http://informahealthcare.com/ijf ISSN: 0963-7486 (print), 1465-3478 (electronic) Int J Food Sci Nutr, 2014; 65(4): 465–469 ! 2014 Informa UK Ltd. DOI: 10.3109/09637486.2013.873890
FOOD COMPOSITION AND ANALYSIS
Quantification of caffeine, trigonelline and nicotinic acid in espresso coffee: the influence of espresso machines and coffee cultivars Giovanni Caprioli1*, Manuela Cortese1*, Filippo Maggi1, Caterina Minnetti1, Luigi Odello2, Gianni Sagratini1, and Sauro Vittori1 School of Pharmacy, University of Camerino, Camerino, Italy, and 2Centro Studi Assaggiatori, Galleria V. Veneto 9, Brescia, Italy
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
1
Abstract
Keywords
Caffeine, trigonelline and nicotinic acid are important bioactive constituents of coffee. In this work, the combination of different water temperatures and pressures in the settings of the espresso coffee (EC) machine was evaluated, to assess how these factors influence how effectively caffeine, trigonelline and nicotinic acid are extracted from both Arabica and Robusta samples. The proposed analytical method, based on a high performance liquid chromatography (HPLC) system coupled to a variable wavelength detector (VWD), showed good linearity (R240.9985) and good recoveries (71–92%); after validation for three monitored compounds, the method was used to analyze 20 commercial samples. The combination of a temperature of 92 C and pressure at 7 or 9 bar seems to be the ideal setting for the most efficient extraction of these compounds and consequently for their intake; the compound extracted in the greatest quantity was caffeine, which was in the range of 116.87–199.68 mg in a 25 ml cup of coffee.
Arabica coffee, caffeine, espresso coffee machine, extraction pressure, extraction temperature, Robusta coffee
Introduction Coffee is a very popular hot drink, highly appreciated all over the world (Butler, 1999; Illy & Viani, 1995; Maeztu et al., 2001). According to the latest statistics from the International Coffee Organization (ICO), about 1.4 billion cups of coffee a day are consumed worldwide. The two most important species of coffee, Coffea arabica and Coffea canephora (Robusta), differ significantly in price, quality and consumer preference. Arabica coffee is more acidic, with a more intense aroma and a richer body than Robusta, which is instead more bitter and characterized by a typical earthy and woody flavor (Illy & Viani, 1995). The preparation of espresso coffee (EC) is influenced by obvious factors such as coffee and water, as well as other technical conditions related to the machine (Andueza et al., 2003; Caprioli et al., 2012). The classic EC machine features a volumetric pump that brings the water to the desired pressure (usually about 9 bar) and temperature (usually between 91 C and 96 C) through a heat exchanger. The water passes through the filter unit containing the pressed coffee, which absorbs some milliliters of water and swells. As the water passes through, the various components are extracted from the espresso (Odello & Odello, 2006).
*These authors contributed equally to this work. Correspondence: Gianni Sagratini, School of Pharmacy, University of Camerino, via S. Agostino 1, 62032 Camerino, Italy. Tel: +390737402238. Fax: +390737637345. E-mail: gianni.sagratini@ unicam.it
History Received 29 October 2013 Revised 27 November 2013 Accepted 7 December 2013 Published online 27 January 2014
As noted by Petracco (1989), and Illy & Viani (1995), there is a dearth of knowledge about the optimal weight of ground coffee and also about the beverage volume and the extraction conditions (water pressure and temperature) that must be used to obtain a high quality EC, judged in terms of aroma and composition (Caprioli et al., 2012; Nunes et al., 1997). Caffeine, trigonelline and nicotinic acid are three important bioactive components of coffee. Caffeine is found in various kinds of food and drink consumed in daily life (Singh & Sahu, 2006), coffee, in particular. It exerts various physiological effects, such as relaxation of bronchial muscle, stimulation of the central nervous system, gastric acid secretion and diuresis (Bolton & Null, 1981). Among the various types of coffee, Arabica has the highest caffeine content, and is considered one of the best quality varieties (Franca et al., 2005). Several chemical and physical methods have been developed to quantify the amount of caffeine in coffee and other beverages. The most widely used methods include various analytical techniques, such as derivative spectrophotometers (Alpdogan et al., 2002) and HPLC (Brannstrom & Edenteg, 2002; Casal et al., 2000; Minawlsawa et al., 2004; Ortega-Burrales et al., 2002). Some authors reported quantification of caffeine, trigonelline (N-methylnicotinic acid) and nicotinic acid using an HPLC/diode-array detector (Casal et al., 1998), and used this method to determine whether nicotinic acid could be used as a discriminant factor, and whether identification of caffeine and trigonelline levels could be useful for distinguishing Arabica from Robusta roasted coffee (Casal et al., 2000). In the present study, we developed a rapid method for measuring caffeine, trigonelline and nicotinic acid using high
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
466
G. Caprioli et al.
performance liquid chromatography system–variable wavelength detector (HPLC-VWD). Our sample preparation is not only quick but practical, with two simple steps of dilution and centrifugation, and the equipment required – the VWD – is common to many laboratories. In addition, we sought to discover the most effective EC machine water temperature and pressure settings for extracting caffeine, trigonelline and nicotinic acid. These experiments were carried out on Arabica and Robusta samples using an Aurelia Competizione EC machine, which can be used at pre-fixed, settable, and constant temperature and pressure: three water pressures (7, 9, 11 bar) and three temperatures (88, 92, 98 C) were tested. Furthermore, this analysis was also conducted on time portions of EC, prepared by taking a portion of EC produced in the first 10 s, and then portions every 5 s, to study the kinetics of extraction during espresso making and its dependence on pressure, temperature and coffee blend used (Arabica and Robusta). The method was then applied to analyze 20 commercially available EC samples from local retailers to determine the levels of caffeine, trigonelline and nicotinic acid.
Methods Reagents and standards The analytical standards of caffeine, trigonelline, and nicotinic acid were purchased from Sigma-Aldrich (Milano, Italy). Individual stock solutions were prepared by dissolving 100 mg of each compound in 100 ml of methanol, and then stored in glassstoppered bottles at 4 C. Standard working solutions at various concentrations were prepared daily by appropriate dilution of aliquots of the stock solutions in methanol. HPLC-grade methanol was supplied by Sigma-Aldrich (Milano, Italy) and HPLC-grade formic acid was supplied by Merck (Darmstadt, Germany). Deionized water (48 M cm resistivity) was obtained from the Milli-Q SP Reagent Water System (Millipore, Bedford, MA). All solvents and solutions were filtered through a 0.45-mm PTFE filter from Supelco (Bellefonte, PA) before use. Coffee types and espresso machine Two types of coffee, Arabica (pure C. arabica from Colombia, 2% moisture content) and Robusta (95:5 blend of C. canephora and C. arabica, 2% moisture content) and two EC machines, the Aurelia Competizione (A) and the Leva Victoria Arduino (B), which work with different specific curves of pressure and temperature, were provided by a local factory (Nuova Simonelli, Belforte del Chienti, Italy). EC sample preparation Ground coffee was obtained using a coffee grinder set so the coffee machine can produce 25 ml of EC in 25 s of extraction. Roasted coffee beans were milled just before each preparation, and exactly 7.5 g of ground coffee were used for each cup. For espresso machine ‘‘A,’’ operating conditions were set by combining three water pressures (7, 9, 11 bar) and three temperatures (88, 92, 98 C), obtaining an array of nine conditions, while the extraction time was kept constant for each test (25 s). Instead, the temperature and pressure settings of espresso machine ‘‘B’’ cannot be modified, as detailed in a previous study (Caprioli et al., 2012). Coffee samples were diluted 50 times in the mobile phase, centrifuged at 10 000 rpm for 10 min, and filtered through a 0.45mm PTFE filter before HPLC-VWD analysis. After in-depth studies of array results on EC quality, we chose 9 bar of pressure and a temperature of 92 C as settings for the
Int J Food Sci Nutr, 2014; 65(4): 465–469
Aurelia EC machine for the analysis of time portions of the EC sample (0–10, 11–15, 16–20, 21–25, 26–30, 31–35, 36–40 s), in order to quantify the three compounds. To obtain statistically significant data, the reported values are the mean of three EC replicates, prepared with the same EC machine conditions. HPLC-VWD analysis Separation was achieved on a Gemini C18 110 A analytical column (250 3 mm I.D., 5 mm) from Phenomenex (Chesire, UK). The mobile phase for HPLC-VWD analysis was water (A) containing 0.3% of formic acid and methanol (B), at a flow rate of 0.4 ml min1. The gradient program was: 0 min, 25% B; 0–10 min, 60% B; 10–15 min, 60% B; 15–20 min, 25% B; held at 25% until the end of the run at 25 min. The injection volume was 10 ml. HPLC-VWD studies were performed using a Hewlett Packard (Palo Alto, CA) HP-1090 Series II, made of an autosampler and a binary solvent pump. HPLC-VWD analyses were performed at two different wavelengths in the same run: 265 nm for trigonelline and nicotinic acid and 270 nm for caffeine.
Results and discussion Method validation The method was validated by determining linearity, recovery at three fortification levels, repeatability and within-reproducibility, limits of detection (LODs) and limits of quantification (LOQs) (instead of CCs alpha and CCs beta). Calibration curves of the analyzed compounds were constructed injecting 10 ml of standard solutions at five different concentrations, i.e. 10, 20, 50, 100, 250 mg l1, in HPLC-VWD. Five replicates for each concentration were performed, and the relative standard deviations (RSDs) ranged from 0.33% to 3.17% for run-to-run precision, and from 1.53% to 2.31% for day-to-day precision. All the calibration curves of the analyzed compounds showed a correlation coefficient greater than 0.9985. In the HPLC-VWD analysis of EC samples, the recoveries, obtained by spiking in the beverage solution a final concentration of 10 and 50 mg l1 with a standard mixture of the three compounds, were in the range 71–92% for all analyzed compounds. The repeatability of the method was calculated on fortified samples at 10 and 50 mg l1 (n ¼ 8), giving RSD% that were in a range 3.53–11.01% and 0.74–4.44%, respectively, for all compounds. The LOD and the LOQ of the three compounds, expressed in mg kg1 and calculated in the matrix, were estimated on the basis of 3:1 and 10:1 S/Ns (signal to noise ratio). LODs and LOQs were in the range of 0.03–0.06 and 0.1–0.2 mg l1, respectively. Retention time stability was utilized to demonstrate the specificity of the method. Reproducibility of the chromatographic retention time for each compound in coffee samples was examined 5 times per day over a 5-d period (n ¼ 25). The retention times using this method were stable with a percent RSD value of 1.52%. Influence of water pressure and water temperature on the extraction of caffeine, trigonelline and nicotinic acid from Robusta and Arabica blends The effects of water temperature and water pressure on EC extraction were investigated. These experiments analyzing both Robusta and Arabica samples were carried out using an Aurelia Competizione EC machine, in which pre-fixed, settable, and constant temperature and pressure can be dictated by combining three water pressures (7, 9, 11 bar) and three temperatures (88, 92, 98 C).
Quantification of caffeine, trigonelline and nicotinic acid
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
DOI: 10.3109/09637486.2013.873890
With the Robusta blend, the extracted amount of caffeine in a 25 ml cup of coffee ranged from 111.21 mg (98 C, 11 bar) to 257.57 mg (92 C, 7 bar). An increase of temperature from 88 C to 92 C led to an increase in the content of caffeine in the cup. On the contrary, at a temperature of 98 C, less caffeine was extracted, regardless of the pressure setting. At constant temperature, increased pressure yielded slightly lower caffeine levels, particularly at 11 bar. We can conclude that the best conditions for caffeine extraction seem to be 92 C at 9 bar and 92 C at 7 bar (Table 1). Regarding trigonelline, the extracted amount in a cup of EC ranged from 21.99 mg (98 C at 11 bar) to 49.44 mg (92 C at 7 bar). With the increase of temperature from 88 to 92 C, there was an increase in the extraction efficiency for trigonelline, which then decreased at 98 C, as also found for caffeine. At constant temperature, increased pressure yielded lower trigonelline levels. The highest efficiency of extraction was at 92 C and 7 bar, though the quantity of trigonelline extracted at this same temperature but at 9 bar of pressure was only slightly inferior (Table 1). For nicotinic acid, the amount in a cup of coffee ranged from 4.62 mg (98 C, 7 bar) to 10.27 (92 C, 9 bar). Data showed that the extracted amount increased when the temperature was raised from 88 C to 92 C, while the least efficient temperature was 98 C (Table 1). With the Arabica blend, the amount of caffeine extracted in a cup of coffee ranged from 88.51 mg (88 C, 11 bar) to 131.74 mg (92 C, 9 bar). Data show that with change of temperature, the extraction trend was similar to that of the Robusta blend. In fact, the amount of caffeine overall increased with the increase of temperature from 88 to 92 C at any tested pressure. However, analyzing the content at a fixed temperature, it was seen that pressure exerted minimal influence on the extraction efficiency of caffeine, with small oscillations. Overall, the best conditions for extracting caffeine appear to be 92 C at 9 bar and 92 C at 11 bar (Table 1). Concerning trigonelline extraction from the Arabica blend, the amount in a cup of coffee ranged from 50.74 mg (88 C, 11 bar) to 72.70 mg (92 C, 9 bar). Once again, with the increase of temperature from 88 C to 92 C there was an increase in the extraction of trigonelline, which was reversed when the
467
temperature was set to 98 C. At 88 C, the best extraction condition for trigonelline was in combination with the pressure setting of 9 bar, while at 92 C both the 9 bar and the 11 bar settings obtained the best extraction. At 98 C, the change in pressure did not affect the efficiency of extraction, but the combinations 92 C at 9 bar and 92 C at 11 bar gave the highest yield of extraction (Table 1). For nicotinic acid from the Arabica blend, the amount in a cup of coffee ranged from 4.20 mg (88 C at 11 bar) to 8.80 (98 C at 9 bar). The best extraction condition seemed to be 9 bar at either 92 C or 98 C (Table 1). In conclusion, for the Robusta blend, the best extraction was at 92 C with 7 and 9 bar of pressure, while for Arabica it was 92 C at 9 and 11 bar. The average amount of caffeine extracted in all EC samples prepared with Arabica (109.95 mg) was much lower than the value obtained with Robusta (198.70 mg). On the contrary, the average content of trigonelline found in EC prepared with Arabica was higher than in EC prepared with Robusta (60.59 mg and 36.08 mg, respectively). For nicotinic acid, there was no significant difference between Robusta and Arabica EC samples. Robusta samples showed a caffeine:trigonelline ratio of 4:1, while that for Arabica samples was 2:1. Hence, the two coffee varieties could be identified through their ratio of trigonelline to caffeine, as also stated by other researchers (Anthony et al., 1993; Casal et al., 2000). Subsequently, we compared the quantity of bioactive compounds extracted in Robusta and Arabica EC prepared with the Aurelia machine set at 9 bar and 92 C against that obtained using the Leva machine (Table 2). The content of these compounds in Robusta EC prepared using the Leva EC machine was lower than in the same blend of coffee prepared using the Aurelia one at optimum settings. The difference was small for nicotinic acid, whereas it was significant for caffeine. This observation is in line with the fact that the Leva machine uses low pressure and high temperature, and that when the Aurelia EC machine is operated at these conditions (98 C, 7 bar), it also has low-extraction efficiency. On the contrary, using the Arabica blend, the Leva machine seemed to yield a higher content of those compounds than did the Aurelia machine.
Table 1. Milligrams of caffeine, trigonelline and nicotinic acid obtained in Robusta (n ¼ 3, RSD% 55) and Arabica (n ¼ 3, RSD% 54) EC samples by setting the EC machine at different water temperature and pressure. Robusta Temperature 88 C 88 C 88 C 92 C 92 C 92 C 98 C 98 C 98 C
Pressure 7 9 11 7 9 11 7 9 11
bar bar bar bar bar bar bar bar bar
Arabica
Caffeine
Trigonelline
Nicotinic acid
Caffeine
Trigonelline
Nicotinic acid
212.61 200.95 229.28 257.57 255.15 201.30 160.81 159.40 111.20
40.71 37.26 29.22 49.44 47.49 39.28 29.22 30.13 21.99
7.15 8.31 7.33 9.84 10.27 7.66 4.62 5.24 4.65
97.76 96.27 88.51 118.74 131.74 130.77 112.79 108.69 104.30
52.50 55.10 50.74 66.18 72.70 72.48 60.31 58.50 56.77
7.73 4.64 4.20 8.03 8.38 8.31 8.09 8.80 6.76
Table 2. Amount of trigonelline, nicotinic acid and caffeine (mg) in a single EC cup (25 ml) made using Robusta (a) and Arabica (b) espresso coffee with the Aurelia and Leva machines (n ¼ 3 RSD% 54.5). Robusta EC Machine Aurelia EC Machine Leva EC Machine
Arabica
Caffeine
Trigonelline
Nicotinic acid
Caffeine
Trigonelline
Nicotinic acid
255.2 227.8
47.5 41.7
10.3 9.5
131.7 144.0
72.7 75.0
8.4 8.7
468
G. Caprioli et al.
Int J Food Sci Nutr, 2014; 65(4): 465–469
Once again, we can observe parallel performance with the two EC machines. The results obtained confirmed that both machines have good extraction efficiency of caffeine at high temperature and low pressure.
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
Quantification of caffeine, trigonelline and nicotinic acid in time portions of EC The above analysis was applied to time portions of EC, prepared by collecting the portion of EC produced in the first 10 s, and then portions every 5 s (i.e. 0–10, 11–15, 16–20, 21–25, 26–30, 31–35, 36–40 s), for a total of seven fractions, to study the kinetics of extraction and its dependence on pressure, temperature and coffee blends used during espresso making. The setting of 9 bar of pressure and a temperature of 92 C was chosen for the Aurelia EC machine for preparation and analysis of time portions of the EC sample. Data are reported in Table 3. For the Robusta sample, it is evident that as extraction continued, there was a gradual decrease in the concentration of the target compounds, in particular, caffeine and trigonelline. Nicotinic acid maintained a moderately decreasing rate throughout the time course of percolation. In percentage terms, we can say that within the first four fractions (0– 25 ml), 86.41% of total trigonelline, 64.90% of total nicotinic acid, and 85.46% of total caffeine were extracted. A prolonged extraction, up to 40 s, does not contribute substantially to obtaining a higher content of these compounds, but merely dilutes the coffee. This confirms the traditional wisdom of coffee making that identifies 25 ml as the correct volume of a certified Italian EC. For Arabica samples, increased extraction time led to decreased concentrations of the compounds. Also in this case, 88.12% of the total trigonelline, 61.56% of total nicotinic acid and 84.31% of the total caffeine were extracted in the first four fractions. Nicotinic acid was extracted in a constant manner in all fractions and did not present the strong decrease of the other two molecules. For Arabica EC samples as well, extraction for 40 s simply led to dilution of the coffee. The concentration of caffeine and nicotinic acid in all seven portions examined was higher in Robusta than in Arabica EC samples; on the contrary, the concentration of trigonelline was always lower in Robusta than in Arabica EC samples. Quantification of caffeine, trigonelline and nicotinic acid in the 20 EC brands The 20 commercial EC samples were analyzed for caffeine, trigonelline and nicotinic acid (Table 4). The Aurelia machine was set at a temperature of 92 C and 9 bar of pressure. The volume of a cup of coffee was 25 ml, according to the volume established for a Certified Italian Espresso (Odello & Odello, 2006).
As expected, the predominant compound in all the analyzed coffees was caffeine. It was detected at the highest level (199.68 mg) in the Cerruti sample, and at the lowest level in the Principe one (116.87 mg). The average amount of caffeine found in the 20 samples was 145.85 mg. As an example, Figure 1 provides the overlapping HPLC-VWD chromatograms for the Principe (a) and Cerruti samples (b). The Principe sample showed higher levels caffeine, trigonelline and nicotinic acid than the Cerruti one. The highest level of trigonelline was detected in the Bar Caracol coffee (65.08 mg) and the lowest in the Mexico sample (28.20 mg), while its average value in all 20 brands was 42.40 mg. As expected, the compound least present was nicotinic acid with an average content of 8.94 mg. The highest level was found in the Mokariko Noir sample (9.90 mg) and the lowest in the Bar Caracol EC one (7.30 mg).
Conclusion Our investigation of the effect of water temperature and water pressure on EC extraction using an Aurelia Competizione EC machine provided important information about espresso-making techniques. The usual espresso machine settings (92 C and 9 bar) are very close to those needed to obtain the best quality espresso, as reported by others (Caprioli et al., 2012), and are also the best choice for the extraction of caffeine, trigonelline and nicotinic acid, especially when using the Robusta blend. The average Table 4. Amount (mg) of caffeine, trigonelline and nicotinic acid in 20 commercial coffee samples used to prepare EC with ‘‘Aurelia Competizione’’ EC machine at 9 bar and 92 C (n ¼ 3, RSD% 55.5). Trade name Bar Caracol Caffen ‘‘il Don Caffe’’ Cartapani Cerruti Corsini Costadoro Dromedario Gourmet Gran Salvador Guglielmo Jolly La Genovese Mexico Mokarico Noir Parana caffe Pellini Top Pelourinho Principe River Zicaffe
Caffeine
Nicotinic acid
Trigonelline
155.04 135.49 126.72 199.68 134.93 120.15 128.59 143.98 138.98 155.16 141.01 143.20 138.40 147.48 179.54 133.11 117.75 116.87 187.46 173.52
7.30 9.48 8.44 9.14 9.04 8.26 9.15 9.03 9.02 9.52 8.42 8.37 9.07 9.90 9.26 9.36 9.76 8.60 9.14 8.57
65.08 33.59 48.10 38.31 45.10 41.53 47.44 51.72 41.74 30.23 44.98 43.41 28.20 46.01 38.81 53.23 44.50 35.07 30.40 40.59
Table 3. Concentration in mg/l of trigonelline, nicotinic acid and caffeine in time portions of EC with Robusta and Arabica blends (92 C, 9 bar) (n ¼ 3, RSD% 510.7). Robusta
Arabica
Time portions (sec.)
Caffeine
Trigonelline
Nicotinic Acid
Caffeine
Trigonelline
Nicotinic acid
0–10 11–15 16–20 21–25 26–30 31–35 36–40
19462.23 11474.03 7029.68 4519.35 3035.10 2315.25 1877.51
5226.01 2699.12 1510.13 936.65 646.09 523.72 461.19
557.26 412.14 323.50 308.08 291.46 287.88 286.84
11457.98 7772.80 4383.65 2733.80 1986.83 1561.63 1355.20
6134.86 3645.08 1785.15 996.21 685.68 537.50 470.64
399.04 364.81 318.81 296.86 288.28 287.11 286.00
DOI: 10.3109/09637486.2013.873890
Quantification of caffeine, trigonelline and nicotinic acid
469
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
Figure 1. Overlapping of a HPLC-VWD chromatograms referred to ‘‘Principe’’ sample (black) and ‘‘Cerruti’’ sample (grey). Legend: (1) trigonelline, (2) nicotinic acid and (3) caffeine.
amount of caffeine extracted in Arabica EC samples was lower than that obtained from the Robusta blend; the opposite was true for trigonelline. The analysis of time portions of EC showed that for both Arabica and Robusta EC samples, continued extraction after 25 s merely dilutes the coffee. Analysis of 20 commercial coffee samples revealed that the average total amount of the three compounds in a 25 ml cup of coffee is about 197.19 mg, assessing the richness of EC samples prepared using the Aurelia Competizione EC machine.
Acknowledgements The authors are grateful to Sheila Beatty for editing the English usage of the text.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.
References Alpdogan G, Karbina K, Sungur S. 2002. Derivative spectrophotometer determination of caffeine in some beverages. Turk J Chem 26:295–302. Andueza S, Maeztu L, Pascual L, Ibanez C, Paz MPD, Cid C. 2003. Influence of extraction temperature on the final quality of espresso coffee. J Sci Food Agric 83:240–248. Anthony F, Clifford MN, Noirot M. 1993. Biochemical diversity in the genus Coffea L.: chlorogenic acids, caffeine and mozambioside contents. Genet Resour Crop Ev 40:61–70. Bolton S, Null G. 1981. Caffeine, psychological effect, use and abuse. J Orthomol Med 10:202–211. Brannstrom M, Edenteg K. 2002. Determination of caffeine in Vietnamese coffee using HPLC. Umea, Sweden: Minor field study. Butler R. 1999. Solubles into the millennium. Tea Coffee Trade J 171: 72–76.
Caprioli G, Cortese M, Cristalli G, Maggi F, Odello L, Ricciutelli M, Sagratini G, et al. (2012). Optimization of espresso machine parameters through the analysis of coffee odorants by HS-SPME–GC/MS. Food Chem 135:1127–1123. Casal S, Oliveira MB, Ferreira MA. 1998. Development of an HPLC/ Diode-array detector method for simultaneous determination of trigonelline, nicotinic acid and caffeine in coffee. J Liq Chromatogr RT 21:3187–3195. Casal S, Oliveira M.B, Ferreira MA. 2000. HPLC/Diode-array applied to thermal degradation of trigonelline, nicotinic acid and caffeine in coffee. J Food Chem 68:481–485. Franca AS, Mendonca JCF, Oliveira SD. 2005. Composition of green and roasted coffees of different cup qualities. Lebensm-Wiss Technol 38: 709–715. Illy A, Viani R. 1995. Espresso coffee: the chemistry of quality. London, England: Academic Press. Ltd. Maeztu L, Andueza S, Iban C, Pen MPD. 2001. Multivariate methods for characterization and classification of espresso coffees from different botanical varieties and types of roast by foam, taste, and mouthfeel. J Agric Food Chem 49:4743–4747. Minawlsawa M, Yoshida S, Takali N. 2004. Determination of biologically active substances in roasted coffee using diode – array HPLC system. J Anal Sci 26:325–328. Nunes FM, Coimbra MA, Duarte AC, Delgadillo I. 1997. Foamability, foam stability and chemical composition of espresso coffee as affected by the degree of roast. Statistics 5:3238–3243. Odello L, Odello C. 2006. Espresso Italiano Tasting. Centro Studi Assaggiatori. Brescia, Italy: Artigianelli Spa. Ortega-Burrales P, Padilla-Weigand R, Molina-Diaz A. 2002. Simultaneous determination of paracetamol and caffeine by flow injection solid phase spectrometry using C18 silica Gel as a sensing support. J Anal Sci 18:1241–1246. Petracco M. 1989. Physico-chemical and structural characterization of espresso coffee brew. Proceedings of the 13th International Colloquium on the Chemistry of Coffee, ASIC, Paris, France. Singh DK, Sahu A. 2006. Spectrophotometer determination of caffeine and theophylline in pure alkaloids and its application in pharmaceutical formulations. J Anal Biochem 349:176–180.
FOOD COMPOSITION AND ANALYSIS
Quantification of caffeine, trigonelline and nicotinic acid in espresso coffee: the influence of espresso machines and coffee cultivars Giovanni Caprioli1*, Manuela Cortese1*, Filippo Maggi1, Caterina Minnetti1, Luigi Odello2, Gianni Sagratini1, and Sauro Vittori1 School of Pharmacy, University of Camerino, Camerino, Italy, and 2Centro Studi Assaggiatori, Galleria V. Veneto 9, Brescia, Italy
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
1
Abstract
Keywords
Caffeine, trigonelline and nicotinic acid are important bioactive constituents of coffee. In this work, the combination of different water temperatures and pressures in the settings of the espresso coffee (EC) machine was evaluated, to assess how these factors influence how effectively caffeine, trigonelline and nicotinic acid are extracted from both Arabica and Robusta samples. The proposed analytical method, based on a high performance liquid chromatography (HPLC) system coupled to a variable wavelength detector (VWD), showed good linearity (R240.9985) and good recoveries (71–92%); after validation for three monitored compounds, the method was used to analyze 20 commercial samples. The combination of a temperature of 92 C and pressure at 7 or 9 bar seems to be the ideal setting for the most efficient extraction of these compounds and consequently for their intake; the compound extracted in the greatest quantity was caffeine, which was in the range of 116.87–199.68 mg in a 25 ml cup of coffee.
Arabica coffee, caffeine, espresso coffee machine, extraction pressure, extraction temperature, Robusta coffee
Introduction Coffee is a very popular hot drink, highly appreciated all over the world (Butler, 1999; Illy & Viani, 1995; Maeztu et al., 2001). According to the latest statistics from the International Coffee Organization (ICO), about 1.4 billion cups of coffee a day are consumed worldwide. The two most important species of coffee, Coffea arabica and Coffea canephora (Robusta), differ significantly in price, quality and consumer preference. Arabica coffee is more acidic, with a more intense aroma and a richer body than Robusta, which is instead more bitter and characterized by a typical earthy and woody flavor (Illy & Viani, 1995). The preparation of espresso coffee (EC) is influenced by obvious factors such as coffee and water, as well as other technical conditions related to the machine (Andueza et al., 2003; Caprioli et al., 2012). The classic EC machine features a volumetric pump that brings the water to the desired pressure (usually about 9 bar) and temperature (usually between 91 C and 96 C) through a heat exchanger. The water passes through the filter unit containing the pressed coffee, which absorbs some milliliters of water and swells. As the water passes through, the various components are extracted from the espresso (Odello & Odello, 2006).
*These authors contributed equally to this work. Correspondence: Gianni Sagratini, School of Pharmacy, University of Camerino, via S. Agostino 1, 62032 Camerino, Italy. Tel: +390737402238. Fax: +390737637345. E-mail: gianni.sagratini@ unicam.it
History Received 29 October 2013 Revised 27 November 2013 Accepted 7 December 2013 Published online 27 January 2014
As noted by Petracco (1989), and Illy & Viani (1995), there is a dearth of knowledge about the optimal weight of ground coffee and also about the beverage volume and the extraction conditions (water pressure and temperature) that must be used to obtain a high quality EC, judged in terms of aroma and composition (Caprioli et al., 2012; Nunes et al., 1997). Caffeine, trigonelline and nicotinic acid are three important bioactive components of coffee. Caffeine is found in various kinds of food and drink consumed in daily life (Singh & Sahu, 2006), coffee, in particular. It exerts various physiological effects, such as relaxation of bronchial muscle, stimulation of the central nervous system, gastric acid secretion and diuresis (Bolton & Null, 1981). Among the various types of coffee, Arabica has the highest caffeine content, and is considered one of the best quality varieties (Franca et al., 2005). Several chemical and physical methods have been developed to quantify the amount of caffeine in coffee and other beverages. The most widely used methods include various analytical techniques, such as derivative spectrophotometers (Alpdogan et al., 2002) and HPLC (Brannstrom & Edenteg, 2002; Casal et al., 2000; Minawlsawa et al., 2004; Ortega-Burrales et al., 2002). Some authors reported quantification of caffeine, trigonelline (N-methylnicotinic acid) and nicotinic acid using an HPLC/diode-array detector (Casal et al., 1998), and used this method to determine whether nicotinic acid could be used as a discriminant factor, and whether identification of caffeine and trigonelline levels could be useful for distinguishing Arabica from Robusta roasted coffee (Casal et al., 2000). In the present study, we developed a rapid method for measuring caffeine, trigonelline and nicotinic acid using high
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
466
G. Caprioli et al.
performance liquid chromatography system–variable wavelength detector (HPLC-VWD). Our sample preparation is not only quick but practical, with two simple steps of dilution and centrifugation, and the equipment required – the VWD – is common to many laboratories. In addition, we sought to discover the most effective EC machine water temperature and pressure settings for extracting caffeine, trigonelline and nicotinic acid. These experiments were carried out on Arabica and Robusta samples using an Aurelia Competizione EC machine, which can be used at pre-fixed, settable, and constant temperature and pressure: three water pressures (7, 9, 11 bar) and three temperatures (88, 92, 98 C) were tested. Furthermore, this analysis was also conducted on time portions of EC, prepared by taking a portion of EC produced in the first 10 s, and then portions every 5 s, to study the kinetics of extraction during espresso making and its dependence on pressure, temperature and coffee blend used (Arabica and Robusta). The method was then applied to analyze 20 commercially available EC samples from local retailers to determine the levels of caffeine, trigonelline and nicotinic acid.
Methods Reagents and standards The analytical standards of caffeine, trigonelline, and nicotinic acid were purchased from Sigma-Aldrich (Milano, Italy). Individual stock solutions were prepared by dissolving 100 mg of each compound in 100 ml of methanol, and then stored in glassstoppered bottles at 4 C. Standard working solutions at various concentrations were prepared daily by appropriate dilution of aliquots of the stock solutions in methanol. HPLC-grade methanol was supplied by Sigma-Aldrich (Milano, Italy) and HPLC-grade formic acid was supplied by Merck (Darmstadt, Germany). Deionized water (48 M cm resistivity) was obtained from the Milli-Q SP Reagent Water System (Millipore, Bedford, MA). All solvents and solutions were filtered through a 0.45-mm PTFE filter from Supelco (Bellefonte, PA) before use. Coffee types and espresso machine Two types of coffee, Arabica (pure C. arabica from Colombia, 2% moisture content) and Robusta (95:5 blend of C. canephora and C. arabica, 2% moisture content) and two EC machines, the Aurelia Competizione (A) and the Leva Victoria Arduino (B), which work with different specific curves of pressure and temperature, were provided by a local factory (Nuova Simonelli, Belforte del Chienti, Italy). EC sample preparation Ground coffee was obtained using a coffee grinder set so the coffee machine can produce 25 ml of EC in 25 s of extraction. Roasted coffee beans were milled just before each preparation, and exactly 7.5 g of ground coffee were used for each cup. For espresso machine ‘‘A,’’ operating conditions were set by combining three water pressures (7, 9, 11 bar) and three temperatures (88, 92, 98 C), obtaining an array of nine conditions, while the extraction time was kept constant for each test (25 s). Instead, the temperature and pressure settings of espresso machine ‘‘B’’ cannot be modified, as detailed in a previous study (Caprioli et al., 2012). Coffee samples were diluted 50 times in the mobile phase, centrifuged at 10 000 rpm for 10 min, and filtered through a 0.45mm PTFE filter before HPLC-VWD analysis. After in-depth studies of array results on EC quality, we chose 9 bar of pressure and a temperature of 92 C as settings for the
Int J Food Sci Nutr, 2014; 65(4): 465–469
Aurelia EC machine for the analysis of time portions of the EC sample (0–10, 11–15, 16–20, 21–25, 26–30, 31–35, 36–40 s), in order to quantify the three compounds. To obtain statistically significant data, the reported values are the mean of three EC replicates, prepared with the same EC machine conditions. HPLC-VWD analysis Separation was achieved on a Gemini C18 110 A analytical column (250 3 mm I.D., 5 mm) from Phenomenex (Chesire, UK). The mobile phase for HPLC-VWD analysis was water (A) containing 0.3% of formic acid and methanol (B), at a flow rate of 0.4 ml min1. The gradient program was: 0 min, 25% B; 0–10 min, 60% B; 10–15 min, 60% B; 15–20 min, 25% B; held at 25% until the end of the run at 25 min. The injection volume was 10 ml. HPLC-VWD studies were performed using a Hewlett Packard (Palo Alto, CA) HP-1090 Series II, made of an autosampler and a binary solvent pump. HPLC-VWD analyses were performed at two different wavelengths in the same run: 265 nm for trigonelline and nicotinic acid and 270 nm for caffeine.
Results and discussion Method validation The method was validated by determining linearity, recovery at three fortification levels, repeatability and within-reproducibility, limits of detection (LODs) and limits of quantification (LOQs) (instead of CCs alpha and CCs beta). Calibration curves of the analyzed compounds were constructed injecting 10 ml of standard solutions at five different concentrations, i.e. 10, 20, 50, 100, 250 mg l1, in HPLC-VWD. Five replicates for each concentration were performed, and the relative standard deviations (RSDs) ranged from 0.33% to 3.17% for run-to-run precision, and from 1.53% to 2.31% for day-to-day precision. All the calibration curves of the analyzed compounds showed a correlation coefficient greater than 0.9985. In the HPLC-VWD analysis of EC samples, the recoveries, obtained by spiking in the beverage solution a final concentration of 10 and 50 mg l1 with a standard mixture of the three compounds, were in the range 71–92% for all analyzed compounds. The repeatability of the method was calculated on fortified samples at 10 and 50 mg l1 (n ¼ 8), giving RSD% that were in a range 3.53–11.01% and 0.74–4.44%, respectively, for all compounds. The LOD and the LOQ of the three compounds, expressed in mg kg1 and calculated in the matrix, were estimated on the basis of 3:1 and 10:1 S/Ns (signal to noise ratio). LODs and LOQs were in the range of 0.03–0.06 and 0.1–0.2 mg l1, respectively. Retention time stability was utilized to demonstrate the specificity of the method. Reproducibility of the chromatographic retention time for each compound in coffee samples was examined 5 times per day over a 5-d period (n ¼ 25). The retention times using this method were stable with a percent RSD value of 1.52%. Influence of water pressure and water temperature on the extraction of caffeine, trigonelline and nicotinic acid from Robusta and Arabica blends The effects of water temperature and water pressure on EC extraction were investigated. These experiments analyzing both Robusta and Arabica samples were carried out using an Aurelia Competizione EC machine, in which pre-fixed, settable, and constant temperature and pressure can be dictated by combining three water pressures (7, 9, 11 bar) and three temperatures (88, 92, 98 C).
Quantification of caffeine, trigonelline and nicotinic acid
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
DOI: 10.3109/09637486.2013.873890
With the Robusta blend, the extracted amount of caffeine in a 25 ml cup of coffee ranged from 111.21 mg (98 C, 11 bar) to 257.57 mg (92 C, 7 bar). An increase of temperature from 88 C to 92 C led to an increase in the content of caffeine in the cup. On the contrary, at a temperature of 98 C, less caffeine was extracted, regardless of the pressure setting. At constant temperature, increased pressure yielded slightly lower caffeine levels, particularly at 11 bar. We can conclude that the best conditions for caffeine extraction seem to be 92 C at 9 bar and 92 C at 7 bar (Table 1). Regarding trigonelline, the extracted amount in a cup of EC ranged from 21.99 mg (98 C at 11 bar) to 49.44 mg (92 C at 7 bar). With the increase of temperature from 88 to 92 C, there was an increase in the extraction efficiency for trigonelline, which then decreased at 98 C, as also found for caffeine. At constant temperature, increased pressure yielded lower trigonelline levels. The highest efficiency of extraction was at 92 C and 7 bar, though the quantity of trigonelline extracted at this same temperature but at 9 bar of pressure was only slightly inferior (Table 1). For nicotinic acid, the amount in a cup of coffee ranged from 4.62 mg (98 C, 7 bar) to 10.27 (92 C, 9 bar). Data showed that the extracted amount increased when the temperature was raised from 88 C to 92 C, while the least efficient temperature was 98 C (Table 1). With the Arabica blend, the amount of caffeine extracted in a cup of coffee ranged from 88.51 mg (88 C, 11 bar) to 131.74 mg (92 C, 9 bar). Data show that with change of temperature, the extraction trend was similar to that of the Robusta blend. In fact, the amount of caffeine overall increased with the increase of temperature from 88 to 92 C at any tested pressure. However, analyzing the content at a fixed temperature, it was seen that pressure exerted minimal influence on the extraction efficiency of caffeine, with small oscillations. Overall, the best conditions for extracting caffeine appear to be 92 C at 9 bar and 92 C at 11 bar (Table 1). Concerning trigonelline extraction from the Arabica blend, the amount in a cup of coffee ranged from 50.74 mg (88 C, 11 bar) to 72.70 mg (92 C, 9 bar). Once again, with the increase of temperature from 88 C to 92 C there was an increase in the extraction of trigonelline, which was reversed when the
467
temperature was set to 98 C. At 88 C, the best extraction condition for trigonelline was in combination with the pressure setting of 9 bar, while at 92 C both the 9 bar and the 11 bar settings obtained the best extraction. At 98 C, the change in pressure did not affect the efficiency of extraction, but the combinations 92 C at 9 bar and 92 C at 11 bar gave the highest yield of extraction (Table 1). For nicotinic acid from the Arabica blend, the amount in a cup of coffee ranged from 4.20 mg (88 C at 11 bar) to 8.80 (98 C at 9 bar). The best extraction condition seemed to be 9 bar at either 92 C or 98 C (Table 1). In conclusion, for the Robusta blend, the best extraction was at 92 C with 7 and 9 bar of pressure, while for Arabica it was 92 C at 9 and 11 bar. The average amount of caffeine extracted in all EC samples prepared with Arabica (109.95 mg) was much lower than the value obtained with Robusta (198.70 mg). On the contrary, the average content of trigonelline found in EC prepared with Arabica was higher than in EC prepared with Robusta (60.59 mg and 36.08 mg, respectively). For nicotinic acid, there was no significant difference between Robusta and Arabica EC samples. Robusta samples showed a caffeine:trigonelline ratio of 4:1, while that for Arabica samples was 2:1. Hence, the two coffee varieties could be identified through their ratio of trigonelline to caffeine, as also stated by other researchers (Anthony et al., 1993; Casal et al., 2000). Subsequently, we compared the quantity of bioactive compounds extracted in Robusta and Arabica EC prepared with the Aurelia machine set at 9 bar and 92 C against that obtained using the Leva machine (Table 2). The content of these compounds in Robusta EC prepared using the Leva EC machine was lower than in the same blend of coffee prepared using the Aurelia one at optimum settings. The difference was small for nicotinic acid, whereas it was significant for caffeine. This observation is in line with the fact that the Leva machine uses low pressure and high temperature, and that when the Aurelia EC machine is operated at these conditions (98 C, 7 bar), it also has low-extraction efficiency. On the contrary, using the Arabica blend, the Leva machine seemed to yield a higher content of those compounds than did the Aurelia machine.
Table 1. Milligrams of caffeine, trigonelline and nicotinic acid obtained in Robusta (n ¼ 3, RSD% 55) and Arabica (n ¼ 3, RSD% 54) EC samples by setting the EC machine at different water temperature and pressure. Robusta Temperature 88 C 88 C 88 C 92 C 92 C 92 C 98 C 98 C 98 C
Pressure 7 9 11 7 9 11 7 9 11
bar bar bar bar bar bar bar bar bar
Arabica
Caffeine
Trigonelline
Nicotinic acid
Caffeine
Trigonelline
Nicotinic acid
212.61 200.95 229.28 257.57 255.15 201.30 160.81 159.40 111.20
40.71 37.26 29.22 49.44 47.49 39.28 29.22 30.13 21.99
7.15 8.31 7.33 9.84 10.27 7.66 4.62 5.24 4.65
97.76 96.27 88.51 118.74 131.74 130.77 112.79 108.69 104.30
52.50 55.10 50.74 66.18 72.70 72.48 60.31 58.50 56.77
7.73 4.64 4.20 8.03 8.38 8.31 8.09 8.80 6.76
Table 2. Amount of trigonelline, nicotinic acid and caffeine (mg) in a single EC cup (25 ml) made using Robusta (a) and Arabica (b) espresso coffee with the Aurelia and Leva machines (n ¼ 3 RSD% 54.5). Robusta EC Machine Aurelia EC Machine Leva EC Machine
Arabica
Caffeine
Trigonelline
Nicotinic acid
Caffeine
Trigonelline
Nicotinic acid
255.2 227.8
47.5 41.7
10.3 9.5
131.7 144.0
72.7 75.0
8.4 8.7
468
G. Caprioli et al.
Int J Food Sci Nutr, 2014; 65(4): 465–469
Once again, we can observe parallel performance with the two EC machines. The results obtained confirmed that both machines have good extraction efficiency of caffeine at high temperature and low pressure.
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
Quantification of caffeine, trigonelline and nicotinic acid in time portions of EC The above analysis was applied to time portions of EC, prepared by collecting the portion of EC produced in the first 10 s, and then portions every 5 s (i.e. 0–10, 11–15, 16–20, 21–25, 26–30, 31–35, 36–40 s), for a total of seven fractions, to study the kinetics of extraction and its dependence on pressure, temperature and coffee blends used during espresso making. The setting of 9 bar of pressure and a temperature of 92 C was chosen for the Aurelia EC machine for preparation and analysis of time portions of the EC sample. Data are reported in Table 3. For the Robusta sample, it is evident that as extraction continued, there was a gradual decrease in the concentration of the target compounds, in particular, caffeine and trigonelline. Nicotinic acid maintained a moderately decreasing rate throughout the time course of percolation. In percentage terms, we can say that within the first four fractions (0– 25 ml), 86.41% of total trigonelline, 64.90% of total nicotinic acid, and 85.46% of total caffeine were extracted. A prolonged extraction, up to 40 s, does not contribute substantially to obtaining a higher content of these compounds, but merely dilutes the coffee. This confirms the traditional wisdom of coffee making that identifies 25 ml as the correct volume of a certified Italian EC. For Arabica samples, increased extraction time led to decreased concentrations of the compounds. Also in this case, 88.12% of the total trigonelline, 61.56% of total nicotinic acid and 84.31% of the total caffeine were extracted in the first four fractions. Nicotinic acid was extracted in a constant manner in all fractions and did not present the strong decrease of the other two molecules. For Arabica EC samples as well, extraction for 40 s simply led to dilution of the coffee. The concentration of caffeine and nicotinic acid in all seven portions examined was higher in Robusta than in Arabica EC samples; on the contrary, the concentration of trigonelline was always lower in Robusta than in Arabica EC samples. Quantification of caffeine, trigonelline and nicotinic acid in the 20 EC brands The 20 commercial EC samples were analyzed for caffeine, trigonelline and nicotinic acid (Table 4). The Aurelia machine was set at a temperature of 92 C and 9 bar of pressure. The volume of a cup of coffee was 25 ml, according to the volume established for a Certified Italian Espresso (Odello & Odello, 2006).
As expected, the predominant compound in all the analyzed coffees was caffeine. It was detected at the highest level (199.68 mg) in the Cerruti sample, and at the lowest level in the Principe one (116.87 mg). The average amount of caffeine found in the 20 samples was 145.85 mg. As an example, Figure 1 provides the overlapping HPLC-VWD chromatograms for the Principe (a) and Cerruti samples (b). The Principe sample showed higher levels caffeine, trigonelline and nicotinic acid than the Cerruti one. The highest level of trigonelline was detected in the Bar Caracol coffee (65.08 mg) and the lowest in the Mexico sample (28.20 mg), while its average value in all 20 brands was 42.40 mg. As expected, the compound least present was nicotinic acid with an average content of 8.94 mg. The highest level was found in the Mokariko Noir sample (9.90 mg) and the lowest in the Bar Caracol EC one (7.30 mg).
Conclusion Our investigation of the effect of water temperature and water pressure on EC extraction using an Aurelia Competizione EC machine provided important information about espresso-making techniques. The usual espresso machine settings (92 C and 9 bar) are very close to those needed to obtain the best quality espresso, as reported by others (Caprioli et al., 2012), and are also the best choice for the extraction of caffeine, trigonelline and nicotinic acid, especially when using the Robusta blend. The average Table 4. Amount (mg) of caffeine, trigonelline and nicotinic acid in 20 commercial coffee samples used to prepare EC with ‘‘Aurelia Competizione’’ EC machine at 9 bar and 92 C (n ¼ 3, RSD% 55.5). Trade name Bar Caracol Caffen ‘‘il Don Caffe’’ Cartapani Cerruti Corsini Costadoro Dromedario Gourmet Gran Salvador Guglielmo Jolly La Genovese Mexico Mokarico Noir Parana caffe Pellini Top Pelourinho Principe River Zicaffe
Caffeine
Nicotinic acid
Trigonelline
155.04 135.49 126.72 199.68 134.93 120.15 128.59 143.98 138.98 155.16 141.01 143.20 138.40 147.48 179.54 133.11 117.75 116.87 187.46 173.52
7.30 9.48 8.44 9.14 9.04 8.26 9.15 9.03 9.02 9.52 8.42 8.37 9.07 9.90 9.26 9.36 9.76 8.60 9.14 8.57
65.08 33.59 48.10 38.31 45.10 41.53 47.44 51.72 41.74 30.23 44.98 43.41 28.20 46.01 38.81 53.23 44.50 35.07 30.40 40.59
Table 3. Concentration in mg/l of trigonelline, nicotinic acid and caffeine in time portions of EC with Robusta and Arabica blends (92 C, 9 bar) (n ¼ 3, RSD% 510.7). Robusta
Arabica
Time portions (sec.)
Caffeine
Trigonelline
Nicotinic Acid
Caffeine
Trigonelline
Nicotinic acid
0–10 11–15 16–20 21–25 26–30 31–35 36–40
19462.23 11474.03 7029.68 4519.35 3035.10 2315.25 1877.51
5226.01 2699.12 1510.13 936.65 646.09 523.72 461.19
557.26 412.14 323.50 308.08 291.46 287.88 286.84
11457.98 7772.80 4383.65 2733.80 1986.83 1561.63 1355.20
6134.86 3645.08 1785.15 996.21 685.68 537.50 470.64
399.04 364.81 318.81 296.86 288.28 287.11 286.00
DOI: 10.3109/09637486.2013.873890
Quantification of caffeine, trigonelline and nicotinic acid
469
Int J Food Sci Nutr Downloaded from informahealthcare.com by Queen's University on 01/03/15 For personal use only.
Figure 1. Overlapping of a HPLC-VWD chromatograms referred to ‘‘Principe’’ sample (black) and ‘‘Cerruti’’ sample (grey). Legend: (1) trigonelline, (2) nicotinic acid and (3) caffeine.
amount of caffeine extracted in Arabica EC samples was lower than that obtained from the Robusta blend; the opposite was true for trigonelline. The analysis of time portions of EC showed that for both Arabica and Robusta EC samples, continued extraction after 25 s merely dilutes the coffee. Analysis of 20 commercial coffee samples revealed that the average total amount of the three compounds in a 25 ml cup of coffee is about 197.19 mg, assessing the richness of EC samples prepared using the Aurelia Competizione EC machine.
Acknowledgements The authors are grateful to Sheila Beatty for editing the English usage of the text.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.
References Alpdogan G, Karbina K, Sungur S. 2002. Derivative spectrophotometer determination of caffeine in some beverages. Turk J Chem 26:295–302. Andueza S, Maeztu L, Pascual L, Ibanez C, Paz MPD, Cid C. 2003. Influence of extraction temperature on the final quality of espresso coffee. J Sci Food Agric 83:240–248. Anthony F, Clifford MN, Noirot M. 1993. Biochemical diversity in the genus Coffea L.: chlorogenic acids, caffeine and mozambioside contents. Genet Resour Crop Ev 40:61–70. Bolton S, Null G. 1981. Caffeine, psychological effect, use and abuse. J Orthomol Med 10:202–211. Brannstrom M, Edenteg K. 2002. Determination of caffeine in Vietnamese coffee using HPLC. Umea, Sweden: Minor field study. Butler R. 1999. Solubles into the millennium. Tea Coffee Trade J 171: 72–76.
Caprioli G, Cortese M, Cristalli G, Maggi F, Odello L, Ricciutelli M, Sagratini G, et al. (2012). Optimization of espresso machine parameters through the analysis of coffee odorants by HS-SPME–GC/MS. Food Chem 135:1127–1123. Casal S, Oliveira MB, Ferreira MA. 1998. Development of an HPLC/ Diode-array detector method for simultaneous determination of trigonelline, nicotinic acid and caffeine in coffee. J Liq Chromatogr RT 21:3187–3195. Casal S, Oliveira M.B, Ferreira MA. 2000. HPLC/Diode-array applied to thermal degradation of trigonelline, nicotinic acid and caffeine in coffee. J Food Chem 68:481–485. Franca AS, Mendonca JCF, Oliveira SD. 2005. Composition of green and roasted coffees of different cup qualities. Lebensm-Wiss Technol 38: 709–715. Illy A, Viani R. 1995. Espresso coffee: the chemistry of quality. London, England: Academic Press. Ltd. Maeztu L, Andueza S, Iban C, Pen MPD. 2001. Multivariate methods for characterization and classification of espresso coffees from different botanical varieties and types of roast by foam, taste, and mouthfeel. J Agric Food Chem 49:4743–4747. Minawlsawa M, Yoshida S, Takali N. 2004. Determination of biologically active substances in roasted coffee using diode – array HPLC system. J Anal Sci 26:325–328. Nunes FM, Coimbra MA, Duarte AC, Delgadillo I. 1997. Foamability, foam stability and chemical composition of espresso coffee as affected by the degree of roast. Statistics 5:3238–3243. Odello L, Odello C. 2006. Espresso Italiano Tasting. Centro Studi Assaggiatori. Brescia, Italy: Artigianelli Spa. Ortega-Burrales P, Padilla-Weigand R, Molina-Diaz A. 2002. Simultaneous determination of paracetamol and caffeine by flow injection solid phase spectrometry using C18 silica Gel as a sensing support. J Anal Sci 18:1241–1246. Petracco M. 1989. Physico-chemical and structural characterization of espresso coffee brew. Proceedings of the 13th International Colloquium on the Chemistry of Coffee, ASIC, Paris, France. Singh DK, Sahu A. 2006. Spectrophotometer determination of caffeine and theophylline in pure alkaloids and its application in pharmaceutical formulations. J Anal Biochem 349:176–180.
