DETERMINATION OF HALOACETIC ACIDS IN BEER

Proceedings of ECOpole
Vol. 2, No. 1 2008
Jiří ČULÍK 1 , Tomáš HORÁK 1 , Marie JURKOVÁ 1 , Pavel ČEJKA 1
and Vladimír KELLNER 1
DETERMINATION OF HALOACETIC ACIDS IN BEER
OZNACZANIE POCHODNYCH KWASU OCTOWEGO W PIWIE
Summary: A method for the determination of chloroacetic acid, bromoacetic acid and iodoacetic acid in beer is
described. Solid phase extraction for the sample clean-up was used. Several SPE columns were tested. Then the
analytes were derivatized using BF3/methanol as methylation agent and determined by capillary gas
chromatographic analysis coupled with electron capture detector. Recovery yields are nearly 100% for all
monohaloacetic acids. Detection limits are lower than 0.1, 0.05 and 0.01 mg/dm 3 for chloroacetic acid,
bromoacetic acid and iodoacetic acid, respectively. This procedure is suitable for routine analyses of haloacetic
acids.
Keywords: haloacetic acids, solid phase extraction, beer
Haloacetic acids are toxic compounds which may be detected in beers after the
inadequate disinfection of the technological equipment during beer production. The content
of these compounds in beer is prohibited by the legislation of most countries [1].
The most of published methods have been based on different extraction procedures and
derivatization followed their determination by gas chromatography [2-5]. The electron
capture detector respectively mass selective detector were used most frequently [2-6].
Liquid-liquid extraction procedures were shown in studies [2-5]. Solid phase extraction
(SPE) is another possibility for isolation of these compounds [7-9].
The extracted haloacetic acids have been esterified by methylation with BF3/MeOH
[7], ethylation with EtOH/SO4H2 [4, 10], buthylation with BF3/ButOH [5] or
pentafluorobenzoylation with PFBBr [5, 10].
In this work an improved method for the determination of haloacetic acids in beer is
reported. Systematic investigations were performed on the off-line clean-up using SPE
columns filled with four different types of sorbents followed derivatization of the
compounds of interest with BF3/methanol derivatization agent.
Experimental
Reagents and standards
Methanol p.a., ethanol p.a., sodium sulphite anhydrous p.a., salt acid p.a. were
purchased from Lachema (Czech Republic), ethylacetate p.a., 20% solution of BF3/MetOH,
hexane, chloroacetic acid, dichloroacetic acid, bromoacetic acid, iodoacetic acid were
obtained from Merck (Germany). Water used was Milli-Q grade Millipore (USA).
SPE columns: Extrelut (Merck, Germany), Separon SGX CN (Tessek, Czech
Republic), Bond Elut SAX (Varian, USA), Separcol Si C18 (Anapron, Slovakia).
1 Research Institute of Brewing and Malting, Brewing Institute Prague, Lípová 15, 120 44 Prague 2, CZ,
email: horak@beerresearch.cz

16
Instrumentation
Jiří Čulík, Tomáš Horák, Marie Jurková, Pavel Čejka and Vladimír Kellner
The GC analysis was carried out using a Carlo Erba HRGC 5300 Mega Series gas
chromatograph equipped with electron capture detector. Three fused capillary columns
were tested: SPB-5, 30 m x 0.32 mm i. d. with 0.25 µm film thickness; DB-1301,
30 m x 0.32 mm i. d. with 0.25 µm film thickness; DB-5, 60 m x 0,32 mm i. d. with 1.0 µm
film thickness.
Sample preparation
Prior to use SPE cartridges were conditioned by 3 cm 3 of methanol followed 3 cm 3 of
purified water. Then the 20 cm 3 of sample was forced through the bed of sorbent using
vacuum. The sorbent was washed with 1 cm 3 of purified water. 2 g of NaCl and 10 mm 3 of
internal standard (dichloroacetic acid) were added to the collected eluate. The pH value of
the sample was changed to the value 0.5 by addition of concentrated HCl and sonication
was carried out for 1 min. Then the haloacetic acids were extracted with 4 x 2.5 cm 3 of
ethylacetate (EtAc) by intensive shaking. After centrifugation at 3900 rpm for 4 min, the
supernatant EtAc layer was transferred into flask. The EtAc solution was dried over
2 g anhydrous Na2SO4. Then EtAc was removed by rotary evaporator and vacuum at 35°C.
The residue was redissolved in 2 cm 3 of methanol followed by 1 cm 3 of BF3/MeOH
and transferred into a screw-capped glass vial. This was cap sealed and placed for 60 min in
a water bath at 85°C.
After methylation, the solution was cooled and extracted with 1.5 cm 3 of hexane.
Hexane layer was collected and either immediately injected into the gas chromatograph or
transferred into a screw-capped glass vial and stored at –18°C for several days.
Results and discussion
The optimal GC conditions were evaluated. The separation of bromoacetic acid and
dichloroacetic acid was not satisfied with SPB-5 column. The better results were obtained
with DB 1301 column. Separation of bromoacetic acid and dichloroacetic acid was perfect
but unfortunately some compound interfered with chloroacetic acid during analyses of real
beer sample. The chromatogram obtained by DB-5 column, 60 m x 0.32 mm i. d. with
1.0 µm film thickness gave satisfactory results. The peaks of all compounds were clearly
free from interference by other GC eluents.
These conditions could be recommended: the GC column was maintained at 60°C for
2 min, ramped at a rate of 3°C/min to 80°C and then ramped at a rate of 40°C/min to
285°C. The split-splitless injector was used and the split vent was opened after 0.5 min.
Temperature of the injector and the electron capture detector were 250 and 320°C,
respectively. The carrier gas was helium 5.0 quality with a column head pressure of 55 kPa
at 60°C.
In the second step the extraction procedure was tested. Results obtained by using
sorbent Extrelut were unsatisfactory (recovery < 5%). Low recovery might be a result of
degradation of sorbent at pH 0.5.
SPE cartridges Separon SGX CN gave similar bad results. Cartridges filled by Bond
Elut SAX produced a little bit better recoveries (about 25% for chloroacetic acid but for

Determination of haloacetic acids in beer
another compounds only about 10%). Separcol Si C18 (Anapron, Slovakia) yielded the best
results. So all validated experiments were done with this sorbent.
The accuracy of the method was investigated by conducting the haloacetic acids
recovery test. The limit of detection was obtained as 3×S/N. The repeatability of the
method was investigated by repeating (five times during the same day) of the same beer
sample. Results are shown in Table 1. Eight Czech beers were analyzed for haloacetic acids
content. Results were negative in all cases.
Limit of detection, recovery yields and RSD for haloacetic acids in beer
Haloacetic acid
Detection limit
[mg/dm 3 ]
Recovery
[%]
RSD
[%]
Chloroacetic acid 0.1 98 7
Dichloroacetic acid 0.05 96 5
Bromoacetic acid 0.05 96 4
Iodoacetic acid 0.01 93 3
Conclusions
17
Table 1
We have demonstrated that haloacetic acids after SPE clean-up step, their extraction
with ethylacetate and derivatization by BF3/MeOH can be determined in tenth respectively
hundredth of mg/dm 3 in beer using gas chromatography with electron capture detector. This
procedure is suitable for routine analyses of haloacetic acids in beer.
Acknowledgements
Authors thank members of the Czech Beer and Malt Association for financial support.
This work is a part of the Research Project of the RIBM No. 6019369701.
References
[1] Council of the European Communities, 1987, Regulation No. 822/87.
[2] Guyot A.M. and Balatre P.: Ann. Falsificat. l`expertise chimique, 1968, 61, 346-359.
[3] Nijboer L.: De Ware(n)-Chemicus, 1985, 15, 18-25.
[4] Fürst P., Krüger C., Habersaat K. and Groebel W.: Z. Lebensmittel Untersuch. Forsch., 1987, 185, 17-20.
[5] Luckas B.: Z. Lebensmittel Untersuch. Forsch., 1987, 184, 30-32.
[6] Amtliche Sammlung von Untersuchungsverfahren nach § 35 LMBG, 1989, L 36.00-10.
[7] Rothenbücher L., Weishaar R. and Köbler H.: Lebensmittel Chem. Gericht. Chem., 1986, 40, 4-5.
[8] Willetts P., Dennis M. J. and Massey R. C.: Food Addit. Contamin., 1991, 8, 119-124.
[9] Sendra J. M. and Todo V.: J. Inst. Brew., 1990, 96, 85-87.
[10] Gilsbach W.: Deutsche Lebensmittel-Rundschau, 1986, 82, 107-111.
OZNACZANIE POCHODNYCH KWASU OCTOWEGO W PIWIE
Streszczenie: Opisano metodę oznaczenia w piwie kwasów chlorooctowego, bromooctowego i jodooctowego. Do
oczyszczania próbek wykorzystano stałą fazę ekstrakcyjną. Przetestowano kilka kolumn SPE. Anality były
metylowane z wykorzystaniem układu BF3/metanol, a ich pochodne metylowe były oznaczane za pomocą
kapilarnej chromatografii gazowej sprzęŜonej z detektorem wychwytu elektronów. Wydajność tego procesu jest
prawie 100% dla wszystkich monohalogenowych pochodnych kwasu octowego. Granice wykrywalności dla
kwasu chlorooctowego, bromooctowego i jodooctowego wynoszą kolejno mniej niŜ 0,1; 0,05 i 0,01 mg/dm 3 .
Metoda ta moŜe być stosowana podczas rutynowych analiz pochodnych kwasu octowego.
Słowa kluczowe: pochodne kwasu octowego, ekstrakcja fazy stałej, piwo