Determination of Some Beer Flavours by Stir Bar Sorptive Extraction and Solvent Back Extraction

Transcription

1 Determination of Some Beer Flavours by Stir Bar Sorptive Extraction and Solvent Back Extraction Tomáš Horák 1,2, Vladimír Kellner 1, Jiří Čulík 1, Marie Jurková 1 and Pavel Čejka 1 ABSTRACT J. Inst. Brew. 113(2), , 2007 This work presents the development of a simple, rapid and low cost method for the determination of some beer flavours as esters such as isoamyl acetate, ethyl caproate, ethyl caprylate, phenyl acetate, ethyl caprate, phenylethyl acetate, ethyl laurate, ethyl myristate and ethyl palmitate using stir bar sorptive extraction (SBSE). The combination of this extraction technique with solvent back extraction and subsequent gas chromatographic analysis with flame ionisation detection was used for the determination of these compounds. The effects of different mixtures of organic solvents and the influence of sampling time and time of solvent back extraction were studied. An optimum procedure for the SBSE method was developed. The method had high repeatability (RSD %), good linearity (the correlation coefficients were higher than for quadratic fit over the concentration range from to 30 mg/l) and acceptable recoveries (78 107%) and precision (RSD < 7.3%). Key words: Beer, beer flavours, solvent back extraction, stir bar sorptive extraction (SBSE). INTRODUCTION It is well known that beer consists of a many volatile and non-volatile compounds which impact beer flavour. Meilgaard 16 discussed the influence of chemical composition on the flavour of beer. Beer flavour, a combination of odour and taste, is a very important factor in the consumer s perception of the quality of a beer. The total sensory profile of beer is the main contributor, with the exception of price, to the success of this product in the market. The long-term stability of the sensory profile is very important because consumers may not be satisfied if they purchase the same product with a different flavour. Thus, the ability to be able to analytically determine the most important flavour groups in beer is important. 1 Research Institute of Brewing and Malting PLC, Brewing Institute Prague, Lípová 15, CZ Prague 2, Czech Republic. 2 Corresponding author. horak@beerresearch.cz This work has been presented as a poster presentation at the 29th International Symposium on Capillary Chromatography, Riva, Italy, Publication no. G The Institute of Brewing & Distilling Esters are one of these groups. A typical flavour characteristic contributed to beer by most esters is a pleasant fruit flavour 1. During the brewing process, ester production is influenced mainly by the wort composition, fermentation parameters and yeast strain 6,17. Arkima 3 has shown that the contribution of esters to the total beer bouquet is approximately one fifth. For the determination of beer flavour compounds, wellestablished methods require sample preparation steps such as static headspace methods, purge-and-trap pre-concentration techniques, distillation procedures, liquid-liquid extraction or solid phase extraction 1,7,10,13,15,23,26 due to sample concentration. During the past years, miniaturization has become a dominant trend in analytical chemistry. Solid phase microextraction (SPME) is a typical example of miniaturization in sample preparation techniques. This method was developed by Arthur and Pawliszyn 4 and this procedure can be used for sample extraction and pre-concentration instead of the classical sample preparation methods. SMPE simply involves immersing a phase-coated fused silica fiber into the liquid sample, or the headspace above the sample, to adsorb the analytes of interest. The adsorbed analytes are then thermally desorbed in the heated injection port of the gas chromatograph and transferred to the capillary column for separation. SPME is a fast, simple and solventless alternative sampling technique 19,27. Some flavour groups e.g. alcohols, esters 14, dimethylsulfide 22, vicinal diketones 12, fatty acids 11 and carbonyl compounds 25 have been determined in beer using SPME. Another development of microextraction sample preparation methods, based on the polydimethylsiloxane phase as an extraction medium, was shown by Baltussen et al. 5 and a new extraction technique known as stir bar sorptive extraction (SBSE) was introduced. The extraction mechanism and advantages are similar to those of SPME, and because much more polydimethylsiloxane phase is used ( µl), the enrichment factor is consequently higher and sensitivity is increased by a factor of 100 to In the technique of SBSE, magnetic stirring rods are incorporated in glass jackets and coated with a layer of polydimethylsiloxane phase. These coated stir bars are commercially known as Twister and are produced by Gerstel GmbH. Several different types of Twisters are available 10 mm length with 0.5 mm respectively 1.0 mm phase thickness or 40 mm length also with 0.5 mm respectively 1.0 mm phase thickness. Typically the 10 mm 154 JOURNAL OF THE INSTITUTE OF BREWING

2 stir bars are used for 1 50 ml sample volumes and the 40 mm stir bars are used for ml sample volumes 21. The extraction procedure is very simple. Twister is placed into the sample in the vial and stirred. During the stirring period analytes are partitioned between the polydimethylsiloxane phase and the aqueous matrix. After extraction the Twister is introduced into a thermal desorption unit. The analytes are desorbed thermally and immediately separated on the capillary column of a gas chromatograph 21. Alternatively, the desorption of the extracted analytes in Twister can be also performed using a small volume of organic solvent and solvent back extraction 20. Headspace sorptive extraction (HSSE) can be used for the determination of volatile compounds. HSSE is a concentration technique that is much more sensitive than traditional static headspace analysis. A Twister is placed in an open adapter inside a closed headspace vial. The Twister is placed in the vapour phase above the sample such that analytes are extracted by, and concentrated into, the polydimethylsiloxane phase of the Twister 24. SBSE methods in beer analyses have been applied for the determination of sunstruck flavour (3-methyl-2-butene- 1-thiol) and other sulphur compounds 8, bitter acids using HPLC analysis 9 and stale-flavour carbonyl compounds 18. The aim of this work was to optimize SBSE for the determination of esters in beer. Due to the great efficiency of SBSE, the solvent back extraction of compounds sorptived in Twister was used instead of thermal desorption. An aliquot of this extract was then injected into a gas chromatograph fitted with a flame ionisation detector. MATERIALS AND METHODS Beer samples All analyzed beer samples were fresh commercial lagers of the Pilsner type, produced and bottled in the Czech Republic. Reagents Analytical reagent grade ethanol (Lach-Ner, Czech Republic), methanol, hexane, dichloromethane (Merck, Germany), purified water (Milli-RO 5plus, Millipore, USA), helium 5.0 quality, hydrogen 5.0 quality and synthetic air (Messer, Czech Republic) were used. Standards of analyzed compounds (isoamyl acetate, ethyl caproate, ethyl caprylate, phenyl acetate, ethyl caprate, phenylethyl acetate, ethyl laurate, ethyl myristate, ethyl palmitate) were purchased from Sigma-Aldrich (USA). Sample preparation Bottled beers were kept cool (4 C) until they were analysed. A solution of 5% (v/v) ethanol was used for evaluation of the method and for the calibration curves. Sample extraction was performed by placing 10 ml of sample amount in a 20 ml glass vial, adding a 10 mm length and 3.2 mm o. d. and 0.5 mm thickness of polydimethylsiloxane coated stir bar ( Twister, Gerstel, Germany). The vial was capped with an aluminium coated septum and stirred at room temperature at 1200 rpm. After extraction the Twister was removed with forceps and rinsed briefly in distilled water. For back extraction the Twister was placed into a 350 µl glass insert containing 200 µl of organic solvent. This insert was immersed into a 2 ml vial and stirred at 1200 rpm again. An aliquot (2 µl) of this extract was injected into the GC column. The Twister was reconditioned in a glass tube at 300 C with 50 ml/min helium for 1 h. The Twister was ready for another use after cooling. GC analysis The GC analysis was carried out using a Chrompack CP 9001 gas chromatograph equipped with autosampler Labio ASG 40. Analytes were separated on 30 m 0.32 mm i.d. fused silica capillary column of Phenomenex ZB- WAX with 0.25 µm film thickness. The GC column was maintained at 80 C for 1 min, ramped at a rate of 8 C/min to 240 C and then held at this temperature for 3 min. The split-splitless injector was used and the split vent was opened after 0.25 min. Temperatures of the injector and the flame ionisation detector were 260 C and 270 C, respectively. The carrier gas was helium 5.0 quality with a column head pressure of 150 kpa at 80 C. RESULTS AND DISCUSSION Development of the method In the first experiment the effect of the solvent back extraction of the Twister stir bar on the determination of esters was examined by the different solvent solutions dichloromethane, a mixture of dichloromethane:methanol 90:10 and a mixture of dichloromethane:hexane 50:50. The results are shown in Fig. 1. The mixture of dichloromethane:methanol 90:10 produced the best result for the sum of responses of all compounds. However in comparison with dichloromethane:hexane 50:50, the responses were more intense (about 36%) for the more volatile analytes such as isoamyl acetate, ethyl caproate and ethyl caprylate but not for less volatile compounds such as ethyl laurate, ethyl myristate and ethyl palmitate. For these semivolatile compounds, which are present in beer in very low concentrations, the solvent back extraction with dichloromethane:hexane 50:50 produced more efficiency, about 10% more than the mixture of dichloromethane: methanol 90:10. Thus all other experiments were carried out using the mixture of dichloromethane:hexane 50:50. In the next experiment, the effect of the sampling time on extraction was tested. As shown in Fig. 2 the responses increased with the sampling time. The responses of volatile flavour components, such as isoamyl acetate, were saturated after 45 min. The longer sampling time, for example 120 min, increased the response of this compound by only 1%. However, in comparison, the response of ethyl palmitate was greater by 42% after a 120 min sampling time, than after a 60 min sampling time. Semi-volatile compounds, evidently partition to the polydimethylsiloxane phase more slowly, and so a longer period is required for equilibrium to be reached. In practise, full equilibration is not necessary for an accurate determination. On the other hand, a relatively short extraction time will not only result in a loss of sensitivity, but also of pre- VOL. 113, NO. 2,

3 Fig. 1. Influence of different solvent solutions used for the solvent back extraction of the Twister on the total response of the esters (A dichloromethane, B dichloromethane: methanol 90:10, C Dichloromethane:hexane 50:50). Fig. 2. Effect of the sampling time on the responses of some esters extracted by the Twister stir bar. Fig. 3. Influence of different solvent back extraction times for the total response of the esters. 156 JOURNAL OF THE INSTITUTE OF BREWING

4 cision. Consequently, a 60 min sampling time was selected for time saving and precision. Fig. 3 demonstrates the influence of different solvent back extraction times to the total responses of the compounds of interest. The solvent back extraction was completed after 40 min. In comparison with a 20 min re-extraction, the isoamyl acetate and ethyl myristate responses increased from 28% to 47%, respectively. However, the responses of isoamyl acetate, ethyl laurate, ethyl myristate, ethyl palmitate obtained by 60 min reextraction gave an increase of only 3% to <1%, in comparison with a 40 min solvent back extraction. Thus a 40 min solvent back extraction time was selected for further experiments. Fig. 4 illustrates the chromatogram of the beer sample after stir bar sorptive extraction. The peaks of all compounds were clearly free from interference by other GC eluents. The Twister extracts were also analyzed after Twister reconditioning. No peaks of the compounds of interest were presented in chromatograms, thus the procedure of reconditioning described above is sufficient. Method validation Validation of the method is summarized in Table I. The nine data point calibration curve for a range of ester concentrations from mg/l to 30 mg/l showed a slightly better fit to a quadratic curve than to a linear curve. For all compounds the correlation coefficients to straight lines were from to and for quadratic curves, fits were from to The accuracy of the method was investigated by conducting a recovery test. The test was performed by measuring the natural concentration of the esters in five different beers, followed by measuring the same beers spiked with a known concentration of each ester (7.5 mg/l). The results showed that recoveries were in the range %. The repeatability of the method was investigated by repeating the SBSE (five times during the same day) of the same beer sample. The results indicated that the method has good repeatability (RSD ranged in %) and that the concentrations of the esters were accurately determined. Limits of detection (LOD s) were determined as three times the standard deviation for six replicates, for analyte concentrations no higher than five times the LOD 2. The LOD s ranged between mg/l. CONCLUSIONS The determination of some flavour esters isoamyl acetate, ethyl caproate, ethyl caprylate, phenyl acetate, ethyl caprate, phenylethyl acetate, ethyl laurate, ethyl myristate and ethyl palmitate in beer using SBSE with solvent back extraction was described. The procedure devised for the practical extraction utilized 10 ml of the sample stirring with Twister (length 10 mm, film thickness 0.5 mm) with 1200 rpm during 1 h at room temperature. Solvent back extraction of Twister used 200 µl of Fig. 4. Chromatogram obtained by SBSE followed by the solvent back extraction of beer. Labelled peaks: 1 isoamyl acetate, 2 ethyl caproate, 3 ethyl caprylate, 4 phenyl acetate, 5 ethyl caprate, 6 phenylethyl acetate, 7 ethyl laurate, 8 ethyl myristate, 9 ethyl palmitate. Table I. Method validation for the determination of esters in beer by SBSE followed by solvent back extraction. Correlation coefficient Recovery Repeatability RSD Limit of detection Compound Linear curve Quadratic curve (%) (%) (mg/l) Isoamyl acetate Ethyl caproate Ethyl caprylate Phenyl acetate Ethyl caprate Phenylethyl acetate Ethyl laurate Ethyl myristate Ethyl palmitate VOL. 113, NO. 2,

5 the solvent mixture dichloromethane:hexane 50:50 with 1200 rpm for 40 min. The recovery of this method was good (>78%) and precision acceptable (RSD < 7.3%). The proposed method has many practical advantages, such as small sample volume and simplicity of extraction and only microliters of solvent consumption. The method can be performed in parallel at room temperature. SBSE followed by solvent back extraction is also low in cost compared to SBSE combined with an expensive thermal desorption device. ACKNOWLEDGEMENTS The authors thank members of the Czech Beer and Malt Association for financial support. This work is a part of the Research Plan of the RIBM No. MSM REFERENCES 1. Alvarez, P., Malcorps, P., Sa Almeida, A., Ferreira, A., Meyer, A. M., Dufour, J. P., Analysis of free fatty acids, fused alcohols and esters in beer: an alternative to CS 2 extraction. J. Am. Soc. Brew. Chem., 1994, 52(3), Anonymous. Code Federal Reference, Title 40, Part 136: Guidelines Establishing Test Procedures for the Analysis of Pollutants, Appendix B to Part 136. Definition and Procedure for the Determination of the Method Detection Limit, Revision 1.11, EPA, Arkima, V. and Jounela-Eriksson, P., Über die Bedeutung der Ester für das Bieraroma. Proceedings of the European Brewery Convention Congress, Berlin, DSW: Dordrecht, 1979, pp Arthur, C. L. and Pawliszyn, J., Solid phase microextraction with thermal desorption using fused optical fibres. Anal. Chem., 1990, 62, Baltussen, E., Sandra, P., David, F. and Cramers, C., Stir bar sorptive extraction (SBSE), a novel extraction technique for aqueous samples: Theory and principles. J. Microcolumn Separation, 1999, 11, Berry, D. R., Products of primary metabolic pathways. In: Physiology of Industrial Fungi. D. R. Berry, Ed., Blackwell Scientific Publications: Oxford, U. K., Chen, E. C.-H., Analysis of volatile beer flavor compounds by a dynamic headspace entrainment technique. J. Am. Soc. Brew. Chem., 1983, 41, David, F., Sandra P., Hoffmann, A., Harms, D. and Nietzsche, F., Elucidation of the hoppy aroma in beers by stir bar and headspace sorptive extraction followed by thermal desorption CGC MS/PFPD, AppNote 4/2001, Gerstel, 2001, Harms, D., Nietzsche, F., Hoffmann, A., David, F. and Sandra, P., The analysis of the bitter and other flavour compounds in beer and wort by stir bar sorptive extraction (SBSE) followed by HPLC, AppNote 5/2001, Gerstel, 2001, Hawthorne, D. B., Kavanagh, T. E. and Clarke, P. J., Determination of low molecular weight organic compounds in beer using capillary gas chromatography. J. Am. Soc. Brew. Chem., 1987, 45, Horák, T., Čulík, J., Jurková, M., Čejka, P. and Kellner, V., Determination of the fatty acids in beer by SPME. Kvasny Prum., 2005, 51, Horák, T., Kellner, V., Čulík, J., Jurková, M. and Čejka, P., Solid phase microextraction the new alternative for the determination of the vicinal diketone in beer. Proceedings of the European Brewery Convention Congress, Budapest, 2001, Fachverlag Hans Carl: Nürnberg, Germany, CD ROM 2001, Contribution Irwin, A. J. and Thompson, D. J., A rapid method for the extraction and analysis of beer flavour components. J. Inst. Brew., 1987, 93, Jelen, H. H., Wlazly, K., Wasowicz, E. and Kaminski, E., Solidphase microextraction for the analysis of some alcohols and esters in beer: Comparison with static headspace method. J. Agric. Food Chem., 1998, 46, Lindsay, R. C., Withycombe, D. A. and Micketts, R. J., Comparison of gas chromatographic methods for analysis of beer flavors. Proc. Am. Soc. Brew. Chem., 1972, Meilgaard, M. C., Prediction of flavor differences between beers from the chemical composition. J. Agric. Food Chem., 1982, 30, Nykänen, L. and Suomalainen, H., Formation of aroma compounds by yeast. In: Aroma of Beer, Wine and Distilled Alcoholic Beverages, Akademie-Verlag: Berlin, Ochiai, N., Sasamoto, K., Daishima, S., Heiden, A., C. and Hoffmann, A., Determination of stale-flavor carbonyl compounds in beer by stir bar sorptive extraction with in-situ derivatization and thermal desorption-gas chromatography-mass spectrometry. J. Chromatogr. A, 2003, 986, Pawliszyn, J., Solid-Phase Microextraction: Theory and Practice, Wiley-VCH, New York, Popp, P., Bauer, C. and Wennrich, L., Application of stir bar sorptive extraction in combination with column liquid chromatography for the determination of polycyclic aromatic hydrocarbons in water samples. Anal. Chim. Acta, 2001, 436, Sandra, P., Baltussen, E., David, F. and Hoffmann, A., A novel extraction technique for aqueous samples: Stir bar sorptive extraction, AppNote 1/2000, Gerstel, 2000, Scarlata, C. J. and Ebeler, S. E., Headspace solid-phase microextraction for the analysis of dimethyl sulfide in beer. J. Agric. Food Chem., 1999, 47, Stenroos, L. E., Grabowski, J., Spearman, J. and Siebert, K. J., The semi-routine use of capillary gas chromatography for analysis of aroma volatiles in beer. J. Am. Soc. Brew. Chem., 1985, 43, Tienpont, B., David, F., Bicchi, C. and Sandra, P., High capacity headspace sorptive extraction. J. Microcolumn Separations, 2000, 12, Veselý, P., Lusk, L., Basařová, G., Seabrooks, J. and Ryder, D., Analysis of aldehydes in beer using solid-phase microextraction with on-fiber derivatization and gas chromatography/mass spectrometry. J. Agric. Food Chem., 2003, 51, Wampler, T., P., Washall, J., W. and Matheson, M., J., Applications of purge-and-trap to the analysis of beers. American Laboratory, 1996, 28(11), 18T 18V. 27. Zhang, Z. and Pawliszyn, J., Headspace solid-phase microextraction. Anal. Chem., 1993, 65, (Manuscript accepted for publication June 2007) 158 JOURNAL OF THE INSTITUTE OF BREWING