APPLICATION OF SITE-SPECIFIC NATURAL ISOTOPE FRACTIONATION (SNIF-NMR) OF HYDROGEN TO THE CHARACTERIZATION OF EUROPEAN BEERS

Transcription

1 J. nt. Brew.. July-ugut,, Vol.,??. - PPLCTON OF TE-PECFC NTUL OTOPE FCTONTON (NF-NM) OF HYOGEN TO THE CHCTEZTON OF EUOPEN BEE By G. J. Martin, M. Benbeknuu and F. Lantier (Laboratoire tie MN el tie eactivite Chimique U CN Univeriie tie Nante. me tie la Houiniere 0 NNTE-Cetlex (France)) eceived ugut More than one hundred ample of European beer have been invetigated by the new NF-NM method which i baed on H NM at the natural abundance level and enable ite-pecific natural iotope fractionation factor in ethanol to be determined. The relative () and abolute (H),, parameter are hown to be characteritic of the country where the beer are brewed and the oberved variation are explained in term of cereal compoition, water reource, and manufacturing proceing (fermentation, yeat, temperature cycle). Thee new parameter find analytical and mechanitic application in the identification of a beer and in the invetigation of a fermentation proce. Key word: euterium NM, ite pecific iotope ratio. European beer, characteriation. ntroduction Quantitative deuterium NM at the natural abundance level i a powerful method lor tudying the ite-pecific natural iotope fractionation (NF) occurring in natural and ynthetic compound.- Thu NF-NM ha been applied to the identification of the origin of ethanol reult ing from the fermentation of variou fruit uice, tarche, cereal or from chemical ynthei. The cae of wine ha received a great attention ince ome prohibited practice uch a chaptaliation or water and ugar addition to the grape before fermentation arc frequently upected in the European Community. t wa intereting to tudy beer from at leat two point of view. Firt, due to the different regulation exiting in the variou countrie of the EEC, the cereal compoition of beer i not uniform. Then, it hould be valuable to ten tatively finding ome claification of the European beer in term of their cereal compoition. The econd apect i more fundamental and i concerned with the mechanim of the NF proce conidered a an analytical tool for detecting adulteration of beverage, and namely for invetigating the cae where exogenou water i added to a fermentable uice. n brewing indutry it i a natural practice to add water to the malt before mahing in order to obtain a beer after fermentation. f we ignore for the preent purpoe all the flavouring ubtance e.g. hop: a pure malt beer will con tain only water and tnalt-ethanol and an adunct beer will contain water and cthanol ariing from malt and maize or rice, cane-ugar etc. The fermentation proce i then con ducted with a wort containing ugar and exogencou water. Exogcneou water mean that the water found in wort i not the ame a the water ued during the photoynthei of barley tarch. Then from the point of view of iotope fractionation, a pure malt beer i equivalent to a watered wine and an adunct beer to a watered and enriched wine. t hould be emphaied however, that in mot countrie of the EEC, excepted in Germany, it i a legal practice to add ome cereal or glucoe (0% in France) to the malt before fermentation. e hall examine to which extent the NF- NM method i capable of characteriing the beer and providing information about it origin and 'hitory'. Theoretical and Experimental a) Theoretical conideration n alcoholic beverage i conidered from an analytical point of view a being a mixture of water and ethanol. far a only the table iotope of hydrogen arc invetigated, the molecular pecie which will be dicued are repec tively: (i) two fully protonated molecule HO and CH.OH (ii) four monodeutcrated molecule HO, CH,CHO, Ch,CHOH,CHCHOH (V) () () () The low abundant multidcutcrated pecie will be ignored ccording to the derivation preented in reference, we define a relative internal ditribution of deuterium in cthanol K = () and a ite-pecific iotope ratio (H), (H),. f. (H) (H)= (H), where i, ft, Fi arc repectively the ignal intenity, the real molar fraction and the tatitical molar fraction of pecie i (,, ) and (H) i the average iotope ratio of ethanol. hen only one unknown contituent of a mixture i earched for, it i ufficient and convenient to conider the relative parameter ; thi i the cae for the detection of the dry chaptaliation of a wine, or for the ditinction of a pure malt beer from an adunct beer prepared from the ame water. n both cae, the deuterium content of the water i aumed not to have changed from the natural to the chaptalied wine, or from the pure malt to the adunct beer: The parameter then give a meaure of the variation in concentration of pecie. However, when wine are upected to have been watered and enriched, or when wine and beer of different geographical area are to be compared the deuterium con tent of both water and ugar are expected to change and two parameter mut be determined. e have hown0 that the deuterium atom which are contained in water and in ugar cramble during the fermen tation proce, and are differently ditributed in the cthanol iotopomer. For a given ugar fermented by accharomyce Ceriviiae under tandard condition, the deuterium content of pecie J, defined by (H), originate from ugar and water in the proportion of about % and % repectively wherea the deuterium content in ite, (H),,, arie mainly from water (a:0%) and only about 0% from the ugar. Moreover, the iotope ratio (H)w of water after fermentation lightly increae ( to 'ppm) ubequently to a deuterium tranfer from the ugar to the medium. Thu the knowledge of (H),, (H),, and in () () ()

2 Vol., ] MTN ETl: PPLCTON OF TE-PECFC NTUL OTOPE FCTONTON Grape Juice Cereal and Glucide Blended whiky dunct Beer Pure Grape ine ater Mall Barley Figure euterium tom from Glucoe euierated Molecule rom ine, Beer. hiky... HO HOCH.CH euterium tom from ater Figure T\ (moww) are directly derived from ignal intenitie Ti = () denote the working tandard and V. MO the international tandard ued for the report of the iotopic ratio. The ite-pecific iotopic ratio (H), are then ome cae (H)u, allow a more precie indcntification calculated from: of an alcoholic beverage to be performed than it would be poible on the ole bai of the parameter. n principle it - T, i more convenient to expre the deuterium ditribution in (H),= () (H) ppm unit, ince thi cale (or it %o analogue) i T(MO) largely ued in ma pectrometry. However it hould be emphaied that the (H), value are obtained through a 0- T,, calibration procedure involving a working tandard and (H),,= () may uffer from light ytematic deviation. comparion of the (H), value i neverthele fully authoried. n prac tice experimental parameter T +,() and T(MO)= t being the alcoholic percentage of the ample.

3 MTN ETL: PPLCTON OF TE-PECFC NTUL OTOPE FCTONTON [J.lnt. Brew Volume % ol ethanol Number of ample Belgium ",, }.0 b) Experimental France Table ^"'-'-''^Mw luxe = _^ Vollbier peciale Figure.0.0 V [.0.0 Figure Germany Einfchbier tarkbier Germany France eight m% ol ethanol The ratio were determined according to Martin & Martin reference and (H),, (H),, and (H),,, were meaured by uing a quantitative reference. The calibration factor of the working tandard with repect to V. MO wa TMO =. Bruker M 00 pectrometer, operating at the fre quency of deuterium ( - M Hz) wa ued for obtaining the natural abundance H pectra of beer ditillate, contained in a mm O cell. mm O coaxial cell, containing the reference ubtance wa fitted into the main cell; to 0 pectra of each ample were recorded uing the following acquiition parameter: acquiition time -, weep width 0 Hz, number of can, proton broad-band decoupling, line-broadening= Hz. The reproducibility of a meaurement i better than 0-% for and -% for T. Uually 00 to 00 ml of beer were carefully ditilled with a rotating band column; to 0 ml of to -% ethanol i collected and the reidual water i retained for ubequent iotope analyi. The alcoholic grade of ethanol wa determined from a Karl-Ficher titration. eult and icuion a) Claification ofthe beer tudied. Belgian beer from three main region and from brewerie, German beer -0 - manufactured in different brewerie ituated in differ ent region (mainly in Bavaria, etphalia, Hee and axe), French beer manufactured in brewerie mainly ituated in lace and North of France, Czecholovakian beer from the Prague area and Bohemia, and other beer from England, cotland, reland and enmark were bought in pecialied hop and not directly from the brewery. n ome cae, two ample of the ame quality of beer from a given brewery, were bought in a period of month, in order either to keep track of the brewery production or to check any climatological influence on the parameter. Until now, there i no uniform regulation for beer in the different countrie of the EEC. However, in mot countrie, a claification of beer i made a a function of their ethanol content. Figure no, which compare the dif ferent categorie of beer in Belgium, France and Germany, exemplifie the ituation. The vertical cale i given either in volume or weight per cent unit of ethanol. t i apparent that for at leat thee three countrie the beer production can be claified into four imilar categorie, according to the ethanol content of the beer. n the table the Belgian ymbolim,, and are ued which avoid confuion of term: >% vol.; <% vol.; <% (Belgium and Germany), -% (France) and all other beer. For Germany, the denomination Pil (t%=- to ) Bock (t%>) or oppelbock (t%> to -) will alo be given -when available on the label. n addition more pecific denomination a regard the brewing procedure will be indicated e.g. bottom, top or pontaneou fermentation, white 'abbaye' or 'on yeat' beer. lthough the nature of the yeat ued during a fermen tation wa hown to have a pecific influence on the deuterium ditribution, the two main factor, which are expected to trongly modify the NF-ratio, are the nature of the water and the kind of cereal ued in the brewing pro ce. For commercial beer, the nature of the cereal and malt i uually not available, and one purpoe of thi work i to tentatively make a ditinction between pure malt (barley) beer and adunct beer where maize or rice ha been added to barley. The cae of wheat malt beer (weibier) will alo be dicued. far a water i concerned, it origin will be characteried by giving the name of the region where the brewery i etablihed, ince the water reource of a given geographical area ha been carefully invetigated by hydrogeologit. n mot cae tudied an etimation of the (H) ratio of the local water can be obtained from the E table () and related publication. b) eult.i The reult are given in Table (German beer) (French beer) (Belgian beer) and V (beer of other countrie). n order to clarify the reult, the hitogram of are alo drawn a a function of the country (Fig. ) and in Table V are collected the average value of, (H), and (H),, for the different region concerned. i) itinction of the beer a a function of the region of production The relative deuterium ditribution parameter,, i a enitive analytical tool for ditinguihing the origin of a beer. Thu it i intereting to note that French and German beer are characteried by ignificantly different ratio (Fig. ). The average value of for French and German beer differ by 0 unity. Four German beer are found to exhibit a value correponding to the domain of the French beer, but it hould be emphaied that three of them are pecially elaborated for exportation. One French beer ha a value greater than -0. f we except ample no,,, the Czccholovakian beer are characteried by a

4 Vol.. ] MTN ETL. PPLCTON OF TE-PECFC NTUL OTOPE FCTONTON TBLE. euterium ditribution in cthanol extracted from German beer. The categorie of beer have been defined in the text (% vol. <: - < % vol. < -0; % vol. <-) (P Pil B Bock oppel-bock cibicr T Top fermentation). The average (H) value were computed according to equation. with (H) = ppm for Bavaria and outh-germany and 0 ppm for the other region of Germany. Thee value have been elected from the data of Fortel () conidering that fermentation induce a light tranfer of deuterium from the glucidc to the water. Characteriation of ample "HNM eult No egion Cat t% Year (H), (H) Bavaria 0 Badcn-urtemberg North Germany Hee (b) etphalia B P P B T P P T ' i ) a) chlcwig Holtein and Nicdcrachen b) Berlin parameter comparable to that of German beer, although ignificantly greater. The 'anomalou' ample no,, again are intended for exportation. The value of the Belgian beer are much more cattered than thoe of the other ample, but their mean lie cloe to -. more pecific ditinction can alo be made if we conider now the nature of the brewery where the beer i elaborated. hen a ufficient number of ample i available (n>) it appear that a given brewery i characteried by a pecific ratio: Brewery: (Germany) = - (00); B (France) =- H(00);C(France) =-(00) f now the abolute iotope ratio (H), of the variou beer invetigated are examined it appear that the differ ence oberved in the value of the German and French beer can be explained by a ignificative difference in the (H), value, which are lower for German beer wherea the (H),, value are imilar for both countrie (Table V). Thi intereting behaviour will be dicued later in term of cereal compoition. The Belgian beer which exhibit value frequently ituated in a range cloe to that of Czecholovakian or even German beer are in fact charac teried by relatively high value of the (H),, parameter, wherea their (H), value uually lie between thoe of German and French beer. The Czecholovakian ample preent the lowet (H), or (H),, value among the whole erie of beer invetigated. Conideration of the (H), value can alo explain ome cattered reult oberved on the parameter. Thu the relatively high (H), value meaured for the German ample no and (Table ) which are exportation beer may be attributed to the contribution of a high deuterium content cereal uch a maize. By contrat the ample no which i a 'wcibier' ha a 'normal' (H), value but a relatively mall deuterium content to ite. imilar ober vation can be made for Czecholovakian beer. hen com pared to the other compound of the erie ample and (Table V) are characteried by high (H), value wherea the (H),, ratio are cloe to thoe of the other ample. gain thi obervation can be explained by the preence of maize glucoe in the wort of ample and, but it i clear that all the product are brewed in the ame country (imilar (H),, value). ii) nfluence of cereal and water on the NF parameter in beer The characteriation of the nature and of the origin of malt or grain ued in brewing i an important problem. t i therefore intereting to examine to what extent the ignifi cant difference in the deuterium ditribution oberved for beer elaborated in variou European countrie offer a ource of information on thi problem. To thi aim the itepecific iotope parameter will now be dicued in term of the nature of the raw material and of the manufacturing procee. f the reult in Table V are examined it appear that the maor contribution to the difference in the value oberved between French and German beer arie from an important change in the (H), value. n term of cereal nature thi behaviour indicate that the glucoe contained in the cereal from which the French beer have been brewed ha a higher deuterium content than the glucoe of the German beer. Conidering that barley hould have relatively mall (H) ratio- it can be concluded that

5 < * MTN ET.: PPLCTON OF TE-P-CFC NTUL OTOPE FCTONTON [J. nt. Brew TBLE. euterium ditribution in cthanol extracted from French beer. The categorie of beer correpond to the following pecification % vol. <-;- <% vol. < -; % vol >-. The (H) value were computed from equation uing (H) value of 0 ppm and ppm for North and lace repectively. Characteriation of ample HNM eult No egion Quality Year (H), (H),, (H) 0 0 lace North i - - ' a) origin not pecified TBLE. euterium ditribution in cthanol extracted from Belgian beer. The categorie of beer correpond to the following pecification: % vol. <:- < % vol. < -; %_vol. > - (T Top fermentation, bbayc, pontaneou fermentation, Y beer with yeat. V white beer). The (H) value were computed from equation uing (H)lll value of 0 ppm. Characteriation of ample :H NM eult No egion Quality Year (H), (H),, (H) Brabant w w Y Y toil Flundrc - - V T nver - - TY liainaut! Y Y i Namur! T Luxembourg!

6 Vol., ] MTN ETL: PPLCTON OF TE-PECFC NTUL OTOP FCTONTON TBLE V. euterium ditribution in cthanol extracted from beer ormiccllancou European countrie. % vol. <-% vol >-. Thc(H) value were computed from equation uing (l ) value of ppm (Czecholovakia) ppm (cotland and enmark) and l0ppm (England and reland). Characteriation of ample H NM eult No egion Quality Year (H), (H),, (H) England reland g 0 cotland - - Mall hiky Mall hiky Mall hiky Mall hiky enmark Czecholovakia North Bohemia outh Bohemia a) ample bought in France. TBLE V. verage deuterium ditribution in cthanol extracted from European beer. X i the mean, t the confidence interval computed Tor a % (n > ) or -% (n > ) certitude, N the number of beer ample of the concerned country and n the number or cae conidered in the calculation. COUNTY (H), (H),, X t n N X t n N X t n N GEMNY Bavaria North etphalia , FNCE 00, lace North , 0-0, BELGUM Brabant Hainaut Namur , 0-0, CZECHOLOVK - 0-0, a) the number ofcae i loo mall the German beer invetigated here (with the exception or ample no, ) contain only malt and that the French beer contain quantitie of the order of -0% (ee below) of high deuterium content cereal. t hould be emphaied that climatic and phyiological (actor alo induce mall variation in the deuterium con tent of barley tarch, which i itelf governed by the iotope ratio of the growing water and by the evapotranpiration cycle of the plant. However thee effect are econd-order perturbation which do not trongly modify the oberved trend. From thi point of view, it i intereting to dicu the behaviour of beer in the light of that of whikie. Thu an important decreae in the value wa oberved when going from pure malt whikie to blended whikie contain ing maize ethanol. imilar difference are in fact oberved between pure malt beer and adunct beer. However from the reult in Table V it appear that the iotopic ratio and the deuterium content in the methyl ite are repectively

7 MTN ETL: PPLCTON Of TE-PECFC NTUL OTOPE FCTONTON [J. nt. Brew omewhat lower and higher in pure malt whikie (=-; (H)=-) than in pure malt beer. There fore the influence of difference in the manufacturing procee hould alo be conidered. The cae of pure and adunct beer may alo be compared to that of natural and enriched wine (cheme ). n term of value the compoition of an ethanol obtained by fermentation of a mixture of two different glucoe can be computed, according to reference, a: t(ad) = t. (mixt) (malt) (ad) (malt) where t(ad) i the volume percentage of ethanol from the adunct cereal and the overall alcohol content of the beer. (t(ad)t i in fact the molar fraction of the adunct ethanol in the beer). etaining for (malt) and (adunct) the value - and -* repectively we obtain for the French beer a typical value t(ad)t=o% which i in accordance with the regulation. far a deuterium ditribution i concerned, no ignifi cative difference are oberved for beer brewed with a mix ture of barley and wheat malt (wcibier) (Table and H). Thi obervation i in agreement with the reported data for wheat and barley ethanol. regard the influence of the water it hould be emphaied that the cae of beer differ from that of wine, ince the malt i not necearily brewed at the place where barley ha been harveted wherea grape are generally fermented in itu. Brewerie ue their local, and in ome cae their own water. already mentioned the deuterium content of thi water will therefore trongly influence the iotope ratio of pecie, (H),,. t i intereting to note that thi parameter doc not vary ignificantly for a given country whatever the cereal ued in the brewing proce, but, in turn, appear to depend on the origin of the water and on the technical treatment undergone during brewing a will be dicued now. Hi) Otherfactor influencing the iotopefractionation in beer n addition to the nature of the cereal and the origin of the water, other factor, which are related with the fermen tation proce itelf, may influence the NF parameter of the beer. Thu we have hown that the nature of the yeat cla can have an important effect on the deuterium di tribution in ethanol. However in the cae of beer, which make ue of accharomyce yeat train, only, mall, although ignificative, difference arc expected in identical () condition of alcoholic production. Thu when a given ugar i fermented eparately with accharomyce martini and with accharomyce cereviie, the latter train i characteried by higher and (H) value "\ For example in the cae of a pure grape uice fermented in tandard con dition the value equal repectively - and - which correpond to a difference of about ppm in the (H), value. nteretingly a imilar behaviour of (H),, i oberved (Table and ) for beer produced by top and bottom fermentation (,) which utilie thee different accharomycc pecie. Moreover the poible occurrence of temperature effect mut be conidered. n thi repect the formation of ugar from malt ha been hown to be trongly temperature dependent () and the temperature cycle adopted in the brewing proce i alo expected to influence the iotopic compoition of the brewing water. regard the role of water it ha been noted that, the H value of ground water in etern Europe doe not change very much at leat when compared to the magnitude of the effect meaured here. Thu an increae of ppm in the (H),, value of ethanol hould correpond to a variation of about ppm in the iotopic content of the fermentation water; uch a variation i much higher than the difference expected on the ole ground of geographical effect but can be explained by an increae in the temperature of the water pool which lead to a lo in the light iotope'0 and, according to cheme, to an increae in the ite pecific (H)u parameter. Concluion The NF-NM method ha been hown to contitute a very powerful ource of information for the characteri ation of variou beer ample brewed in different European countrie: Germany (), France (), Belgium () and variou region (). The relative deuterium ditribution defined by the parameter appear to be ignificantly dependent on the country where the beer i brewed. The oberved trend are explained in term of variation in the deuterium content of ite (CHCHOH), preferentially related to the glucoe ource and ite (CHCHOH), preferentially related to the fermentation water. n practice the change in the three parameter, (H), and (H),,, can be ued for inducing ueful information on the nature of the raw material and on ome of the technical condition of brewing a hown below: Material NF-NM Cereal ater Parameter malt (barley) adunct (maize) ground (light) recycled (heavy) () (H)ll O =0 Brewing NF-NM Temperature Yeat (acch.) Parameter low high Uvarum Ccrcviiac - : 0 0

8 Vol..] MTN ET. PPLCTON Or TE-PECFC NTUL OTOPE FCTONTON The ign + (increae) and (decreae) compare the effect of a given factor (cereal, water, temperature, yeat) in each pair ofcae conidered, indicate mall change. cknowledgement!!. Thi work wa upported by a grant from the 'irection de ndutrie gricolc ct limentaire' ou-irection de 'lnnovalion which i acknowledged. e alo thank r. Flayeux and r. Moll (TEPL) for valuable dicuion. eference. Forlcl. H. & Hulzcn. H.. Nuiun: 0.. p. U>.. Gonfianlini... Nature..... Hough. J... Brigg.. E. & Young. T... Mulniix mid Brewing cience Vol.. pp. Ed. Chapman London X. Maouhe. M..J Cliim. Pin Martin.G. J & Marlin. M L.. Tetrahedron letier\... Marlin. G. J & Marlm. M. L.. C...hail. t. Pun.... Marlin. G.J.. Marlin. M L... Mabon. F. &Michon. M. J.. nalyti cal Chemitry.,. 0.. Marlin. G. J.. Martin. M. L.. Mabon. F. & Michon. M. J.. Jowmil vriclillliriil Food C Uemilr\... l... Marlin. G. J. & Marlin, M. L... J. Chim. Pin.. 0,.. 0. Marlin. G. J.. Zhang. ). L.. Martin. M. L. & upuy. P. Bhrliem. Biophy. e. Coimn.. ll,. 0. Moll. M.. Huycux.. & Marc... European Brewery Convention. Proceeding of the th Congre Copenhagen... Montcmhaull... utrun. J. C.. Joudrier. P. & Moll. M.. C.. cad. i. Pari.... chmucker. K. L.. Bnumvlt table otope in Hydrology. euterium and oxygen in the water cycle. Technical eport erie no ; 0. E. VENN.. Vcrmeylen. J.. Guide Pratique de Braerie,. f.. Zhang. B. L.. iertation. NNTE.