Investigating the Variation of Volatile Compound Composition in Maotai-Flavoured Liquor During Its Multiple Fermentation Steps Using Statistical Methods

Abstract

The use of multiple fermentations is one of the most specific characteristics of Maotai-flavoured liquor production. In this research, the variation of volatile composition of Maotai-flavoured liquor during its multiple fermentations is investigated using statistical approaches. Cluster analysis shows that the obtained samples are grouped mainly according to the fermentation steps rather than the distillery they originate from, and the samples from the first two fermentation steps show the greatest difference, suggesting that multiple fermentation and distillation steps result in the end in similar volatile composition of the liquor. Back-propagation neural network (BNN) models were developed that satisfactorily predict the number of fermentation steps and the organoleptic evaluation scores of liquor samples from their volatile compositions. Mean impact value (MIV) analysis shows that ethyl lactate, furfural and some high-boiling-point acids play important roles, while pyrazine contributes much less to the improvement of the flavour and taste of Maotai-flavoured liquor during its production. This study contributes to further understanding of the mechanisms of Maotai-flavoured liquor production.

Headnote

Summary

The use of multiple fermentations is one of the most specific characteristics of Maotai-flavoured liquor production. In this research, the variation of volatile composition of Maotai-flavoured liquor during its multiple fermentations is investigated using statistical approaches. Cluster analysis shows that the obtained samples are grouped mainly according to the fermentation steps rather than the distillery they originate from, and the samples from the first two fermentation steps show the greatest difference, suggesting that multiple fermentation and distillation steps result in the end in similar volatile composition of the liquor. Back-propagation neural network (BNN) models were developed that satisfactorily predict the number of fermentation steps and the organoleptic evaluation scores of liquor samples from their volatile compositions. Mean impact value (MIV) analysis shows that ethyl lactate, furfural and some high-boiling-point acids play important roles, while pyrazine contributes much less to the improvement of the flavour and taste of Maotai-flavoured liquor during its production. This study contributes to further understanding of the mechanisms of Maotai-flavoured liquor production.

Key words: Maotai-flavoured liquor, multiple fermentations, volatile compounds, statistical analysis, back-propagation neural network

Introduction

Maotai-flavoured liquor, generally described as a highly complex-flavoured, sweet and refreshing soy sauce aroma style alcoholic drink, is one of the most popular and representative liquors in China. The formation of the special flavour of Maotai-flavoured liquor can be largely attributed to its unique and complicated production techniques. The process of Maotai-flavoured liquor production differs from those of other liquors in many aspects including starter preparation, grain (mainly sorghum and wheat) piling and liquor distillation. Briefly, the production of Maotai-flavoured liquor consists in nine fermentation steps and the whole process lasts almost a year. Each fermentation step includes starter addition, piling (putting the mixture of cooked grains and starter powder on the ground, making it into a small hill, and then undergoing fermentation for 4-5 days), fermentation in a pit and distillation. After each fermentation step, the fermented mixture is distilled, the liquor is collected, and the fermented grains are used as the material for the next step. The liquor from the first two fermentations, due to its coarse taste, is poured back on the piled mixture, while the liquor from the other seven fermentations is stored separately for further blending to form the final product.

Much effort has been made in recent years to find the complicated flavour composition of Maotai-flavoured liquor, how it is formed and changes during the brewing process. The research includes microorganism composition analysis (1,2), the isolation and characterization of functional strains (3), flavour component determination (4,5), and the analysis of the relationships between the microorganisms and flavour compounds (6,7). However, the utilization of multiple fermentation and distillation steps, a very special technique for Maotai- fl avoured liquor manufacturing, has not received enough academic attention so far.

The flavour of this Chinese liquor is rather complex and it is generally presented in the form of numerous gas chromatography (GC) or gas chromatography-mass spectrometry (GC-MS) data, for which statistical approach is a necessity to process and analyze the data. Cluster analysis (CA), discriminant analysis (DA), principal component analysis (PCA) and partial least square (PLS) regression have been widely adopted statistical methods in recent years. All these approaches have shown good performance in many cases of liquor flavour research, such as spectral analysis (8), artificial nose (9), liquor discrimination and identification (10-12). However, most of these methods are linear in nature, thus may not be capable of describing non-linear systems satisfactorily. As a promising alternative, artificial neural network (ANN) has a lot of advantages in parallel processing, classification, learning and pattern recognition. ANN has also been successfully used in researching the productions of wine and beer, such as prediction of process problems (13), sensory evaluation (14) and process optimization (15). As far as we know, ANN has seldom been applied in flavour research of traditional Chinese liquor, especially Maotaiflavoured liquor, whose flavour composition is regarded to be the most complicated among Chinese liquors.

The aim of this work, therefore, is to analyze the variations of the flavour composition of Maotai-flavoured liquor during its multiple fermentation process with statistical approaches, and to provide useful information for better understanding of the formation of its flavour style.

Material s and Methods

Liquor samples

The raw liquor samples were collected from nine distilleries of Langjiu Group Co., Ltd., Sichuan, PR China. Raw liquor was sampled after each of the seven fermentation and distillation steps. A total of 63 liquor samples were used for flavour compound analysis.

Analytical methods

Gas chromatography (GC) analyses of liquor samples were performed on an Agilent 7890A gas chromatograph (Agilent Technologies Co. Ltd., Santa Clara, CA, USA) equipped with automatic sampler and flame ionization detector. Samples were analyzed on a CP-Wax 57 CB column (50 m×0.25 mm×0.2 μm). The injector, detector and column temperatures were set at 125, 120 and 90 °C, respectively. The carrier gas was N2. The flow rates of N2, H2 and air were set at 20, 20 and 230 mL/min, respectively. A total of 68 flavour compounds were determined by comparing their peak areas to those of the standards. All chemicals used in the analyses were of chromatographic grade.

Organoleptic evaluation

Organoleptic evaluation of liquor samples was conducted according to a literature method (16). All liquor samples were evaluated by ten tasters, and the average score of the flavour and taste of each sample was calculated.

Data analysis

Student-Newman-Keuls test, correlation analysis and principal component analysis (PCA) were performed using SAS v. 8.1 software (17). Cluster analysis, neural network model development and the calculation of each input neuron mean impact value (MIV) were carried out using MATLAB v. 7.1 software (18).

Cluster analysis was used to group the liquor samples according to Euclidean distances between the samples based on their volatile compound compositions. PCA was applied to reduce the dimensionality of the original data matrix and allow the visualization of liquor samples with different origins in a lower dimensional space.

Back-propagation neural network (BNN) models were established to predict fermentation steps and organoleptic evaluation scores of liquor samples based on their volatile compound compositions. The architecture of the neural network consisted of an input layer, a hidden layer and an output layer. The input nodes were the concentrations of the 68 volatile compounds of a liquor sample or corresponding principal components (PCs). The output is the fermentation step or the organoleptic evaluation score of the liquor sample. The number of the nodes in the hidden layer was selected according to the following equation:

n=n1+ n2+a /1/

where n is the number of the nodes in the hidden layer, n1 is the number of the input nodes, n2 is the number of the output nodes, and a is a constant between 0-10.

The total dataset of 63 liquor samples was randomly split into two subdatasets, 48 samples for training and 15 samples for testing. The input variables for training and testing were standardized by using 'prestd' and 'trastd' functions, respectively, while the output variables were postprocessed by using a 'poststd' function in the neural network toolbox of MATLAB v. 7.1. For fermentation step prediction, the output variable was rounded to the nearest integral number. Bayesian regularization was adopted in training the neural network to avoid overtraining, and this was realized by using a 'trainbr ' function in the neural network toolbox of MATLAB v. 7.1. After BNN models with satisfactory predictive ability were established, MIVs were calculated to screen the most influential volatile compounds (19).

Results and Discussion

Variations of volatile compound composition of liquor samples from different fermentation steps

The average concentrations of the volatile compounds in liquor samples from different fermentation steps in the nine distilleries are listed in Table 1. It can be seen that most of the volatile compounds underwent significant concentration change during the multiple fermentation steps. Some components, like acetic acid, ethyl acetate, 2,3-butanedione, 2-butanol and n-propanol, decreased dramatically after one or two fermentations, while other components, like ethyl caproate, ethyl lactate, ethyl palmitate, ethyl oleate and ethyl linoleate, increased gradually dur-ing the whole process. Organoleptic evaluation showed that the liquor after 3-6 fermentations had softer and sweeter flavour compared to those after fermentations 1 and 2. This implies that some of the components that decreased significantly during the multiple fermentations very possibly stimulate flavour and taste of the liquor.

Correlation analysis suggests close relationships among some compounds (Table 2). These relationships can be partly explained by the sharing of common metabolic pathways or enzymes used for the formation of different compounds (acetaldehyde and acetal, n-propanal and n-propanol, 2-butanone and 2-butanol, for example). However, the close relationships among some other components (ethyl acetate and methanol, for example) are not fully understood and thus need further research in the future.

Cluster and principal component analyses of liquor samples after different fermentation steps

Cluster analysis was performed to find similarities among liquor samples after different fermentation steps and distilleries based on their volatile compound compositions. The results show that except for several samples from the first two fermentations, most samples from a same fermentation are clustered together (details not shown), suggesting that the fermentation step plays a more important role in forming the liquor style than the distillery where it is produced.

Comparing the samples from the same distillery, those after 3 to 7 fermentations are similar to each other, while those after the first two rounds, especially the first one, show much difference (Fig. 1). This means that although there may be considerable differences in the beginning, multiple fermentations and distillations lead to similar flavour compound composition of the liquor after several steps.

In order to reduce the data dimensionality and visualize different liquor samples in a lower dimensional space, PCA was conducted on the data matrix of 63 liquor samples×68 volatile compounds. The results reveal that the first ten principal components (PCs) extracted are needed to account for 86 % of the total variance in the data matrix. The first three PCs (PC1, PC2 and PC3), however, explain only 15.8, 14.1 and 9.4 % of the total variance, respectively. The three-dimensional plot of the PCA (Fig. 2) shows that liquor samples taken after fermentations 1 and 7 can be separated appropriately based on these three PCs, while other samples, especially those taken after fermentations 3-5, are very closely located. Samples from the same fermentation step from different distilleries also failed to be separated satisfactorily from each other (details not shown). On the whole, the PCA here does not provide much insight for understanding the differences among the liquor samples.

Developing BNN models for predicting the number of fermentation steps and organoleptic evaluation score of liquor samples and variable screening

BNN models were developed to predict the number of fermentation steps and organoleptic evaluation scores of the liquor samples based on their volatile compositions and PCs, respectively. After a trial of the topological structure, it was found that BNN models with nine nodes in the hidden layer could provide satisfactory prediction when 68 volatile compound concentrations were used as inputs, while six nodes in the hidden layer were appropriate when ten PCs were used as inputs. Some representative predictions in the test are shown in Fig. 3. The accuracy of brewing round prediction in the test was between 80 and 100 % when 68 volatile compound concentrations were used as inputs, while 60-90 % of accuracy was obtained with ten PC inputs. For organoleptic evaluation score prediction, BNN model using 68 volatile compound concentrations as inputs also had better performance (R2 value between 0.80 and 0.95) than those using the ten PCs as inputs (R2 value between 0.70 and 0.90). This result was not entirely unexpected since the ten PCs account for only 86 % of the data variance and the linear PCA may lose some non-linear information of the investigated system.

As the BNN model with 68 volatile compound concentrations as inputs represents well the relationship between the volatile compositions and organoleptic evaluation results of the liquor samples, mean impact value (MIV) analysis was adopted to find which volatile compounds play more important roles in forming the liquor flavour style (Table 3). High MIVs of many alcohols and esters were observed. However, most of these MIVs are negative, suggesting that high concentration of these compounds may degrade the flavour and taste of the liquor. Noticeable volatile compounds that showed positive and relatively high MIVs are ethyl lactate, furfural and several acids including valeric acid, heptanoic acid, isobutyric acid and nonanoic acid. This implies that these compounds contribute greatly to the formation of the flavour of the liquor. The concentrations of almost all these compounds increased with the number of fermenations (Table 1), suggesting multiple fermentations are vital in forming the liquor flavour.

Major flavour components in Maotai-flavoured liquor have been discussed extensively in recent years but no consistent opinion has been obtained so far (20,21). Furfural (22), acids with high boiling point (23), and pyrazines (24) have all been suggested to be the major flavour compounds in Maotai-flavoured liquor. Our results show that ethyl lactate, furfural and some acids with high boiling points do play important roles in forming the liquor style. Pyrazines, however, seem to contribute less or even negatively according to their MIVs in Table 3. This is supported by a previous report where the concentration of pyrazines varied significantly in different Maotai-flavoured liquor samples (20). However, as the analyses here are based exclusively on Maotai-flavoured liquor samples, we cannot assert that the components with moderate or low MIVs are unimportant or even unnecessary in form-ing the liquor flavour. Undoubtedly, further elucidation of major flavour components in Maotai-flavoured liquor requires more samples with different flavour characteristics.

Conclusion

The results of this research show that fermentation steps exert much more influence on the volatile composition of Maotai-flavoured liquor than the distillery. Although there may be considerable differences in the volatile com- position among the liquor samples at the beginning, multiple fermentations and distillations ultimately lead to similar volatile composition of the liquor. Based on the statistical analyses, we suggest that ethyl lactate, furfural and some high-boiling-point acids make relatively high contribution in forming the special flavour of Maotai-flavoured liquor.

Acknowledgements

This work was financially supported by science and technology support project of Science and Technology Department of Sichuan province (2011GZ0020), PR China, and open fund for liquor making biotechnology and application key laboratory of Sichuan province, PR China (2012LF3010).

References

References

1 . Liu X, Guo KL, Zhang HX. Determination of microbial diversity in Daqu, a fermentation starter culture of Maotai liquor, using nested PCR-denaturing gradient gel electrophoresis. World J Microbiol Biotechnol. 2012;28:2375-81. http://dx.doi.org/10.1007/s11274-012-1045-y

2. Wang CL, Shi DJ, Gong GL. Microorganisms in Daqu: a starter culture of Chinese Maotai-flavor liquor. World J Microbiol Biotechnol. 2008;24:2183-90. http://dx.doi.org/10.1007/s11274-008-9728-0

3. Wu Q, Xu Y, Chen L. Diversity of yeast species during fermentative process contributing to Chinese Maotai-flavour liquor making. Lett Appl Microbiol. 2012;55:301-7. http://dx.doi.org/10.1111/j.1472-765X.2012.03294.x

4. Zhu SK, Lu X, Ji KL, Guo KL, Li YL, Wu CY, Xu GW. Characterization of flavour compounds in Chinese liquor Moutai by comprehensive two-dimensional gas chromatography/ time-of-flight mass spectrometry. Anal Chim Acta. 2007;597: 340-8. http://dx.doi.org/10.1016/j.aca.2007.07.007

5. Peng QR, Dong R, Xun SY, Yang M, Feng Y, Sun D, Geng PL. Determination of volatile phenols in Chinese liquors by high-performance liquid chromatography associated with β-cyclodextrin and a protective barrier layer. Flavour Fragr J. 2013;28:137-43. http://dx.doi.org/10.1002/ffj.3139

6. Huang YG, Wu Q, Xu Y. Isolation and identification of a black Aspergillus strain and the effect of its novel protease on the aroma of Moutai-flavoured liquor. J Inst Brew. 2014; 120:268-76. http://dx.doi.org/10.1002/jib.135

7. Zhang R, Wu Q, Xu Y. Aroma characteristics of Moutai-flavour liquor produced with Bacillus licheniformis by solid-state fermentation. Lett Appl Microbiol. 2013;57:11-8. http://dx.doi.org/10.1111/lam.12087

8. Jiang A, Peng JT, Peng SL, Wei JP, Li CW. Analysis of liquor flavour spectra and pattern recognition computation. Guang Pu Xue Yu Guang Pu Fen Xi. 2010;30:920-3 (in Chinese). http://dx.doi.org/10.3964/j.issn.1000-0593(2010)04-0920-04

9. Ya Z, He K, Lu ZM, Yi B, Hou CJ, Shan S, et al. Colorimetric artificial nose for baij iu identification. Flavour Fragr J. 2012; 27:165-70. http://dx.doi.org/10.1002/ffj.3081

10 . Fan WL, Shen HY, Xu Y. Quanti fi cation of volatile compounds in Chinese soy sauce aroma type liquor by stir bar sorptive extraction and gas chromatography-mass spectrometry. J Sci Food Agric. 2011;91:1187-98. http://dx.doi.org/10.1002/jsfa.4294

11. Van-Diep L, Zheng X, Ma K, Chen J, Han B, Nout MJR. Characterization of Fen-Daqu through multivariate statistical analysis of H-1 NMR spectroscopic data. J Inst Brew. 2011; 117:516-22. http://dx.doi.org/10.1002/j.2050-0416.2011.tb00499.x

12 . Zheng J, Liang R, Wu CD, Zhou RQ, Xiao XP. Discrimination of different kinds of Luzhou-flavour raw liquors based on their volatile features. Food Res Int. 2014;56:77-84. http://dx.doi.org/10.1016/j.foodres.2013.12.011

13 . Román RC, Hernández OG, Urtubia UA. Prediction of problematic wine fermentations using artificial neural networks. Bioprocess Biosyst Eng. 2011;34:1057-65. http://dx.doi.org/10.1007/s00449-011-0557-4

14 . Dong JJ, Li QL, Yin H, Zhong C, Hao JG, Yang PF, et al . Predictive analysis of beer quality by correlating sensory evaluation with higher alcohol and ester production using multivariate statistics methods. Food Chem. 2014;161:376-82. http://dx.doi.org/10.1016/j.foodchem.2014.04.006

15. Xi J, Xue YJ, Xu YX, Shen YH. Artificial neural network modeling and optimization of ultrahigh pressure extraction of green tea polyphenols. Food Chem. 2013;141:320-6. http://dx.doi.org/10.1016/j.foodchem.2013.02.084

16 . Lai DY. A study of the relationship between the quality of distilled liquor and the acetaldehyde and acetal. Liquor Making. 2001;28:53-5 (in Chinese).

17 . SAS so ft ware, v. 8.1, SAS Institute Inc, Cary, NC, USA; 2000.

18. MATLAB, v. 7.1, Mathworks Inc., Natick, MA, USA; 2005.

19 . Mukherjee A, Deshpande JM. Application of arti fi cial neural networks in structural design expert systems. Comput Struct. 1995;54:367-75. http://dx.doi.org/10.1016/0045-7949(94)00342-Z

20. Li X, Fang SL, Liu C, Li RL, Chen ML. Research progress of the main flavour substances in Maotai-flavor. Liquor Making. 2012;39:19-22 (in Chinese).

21 . Liu JW, Jiang YL, Shen Y, Liu YC, Luo AM. Current research of the formation of special flavour substances in Maotai-flavor liquor. Liquor-Making Sci Technol. 2013;227:85-9 (in Chinese).

22. Hu GD, Cheng JS. Investigating the compounds of furanone and pyrone in Maotai-flavor. Liquor-Making Sci Technol. 1997;82:31-4 (in Chinese).

23. Fan WL, Xu Y. The review of the research of aroma compounds in Chinese liquors. Liquor Making. 2007;34:31-7 (in Chinese).

24. Guo C, Ouyang P, Xie H. Analysis of volatile compounds produced by Bacillus subtilis E20 with GC/MS technique. China Brew. 2010;222:153-5 (in Chinese).