3 min for Turkish coffee, 2 min for filter coffee and 25–30 s for espresso. The particle size of grounded coffee seed used for Turkish coffee was fine powder which smaller than other two method and, ratio of water/grounded coffee seed used was also different. Long extraction time, higher amount of water used and smallest particle size in Turkish coffee may be reason for lowest antioxidant level than other two method.

In the study of López et al. (2016), they explained that there is significant effect of applied pressure and water temperature between espresso brewing techniques on extraction kinetic. High pressure (11 bar) increased volatiles with increased water temperature (López et al. 2016). Pressure creates driving force for flow of water through compact coffee cake and make easy for extraction of the soluble particles and oil droplets from grounded coffee beans. They also explained that most polar compounds were extracted within the first seconds and at longer extraction time, decrease in their concentrations were observed. Ludwig et al. (2012) found the higher phenolic compounds in filter coffee than espresso. Reason was explained as contact time of water with grounded coffee and they concluded that brewing time has one of the key factors determining the antioxidants level in coffee beverage (Ludwig, et al. 2012).

Conclusion

In our work, we presented a new electrochemical methods such as square wave stripping voltammetry (SWSV), differential pulse stripping voltammetry (DPSV) and cyclic voltammetry (CV) for constructing antioxidant capacity of coffee brews prepared with three different brewing techniques at three roasting levels. These methods are fast, reliable, fully validated and may be used for determination of antioxidant capacity of coffee beverages in the future.

Rutin and caffeic acid were determined as antioxidant level of coffee brews. They are important metabolites of coffee and also have health benefits for coffee. Our data showed that roasting temperature, brewing method, extraction time, ratio of grounded coffee seed/water, effect considerably the antioxidant capacity of coffee brews and also influence the potential health benefits of coffee.

In this study, it was difficult to use the results of antioxidant activity data of coffee beverages in the most researches for comparison, because of various types of methods with inconsistent data’s. Differences between data may be explained by the complex structure, variability in amount of coffee components according to conditions and specific techniques of the brews used in the studies and unstandardized methods. Amount of grounded coffee and water ratio, the extraction time, temperatures of roasting, type of coffee seeds, variation between brewing techniques, type of data representation with used units such as w/v, v/w, w/cup, w/dose changes and makes it difficult to compare and make correlation between results. But this inconsistent results provide new opportunities to make further researches.

Authors’ contribution

SY: methodology and formal analysis. ED: designed, planned, supervised the study, writing the original draft. Ilkay Gok: data curation and validation, HYAE: reviewing and editing the final draft and carried the necessary changes in format, grammar and English.

Funding

None.

Declarations

Conflict of 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. The authors declare no conflict of interest.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.