PHYSICOCHEMICAL EVALUATION OF THE ORGANOLEPTIC QUALITY OF FLAVORED VODKAS

Józef Mazurkiewicz¹, Hanna M. Baranowska², Bohdan Achremowicz³, Piotr Tomasik^4
1 Department of Physics, University of Agriculture in Cracow, Poland
² Department of Physics, University of Life Sciences in Poznań, Poland
³ Department of Carbohydrate Technology, University of Agriculture in Cracow, Poland
⁴ Nantes Nanotechnological Systems, Bolesławiec, Poland

ABSTRACT

Changes of viscosity of 40% aqueous ethanol (vodka) flavored with commercially available essences of vanilla, lemon, orange, rum, and cream applied at concentration of 0.1 and 0.5 w% had no effect upon the macrostructure of vodka. These admixtures changed solely the flavor of that drink. Such result was confirmed by changes of the NMR spin-lattice, T₁, and spin-spin, T₂₁ and T₂₂ relaxation times and computer simulations of the energy of interactions of the molecules of particular essences in aqueous solutions of ethanol of varying concentration.

Key words: cream essence, computer simulations, lemon essence, mouth feel, orange essence, relaxation times, rum essence, vanilla essence, viscosity.

INTRODUCTION

According to the mouthfeel theory [2,3,4,5,9], taste depends on the organization of the macrostructure of the substance causing the sensory impression. More viscous liquids, that is, these with more ordered solution macrostructure should offer better taste. Therefore, in our former paper [8] based on rheological and nuclear magnetic studies we presented evidences that vodka has superior taste at about 40% (80 proof) aq. ethanol as such solutions had the highest viscosity. On the market there are also liqueurs being infusions of herbs and/or fruits and even propolis [10] in alcohol, cordials and meads and commonly these products are sweetened. Some consumers like flavored vodkas being 40% ethanol. Flavoring additives are either natural or synthetic. In this paper we checked how addition of commercial natural rum, vanilla, lemon, orange and cream essential oils and one synthetic vanilla essence affected macrostructure of 40% aqueous ethanol. The studies involved measurements of viscosity and nuclear magnetic resonance relaxation times of protons. The experimental results were backed by computer simulations of energy of intermolecular interactions.

MATERIAL AND METHODS

Materials. 96% ethanol, analytical grade was purchased from POCh Gliwice, Poland. Water was redistilled. Essences: lemon, lemon L, orange, orange L, cream, cream concentrated, cream concentrated B, rum, rum C, vanilla, vanilla D, and synthetic vanilla were kindly offered by the Jaskulski Aromaty "JAR" Enterprise, Warsaw, Poland. They were admixed to 40% vodka on the level of 0.1 and 0.5 weight%.

Viscosimetric measurements. Measurements for aqueous solutions of 15–70% vodka and 40% vodka prior and after flavoring were carried out at 20 ± 0.05°C with a fully computerised Zimm-Crothers rotary viscometer [6].

Relaxation time measurements. Relaxation spin-lattice, T₁, and spin-spin, T₂, times for 0.2 mL samples of vodka prior and after flavoring were recorded at 25 ± 1°C with NMR PS15T (ELLAB, Poznań, Polska) spectrometer operating at 15 MHz. T₁ were measured involving inversion-recovery sequence whereas T₂ were available from the sequence of CPMG pulses. The measured systems were characterized with single T₁ relaxation time and two components of T₂ relaxation time.

Computations. Model for computations was arranged in the following manner. In consecutive calculations, a theoretical 20×20×20Å cells filled with 0, 10, 20, 30, 40, and 50 ethanol molecules. Water molecules were added to those cells to fill these cells completely. The MM+ Monte Carlo approach was applied (HyperChem 7.5 with Amber 99 software). After admixing five molecules of the additive (benzene cyclohexane, 1,4-dioxane, methanol, 1-propanol, xylitol, D-glucose, ethyl methanoate, limonene, citral A and vanillin) into aqueous ethanol the system was agitated involving Molecular Dynamics followed by computation of the optimization energy for the whole system without molecules of additive and for molecules of additives. That approach provided energy difference for isolated and dissolved molecules of additives. The results of the lowest optimization energy were selected. The total number of molecules in the cells varied from 265 in case of pure water up to 153 in case of the most concentrated, 33.3 mol/mol% aqueous ethanol. There were 500 calculating steps. In such manner, an average potential energy became available for the systems under study. Computations were performed with two-processor 2.5 GHZ computer.

RESULTS AND DISCUSSION

Fig. 1 presents viscosity of 15, 25, 30, 40, and 70% aqueous solutions of ethanol and effect of admixture of 0.5 w% vanillin D and lemon essences to them. One can see that only in case of 15 and 25% ethanol admixture of lemon essence produced a subtle decrease in viscosity.

Certainly, limonene (I), an cycloalkene, is a dominating component of lemon essence. Dominating components of vanilla is vanilin (II), an aromatic phenoloaldehyde, and in orange essence dominate two isomeric aliphatic aldehydes: neral (III) and citral A (geranial) (IV). In the rum essence several esters of fatty acids with ethyl butanoate (V) are responsible for rum aroma.

Thus, observed effect of the lemon essence upon viscosity of diluted solutions of ethanol could be produced by a deficiency of ethanol for full solvation of hydrophobic/lipophilic limonene. In remained cases admixture of either limonene or vanillin did not reduce the original viscosity. Hydrophilic vanillin could be solvated by both components of aqueous ethanol. As an aldehyde it could also form hemiacetal and hydrate, both residing in the solution in an equilibrium.

Fig. 2 presents changes in the viscosity taken at 20±0.1°C for 40% ethanol (vodka) prior and after flavoring with addition of 0.1 and 0.5 w% of particular essences. One can see that none of admixed essences reduced viscosity of vodka but an increase in the viscosity was negligible although positively dependent on the level of admixture. The viscosity rose with increasing concentration of admixture. Among applied essences at 0.1% concentration the lemon L aroma provided the most remarkable increase in the viscosity whereas at concentration of 0.5 w% the vanilla D essence the most efficiently contributed to the organization of the macrostructure of vodka. Both, lemon and orange essences provided the least increase in the viscosity. Likely, differences in the reaction of the vodka to the essence could be related to the different number of hydrophilic and lipophilic centers in admixed essences influencing the organization of the macrostructure of the solutions.

Extended studies of the effect of concentration were carried out for vanilla and lemon essences. Concentrations in both cases increased up to 1.5%. Viscosity of vodka flavored with vanilla essence almost linearly increased with concentration whereas increase in the concentration of the lemon essence had no effect upon the viscosity of vodka (Fig. 3).

Analysis of the results presented in Figs. 2 and 3 suggested that the role of interaction of ethanol and water with the components of the essences at low concentration were unessential. As the concentration of the admixed essences rose, such role of interactions seemed to increase. Increasing dose of limonen, an alkene, had no influence upon viscosity of vodka. Because neral and citral as well as ester groups of the components of the rum essence produced relatively slight increase in the viscosity of vodka one could anticipate that the phenolic hydroxyl group in vanilin provided major ordering effects to the macrostructure of vodka.

The NMR spin-lattice, T₁, and spin-spin, T₂, relaxation times provided more information on factors influencing the viscosity of flavored vodka.

Commonly, T₁ is reciprocally proportional to viscosity as an increase in viscosity accelerated strained relaxation transitions on the microscopic level [1]. All essences admixed to 40% vodka increased initial viscosity and, indeed, on increase in the concentration of admixed essences all of them but synthetic vanilla essence induced decrease in the spin-lattice relaxation times, T₁. In the latter case that increase could hardly be recorded (Table 1).

The spin-spin relaxation offered another way of the transfer of excessive energy absorbed. In 40% vodka two components of the spin-spin relaxation time, T₂ could be noted. These two components, a short one, T₂₁ and a long one, T₂₂, were also observed in all flavored vodkas. The short component was within the range of the measurement error and, therefore, one could assume that it was identical for all solutions. The second component varied from one admixture into another but always decreased with an increase in the concentration of the additive.

Because viscosity decreased with increase in temperature, T₁ increased with temperature (Table 2) and that increase was fairly remarkable. Also T₂₂ increased with temperature but not so remarkably as T₁. In contrast to T₂₂, T₂₁ slightly decreased under such circumstances and such result is beyond any anticipation. That decrease could, eventually, be accounted for volume effects.

Temperature studies provided estimation of the activation energy, ΔE, based on the spin-lattice relaxation times assuming – Equation 1.

(1)

where rate of spin lattice relaxation R = 1/T₁ [6]. Some values are presented in Table 3. The activation energy could be interpreted as the energetic rotational barrier of the molecules of essences.

One could see that activation energies were only slightly dependent on the type and concentration of flavoring essences. Likely, at the concentrations applied, there were rather unessential intermolecular interactions in the solutions and increase in the concentration of the essence solvation of the essence water and ethanol molecules did not change on any essential way.

Numerical simulations were performed for a wide variety of solutes (Fig. 4). Fig. 4 presents relationship between changes of energy of interaction of the molecules of additives with molecules of the ethanol solution at varying concentration of the solutions.

One could see that vanilin, citral and limonene relatively strongly interacted with the solvent and that interaction turned stronger as the concentration of ethanol in solution increased. That result fitted experimental results. These interactions appeared as strong as these in case of xylitol and 1,4-dioxane [7] and stronger than interactions of D-glucose, methanol, 1-propanol, and ethyl methanoate.

Fig. 5 presents interaction energy per one molecule of the solution. Changes of energy are stronger in 20 to 30 mole/mol% ethanol solutions regardless on whether an essence was dissolved in it or ethanol solutions free of admixtures were considered.

Changes of viscosity depended not only on the energy of interactions but also on molecular mass of the additive (as disaccharides, sucrose and maltose had the highest molecular mass) and its either hydrophilic or hydrophobic character. One could see that energy of interactions begun to decrease in about 17 mol/mol% aqueous ethanol in order to reach minimum in about 23 mol/mol% ethanol. This result fitted our former estimations for aqueous ethanol [8]. One could see from Fig. 5 that only sucrose and maltose provided slightly higher energy that that calculated for aqueous ethanol without any admixture. D-Glucose and xylitol did not introduce any essential change of energy whereas all other admixtures slightly decreased the energy and the effect of those admixtures was practically like one another. An increase in the interaction energy beginning from 25 mol/mol% ethanol could result from changes in packing molecules in solution.

CONCLUSIONS

All experiments and computer simulations lead to conclusion that flavoring additives applied added flavor to vodka without affecting its mouthfeel.

ACKNOWLEDGMENT

Authors feel very much indebted to Jaskulski Aromaty JAR Enterprise for providing aromas used in this research.

REFERENCES

1. de Certaines, J.D., Bovee, W.M.M.J., Podo, F., 1992. Magnetic resonance spectroscopy in biology and medicine. Pergamon Press Ltd. New York.

2. Ferry A.-L., Hort, J., Mitchell J.R., Valles Pamies B., Lagarrigue S., 2006. Viscosity and flavour perception. Why is starch different from hydrocolloids. Food Hydrocoll., 20, 855-862.

3. Hill M.A., Mitchell J.R., Sherman P., 1996. The relationship between the rheological and sensory properties of a lemon pie filling. J Text. Stud., 26, 457-470.

4. Kokini J.L., Kadane J.B., Cussler E.L., 1977. Liquid texture perceived in the mouth. J. Text. Stud., 8, 195-218.

5. Lai V.M.-F., Lii, C.Y., 2004. Role of saccharides in texturization and functional properties of foodstuffs. [In] P. Tomasik (Ed.), Chemical and functional properties of food saccharide. Boca Raton, CRC Press, 159-179.

6. Mazurkiewicz J., Tomasik P., 1996. Perestroika effect. A novel example of electroviscosity. Bull. Chem. Soc. Belg., 105, 173-180.

7. Mazurkiewicz, J., Tomasik, P., 2006. Why 1,4-dioxane is a water structure breaker? J. Mol. Liq., 126, 111-116.

8. Mazurkiewicz J., Baranowska H.M., Wojtasik, M., Tomasik P., 2007. Macrostructure of aqueous solutions of ethanol and its implications. EJPAU, Agricultural Engineering, 10, 2, www.ejpau.media.pl.

9. Szczesniak A., Farkas E., 1962. Objective characterization of the mouthfeel of gum solutions. J. Food Sci., 27, 381-385.

10. Tuszyński, T., Tarko, T., Sposób otrzymywania wódek gatunkowych aromatyzowanych propolisem [The way to obtain vodkas flavored with propolis]. Polish Patent PL 193906 B1 (2007).

Accepted for print: 8.07.2009

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.