Effect of the spontaneous fermentation and the ageing on the chemo-sensory quality of Brazilian organic cachaca/ Efeito da fermentacao natural e do envelhecimento sobre a qualidade quimica e sensorial de cachaca organica Brasileira.
Around 30% of the annual production in Brazil, which is estimated to be 1.3 billion litres, constitutes the artisanal cachaca, in which the fermentation is initiated by autochthonous yeasts from the sugar cane juice, the so-called natural ferment (PATARO et al., 2000). The preparation of this traditional starter ferment does not follow a standard recipe and basically consists of adding sugar cane juice to crushed corn and powdered rice (ROSA et al., 2009). It is a spontaneous fermentation using wild yeasts (non-Saccharomyces) and Saccharomyces cerevisiae, with a predominance of the latter species (MORAIS et al., 1997; PATARO et al., 1998). As a consequence, artisanal cachacas have different sensory qualities due to the metabolites and volatiles produced by the diversity of yeasts present during the fermentation process. Hence yeast composition is a key factor in cachaca quality (NOVA et al., 2009). Non-Saccharomyces yeasts studied by OLIVEIRA et al. (2005) in small-scale sugar cane fermentations did not exert a negative influence on the sensory quality of the cachaca. Wild yeasts (Pichia silvicola, Pichia anomala and Dekkera bruxellensis) were also adequate for high-quality cachaca production, because these microorganisms produce acceptable concentrations of secondary compounds (DATO et al., 2005).
Studies on the chemo-sensory quality of artisanal organic cachaca produced with natural ferment are scarce. Besides, although the ageing is optional for cachaca, it is a common technique for distilled beverages in several countries resulting in a better sensory quality, and constitutes the final step in the production of distilled beverages of differing qualities (FARIA et al., 2003). In this context, this work aimed to evaluate the influence of the spontaneous fermentation on the chemo-sensory characteristics of artisanal organic cachaca, comparing to the commercial ferment. The effect of maturation (ageing) in oak barrels was also evaluated, as a contribution to those small producers who want to create a different quality beverage.
MATERIAL AND METHODS
Initially the natural yeast inoculum (NF) was prepared by mixing 2kg rice bran, 1kg breadcrumbs and 1kg corn flour, which were enclosed in a cotton tissue into a container with around 30L of diluted sugar cane juice (6-7[degrees]Brix) and then heated at 28[degrees]C-35[degrees]C. Five additions of 15L of diluted sugar cane juice were made within a 10-day period until a volume of approximately 100L was obtained to be added to the fermentation tanks. The juice was extracted from the sugar cane variety RB855156 cultivated under organic management in Araras, Sao Paulo State, Brazil. The harvest was made manually without burning and the sugar cane stalks were crushed on the day of harvesting.
The fermentations were carried out in four 1000-L tanks (200-L working volume), two for each ferment (NF; and CF as fresh baking yeast Fleishmann[R] in the proportion of 8g [L.sup.-1] hydrated in warm water before inoculation in the tank), in a cell-recycle process for three consecutive cycles, at room temperature (30[degrees]C32[degrees]C). The initial cell concentration was [10.sup.6]-[10.sup.7] living cells [mL.sup.-1] for both inocula. After the fermentation completion, the must was distilled and around 20% vol vol-1 of the ferment remained in the tank for the next cycle.
The distillation (150-200L of fermented must) was conducted in 300-litre copper alembics with pre-heater and cooler, discarding the "head" and "tail" fractions (around 10% of the initial and final distillate, respectively). The "heart" fraction was diluted in distilled water and distilled again according to FRANCO (2008). The distilled samples were separated in "head", "tail" and "heart" fractions again and the last fraction was kept in glass containers for chemosensory analysis (at 40[degrees]GL) and stored (at 42[degrees]GL) in 5litre oak barrels for 45 days, based on the work of BORRAGINI (2009).
Cachaca samples (at 42[degrees]GL) were analysed for copper, with direct measuring by atomic absorption spectrometry--Perkin-Elmer 373 (AOAC, 1995); acetaldehyde, ethyl acetate, methanol, n-propanol, isobutanol and isoamyl alcohol, using gas chromatography with a flame ionization detector (INSTITUTO ADOLFO LUTZ, 1985); pH, with direct measuring in a digital pH-meter; alcohol content ([degrees]GL), using a digital densimeter; total, fixed and volatile acidity, by titration (AMORIM et al., 1979). The concentration of higher alcohols (sum of n-propanol, isoamyl alcohol and isobutanol) and total volatile compounds (sum of volatile acidity, higher alcohols, acetaldehyde and ethyl acetate) were estimated. Samples of the natural ferment (during preparation), must (at the start of the fermentative cycle) and ferment were taken out and analysed for pH and total acidity; and soluble solids ([degrees]Brix), using a refratometer. For microbiological analysis, samples of the must and ferment were centrifuged at 3,000rpm for 5 minutes, and the washed supernatant was serially diluted and plated onto WLN, WLD and Lysine Agar for yeast and Nutrient Agar for bacteria countings (CECCATOANTONINI, 2010). The plates were incubated at 30[degrees]C for 2 and 3 days (for bacteria and yeasts, respectively).
For the sensory analysis of acceptability (affective test), a panel of 62 non-trained judges evaluated the samples with respect to aroma, flavour, colour, body (adstringency) and global impression, using a 7-point hedonic scale ranging from "dislike very much" to "like very much". The cachaca samples (20mL) were served in transparent, colourless containers labelled with a three-digit code and presented in a monadic manner (MacFIE et al., 1989). White light was used in the individual booths, with a balanced presentation of the cachaca samples according to a completely randomized block design. The sensory tests were approved by the University Ethics Committee (CAAE 0026.0.135.000-09).
The analysis of variance and Tukey's test (P<0.05) were utilized for the comparisons but for the sensory analysis data the non-parametric test of Kruskall-Wallis was employed.
RESULTS AND DISCUSSION
During the ferment preparation, the soluble solids values were kept constant due to the periodical addition of sugar cane juice; in contrast, the total acidity increased significantly probably due to the organic acid production, but without a relevant decrease in pH after 2 days of preparation. A progressive increase in the total yeast count (from WLN medium), a stable number of wild yeasts resistant to actidione (from WLD medium) and a low number of non-Saccharomyces (from Lysine Agar) demonstrated that the number of Saccharomyces increased considerably with the addition of sugar cane juice (Figure 1). The yeast count increased 5 log cycles in a 10-day period, suggesting that a yeast selection occurred during the preparation of the NF, especially Saccharomyces, which are more adapted to the high sugar concentrations from the sugar cane juice. The high must acidification and alcohol content also contribute to the selection of yeasts that prevail in the cachaca production (MORAIS et al., 1997; PATARO et al., 2000). The high number of bacteria at the end of the preparation would be a negative point for the NF, but the microbiological analysis of the must and the ferment inside the fermentation tank in the first fermentative cycle showed bacterial count much lower than that observed in the NF at the end of preparation. A dilution effect by the introduction of the ferment into the tank or bacterial death caused by the high sugar concentration could explain this fact. Moreover, the NF does not constitute a primary source of bacterial contamination (Figure 1).
The microbiological analysis of the ferment after the fermentative cycles revealed that a higher number of yeasts was found in the CF, but for both ferments there was a decrease along the cycles. The number of wild yeasts leaned to decrease along the fermentative cycles with the NF while for the CF an increase or stabilization was verified (Figure 2). The characteristic nature of the ferments could explain these results once in the CF the yeast population is exclusively S. cerevisiae, but with the cell recycle and new sugar cane juice being introduced into the tanks, the ferment is contaminated with indigenous yeasts. With the NF, although these facts had also occurred, the yeast population is more stabilized because there was a previous adaptation and selection during the propagation of it. Thus the NF would be more refractory to the contamination by wild yeasts. The NF acidity was significantly higher and the residual solids (Brix) for both ferments were very low, as expected (Figure 3).
The NF-must presented a higher number of wild yeasts while there was no difference in the total number of yeasts and bacteria comparing to the CF-must (Figure 2). The values of pH were significantly lower in the NF-must but despite of the high total acidity in this initial inoculum, the values decreased significantly for both ferments during the fermentation. The soluble solids values in the CF-must was higher, however there was only significant difference in the last cycle (Figure 3).
The results of the chemical analysis for the redistilled cachaca are shown in table 1. The cachaca pH differed significantly only in response to maturation, with significant decrease in values as already observed (PARAZZI et al., 2008). There were no significant variations in the copper values and they did not exceed the limit of 5mg [L.sup.-1] (CARDOSO et al., 2003), which was due to the redistillation (values decreased from 6.5 to 0.4mg [L.sup.-1]). Maturation was found to have a significant impact on the total and fixed acidity, causing a 3-fold increase in total acidity. After maturation, the volatile acidity increased with the CF, while the fixed acidity was significantly higher with the NF. The influence of the yeast composition (ferment) on the cachaca quality was evident, but depended on the maturation of the beverage in oak barrels. The alcohol content and methanol level of the beverage remained within the limits of Brazilian laws (BRASIL, 2005). Neither the ferment nor the maturation influenced the concentration of methanol, ethyl acetate and isoamyl alcohol in the cachacas. The concentrations of n-propanol and isobutanol, and consequently of higher alcohols, were significantly enhanced by the maturation, but not differ regarding the ferment, except for the total volatile compounds, whose concentration increased significantly in the cachaca produced with the NF after maturation.
It seemed that the choice for a natural or a commercial ferment does not impact considerably the fermentation parameters and consequently the physico-chemical composition of the cachaca significantly, although more wild yeasts were found when natural ferment was utilized. Greater alterations in the cachaca composition were found to be more exclusively related to the maturation period inside the oak barrels. However, considering that cachaca is a beverage and for this reason, the sensory characteristics are also relevant, the evaluation of what extension the utilization of natural ferment alters these characteristics is crucial.
[FIGURE 1 OMITTED]
There was a significant loss of aroma, flavour and global impression after maturation but only in cachaca produced with the CF. Maybe an extensive scanning of compounds (other than those established by Brazilian legislation) produced by the microorganisms could help to substantiate this result, once the ageing (maturation) commonly brings about an increase in the sensorial quality of the beverages. For colour, there was a pronounced (and negative) effect of the maturation regardless the ferment, which is probably explained by the fact that the cachaca gets coloured (brownish colour) due to the interaction with the barrel wood, which may impress the judges used to prove uncoloured (not aged) cachaca (Table 2). The results revealed a strong interaction between ferment and ageing (maturation) of the beverage, suggesting that unknown substances produced by the microorganisms from different inocula during fermentation reacted differently with the components of wood barrels influencing the sensory attributes. Studies have demonstrated that cachacas produced by starter strains of S. cerevisiae isolated as indigenous cultures from natural fermentations and from distillery received better sensorial evaluations (SILVA et al., 2009a, b).
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
A point to be considered is regarding the volume of barrels and time of ageing in the present work. The ideal time for the cachaca storage varies according to the barrel characteristics (type of wood, age and size) and to the environmental conditions of storage (temperature and moisture). Volumetric reductions may occur during the ageing process and the greater the barrel, the lower the loss by evaporation. It is also assumed that the shorter the barrel, the best is the extraction of the substances from the wood because of the closer contact of the beverage and the wood (higher specific surface). To avoid excessive volume reduction in the 5-litre barrel, a period of 45 days was here considered for ageing, best named as maturation. BORRAGINI (2009) has showed an increase in secondary compounds and acidity with longer time of ageing however the sensory quality of the cachaca did not differ substantially at 30 or 60 days of ageing in 5litre barrels comparing to longer periods until 180 days for the majority of the attributes.
A profound investigation on the chemical composition of cachacas besides the parameters considered in the legislation could substantiate the reason why the cachaca maturation may result in a different product depending on the type of inocula, because many unknown compounds should have been produced by the diversity of microorganisms, especially when NF is used. Although lack of control and homogeneity in the preparation can negatively impact this kind of ferment (natural), the results found in this study and others should not be ignored. Technological strategies to support and qualify the cachaca producers are required if this important agro-industry is to be stimulated and extended, ensuring the competitiveness and survival of small agricultural producers.
The type of the ferment (natural or commercial) did not significantly influence the physicochemical composition of cachaca, which was markedly altered by the maturation period in oak containers. However a loss of acceptability of the aged beverage produced with commercial ferment was verified, which did not occur with the cachaca from spontaneous fermentation of sugar cane juice.
The authors wish to thank Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP) for the financial support (process n.07/06979-8).
AMORIM, H.V. et al. Metodo rapido para controle da fermentacao e destilacao. Saccharum/ STAB, v.2, n.4, p.3134, 1979.
AOAC. Association of Official Analytical Chemists. Official methods of analysis. 16.ed. Arlington, 1995. V. 1, p.31-65.
BORRAGINI, M.C.C.B. Envelhecimento da cachaca com circulacao forcada e aeracao. 2009. 90f. Dissertacao (Mestrado em Ciencia dos Alimentos)--Faculdade de Ciencias Farmaceuticas, Universidade Estadual Paulista "Julio de Mesquita Filho", SP. Available from: <http://www.fcfar.unesp.br/ posgraduacao/alimentosenutricao/Disertacao/2009/ Michelle_CaiadoME.pdf>. Accessed: Nov. 26, 2011.
BRASIL. Instrucao Normativa n.13 (2005, 29 de junho). Regulamento tecnico para fixacao dos padroes de identidade e qualidade para aguardente de cana e para cachaca. Diario Oficial da Uniao, Secao 1. 2005. 7p.
CARDOSO, D.R. et al. Influencia do material do destilador na composicao quimica das aguardentes de cana. Parte II. Quimica Nova, v.26, n.2, p.735-739, 2003. Available from: <http:// quimicanova.sbq.org.br/qn/qnol/2003/vol26n2/03.pdf>. Accessed: Nov. 26, 2011. doi: 10.1590/S0100 40422003000200004.
CECCATO-ANTONINI, S.R. Microbiologia da fermentacao alcoolica--a importancia do monitoramento microbiologico em destilarias. Sao Carlos: EDUFSCAR, 2010. 105p.
DATO, M.C.F. et al. Analysis of the secondary compounds produced by Saccharomyces cerevisiae and wild yeast strains during the production of "cachaca". Brazilian Journal of Microbiology, v.36, n.1, p.70-74, 2005. Available from: <http://www.scielo.br/ scielo.php?pid=S1517-83822005000100014&script=sci_arttext>. Accessed: Nov. 26, 2011. doi: 10.1590/S1517 83822005000100014.
FARIA, J.B. et al. Cachaca, pisco and tequila. In: LEA, A.G.H.; PIGGOT, J.R. (Eds.). Fermented beverage production. New York: Kluwer Academic, 2003. p.335-363.
FRANCO, A.C. Redestilacao da cachaca. 2008. 52f. Dissertacao (Mestrado em Nutricao)--Faculdade de Ciencias Farmaceuticas, Universidade Estadual Paulista, SP. Available from: <http://www.fcfar.unesp.br/posgraduacao/ alimentosenutricao/Disertacao/2008/ Alessandra_Franco_Completo.pdf.>. Acessed: Nov. 26, 2011.
INSTITUTO ADOLFO LUTZ. Normas Analiticas do Instituto Adolfo Lutz: metodos quimicos e fisicos para a analise de alimentos. Sao Paulo, 1985. V. 1, 533p.
MacFIE, H.J.H. et al. Designs to balance the effect of order of presentation and first-order carry-over effects in hall tests. Journal of Sensory Studies, v.4, p.129-148, 1989.
MORAIS, P.B. et al. Characterization and succession of yeast populations associated with spontaneous fermentations during the production of Brazilian sugar cane. World Journal of Microbiology and Biotechnology, v.13, p.241-243, 1997. Available from: <http://www.springerlink.com/content/ w312436412302123/fulltext.pdf>. Accessed: Nov. 26, 2011. doi: 10.1023/A:1018558302062.
NOVA, M.X.V. et al. Yeast species involved in artisanal cachaca fermentation in three stills with different technological levels in Pernambuco, Brazil. Food Microbiology, v.26, p.460-466, 2009. Available from: <http://www.sciencedirect.com/science/ article/pii/S0740002009000495>. Accessed: Nov. 26, 2011. doi: 10.1016/j.fm.2009.02.005.
OLIVEIRA, E.S. et al. The influence of different yeast fermentation, composition and sensory quality of cachaca. World Journal of Microbiology and Biotechnology, v.21, p.707-715, 2005. Avaliable from: <http://www.springerlink.com/ content/kj3hk7041801n772/fulltext.pdf>. Acessed: Nov. 26, 2011. doi: 10.1007/s11274-004-4490-4.
PARAZZI, C. et al. Avaliacao e caracterizacao dos principais compostos quimicos da aguardente de cana-de-acucar envelhecida em toneis de carvalho (Quercus sp.). Ciencia e Tecnologia de Alimentos, v.28, n.1, p.193-199, 2008. Available from: <http://www.scielo.br/pdf/cta/v28n1/27.pdf>. Accessed: Nov. 26, 2011. doi: 10.1590/S0101 20612008000100028.
PATARO, C. et al. Physiological characterization of yeasts isolated from traditional fermentation in a cachaca distillery. Revista de Microbiologia, v.29, p.69-73, 1998.
PATARO, C. et al. Yeast communities and genetic polymorphism of Saccharomyces cerevisiae strains associated with artisanal fermentations in Brazil. Journal of Applied Microbiology, v.88, p.24-31, 2000. Available from: < http:// onlinelibrary.wiley.com/doi/10.1046/j.13652672.2000.01092.x/pdf>. Accessed: Nov. 27, 2011. doi: 10.1046/j.1365-2672.2000.01092.x.
ROSA, C.A. et al. Cachaca production. In: INGLEDEW, W.M. (Ed.). The alcohol textbook. Nottingham: Nottingham University, 2009. p.484-497.
SILVA, C.L.C. et al. Selection, growth, and chemo-sensory evaluation of flocculent starter culture strains of Saccharomyces cerevisiae in the large-scale production of traditional Brazilian cachaca. International Journal of Food Microbiology, v.131, p.203-210, 2009a. Available from: <http:// www.sciencedirect.com/science/article/pii/ S0168160509001378>. Accessed: Nov. 27, 2011. doi: 10.1016/ j.ijfoodmicro.2009.02.027.
SILVA, P.H.A. et al. Avaliacao cromatografica de compostos volateis de cachacas produzidas com leveduras de diferentes procedencias. Ciencia e Tecnologia de Alimentos, v.29, n.1, p.100-106, 2009b. Available from: <http://www.scielo.br/ pdf/cta/v29n1/v29n1a16.pdf>. Accessed: Nov. 27, 2011.
Afra Vital Matos Dias Gabriel, Marta Regina Verruma-Bernardi, Luiz Antonio Correa Margarido, Maria Teresa Mendes Ribeiro Borges, Renata Tieko Nassu, Norberto Antonio Lavorenti (II) Sandra Regina Ceccato-Antonini (I,II) *
(I) Programa de Pos-graduacao em Agroecologia e Desenvolvimento Rural, Universidade Federal de Sao Carlos (UFSCar), Campus de Araras, Araras, SP, Brasil.
(II) Departamento de Tecnologia Agroindustrial e Socioeconomia Rural, UFSCar, Campus de Araras, 13600-970, Araras, SP, Brasil. E-mail: email@example.com. *Autor para correspondencia.
(III) Embrapa Pecuaria Sudeste, Sao Carlos, SP, Brasil.
Received 05.08.11 Approved 12.19.11 Returned by the author03.25.12 CR-5302
Table 1--Chemical analysis of cachacas produced with commercial and natural ferments. Commercial ferment Natural ferment Parameter Non-aged Aged Non-aged Aged Fixed acidity (1) 1.14a 6.76b 1.13a 9.39c Volatile acidity (1) 3.29a 4.89b 3.27a 3.63ab Total acidity (1) 4.43a 11.65b 4.40a 13.02b pH 6.44b 4.70a 6.92b 4.27a Alcohol ([degrees]GL) 42.0a 40.2b 42.0a 40.5ab Copper (2) 0.39a 0.31a 0.44a 0.37a Acetaldehyde (2) 6.42ab 10.02b 5.62a 8.85ab Methanol (2) 9.31a 10.70a 8.50a 10.23a Ethyl acetate (2) 8.35a 10.63a 6.34a 8.84a N-propanol (2) 40.58a 50.99b 37.64a 49.35b Isobutanol (2) 45.50a 50.52b 36.26a 48.92b Isoamyl alcohol (2) 131.01a 145.60a 120.83a 134.68a Higher alcohols (2) 217.08a 247.11b 194.73a 232.95b Total volatile compounds (2) 235.13ab 272.65b 209.97a 254.27b Different letters in the lines indicate significant difference (P<0.05 by Tukey's test). Results expressed (1) in mg 100[mL.sup.-1] acetic acid or (2) in mg 100[mL.sup.-1] anhydrous alcohol. Table 2 -Averages and ranks of sensory acceptability for aged and non-aged cachaca, produced with natural ferment and commercial ferment. Commercial ferment Attribute Non-aged Aged Colour 5.66 818.02a 5.16 653.77b Aroma 5.36 798.74a 4.91 669.71b Flavour 4.62 802.45a 4.06 662.89b Body 4.77 763.31a 4.47 692.67a Global impression 4.82 783.99a 4.31 652.31b Natural ferment Attribute Non-aged Aged Colour 5.62 807.80a 5.29 698.41b Aroma 5.36 766.34a 5.29 743.20ab Flavour 4.53 786.49a 4.31 726.17ab Body 4.76 762.95a 4.72 759.07a Global impression 4.77 780.50a 4.69 761.21a Different letters in the lines for ranks indicate significant difference (P<0.05 by Kruskall-Wallis non-parametric test); n=62 judges (1="dislike much" to 7="like much").