BG108556A - One-side mechanical safe lock - Google Patents

Abstract

The invention finds application as an additional locking mechanism for residential and public buildings or as independent lock for small and medium-sized safes, wherein its latch directly locks or blocks the more robust locking mechanisms. It comprises a blocking mechanism with a plastic body (9) wherein a square-shaped duct (10) is machined where the latch (18) moves, round holes (12) for springs (15) and the heads of buttons (17) and lateral slots (11) wherein the blocking plates (16) are moving, which are meshed in ducts (19) of latch (18) and are blocking it. A gear of a driving mechanism is meshed in the ducts (19) at the head of latch (18). In the gear (21), a metal shaft (23) is axially fitted with a nest for a shot (26) lying, together with a spring (27) in a radial hole (22) of gear (21). Shot (26), spring (27) and bracket (28) of engaging and pressing form a connection device. The metal shaft (23) is connected to a button (25) by means of an extension tube (24). When button (25) is revolved for driving the latch (18) (unlocking), shot (26) is being raised, wherein the metal shaft (23), tube (24) and button (25) revolve at empty runs. Only after original card (29) with perforations (30) in the blocking mechanism is fitted, buttons (17) are pressed to such depth, that when button (25) revolves, latch (18) is released. This combination of locking and driving mechanism connected and interdependently depending on latch (18) can secure exceptional safety of the lock.

Description

The present invention relates to the preparation of alcoholic beverages, in particular beer, using in particular combined processes involving continuous and intermittent fermentation process steps.

BACKGROUND OF THE INVENTION

The large number of new publications in this area illustrates the great interest of the brewing industry in immobilization. In several reviews (Enari, 1995; Iserentant, 1995; Masschelein, 1997; Mensouret al., 1997; Stewart, 1996; Virkajarvi & Linko, 1999), the general state of the brewing industry has highlighted the possible revolutionary role of immobilization in beer production. In addition to the groups mentioned in Sections 3.1 to 3.5, many other institutions are involved in the application of immobilized R&D cells in brewing beer. The Miller Brewing Company (Duncombe et al., 1996; Tata et al., 1999) of the United States conducts some preliminary evaluations of the Meura Delta test rig and of the fluidized bed bioreactor with distributor Schott Engineering. Coors Brewery also performs preliminary experiments with the Meura Delta system.

The Slovak Technical University in collaboration with Heineken is

studied the use of calcium pectate gel in an air lift system for beer production (Domeny, 1996). The Slovak Technical University Group has recently published several articles on current research (Smogrovicova et al., 1997; Smogrovicova & Domeny,

1999). Guinness initially investigated the use of various absorption media for immobilization and subsequent fermentation in a fluidized bed bioreactor (Donnelly, 1998). In 1999, Donnelly and colleagues published a report describing the kinetics of sugar metabolism in a fluidized bed bioreactor (Donnely et al., 1999). Their experimental scheme involves the use of Siran porous glass granules as an immobilizer for fermenting yeast. Guinness has become part of the Immocon Consortium, which is described in Section 2.2.5.

Holsten Brauerei AG and Lurgi AG, Germany, have jointly developed and commissioned a pilot plant for the continuous production of alcohol-free beer (Dziondziak, 1995). Calcium alginate granules were used in a fluidized bed fermenter (130 L) in a one-stage loop, whereby a column with lattice bottoms (7 lattice bottoms) was used to de-alcoholize the beer. The total production time according to this process is 8.5 hours.

A group from Sapporo Breweries Ltd., a Beer Research Laboratory located in Japan, examines the use of immobilized cells in a fluidized bed reactor for main beer fermentation. Their studies include the use of polyvinyl alcohol gel granules (Shindo & Kamimur, 1990), calcium alginate granules (Shindo etal., 1994a), two-layer gel filaments (Shindo etal., 1994) and chitosan gel granules (Shindo etal., 1994). ., 1994c) as immobilizing matrices. In a recent study, a 1 liter bioreactor containing 25% by volume Chitopearle® Type II granules (chitosan granules) was operated in continuous mode with non-fermented glucoamylase treated beer. This enzyme treatment ensures the formation of acetate ester in the

immobilized cell to resemble that of conventional batch fermentation and thus closer to the sample taken.

The Sapporo research team is currently focusing its attention on the study of chitosan granules used in a recurrent batch fluidizer. The system operated for 75 days without major problems and the beer obtained was similar in quality to a commercial product. A non-flocculating Anon strain has been shown to be much more effective than a flocculating strain (Maeba et al., 2000; Umemoto et al., 1998).

Two other diacetyl-reducing approaches were presented at the EMC congress held in Mannstreet in 1997. Researchers from France (Dulieu et al., 1997) envisage the use of decarboxylase encapsulated α-acetolactate to quickly convert acetolactate to acetone . Meura Delta presents preliminary results from the use of alumosilicate zeolite as a catalyst for the direct conversion of cold acetolacate to acetone (Andries et al., 1997). If such treatment proves effective and acceptable to the consumer, alternatives to the lower cost of the aging systems offered by Cuitor and Alfa Laval could become a reality.

Another experimental study in the field of beer immobilization is that of the Institute of Standards and Industrial Research of Singapore, which examined the use of filamentous alginate gel particles in a compacted reactor and found it more advantageous in comparison with alginate granules (Que, 1993).

Mafra and colleagues at Universidade do Minho from Portugal have discussed the use of superflocculent yeast strain for

continuous brewing of beer (Mafra et al., 1997). This research group has also published work related to the use of their flocculating yeast in an air lift (airlift type) bioreactor for ethanol production (Vicente et al., 1999; Domingues et al., 2000).

Recently, studies at various academic institutions have also been linked to the use of immobilized cells for beer production (Argiriou et al., 1996; Bardi et al., 1996; Cashin, 1996; Moll & Duteurtre, 1996; Nedovic et al., 1996a ; Nedovic et al., 1996b; Norton et al., 1995; Scott et al., 1995; Wackerbauer et al., 1996a; Wackerbauer etal., 1996b). Research Groups from China (Chao etal., 1990; Yuan, 1987; Zhang etal., 1988), Russia (Kolpachki etal., 1980; Sinitsyn etal., 1986) and Czechoslovakia (Chladek etal., 1989; Curin etal., 1987 Polednikova etal, 1981) also incorporate immobilized cell technology and published the results in 1980.

A large number of sources used have been taken into account in the preparatory work of the study on which the present invention is based. They include:

Abbott, W. J. 1978. Immobilized cells. In: Annual Reports on Fermentation Processes, Ed. Perlman, D., New York: Academic Press, 2:91. Anon. 1994. Maturex® L. New Nordisk Pubication. B 560C-GB. Anon. 1996. Table Curve 2D. User's Manual, Jandel Scientific.

Anon. 1997. Alpha Laval Brewery Systems. Brewers' Guardian 126: 26. Anon. 1998. Digox 5 Operating Manual. Dr.Theidig publication.

Aquilla, T. 1997. The biochemistry of yeast: debunking the myth of yeast respiration and putting oxygen in its proper place. Brewing Techniques 50. Aschengreen, NH, Jepsen, S. 1992. Use of acetolactate decarboxylase in brewing fermentations. Proceedings of the 22 ^{nd Convention of the Institute of Brewing (} Australia and New Zealand Section), Melbourne 80.

Atkinson, B. 1986. Immobilized cells, their application and potential. In: Process engineering aspects of immobilized cell systems. Ed. Webb.C., Black, G. M, Atkinson, B. Manchester: Institution of Chemical Engineers 3.

Audet, P., Paquin, C., Lacroix, C. 1988. Immobilized growing lactic acid bacteria with kappa-carrageenan-locust bean gel. Applied Microbiology and Biotechnology29: 11.

Austin, G. D., Watson, RW. J., Nordstrom, P. A., D'Amore, T. 1994. An on-line capacitance biomass monitor and its correlation with viable biomass. MBAA Technical Quarterly 31: 85.

Axcell, V.C., O'Connor-Cox, E.S.C. 1996. The concept of yeast vitality-an alternative approach. Proceedings of the Brewing Institute Convention (Asia Pacific Sect.). Singapore 24: 64.

Axelsson, A., Sisak, C., Westrin, B. A., Szajani, B. 1994. Diffusion characteristics of a swelling gel and its consequences for bioreactor performance. The Chemical Engineering Journal 55: B35.

Bailey, J. E., Ollis, D. F.1986. Biochemical Engineering Fundamentals. New York: McGraw-Hill, Inc.

Bancel, S., Hu, W. 1996. Confocal laser scanning microscopy examination of cell distribution in macroporous microcarriers. Biotechnology Progress 12: 398.

Barker, M. G, Smart, K. A. 1996. Morphological changes associated with cellular aging of a brewing yeast strain. Journal of the American Society of Brewing Chemists 54 (2): 121.

Bejar, P., Casas, C, Godia, F., Sola, C. 1992. Influence of physical properties on the operation of a three-phase fluidized bed fermenter with yeast cells immobilized in calcium alginate. Applied Biochemistry and Biotechnology

34: 467.

Bickerstaff, G. F. 1997. Immobilization of enzymes and cells. In: Immobilization of Enzymes and Cells, Ed.

Bickerstaff, G. F., New Jersey, U. S. A.: Humana Press, Inc. 1.

Birnbaum, S., Pendleton, R., Larson, P „Mosbach, K.1981. Covalent stabilization of alginate gel for the entrapment of living whole cells. Biotechnology Letters 3: 393.

Budac, D., Margartis, A. 1999. Personal communication.

BuyOkgungor, H. 1992. Stability of Lactobacillus bulgaricus immobilized in kappa-carrageenan gels. Journal of Chemical Technology and Biotechnology 53: 173.

Chahal, P. S. 1992. Fluorosensor controlled fed-batch production of Cyclosporin-A from Beauveria nivea.

Ph. D. Thesis. University of Western Ontario.

Clean, Μ. Y.1989. Airlift bioreactors. London: Elsevier Applied Science.

Chisti, Y „Moo-Young, M. 1993. Improvement of the performance of airlift reactors. Chemical Engineering Progress 6: 38.

Cho, G. H., Choi, C. Y., Choi, Y. D., Han, Μ. H. 1982. Ethanol production by immobilized yeast and its carbon dioxide gas effects in a packed bed reactor. Journal of Chemical Technology and Biotechnology 32: 959.

Coutts, M. W. 1956. Britain Patent No. 872, 391.

Curin, J., Pardonova, B., Polednikova, M., Sedova, H., Kahler, M. 1987. Beer production with immobilized yeast. European Brewing Convention Congress, Madrid 433.

Dale, C. J., Hough, J. S., Young, T. W. 1986. Freaction of high and low molecular weight components from wort and beer by adsorption chromatography using Sephadex LH20 gel. Journal of the Institute of

Brewing 92 (5): 457.

Daoud, I. S., Searle, B. A. 1986. Yeast vitality and fermentation performance. Monograph-XII European Brewery Convention - Symposium on BrewersYeast, Helsinki 108.

de Backer, L., Willaert, R. G., Baron, G. V. 1996. Modeling of immobilized bioprocesses. In: Immobilized Living Cell Systems Modeling and Experimental Methods, Ed. Willaert, R. G., Baron, G. V., and de Backer, L., Toronto: John Wiley and Sons. 47.

de Beer, D., Van den Heuvel, J. C., Ottengraaf, S. P. P. 1993. Microelectrode measurements of activity distribution in nitrifying bacterial aggregates. Applied Environmental Microbiology 59: 573.

Debourg, A., Laurent, M., Goossens, E., Borremans, E “Van De Winkel, L., Masschelein, C. A. 1994. Wort aldehyde reduction potential in free and immobilized yeast systems. Journal of the American Society of Brewing Chemists 52: 100.

Dillenhofer, W., Ronn, D. 1996. Secondary fermentation of beer with immobilized yeast. Brauwelt International 14: 344.

Doran, P. M., Bailey, J. E. 1986a. Effects of hydroxyurea on immobilized and suspended yeast fermentation rates and cell cycle operation. Biotechnology andBioengi7tee7ing 28: 1814.

Doran, P. M., Bailey, J. E. 1986b. Effects of immobilization on growth, fermentation properties, and macromolecular composition of Saccharoinyces cerevisiae attached to gelatin. Biotechnology and Bioengineering 28: 73. dos Santos, V. A. P. M, Bruijnse, M., Tramper, J., Wijffels, R. H. 1996. The magic-bead concept: an integrated approach to nitrogen removal with coimmobilized micro-organisms. Applied Microbiology and Biotechnology 45:

447.

Driessen, W., Habets, L., Vereijken, T. 1997. Novel anaerobic and aerobic process to meet strict effluent plant design requirements.Proceedings of the Brewing Institute (Asia Pacific Section), Aukland 148.

Dulieu, C., Boivin, P., Dautzenberg, H., Poncelet, D. 1996. Immobilized enzyme system to avoid diacetyl formation: a new tool for accelerated beer maturation.International Workshop on Bioencapsulation, Potsdam 22.

Dunbar, J., Campbell, S. L., Banks, D. J., Warren, D. R. 1988. Metabolic aspects of a commercial continuous fermentation system.Proceedings of the Brewing Institute Convention, Brisbane151.

Estape, D., Godia, F., Sola, C. 1992. Determination of glucose and ethanol effective diffusion coefficients in Caalginate gel. Enzyme and Microbial Technology 14: 396.

Evans, H. A. V., Cleary, P. 1985. Direct Measurement of Yeast and Bacterial Viability.

Journal of the Brewing Institute 91: 73.

Fan, L.-S. 1989. Gas-Liquid-Solid Fluidization Engineering. Boston: Butterworths.

Fernandez, E. 1996. Nuclear magnetic resonance spectroscopy and imaging. In: Immobilized Living Cell Systems: Modeling and Experimental Methods. Toronto: John Wiley and Sons, Chapter 6.

Garcia, A. I..Garcia, L. A., Diaz, M. 1994. Prediction of ester production in industrial beer fermentation. Enzyme and Microbial Technology 16 (1): 66. Geankoplis, C. J. 1993. Transport Processes and Unit Operations. New Jersey: Prentice Hall P.T.R.

Gee, D. A., Ramirez, W. F. 1994. A flavor model for beer fermentation. Journal of the Brewing Institute 100: 321.

Geiger, K. H., Compton, J. 1957. 545, 867.

Gekas, V. C. 1986. Artificial membranes as carriers for the immobilization of

biocatalysts. Enzyme and Microbial Technology 8: 450.

Gift, E. A., Park, H. J., Paradis, G. A., Demain, A. L., Weaver, J. C. 1996. FACSbased isolation of slowly growing cells: double encapsulation of yeast in gel microdrops. Nature Biotechnology 14: 884.

Gikas, P., Livingston, A. G. 1993. Use of ATP to characterize biomass viability in freely suspended and immobilized cell bioreactors. Biotechnology and Bioengineering 42: 1337.

Gikas, P., Livingston, A. G. 1996. Viability of immobilized cells: use of specific ATP levels and oxygen uptake rates. Progress in Biotechnolgy 11, Immobilized Cells: Basics and Applications, Noordwijkerhout 11: 264.

Gilson, C. D., Thomas, A. 1995. Ethanol production by alginate immobilized yeast in a fluidized bed bioreactor. Journal of Chemical Technology and Biotechtology 62:38.

Godia, F., Casa, C., Castellano, B., Sola, C. 1987. Immobilized cells: behavior of carrageenan entrapped yeast during continuous ethanol fermentation. Applied Microbology and Biotechnology 26: 342.

Gopal, C. V., Hammond, J. R. M. 1993. Application of immobilized yeasts for fermenting beer. Brewing and Distilling International 24: 72.

Hannoun, B. J. M., Stephanopoulos, G. 1986. Diffusion coefficients of glucose and ethanol in cell-free and cell-occupied calcium alginate membranes. Biotechnology and Bioengineering 28: 829.

Hardwick, W. A. 1995. Handbook of Brewing. New York: Marcel Dekker, Inc. Hayat, M. A. 1972. Basic Electron Microscopy Techniques. Toronto: van Nostrand Reinhold Co. 96.

Heijnen, J. J., van Loosdrecht, M. C. M., Mulder, R., Weltevrede, R., Mulder, A. 1993. Development and scale-up of an aerobic biofilm air-lift suspension reactor. Water Science and Technology 27: 253.

Higbie, R. 1935. The role of pure gas absorption into a still liquid during short periods of exposure. Transactions of the American Institute of Chemical Engineers 31: 365.

Hines, A. L „Maddox, R. N. 1985. Mass Transfer Fundamentals and

Applications. U. S. A.: Prentice-Hall, Inc.

Hinfray, C., Jouenne, T., Junter.G. 1994. Ethanol production from glucose by free and agar-entrapped batch cultures of Saccharornyces cerevisiae at different oxygenation levels. Biotechnology Letters 16: 1107.

Hoekstra, S. F. 1975. Wort composition, a review of known and unknown facts. Proceedings of the European Brewery Convention, Nice 465.

Hooijmans, C. M., Ras, C., Luyben, K. Ch. A. M. 1990. Determination of oxygen profiles in biocatalyst particles by means of a combined polarographic oxygen microsensor. Enzyme and Microbial Technology 12: 178.

Hough, J. S., Briggs, D. E., Stevens, R., Young, T. W. 1982. Wort metabolism by yeast. In: Malting and Brewing Science Volume 2 Hopped Wort and Beer, Ed. Hough, J. S., Briggs, D. E., Stevens, R., and Young, T. W., London, U. K.: Chapman and Hall. 566.

Hiisken, L. E., Tramper, J., Wijffels, R. 1996. Growth and eruption of gelentrapped microcolonies. In: Progress in Biotechnology 11, Immobilized Cells: Basics and Applications Ed. Wijffels, R. H., Buitelaar, R. M .., Bucke, C., Tramper, J., Amsterdam: Elsevier Science 336.

Hutter, K. J. 1996. Flow-cytometric analyzes for the evaluation of the fermentative ability of various yeasts. Brauwelt International 1: 52.

Hwang, S.-J., Fan, L.-S. 1986. Some design considerations of a draft tube gasliquid-solid spouted bed. The Chemical Engineering Journal 33: 49.

Imai, T. 1996. Recent advances in the determination of yeast vitality. Proceedings of the Congress of the Brewing Institute (Asia Pacific Section), Singapore 24: 60.

Inloes, D. S., Taylor, D. P., Cohen, S. N., Michaels, A. S „Robertson, C. R. 1983. Ethanol production by Saccharomyces cerevisiae immobilized in hollowfiber membrane bioreactors. Applied and Environmental Microbiology 46: 264. Inoue, T. 1987. Possibilities opened by new biotechnology and application of immobilized yeast to beer brewing. Kirin Brewery Co., Ltd. Research Reports 7.

Inoue, T. 1992. A review of diacetyl control technology. Proceedings of the 22 ^{nd Convention of the Institute of Brewing (} Australia and New Zealand Section), Melbourne 76.

Jepsen, S. 1993. Using ALDC to speed up fermentation. Brewers' Guardian. 55. Jones, M., Pierce, J. S. 1964. Absorption of amino acids from wort by yeasts. Journal of the Brewing Institute 70: 307.

Jones, R. P., Greenfield, P. F. 1984. A review of yeast ionic nutrition-Part I: growth and fermentation requirements. Process Biochemistry 19 (2): 48.

Jones, R. P., Pamment, N., Greenfield, P. F. 1981. Alcohol fermentation by the yeast-effect of environmental and other variables. ProcessBiochemist7y 16 (3): 42.

Kara, V. V., David, I., Searle, B. A. 1987. Assessment of yeast quality.

Proceedings of the Congress of the European Brewery Convention, Madrid, 21: 409.

Karamanev, D. G.1991. Model of the biofilm structure of Thiobacillus ferrooxidans. Journal of Biotechnology 10: 51.

Karel, S.F., Libicki, S.B., Robertson, C.R. 1985. The immobilization of whole cells: engineering principles. Chemical Engineering Science 40: 1321.

Kasten, F. H. 1993. An Introduction to Fluorescent Probes: Properties, History, and Applications. In: Fluorescent and Luminescent Probes for Biological Activity, Ed. Mason, W. T. London, U. K.: Academic Press Limited 12.

Klopper, W. J. 1974. Wort composition, a survey. European Brewery Convention

Monograph 1. Wort Symposium, Zeist 8.

Korgel, B. A., Rotem, A., Monbouquette, H. G. 1992. Effective diffusivity of galactose in calcium alginate gels containing immobilized Zyrnomonas mobilis. Biotechnology Progress 8: 111.

Kreger-Van Rij, N. 1984. The Yeasts: A Taxonomic Study. Amsterdam: Elsevier Science Publishers V.V.

Kronlof, J., Virkajarvi, 1.1996. Main fermentation with immobilized yeast-pilot scale. Proceedings of the European Brewery Convention Brewing Science Group, Berlin 94.

Kunze, W. 1996. Technology Brewing and Malting. Berlin: VLB.

Kuriyama, H.Jshibashi, H., Umeda, I. Μ. T., Kobayashi, H. 1993. Control of yeast flocculation activity in continuous ethanol fermentation. Journal of Chemical Engineering Japan 26 (4): 429.

Kurosawa, H., Matsamura, M., Tanaka, H. 1989. Oxygen diffusivity in gel beads containing viable cells. Biotechnology and Bioeragineering 34: 926.

Kurosawa, H., Tanaka, H. 1990. Advances in immobilized cell culture: development of a co-immobilized mixed culture system of aerobic and anaerobic micro-organisms. Process Biochemistry International 25: 189.

Kurtzman, C. P., Fell, J. W. 1998. The Yeasts, A Taxonomic Study, Fourth Edition. Amsterdam: Elsevier 361.

Kyung, K. H., Gerhardt, P. 1984. Continuous production of ethanol by yeast immobilized in a membranecontained fermenter. Biotechnology and Bioengineering 26: 252.

Lee, S. S., Robinson, F. M., Wang, Η. Y. 1981. Rapid determination of yeast viability. Biotechnology and Bioengineering Symposium No. 1 11, Gatlingburg, USA641.

Lentini, A. 1993. A review of the various methods available for monitoring the physiological status of yeast: yeast viability and vitality. Ferment 6: 321.

Lentini, A., Takis, S., Hawthorne, DB, Kavanagh, TE 1994. The influence of trub on fermentation and flavour development.Proceedings of the 23 ^{rd Convention of the Institute of brewing (} Asia Pacific Section), Sydney 89. Leudeking , R. 1967. Fermentation process kinetics. In: Biochemical and Biological Engineering, Ed. Blakebrough, N., London: Academic Press, Inc. 203.

Leudeking, R., Piret, E. L. 1959. A kinetic study of lactic acid fermentation. Journal of Biochemical and Microbiolial Technology and Engineering 1: 393.

Lewandowski, Z 'Altobelli, A., Fukushima, E. 1993. NMR and microelectrode studies of hydrodynamics and kinetics in biofilms. Biotechnology Progress 9: 40.

Lewandowski, Z., Stoodley, P., Altobelli, S. 1995. Experimental and conceptual studies on mass transport in biofilms. Water Science Technology 31: 153. Lewis, M., Young, T. 1995. Brewing. London: Chapman and Hall.

Li, J., Humpluey, A. E.1991. Use of fluorometers for monitoring and control of a bioreactor. Biotechnology and Bioengineering 37: 1043.

Lim, H.-S., Han, B.-K., Kim, J.-H., Peshwa, Μ. V, Hu, W.-S. 1992. Spatial distribution of mammalian cells grown on macroporous microcarriers with improved attachment kinetics. Biotechnology Progress 8: 486.

Linko, M „Virkajarvi, l„ Pohjala, N. 1996. Effect of flocculation characteristics on immobilized yeast performance. European Brewery Convention Brewing Science Group, Berlin 102.

Lloyd, D., Moran, C. A., Suller, Μ. T. E., Dinsdale, M. G. 1996. Flow cytometric monitoring of rhodamine 123 and a cyanine dye uptake by yeast during cider fermentation. Journal of the Brewing Institute 102: 251.

Lundberg, P., Kuchel, PW 1997. Diffusion of solutes in agarose and alginate gels: ¹ H and ^{Z3 on} PFGSE and ^{23 on} TQF NMR studies. Magnetic Resonance in Medicine 37: 44.

Margaritis, A., and Bokkel, D.W. "El Kashab, M. 1987. Repeated batch fermentation of ethanol using immobilized cells of Saccharomyces cerevisiae in a fluidized bioreactor system. In: Biological Research on Industrial Yeasts, Ed. Stewart, G. G., Russell, I., Klein, R. D., Hiebsch, R. R., Boca Raton: CRC Press 121.

Margaritis, A., Wallace, J. B. 1982. The use of immobilized cells of Zymomonas mobilis in a novel fluidized bioreactor to produce ethanol. Biotechnology and Bioengineering Symposium 147.

Margaritis, A., Wilke, C.R. J. 978a. The rotorfermentor. Part I: description of the apparatus, power requirements, and mass transfer characteristics. Biotechnology and Bioengineering 20: 709.

Margaritis, A., Wilke, C. R.1978b. The rotorfermentor. Part II: application to ethanol fermentation. Biotechnology and Bioengineering 20: 727.

Marrs, W. M. 1998. The stability of carrageenans to processing. In: Gums and Stabilizers for the Food Industry 9, Cambridge: Royal Society of Chemistry, 218: 345.

Martens, F. B., Egberts, G. T. C., Kempers, J., Robles de Medina, M.H. L., Welton, H. G. 1986. European Brewing Convention Monograph XII, Symposium on Brewing Yeast, Helsinki 339.

Masschelein, C. A. 1990. Yeast metabolism and beer flavor. Proceedings of the Third Aviemore Conference on Malting, Brewing and Distilling 103.

Masschelein, C. A., Carlier, A., Ramos-Jeunehomme, C., Abe, 1.1985. The effect of immobilization on yeast physiology and beer quality in continuous and discontinuous systems. Proceedings of the 20th European Brewery Convention Congress, Helsinki 339.

Masschelein, C. A., Ramos-Jeunehomme, C. 1985. The potential of alginate immobilized yeast in Brewery fermentations. Brewing Institute Central and Southern African Section Proceedings of the 1st st Scientific and Technical Convention, Johannesburg 392.

Masschelein, C. A., Vandenbussche, J. 1999. Current outlook and future perspectives for immobilized yeast technology in the brewing industry. Brewers' Guardian 28 (4): 35.

Masters, B.R., Thaer, A. A. 1994. Real-time scanning of slit confocal microscopy of in vivo human cornea. Applied Optics 33: 695.

Meilgaard, M. 1982. Prediction of flavor differences between beers from their chemical composition, Brygmestere71 5.

Mensour, N., Margaritis, A., Briens, C. L., Pilkington, H., Russell,!. 1996. Application of immobilized yeast cells in the brewing industry. In: Progress in Biotechnology 11, Immobilized Cells: Basics and Applications Ed.

Wijffels, R. H., Buitelaar, R. M .., Bucke, C., Tramper, J., Amsterdam: Elsevier Science 661.

Mensour, N., Margaritis, A., Briens, C. L., Pilkington, H., Russell, 1.1997. New Developments in the Brewing Industry Using Immobilized Yeast Cell Bioreactor Systems. Journalof the Brewiiag Institute 103: 363.

Merchant, F. J. A. 1986. Diffusivity Characteristics of Glucose and Alginate Immobilization Matrices. Ph. D. Thesis. University of Western Ontario.

Merchant, F.J. A, Margaritis, A., Wallace, J. B. 1987. A novel technique for measuring solute diffusivities in entrapment matrices used in immobilization. Biotechnology and Bioengineering 30: 936.

Mochaba, F. M. 1997. A novel and practical yeast vitality method based on magnesium ion release. Journal of the Brewing Institute 103: 99.

Mochaba, F., O'Connor-Cox, E. S. C., Axcell, V. C. 1998. Practical procedures to measure yeast viability and vitality prior to pitching. Journal of the American

Society of Brewing Chemists 56 (1): 1.

Muhr, A. H., Blanshard, J. Μ. V. 1982. Diffusion and gels. Polymer 23: 1012. Mulder, Μ. Η. V., Smolders, C. A. 1986. Continuous ethanol production controlled by membrane processes. Process Biochemistry 21: 35.

Nakanishi, K., Murayama, H., Nagara, A., Mitsui, S. 1993. Beer brewing using an immobilized yeast bioreactor system.Bioprocess Technology 16: 275. Nakanishi, K., Onaka, T., Inoue, T. 1986. A new immobilized yeast reactor system for rapid production of beer. Kirin Brewing Company Research Laboratory Reports 13.

Nakatani, K., Takahashi, T., Nagami, K., Kumada, J. 1984a. Kinetic study of vicinal diketones in brewing (I) formation of total vicinal diketones. MBAA Technical Quarterly 21 (2): 73.

Nakatani, K., Takahashi, T., Nagami, K., Kumada, J. 1984b. Kinetic study of vicinal diketones in brewing (II) theoretical aspect for the formation of total vicinal diketones. MBAA Technical Quarterly 21 (4): 175.

Nakatani, K., Fukui, N., Nagami, K., Nisliigaki, M. 1991. Kinetic analysis of ester formation during beer fermentation. Journal of the American Society of Brewing Chenaists 49 (4): 152.

Narzip, L., Miedaner, H., Graf, P., Eichhorn, P., and Lustig, S. 1993. A technological approach to improve flavor stability. MBAA Technical Quarterly 30: 48.

Nava Saucedo, J.E., Roisin, C., Barbotin, J.-N. 1996. Complexity and heterogeneity of microenvironments in immobilized systems. Progress in Biotechnology 11, Immobilized Cells. Basics and Applications, Noordwijkerhout 39.

Nedovic, A. N., Vunjak-Novakovc, G., Leskosek-Cukalovic, I., Cutkovic, M. 1996. A study on significantly accelerated fermentation of beer using an airlift bioreactor with calcium alginate entrapped yeast cells. Fifth World Congress of Chemical Engineering 2: 474.

Neufeld, R. J., Poncelet, D. J., Norton, S. D. 1996. Canadian Patent Application no. 2133789.

Norton, S., DAmore, T. 1994. Physiological effects of yeast cell immobilization: applications for brewing. Enzyme and Microbial Technology 16: 365.

Norton, S., Watson, K., DAmore, T. 1995. Ethanol tolerance of immobilized brewer's yeast cells. Applied Microbiology and Biotechnology 43:18. O'Connor-Cox, E „Mochaba, F. M., Lodolo, E. J., Majara, M., Axcell, B. 1997. Methylene blue staining: use at your own risk. MBAA Technical Quarterly 34 (1): 306.

O'Reilly, A. M., Scott, J. A. 1995. Defined coimmobilization of mixed microorganism cultures. Enzyme and Microbial Technology 17: 636.

Okazaki, M., Hamada, T., Fujji, H., Mizobe, A., and Matsuzawa, S. 1995. Development of poly (vinyl alcohol) hydrogel for waste water cleaning. I. Study of poly (vinyl alcohol) gel as a carrier for immobilizing organisms. Journal of Applied Polymer Science 58: 2235.

Oldshue, J. Y., Herbst, N. R. 1992. A Guide to Fluid Mixing. New York: Lightnin. Opekarova, M., Sigler, K. 1982. Acidification power: an indicator of metabolic activity and autolytics changes in Saccharomyces cerevisiae (yeast). Folia Microbiologia 27: 395.

0yaas, J., Storro, I., Svendsen, H., Levine, D. W. 1995. The effective diffusion coefficient and distribution constant for small molecules in calcium-alginate gel beads. Biotechnology and Bioengineering 47: 492.

Paiva, T. C. B., Sato, S., Visconti, A. E. S., Castro, L. A. B. 1996. Continuous alcoholic fermentation process in a tower reactor with recycling of flocculating yeast. Applied Biochemistry and Biotechnology 57-58 (0): 55.

Pajunen, E., Makinen, V., Gisler, R. 1987.Secondary fermentation with immobilized yeast. European Brewery Convention Congress, Madrid441. Parascandola, P., de Alteriis, E. 1996. Patterns of growth and respiratory activity of Saccharomyces cerevisiae (copper's yeast) cells growing entrapped in an insolubilized gelatin gel. Biotechnology and Applied Biochemistry 23: 7. Perry, R. H., Green, D. W. 1984. PenysClaemical Erzgiiaeers'Handbook. New York: McGraw-Hill Book Company.

Pilkington, P. H., Margaritis, A., Mensour, N. A. 1998a. Mass transfer characteristics of immobilized cells used in fermentation processes. Critical Reviews in Biotechnology 18 (2 & 3): 237.

Pilkington, P. H., Margaritis, A., Mensour, N. A., Russell.l. 1998b. Fundamentals of immobilized yeast cells for continuous beer fermentation: a review. Journal of the Brewing Institute 104: 19.

Pilkington, H., Margaritis, A., Mensour, N., Sobczak, J., Hancock, I., Russell.l. 1999. Kappa-carrageenan gel immobilization of lager brewing yeast. Journal of the Brewing Institute 105 (6): 398.

Polson, A. 1950. Some aspects of diffusion in solution and a definition of a colloidal particle. Journal of Physical and Colloidal Chemistry 54: 649.

Power, D.A., McCuen, P. J. 1988. Manual of BBL® Products and Laboratory Procedures, Sixth Editiorl. Maryland: Becton Dickinson Microbiology Systems 249.

Priest, F.G., Campbell, 1.1996.Brewing Microbiology 2nd Ed., UK: Chapman and Hall.

Rees, D. A. 1972. Polysaccharide gels: a molecular view.Chemisry and Industry 19: 630.

Roca, E., Camesselle, C., Nunez, M. 1995. Continuous ethanolic fermentation by Saccharomyces cerevisiaeimmobilized in Ca-alginate beads hardened with Al ³⁺ . Biotechnology Letters 9: 815.

Roukas, T. 1994. Continuous ethanol production from carob under extract by immobilized Saccharomyces cerevisiae in a packed bed reactor. Journal of Chemical Technology and Biotechnology 59: 387.

Russell, I., Stewart, G. G. 1992. Contribution of yeast and immobilization technology to flavour development in fermented beverages. Food Technology 148.

Ryder, D. S. 1985. The growth process of brewing yeast and the biotechnological challenge. MBAA Technical Quarterly 22: 124.

Ryu, D. D., Kim, Y. J., Kim, J. H. 1984. The effect of air supplementation on the performance of a continuous ethanol fermentation system. Biotechnology and Bioengineering 26:12.

Salmon, P. M., Robertson, C. R. 1987. Mass transfer limitations in gel beads containing growing immobilized cells. Journal of Theoretical Biology 125: 325. Schumpe, A., Quicker, G., Deckwer, W.-D. 1982. Gas solubilities and microbial culture media. Advances in Biochemical Engineering 24: 1.

Shindo, S., Sahara, H., Koshino, S. 1994. Suppression of alpha-acetolactate formation in brewing with immobilized yeast. Joursaal of the Bnewing Institute 100: 69.

Smart, K. A. 1995. The importance of the brewing yeast cell wall. Brewers'Guardian 124 (4): 44.

Smart, K. A. 1999. Ageing and brewing yeast. Brewers'Guardian 19.

Smart, K. A., Chambers, K. M., Lambert, I., Jenkins, C. 1999. Use of methylene violet procedures to determine yeast viability and vitality.Journal of the American Society of Brewing Chemists 57 (1): 18.

Sharpe, F. R. 1988. Assessment and control of beer flavor. Journal of the Brewing Institute 95: 301.

Stewart, G. G. 1977. Fermentation-yesterday, today and tomorrow. MBAA Technical Quarterly 14: 1.

Stewart, G. G., Russell, 1.1986. The relevance of the flocculation properties of yeast in today's brewing industry.

European Brewery Convention Symposium on Brewers'Yeast, Helsinki 53.

Stewart, G. G., Lyness, A., Younis, 0.1999. The control of ester synthesis during wort fermentation. MBAA Technical Quarterly 36 (1): 61.

Takahashi, S. and Kimura, Y. 1996. Effect of main fermentation parameters on stale flavor. Brauwelt International 14 (3): 253.

Taylor, D. G. 1989. Influence of brewhouse practice on wort composition.

Brewer's Guardian 118 (2): 30. Technical Committee and Editorial Committee of the American Society of Brewing Chemists (ASBC). 1992.

Methods of Analysis. 8 Edition. Minnesota: ASBC.

Uttamlal, M., Walt, D. R. 1995. A fiber-optic carbon dioxide sensor for fermentation monitoring. Biotechnology 13: 597.

Venancio, A., Teixeira, J. A. 1997. Characterization of sugar diffusion coefficients in alginate membranes. Biotechnology Techniques 11: 183.

Vilach, C., Uhlig, K. 1985. The measurement of low oxygen levels in bottled beer. Brauwelt International 70.

Virkajarvi, I., Kronlof, J. 1998. Long-term stability of immobilized yeast columns in primary fermentation. Journal of the American Society of Brewing Chemists 56 (2): 70.

Vives, C., Casas, C., Gbdia, F., Sola, C. 1993. Determination of intrinsic fermentation kinetics of Saccharomyces cerevisiae cells immobilized in calcium alginate beads and observation on their growth. Applied Microbiology and Biotechnology 38: 467.

Wada, M., Kato, J., Chibata.1.1979. A new immobilization of microbial cells. European Journal of Applied Microbiology and Biotechnology 8: 241.

Wang, Η. Y., Lee, S.S., Takach, Y, Cawthon, L. 1982. Maximizing microbial cell loading in immobilized-cell systems. Biotechnology and Bioengineering Symp. No. 12139.

Westrin, B. A., Axelsson, A. 1991. Diffusion in gels containing immobilized cells: a critical review. Biotechnology and Bioengineering 38: 439.

Wheatcroft, R., Lim, Y. M., Hawthorne, D. B., Clarke, BJ, Kavanagh, TE 1988. An assessment of the use of specific oxygen uptake measurements to predict the fermentation performance of brewing yeasts.Table Institute of Brewing (Australia and New Zealand Section) Proceedings of the Twentieth Convention, Brinsbane 193 .

White, F. H., Portno, A. D. 1978. Continuous fermentation by immobilized brewers yeast. Journal of the Brewinig Institute 84: 228.

Wijffels, R. H., de Gooijer, C. D., Schepers, A. W., Tramper, J. 1996. Immobilized cell growth: diffusion limitation in expanding micro-colonies. In: Progress in Biotechnology 11, Immobilized Cells: Basics and Applications Ed. Wijffels, R. H., Buitelaar, R. M .., Bucke, C., Tramper, J., Amsterdam: Elsevier Science 249.

Wijffels, R. H., Englund, G., Hunik, J. H., Leenen, J. T. M., Bakketun, A. et al.

1995. Effects of diffusion limitation on immobilized nitrifying microorganisms at low temperatures. Biotechnology and Bioenghleering 45: 1.

Aivasidis, A., 1996. Another Look at Immobilized Yeast Systems. Cerevisia Belgian Journal of Brewing & Biotechnology, 21.27.

Aivasidis, A., Wandrey, C., Eils, H.-G. & Katzke, M., 1991. Continuous Fermentation of Alcohol-Free Beer with Immobilized Yeast in Fluidized Bed Reactors. Proceedings of the European Brewery Convention Congress, Lisbon, 1991,569.

Akiyama-Jibiki, M., Ishibiki, T., Yamashita, H. & Eto, M., 1997. A Rapid and Simple Assay to Measure Flocculation in Brewer's Yeast. Master Brewers of the Americas Technical Quarterly, 34,278.

Alpha Laval Brewery Systems, 1997. Immobilized Yeast System Reduces Maturation Time. Brewers' Guardian, 126, 26.

Alpha Laval Brewery Systems, 1996. Continuous Maturation of Beer with Immobilized Yeast, Company Report.

Al Taweel, A. M. & Walker, L. D., 1983. Liquid Dispersion and Static In-line Mixers. Canadian Journal of Chemical Engineering, 61,527.

Andries, M., Derdelinckx, G., lone, K. G., Delvaux, F., van Beveren, P. C. & Masschelein, C. A., 1997a. Zeolites as Catalysts for the Cold and Direct Conversion of Acetolactate into Acetoin. Proceedings of the European Brewery Convention Congress, Maastricht, Poster 48.

Andries, M „VanBeveren, P. C., Coffin, O. & Masschelein, C. A., 1995. Design of a Multi-Purpose Immobilized Yeast Bioreactor System for Application in the Brewing Process. European Brewery Convention Symposium: Immobilized Yeast Applications in the Brewing Industry, Espoo, Finland, 134.

Andries, M., Van Beveren, P. C., Goffin.O. & Masschelein, C. A., 1996a. Design and Application of an Immobilized Loop Bioreactor for Continuous Beer Fermentation, in Immobilized Cells: Basics and Applications. (R. H. Wijffels, R. M. Buitelaar, C. Bucke & J. Tramper, eds.) Amsterdam: Elsevier Science, 672.

Andries, M., Van Beveren, PC, Goffin, 0., Rajotte, P. & Masschlein, CA, 1996b. Design and Applications of an Immobilized Loop Bioreactor to Continuous Fermentation of Beer.Proceedings of the 6 ^th Inteniational Brewing Technology Confenence, Harrogate, 380.

Andries, M., Van Beveren, P. C., Goffin.O., Rajotte, P. & Masschlein, C. A., 1997b. Practical Resuits Using the Meura-Delta Immobilized Yeast Fermentation System. Brewers' Guardian, 26.

Andries, M., Van Beveren, P. C., Goffin, 0., Rajotte, P. & Masschlein, C. A., 1997c. First Results on Semi Industrial Continuous Fermentation with Meura-Delta Immobilized Yeast Fermentor. Master Brewers Association of the Americas Technical Quarterly, 34,119.

Argiriou, T., Kanellaki, M., Voliotis, S. & Koutinas, A. A., 1996. Kissiris-Supported Yeast Cells: High Biocatalytic Stability and Productivity Improvement by Successive Preservatives at C. Journal of Agriculture and Food Chemistry.

44,4028.

Audet, P. & Lacroix, C., 1989. Two Phase Dispersion Processes for the Production of Biopolymer Gel Beads: Effects of Barious Parameters on Bead Size and Their Distribution. Process Biochemistry, 24,217.

Axelsson, A. & Persson, B., 1988. Determination of Effective Diffusion Coefficients in Calcium Alginate Gel Plates with Varying Yeast Cell Content. Applied Biochemistry and Biotechnology, 18,231.

Bardi, E. P., Koutinas, A. A., Soupioni, M. J. & Kanellaki, Μ. E., 1996. Immobilization of Yeast on Delignified Cellulosic Material for Low Temperature Brewing. Journal of Agriculture and Food Chemistry, 44,463.

Berkman, P, D. & Calabrese, R, V., 1988. Dispersion of Viscous Liquids by Turbulent Flow in a Static MixerAmerican Institute of Chemical Engineering Journal. 34, 602.

Borremans, E., 1997. Secondary Fermentation of Beer with Immobilized Yeast. Cerevisia Belgian Journal of Brewing & Biotechnology, 22.33.

Bower, J. L. & Christensen, C. M., 1995. Disruptive Technologies: Catching the Wave. Harvard Business Review, 73.43.

Breitenbiicher, K. & Mister, M., 1995. Fluidized Bed Fermenters for Continuous Production of Non-Alcoholic Beer with Open-Pore Sintered Glass Carriers. European Brewery Convention Symposium: Immobilized Yeast Applacations in the Brewing Industry, Espoo, Finland, 77.

Brewers Association of Canada, 1988. About the Beer And The Brewing Industry, Ottawa, Ontario.

Broderick, Harold M., 1979. The Practical Brewer: A Manual for the Brewing Industry, 2nd Ed. Impressions Inc., Wisconsin.

Burns, J. A., 1937. Journal of the Brewing Institute, 43, 31.

Calleja, G. B. & Johnson, B. F., 1977. A Comparison of Quantitative Methods for Measuring Yeast Flocculation. Canadian Journal of Microbiology, 23.68.

Carberry, J. J., 1976.Chemical and Catalytic Reaction Engineering, McGrawHill.

Carberry, J. J. & Varma, A., 1987.Chemical Reaction and Reactor Engi7leeri7zg, Marcel Dekkerlnc., New York.

Cashin, M.-M., 1996. Comparative Studies of Five Porous Supports for Yeast Immobilization by Adsorption / Attachment. Journal of the Brewings Institute 02, 5.

Champagne, C. P „1994. Overview: Immobilized Cell Technology in Food Processing. Proceedings of the Bioencapsulation Research Group IV, Quebec, Canada, 33.

Chang, C. M., Lu, W. J., Own, K. S. & Hwang, S. J., 1994 Comparison of Airlift and Stirred Tank Reactors for Immobilized Enzyme Reactions. Process Biochemistry 29,133.

Chao, X., Ye, G. & Shi, S., 1990. Kinetic Study of the Continuous Fermentative Process of Green Beer Production with Immobilized Yeast. Huagong Jixie, 17.18.

Chibata.I., 1979. Immobilized Microbial cells with Polyacrylamide Gel and Carrageenan and their Industrial Applications. American Chemical Society Series, 106,187.

Chisti, M.Y., 1991. Airlift Bioreactors, Elsevier Applied Science, New York.

Chisti, Y. & Moo-Young, M., 1993. Improve the Performance of Airlift Reactors. Chemical Engineering Progress, 6,38.

Chladek, L., Voborsky, J., Sima, J. & Hosek, Z., 1989. Prumysl Potravin, 40,590.

Coe, H. S. & Clevenger, G. H., 1916. Methods for Determining the Capacities of Slime Thickening Tanks. Transcripst of the American Institute of Mechanical Engineering, 55, 356.

Curin, J., Pardonova, B., Polednikova, M., Sedova, H. & Kahler, M., 1987. Beer

Production with Immobilized Yeast. Proceedings of the European Brewery

Convention, Madrid, 433.

Decamps, C. & Norton, S., 1994. A New Emulsion Process Using Static Mixer for the Production of k-Carrageenan Gel Beads. Labatt Breweries of Canada Internal Report.

Del Pozo, M., Briens, C. L. & Wild, G., 1994. Effect of Liquid Coalescing Properties on Mass Transfer, Heat Transfer and Hydrodynamics in a ThreePhase Fluidized Bed. The Chemical Engineering Journal, 55.1.

Dillenhofer, W. & Ronn, D., 1996a. Alpha Laval / Schott System of Secondary Fermentation with Immobilized Yeast. Brewing & Distilling International, 27, 35. Dillenhofer, W. & Ronn, D., 1996b. Secondary Fermentation of Beer With Immobilized Yeast.Brauwelt International, 14, 344.

Dillenhoffer, W. & Ronn, D., 1996c Continuous Maturation of Beer. Beverage World International, 34.

Domeny, Z 'Smogrovicova, D., Gemeiner, P., Malovikova, A. & Sturdik, E.,

1996. Calcium Pectate Gel to Immobilize Yeast for Continuous Beer Production. Proceedings of the Bioencapsulation Research Group V, Potsdam, Germany, Poster 12.

Domingues, L., Lima, N. & Teixeira, J. A., 2000. Contamination of a High-CellDensity Continuous Bioreactor. Biotechnology and Bioengineering, 68, 584. Donnelly, D., 1998. Kinetics of Sugar Metabolism in a Fluidized Bed Bioreactor for Beer Production. Master Brewers Association of the Americas Annual Convention, Minneapolis, Poster 8.

Donnelly, D., Bergin, J., Gardiner, S. & Cahill, G., 1998. Kinetics of Sugar Metabolism in a Fluidized Bed Bioreactor for Beer Production. Master Brewers Association of the Americas Technical Quarterly, 36,183.

Dulieu, C., Boivin, P., Malanda, M., Dautzenberg, H. & Poncelet, D., 1997. Encapsulation of α-Acetolactate Decarboxylase to Avoid Diacetyl Formation. Proceedings of the European Brewery Convention Congress, Maastricht,

Poster 44.

Duncombe, D., Bower, P., Bromberg, S., Fehring, J., Lau, V. & Tata, M., 1996. The Contribution of Free Cells in an Immobilized Yeast System. Journal of the American Society of Btewing Chemists, Poster Presentation.

Dziondziak, K. & Seiffert, T., 1995. Process for Continuous Production of Alcohol-Free Beer. Proceedings of the European Brewery Convention Congress, Brussels, 301.

Enari, T.-M., 1995. State of the Art of Brewing Research. Proceedings of the European Brewery Convention Congress, Brussels, 1.

Fix, G., 1989. Principles of Brewing Science. Brewers Publications, USA.

Fouhy, K. & Parkinson, G., 1996. Brewers Break with Tradition. Chemical Engineering, 103.45.

Gil, G. H., 1991. Continuous Ethanol Production in a Two-Stage, Immobilized and Suspended Cell Bioreactor (Alcohol Dehydrogenase). Ph. D. Thesis, Georgia Institute of Technology.

Gilliland, R. B., 1951. The Flocculation Characteristics of Brewing Yeasts During Fermentation. Proceedings of the European Brewing Convention, 8.35.

Groboillot, A., D. K. Boadi, D. Poncelet & Neufeld, R. J., 1994. Immobilization of Cells for Application in the Food Industry. Critical Reviews in Biotecltnology, 14.75.

Haikara, A., Virkajarvi, I., Kronlof, J. & Pajunen, E., 1997. Microbial Contaminations in Immobilized Yeast Bioreactors for Primary Fermentations.

Proceedings of the European Brewery Convention Coagness, 439:

Heggart, Η. M „Margaritis, A., Pilkington, H., Stewart, R. J., Dowhanick, T. M. & Russell.l., 1999. Factors Affecting Yeast Viability and Vitality Characteristics: A Review. Master Brewers Association of the Americas Technical Quarterly, 36,383.

Heijnen, J. J., Mulder, A., Enger, W. & Hoeks, F., 1989. Review on the

Application of Anaerobic Fluidized Bed Reactors in Waste-Water Treatment.

The Chemical Engineering Journal, 41, B 37.

Heijnen, J. J., Mulder, A., Weltevrede, R „Hols, J. & van Leeuwen, H. L. J. M., 1991. Large Scale Aerobic Treatment of Industrial Waste-Water Complex Using Biofilm Reactors. Water Science Technology, 23.1427.

Heijnen, J. J., Van Loosdrecht, M. C. M., Mulder, RW. R. & Mulder, A., 1993. Development and Scale-Up of an Aerobic Biofilm Air-Lift Suspension Reactor. Water Science Technology, 27, 253.

Helm.E., Nohr, B. & Thorne, R. S.W., 1953. The Measurement of Yeast

Flocculence and its Significance in Brewing. Wallerstein Laboratories Communications, 16,315.

Horitsu, H „Wang, Μ. Y. & Kawai, K., 1991. A Modified Process for Soy Sauce Fermentation by Immobilized Yeasts. Agricultural and Biological Chemistry, 55, 269.

Hryclik, Kevin, Gerry Ginter, Jim Helmke, & Jim Spiers, 1987. The How's And Why's of Brewing, Revision &# 2, LBOC.

Hunik, J. H., Tramper, J. & Wijffels, R. H., 1994. A Strategy for Scale Up Nitrification Processes with Immobilized Cells of Nitrosomonas europaea and Nitrobacter agilis. Bioprocess Engineering, 11, 73.

Hwang, S.-J. & Fan, L.-S., 1986. Some Design Considerations of a Draft Tube Gas-Liquid-Solid Spouted Bed. The Chemical Engineering Journal, 33, 49. Hyttinen, I., Kronlof, J. & Hartwall, P., 1995. Use of Porous Glass at Hartwall Brewery in the Maturation of Beer with Immobilized Yeast. European Brewey Convention Symposium: Immobilized Yeast Applications in the Brewing Industry, Espoo, Finland, 55.

Ibrahim, Y. A. A., Briens, C. L., Margaritis, A. & Bergougnou, M. A., 1996. Hydrodynamic Characteristics of a Three-Phase Inverse Fluidized Bed Column. Journal of the American Institute of Chemical Engineering, 42.1889.

Inoue, T „1995. Development of a Two-Stage Immobilized Yeast Fermentation System for Continuous Beer Brewing. Proceedings of the European Brewery Convention Congress, Brussels, 25.

Iserentant, D., 1995. Beers: Recent Technological Innovations in Brewing. Fermented Beverage Production. (A. G. H. Lea & J. R. Piggott, eds.) London: Chapman & Hall, 45.

Kaplan, R. S. & Norton, D. P., 1996. The Balanced Scorecard. President and Fellows of Harvard College, Harvard Business School Press, Massachusetts, USA.

Karamanev, D. G „Nagamune, T. & Endo, I., 1992. Hydrodynamic and Mass Transfer Study of a Gas-Liquid Solid Draft Tube Spouted Bed Bioreactor. Chemical Engineering Science, 47, 3581.

Karel, S. F., Libicki, S. B. & Robertson, C. R., 1985. The Immobilization of Whole Cells: Engineering Principles. Chemical Engineering Science, 40,1321. Katzbauer, B., Narodoslawsky, M. & Moser, A., 1995. Classification System for Immobilization Techniques. Bioprocess Engineering, 12,173.

Kennard, M. & Janekeh, M., 1991. Two-and Three-Phase Mixing in a Concentric Draft Tube Gas-Lift Fermentor. Biotechnology and Bioengineering, 38.1261.

Colot, F. B., 1988. Immobilized Microbial Systems. Principles, Techniques and Industrial Applications.

Robert E. Krieger Publishing Company, Florida.

Kolpachki, A. P., Isaeva, V. S., Kazantsev, E. N. & Fertman, G. 1., 1980.

Intensification of Wort Fermentation with Immobilized Yeasts. Fermeiztiiaya I Spirtovaya Prornyshlennost, 9.

Krikilion, Ph., Andries, M., Coffin, 0., van Beveren, P. C. & Masschelein, C. A., 1995. Optimal Matrix and Reactor Design for High Gravity Fermentation with Immobilized Yeast. Proceedings of the European Brewery Convention

Congress, Brussels, 419.

Kronlof, J., 1994. Immobilized Yeast in Continuous Fermentation of Beer. Ph. D. Thesis, VTT Publications, 167.

Kronlof, J., Linko, M. & Pajunen, E., 1995. Primary Fermentation with a TwoStage Packed Bed System Pilot Scale Experiences. European Brewery Convention Symposium: Immobilized Yeast Applications in the Brewing Industry, Espoo, Finland, 118.

Kronlof, J. & Virkajarvi, 1st, 1996. Main Fermentation with Immobilized Yeast-Pilot Scale of European Brewery Convention Brewing Science Group Bulletin, Zoeterwoude, 94.

Kronlof, J., Virkajarvi, I., Storgards, E. L., Londesborough, J. & Dymond, G., 2000. Combined Primary and Secondary Fermentation with Immobilized Yeast. World Brewing Congress, Poster # 56.

Ku, W. Y „1982. Fermentation Kinetics for Ethanol Production by Immobilized Yeast Cells (Biomass).

Ph. D. Thesis, The Louisiana State University and Agricultural and Mechanical College.

Kunze, W., 1999. Technology Brewing and Malting. VLB Berlin, Germany. Kynch, G. J., 1952. Transcripts of the Faraday Society, 48,166.

Lacroix, C., Paquin, C. & Arnaud, J. P., 1990. Batch Fermentation with

Entrapped Growing Cells of Lactobacillus casei: Optimization of the Rheological Properties of the Entrapment Gel Matrix. Applied Microbiology and Biotechnology, 32,403.

Levenspiel.O., 1972. Chemical Reaction Engineering. John Wiley & Sons, New York.

Linko, M., Virkajarvi, I., Pohjala, N., Lindborg, K., Kronlof, J. & Pajunen, E.,

1997. Main Fermentation with Immobilized Yeast-A Breakthrough? .Proceedings of the European Brewery Convention, Maastricht, 385.

Livingston, A. G. & Chase, H. A., 1990. Liquid-Solid Mass Transfer in a Three Phase Draft Tube Fluidized Bed Reactor. Chemical Engineering Communications, 92,225.

Livingston, A. G. & Zhang, S. F., 1993. Hydrodynamic Behavior of ThreePhase (Gas-Liquid-Solid) Airlift Reactors. Chemical Engineering Science, 48, 1641.

Lommi, H., 1990. Immobilized Yeast for Maturation and Alcohol-Free Beer.

Brewing and Distilling International, 21.23.

Lommi, H., Gronqvist, A., & Pajunen, E., 1990. Immobilized Yeast Reactor Speeds Beer Production. Food Technology, 5,128.

Lorenzen, K., 1996. Immobilized Yeast Plants for Alcohol-Free Beer Production and Rapid Maturation. Proceedings of the Brewing Convention Institute, Singapore, 244.

Maeba, H., Umemoto, S., Sato, M. & Shinotsuka, K., 2000. Primary

Fermentation with Yeast Immobilized in Porous Chitosan Beads-Pilot Scale

Trial. Proceedings of the 26 ^th Convention of the Brewing Institute-Asia Pacific Section, 82.

Mafra, I., Machado Cruz, J. M. & Teixeira, J. A., 1997. Beer Maturation in a Continuously Operating Bioreactor Using a Flocculating Brewer's Yeast Strain, Proceedings of the European Brewery Convention, 509.

Maiorella, B. L., 1983. Fermentative Ethanol Production (Alcohol, Distillation,

Economics). Ph. D. Thesis, University of California, Berkeley.

Margaritis, A., 1975. A Study of the Rotorfermentor and the Kinetics of Ethanol Fermentation. Ph. D. Thesis, University of California, Berkeley.

Margaritis, A., & Bajpai, P., 1982a. Continuous Ethanol Production from

Jerusalem Artichoke Tubers, Part I. Use of Free Cellsof Kluyveronzyces maixiafzus. Biotechnology and Bioeragineering, 24,1473.

Margaritis, A., & Bajpai, P „1982b. Continuous Ethanol Production from

Jerusalem Artichoke Tubers, Part II. Use of Immobilized Cellsof Kluyveronryces marxiafaus. Biotechnology and Bioengineering, 24.1483.

Margaritis, A & Merchant, F., 1984. Advances in Ethanol Production Using Immobilized Cell Systems. Critical Reviews in Biotechnology, 1,339.

Margaritis, A. & Rowe, G. E., 1983. Ethanol Production Using Zymoiizoitas mobilis Immobilized in Different Carrageenan Gels. Developments in Industrial Microbiology, 24,329.

Margaritis, A., and Bokkel, D. & El Kashab, M., 1987. Repeated Batch Fermentation of Ethanol Using Immobilized Cells of Saccharomyces cerevisiae in a Fluidized Bioreactor Systesm. Biological Research on Industrial Yeasts. Volume I, (G. G. Stewart, I. Russell, R. D. Klein & R. R. Hiebsch, eds.) Boca Raton: CRC Press, 121.

Margaritis, A. & Wallace, J. B., 1982. The Use of Immobilized Cells of Zymomonas mobilis in a Novel Fluidized Bioreactor to Produce Ethanol. Fourth Symposium on Biotechnology in Energy Production and Conservation, Gatlinburg, Tennessee, 12,147.

Margaritis, A. & Wallace, J. B., 1984. Novel Bioreactor Systems and their Applications.Bio/Technology, 2, 447.

Margaritis, A. & Wilke, C. R., 1978a. The Rotorfermentor. Part I: Description of the Apparatus, Power Requirements, and Mass Transfer Characteristics. Biotechnology & Bioengineering, 20,709.

Margaritis, A. & Wilke, C. R., 1978b. The Rotorfermentor: Part II: Application to Ethanol Fermentation. Biotechnology & Bioengineering, 20,727.

Masschelein, C. A., 1997. A Realistic View of the Role of Research in the Brewing Industry Today. Journal of the Brewing Institute, 103,103.

Masschelein, C. A., 1994. State-of-the-Art and Future Developments in Fermentation. Journal of the American Society of Brewirag Chemists, 52.1. Masschelein, C. A. & Andries, M., 1996a. The Meura-Deltalmmobilized Yeast

Fermenter for the Continuous Production of Beer. Cerevisia Belgian Journal of Brewirag & Biotechnology, 21.28.

Masschelein, C. A. & Andries, M., 1996b. Meura-Delta's Immobilized Yeast Fermenter for Continuous Beer Production. Brewing & Distilling International, 27.16.

Masschelein, C. A., Andries, M., Franken, F., Van de Winkel, L. & Van Beveren, P. C., 1995. The Membrane Loop Concept: A New Approach for Optimal Oxygen Transfer Into High Cell Density Pitching Yeast Suspensions. Proceedings of the EuropeanBrewery Convention Congress, Brussels, 377.

Masschelein, C. A., Ryder, D. S. & Simon, J-P., 1994. Immobilized Cell Technology in Beer Production. Critical Reviews in Biotechnology, 14,155. Matsuura, K., Hirotsune, M., Nakada, F. & Hamachi, M., 1991. A Kinetic Study on Continuous Sake Fermentation. Hakkokogaku Kaishi, 69, 345.

McCabe, J. T., 1999. The Practical Brewer: A Manual for the Brewing Industry. Master Brewers Association of the Americas, USA.

Mensour, N. A., Margaritis, A., Briens, C. L., Pilkington, H. & Russell,!., 1997. New Developments in the Brewing Industry Using Immobilized Yeast Cell Bioreactor Systems. Journal of the Brewing Institute, 103, 363.

Mensour, N. A., Margaritis, A., Briens, C. L., Pilkington, H. & Russell,!., 1996. Application of Immobilized Yeast Cells in the Brewing Industry. Immobilized Cells: Basics and Applications. (R. H. Wijffels, R. M. Buitelaar, C. Bucke & J. Tramper, eds.) Amsterdam: Elsevier Science, 661.

Mensour, N., Margaritis, A., Briens, C. L., Pilkington, H. & Russell, I., 1995. Gas Lift Systems for Immobilized Cell Fermentation. European Breweiy Convention Symposium: Immobilized Yeast Applications in the Brewing Industry, Espoo, Finland, 125.

Mensour, N., Margaritis, A., Russell, I., Briens, C. L., Decamps, C. & Norton, S., 1994. Beer Production with Immobilized Yeast Cells in a Pilot Plant Scale Airlift

Reactor. Proceedings of the Bioencapsulation Research Group IV, Quebec, Canada, 49.

Middleman, S., 1974. Drop Size Distributions Produced by Turbulent Pipe Flow of Immiscible Fluids Through a Static Mixer. Industrial Engineering and Chemical Process Design and Development, 13, 78.

Mieth, H. 0., 1995. Immobilized Yeast Plants for Alcohol Free Beer Production and Rapid Maturation. Proceedings of the Institute of Bnewing Convention, Victoria Falls, 166.

Mister, M., Breitenbucher, K. & Jaeger, R., 1995. Continuous Fermentation of Beer with Yeast Immobilized on Porous Glass Carriers. Brewers Digest, 70, 48. Moll, M. & Duteurtre, B., 1996. Fermentation and Maturation of Beer with Immobilized Microorganisms. Brauwelt International, 3,248.

Motai, H., Hamada, T. & Fukushima, Y., 1993. Application of a Bioreactor System to Soy Sauce Production, and Industrial Application of Immobilized Biocatalysts. (A. Tanaka, T. Tosa & T. Kobayashi, eds.) New York: Marcel Dekker, 315.

Muhr, A. H. & Blanchard, Μ. V., 1982. Diffusion and Gels. Polymer, 23,1012.

Mwesigye, P. K & Barford, J. P., 1994. Transport of Sucrose by Saccharomyces cerevisiae. Journal of Fermentation and Bioengineering, 77, 687.

Nakanishi, K., Murayama, H., Nagara, A. & Mitsui, S., 1993. Beer Brewing Using an Immobilized Yeast Bioreactor System, in Industrial Applications of Immobilized Biocatalysts. (A. Tanaka, T. Tosa & T. Kobayashi, eds.) New York: Marcel Dekker, 275.

Nakanishi, K., T. Onaka, T. Inoue & Kubo, S., 1985. A New Immobilized Yeast Reactor System for Rapid Production of Beer. Proceedings of 20 ^th European Brewing Convention Congress, Helsinki, 331.

Nakatani, K., Takahashi, T., Nagami, K. & Kumada, J., 1984. Kinetic Study of Vicinal Diketones in Brewing (I) Formation of Total Vicinal Diketones.Master

Brewers Association of the Americas Technical Committee, 21.73.

Nedovic, VA, Leskosek-Cukalovic, 1. & Vunjak-Novaki, G., 1996a. Short-Time Fermentation of Beer in Yeast Air-Lift Bioreactor Immobilization.

Proceedings of the Brewing Convention Institute, Singapore, 244.

Nedovic, V. A., Vunjak-Novakovic, G., Leskosek-Cukalovic, 1. & Cutkovic, M., 1996b. A Study on Considerably Accelerated Fermentation of Beer Using an Airlift Bioreactor with Calcium Alginate Entrapped Yeast Cells. Proceedings of the 5th World Congress of Chemical Engineering, San Diego, 474.

Neufeld, R. J., Norton, S. & Poncelet, D. J. C. M., 1994. Immobilized-Cell Carrageenan Bead Production and a Brewing Process Utilizing Carrageenan Bead Immobilized Yeast Cells. Canadian Patent Application 2,133, 789. Nguyen, A. L. & Luong, J.H. T., 1986. Diffusion and k-Carrageenan Gel Beads. Biotechnology and Bioengineering, 28.1261.

Nielsen, J. & Villadsen, J., 1994. Bioreactor Modeling. Bioreaction Engineering Principles, Plenum Press, New York, 9.

Norton, S. & D'Amore, T., 1994. Physiological Effects of Yeast Cell Immobilization: Applications for Brewing. Enzyme and Microbial Technology, 16,365.

Norton, S., Neufeld, R. J. & Poncelet, D. J. C. M „1994. Immobilized-Cell Carrageenan Bead Production and a Brewing Process Utilizing Carrageenan Bead Immobilized Yeast Cells. Canadian Patent Application 2,133,789.

Norton, S., Mensour, N., Margaritis, A., Briens, C. L., Decamps, C. & Russell, I., 1994. Pilot Scale of Primary Beer Fermentation with a Gaslift Immobilized Yeast Reactor. Proceedings of the European Brewing Convention, Sub-Committee, Dublin.

Norton, S., Watson, K. & D'Amore, T., 1995. Ethanol Tolerance of Immobilized Brewers'Yeast Cells. Applied Microbiology & Biotechnology, 43.18.

Nothaft, A., 1995. The Start-Up of an Immobilized Yeast System for Secondary

Fermentation at Brahma. European Brewery Symposium Convention: Immobilized Yeast Applications in the Brewing Industry, Espoo, Finland, 41.

Nunokawa, Y. & Hirotsune, M., 1993. Production of Soft Sake by an Immobilized Yeast Reactor System, in Industrial Application of Immobilized Biocatalysts. (A. Tanaka, T. Tosa & T. Kobayashi, eds.) New York: Marcel Dekker, 235.

Pajunen, E., 1996a. The Behavior of Immobilized Yeast Cells. Cerevisia BelgianJournal of Brewing & Biotechnology, 21, 33.

Pajunen, E., 1996b. The Immobilized System in the Brewing Industry. Proceedings of the Brewing Convention Institute, Singapore, 38.

Pajunen, E., 1995. Immobilized Yeast Lager Beer Maturation: DEAE-Cellulose at Sinebrychoff. European Brewery Convention Symposium: Immobilized Yeast Applications in the Brewing Industry, Espoo, Finland, 24.

Pajunen E. & Gronqvist A., 1994. Immobilized Yeast Fermenters for Continuous Lager Beer Maturation. Proceedings for the Brewing Convention Institute, Sydney, 101.

Pajunen, E., A. Gronqvist & Lommi, H., 1989. Continuous Secondary

Fermentation and Maturation of Beer in an Immobilized Yeast Reactor. MBAA Technical Quarterly, 26,147.

Pajunen, E., Gronqvist, A., Simonsen, B. & Lomrnij H., 1994. Immobilized Yeast Fermenters for Continuous Lager Beer Maturation. ALAFACE Annual Meeting, Quito, 13.

Pajunen, E., Ranta, B., Andersen, K., Lommi, H., Viljava, T., Bergin, J. & Guercia, H., 2000a. Novel Process for Beer Fermentation with Immobilized Yeast. Proceedings of the 26'7'Convention of the Brewing Institute, AsiaPacific Section, 91.

Pajunen, E., Viljava, T. & Lommi, H., 2000b. Novel Primary Fermentation with

Immobilzied Yeast System.World Brewing Congress, Oral presentation.

Paul, F. & Vignais, P. M., 1980. Photophosphorylation in Bacterial Chromatophores Entrapped in Alginate Gel: Improvement of the Physical and Biochemical Properties of Gel Beads with Barium as a Gel-Inducing Agent. Enzyme and Microbial Technology, 2,281.

Peach, M., 1996: Fermenting Faster Pints. New Scientist, 2058, 23.

Pilkington, P. H, Margaritis, A. & Mensour, N. A., 1998a. Mass Transfer Characteristics of Immobilized Cells Used in Fermentation Processes. Critical Reviews in Biotechnology, 18, 237.

Pilkington, P. H., Margaritis, A., Mensour, N. A & Russell, I., 1998b.

Fundamentals of Immobilized Yeast Cells for Continuous Beer Fermentation: A

Review. Journal of the Brewing Institute, 104.19.

Pilkington, H., Margaritis, A., Mensour, N .. Sobczak, J., Hancock, I. & Russell.l., 1999. Kappa-Carrageenan Gel Immobilization of Lager Brewing Yeast. Journal of the Ifzstitute of Brewirzg, 105,398.

Pirt, S.J., 1975. Principles of Microbe and Cell Cultivation, Blackwell Scientific, Oxford.

Pittner, H. & Back, W., 1995. Continuous Production of Acidified Wort for

Alcohol Free Beer Using Immobilized Lactic Acid Bacteria.Master Brewers Association of the Americas Technical Quarterly, 32,163.

Pittner, H., W. Back, W. Swinkels, E. Meersman, B. van Dieren & Lommi, H., 1993.

Continuous Production of Acidified Wort for Alcohol-Free Beer Using Immobilized Lactic Acid Bacteria. Proceedings of the 24tlz European Brewing Convention Congress, Oslo, 323.

Polednikova, M., Sedova, H. & Kahler, M., 1981. Immobilized Brewing

Yeast. Kvasny Prumysl, 27,193.

Poncelet, D., Lencki, R., Beaulieu, C., Halle, J, P., Neufeld, R. J. & Fournier, A.,.

1992. Production of Alginate Beads by Emulsification / lntemal Gelation. Applied Microbiology & Biotechnology, 38, 39.

Poncelet, D., Poncelet de Smet, B., Beaulieu, C. & Neufeld, R.J., 1993. Scale Up of Gel Bead and Microcapsule Production in Cell Immobilization, in Fundamentals of Animal Cell Encapsulation and Immobilization, Goosen, MFA, Ed. ., CRC Press Inc., Boca Raton, Florida.

Prakash, A., Briens, C. L. & Bergougnou, M. A., 1987. Mass Transfer Between Solid Particles and Liquid in a Three Phase Fluidized Bed. The Canadian Journal of Chemical Engineering, 65,228.

Prasad, K. Y. & Ramanujam, T. K „1995. Overall Volumetric Mass Transfer Coefficient in a Modified Reversed Flow Jet Loop Bioreactor with Low Density Particles.Bioprocess Engineering, 12,214.

Pritchett, Price. 1993. Culture Shift. Texas: Pritchett & Associates, Inc.

Que, F., 1993. Using a Thread Type of Alginate Gel Particles as CellImmobilized Support and Some Concept of Packed Bed Fermenter Design.Biotechnology Techniques, 7, 755.

Rajotte, P., 1998. Continuous Fermentation with Immobilized Yeast Cells. American Brewer, 76, 42.

Rajotte, P., 1997. Jumping Into the Next Millennium Canadian Style.American Brewer, 7S, 42.

Russell, I., Norton, S., Mensour, N., Margaritis, A. & Briens, C., 1995.

Immobilized Yeast Cells: Applications for Brewing.Proceedings of the Brewing Convention Institute, Victoria Falls, 159.

Ryder, D. S. & Masschelein, C. A., 1985. The Growth Process of Brewing Yeast and the Biotechnological Challenge.Journal of the American Society of Brewing Chemists, 43,66.

Satterfield, C. N „1970. Mass Transfer in Heterogeneous Catalysts, MIT Press, Cambridge.

Scott, J. A., O'Reilly, A. M. & Kirkhope, S., 1995. A Fibrous Sponge Matrix to Immobilized Yeast for Beverage Fermentations. Biotechnology Techniques, 9, 305.

Shindo, S. & Kamimur, M., 1990 Immobilization of Yeast with Hollow PVA Gel Beads. Journal of Fermentation and Bioengineering, 70,232.

Shindo, S., Sahara, H. & Koshino, S., 1994a. Suppression of α-Acetolactate Formation in Brewing with Immobilized Yeast. Journal of the Brewing Institute, 100.69.

Shindo, S., Sahara, H., Koshino, S. & Tanaka, H., 1993. Control of Diacetyl Precursor [α-acetolactate] Formation During Alcohol Fermentation with Yeast Cells Immobilized in Alginate Fibers with Double Gel Layers.Journal of Fermentation and Bioengineering, 76,199,

Shindo, S, Sahara, S, Watanabe, N. & Koshino, S '1994b. Main Fermentation with Immobilized Yeast Using Fluidized-Bed Reactor. Proceedings of the Brewing Convention Institute, Sydney, 109.

Sinitsyn, A. P., Rajnina, E. I., Efremov, A. B., Gracheva, I. M. & Gernet, Μ. V., 1986. Fermentation of Hydrolysed Mash by Yeasts Immobilized on Borosilicate Carriers.Bermentnaya I Spirtovaya Promyshlennost, 31.

Smogrovicova, D. & Domeny, Z., 1999. Beer Volatile By-Product Formation at Different Fermentation Temperature Using Immobilized Yeasts. Process Biochemistry, 34,785.

Smogrovicova, D., Domeny, Z. Gemeiner, P. Malovikova, A. & Sturdik, E., 1997. Reactors for Continuous Primary Beer Fermentation Using Immobilized Yeast, Biotechnology Techniques, 11, 261.

Sodini, I., Boquien, C. Y „Corrieu, G. & Lacroix, C., 1997. Use of an

Immobilized Cell Bioreactor for Continuous Milk Inoculation in Fresh Cheese Manufacturing. Journal of Industrial Microbiology & Biotechnology, 18,

56.

Speers, R. A. & Ritcey, L. L., 1995. Towards an Ideal Flocculation Assay. Journal of the American Society of Brewing Chemists, 174.

Stewart, G. G., 1996. Brewing Technology for the Future.The Brewer, 82,348.

Stewart, G. G. & Russell,!., 2000. Brewer's Yeast. The Brewing Institute, England.

Stewart, G. G. & Russell, 1st, 1981. Yeast Flocculation, and Brewing Science, Ed. J. R. A. Pollock, Academic Press, New York.

Stratford, M., 1996. Yeast Flocculation: Restructuring the Theories in Line with Recent Research, Cerevisiae Belgium Journal of Biotechnology, 38.

Sumino, T., Nakamura, H. & Mori, N., 1993. Development of a High-Efficiency Wastewater Treatment System Using Immobilized Microorganisms.in Industrial Application of Linmobilized Biocatalysts. (A. Tanaka, T. Tosa & T. Kobayashi, eds.) New York: Marcel Dekker, 377.

Tata, M., Bower, P., Bromberg, S., Duncombe, D., Fehring, J., Lau, V., Ryder, D. & Stassi, P. 1999. Immobilized Yeast Bioreactor Systems for Continuous Beer Fermentation. Biotechnology Progress, 15,105.

Technical Committee and Editorial Committee of the American Society of Brewing Chemists. 1992. Methods of Analysis.8i Edition, Minnesota, ASBC.

Teixera, J. M., Teixera, J. A., Mota, M., Manuela, M., Guerra, B., Machado Cruz, J. M. & SaAlmeida, A. M., 1991. The Cell Wall Influence Composition of a Brewer's Flocculent Lager Yeast on Sedimentation During Successive Industrial Fermentations. Proceedings of the European Brewery Congress, 241. Umemoto, S., Mitani, Y. & Shinotsuka, K., 1998. Primary Fermentation with Immobilized Yeast in a Fluidized Bed Reactor. Master Brewers Association of the Americas Technical Quarterly, 35.58.

Van de Winkel, L., 1995. Design and Optimization of a Multipurpose Immobilized Yeast Bioreactor for Brewery Fermentations. Cerevisia Belgian

Journal of Brewing & Biotechnoloy, 20.77.

Van de Winkel, L. & DeVuyst, L., 1997. Immobilized Yeast Cell Systems in Today's Breweries and Tomorrow's. Cerevisia Belgian Journal of Brewing & Biotechnology, 22, 27.

Van de Winkel, L., McMurrough, I., Evers, G., Van Beveren, P. C. &

Masschelein, C. A., 1995. Pilot-Scale Evaluation of Silicon Carbide Immobilized Yeast Systems for Continuous Alcohol-Free Beer Production. European Brewery Convention Symposium Jrntnobilized Yeast Applications in the Brewing Industry, Espoo, Finland, 90.

Van de Winkel, L., P. C. van Beveren & C. A. Masschelein, 1991a. The

Application of an Immobilized Yeast Loop Reactor to Continuous Production of Alcohol-Free Beer. Proceedings of the 23rd European Brewing Convention Congress, Lisbon, 577.

Van de Winkel, L., P. C. van Beveren, E. Borremans, E. Goosens & C.A.

Masschelein, 1991b. High Performance Immobilized Yeast Reactor Design for Continuous Beer Fermentation. Proceedings of the 24 ^th European Brewing Convention Congress, Oslo, 307.

Van Dieren, B., 1995. Yeast Metabolism and the Production of Alcohol-Free

Beer. European Brewery Convention Symposium: Immobilized Yeast Applications in the Brewing Industry, Espoo, Finland, 66.

Vanlersel, M. F. M., Meersman, E., Swinkels, W., Abee, T. & Rombouts, F. M., 1995 .. Continuous Production of Non-Alcohol Beer by Immobilized Yeast at Low Temperature. Journal of Industrial Microbiology, 14,495.

Van Loosdrecht, M. C. M. & Heijnen, J. J., 1993.Biofilm Bioreactors for WasteWater Treatment. Trends in Biotechnology, 11,117.

Vicente, A. A., Dluhy, M. & Teixeira, J. A., 1999. Increase in Ethanol Productivity in an Airlift Reactor with a Modified Draught Tube. The Canadian Journal of Chemical Engineering, 77,497.

Virkajarvi, I. & Linko, M., 1999. Immobilization: A Revolution in Traditional Brewing.Naturwissehschaften, 86,112.

Virkajarvi,!. & Kronlof, J., 1998. Long Term Stability of Immobilized Yeast Columns in Main Fermentation. Journal of the American Society of Brewing Chemists, 56, 70.

Virkajarvi, I. & Pohjala, N., 1999. Profiting from Immobilized Fermentation. Proceedings of the 5 ^th Avietnore Conference on Malting, Brewing and Distilling, 290.

Wackerbauer, K „Fitzner, M. & Gunther, J., 1996a. Technisch-Technologische Moglichkeiten mit Immobilisierter Hefe. Brauwelt, 136.2140.

Wackerbauer, K., Fitzner, M. & Lopsien, M., 1996b. Untersuchungenmit dem Neuen MPI-Bioreactor-System. Brauwelt, 136,222.

Webb, C., G. M. Black & B. Atkinson, 1986. Process Engineering Aspects of the Immobilized Cell System. England: The Institution of Chemical Engineers. Westrin, B. A. & A. Axelsson, 1991. Diffusion in Gels Containing Immobilized Cells: A Critical Review. Biotechnology and Bioengineering, 38,439.

Wu, W. T., Wu, J. Y. & Jong, J. Z., 1992. Mass Transfer in an Airlift Reactor with a Net Draft Tube. Biotechnology Progress, 8,465.

Yamane, T., 1981. On Approximate Expressions of Effectiveness Factors for Immobilized Biocatalysts. Journal of Ferrhentation Technology, 59,375.

Yamauchi, Y. & Kashihara, T., 1995a. Kirin Immobilized System. European Brewery Convention Symposium Immobilized Yeast Applications in the Brewing Industry, Espoo, Finland, 99.

Yamauchi, Y., Kashihara, T., Murayama, H., Nagara, A., Okamoto, T. & Mawatari, M., 1994. Scaleup of Immobilized Yeast Bioreactor for Continuous Fermentation of Beer. Master Brewers Association of the Americas Technical Quarterly, 31.90.

Yamauchi, Y., Okamoto, T., Murayama, H., Kajino, K., Amikura, T., Hiratsu, H.,

Nagara, A ,, Kamiya, T. & Inoue, T., 1995b. Rapid Maturation of Beer Using an Immobilized Yeast Bioreactor. 1. Heat Conversion of α-Acetolactate. Journal of Biotechnology, 38,101.

Yamauchi, Y, Okamoto, T., Murayama, H., Kajino, K., Nagara, A. & Noguchi, K., 1995c. Rapid Maturation of Beer Using an Immobilized Yeast Bioreactor. 2. Balance of Total Diacetyl Reduction and Regeneration. Journal of Biotechnology, 38,109.

Yamauchi, Y., Okamoto, T., Murayama, H., Nagara, A. & Kashihara, T., 1995d. Rapid Fermentation of Beer Using an Immobilized Yeast Multistage Bioreactor System: Control of Sulfite Formation. Applied Biochemistry and Biotechnology, 53,277.

Yamauchi, Y., Okamoto, T., Murayama, H „Nagara, A., Kashihara, T., Yoshida, M. & Nakanishi.K., 1995e. Rapid Fermentation of Beer Using an Immobilized Yeast Multistage Bioreactor System: Balance Control of Extract and Amino Acid Uptake. Applied Biochemistry and Biotechnology, 53,245.

Yamauchi, Y., Okamoto, T., Murayama, H., Nagara, A., Kashihara, T., Yoshida, M., Yasui, T. & Nakanishi, K., 1995f. Rapid Fermentation of Beer Using an Immobilized Yeast Multistage Bioreactor System: Control of Minor Products of Carbohydrate Metabolism.Applied Biochemistry and Biotechnology, 53,261. Yuan, X., 1987. Application of 4mmobilization Technique in the Brewing Industry. Shipin Kexue, 94, 8.

Zhang, Z., Su, E. & Yu, J., 1988 .. Studies on Continuous and Rapid Fermentation of Beer by Immobilized Yast. Qongye Weis Alengwu, 18.11.

BRIEF DESCRIPTION OF THE INVENTION

The present invention relates to a method for producing alcoholic beverages, comprising a continuous fermentation step, which is used to achieve a certain degree of fermentation and / or at least initial fermentation of beer must containing fermentable sugars.

In particular, a preferred method is provided wherein continuous fermentation is carried out in a gas lift (airlift type) bioreactor using flocculant (and especially high-flocculating or super-flocculating) yeast and applying rigorous oxygen control.

In a particularly preferred embodiment of the present invention, the "at least partially fermented" feedstock from the continuous process is fed to a batch processing step (which in the context of the claims of the present invention can ensure the completion of the fermentation process by converting the fermentable carbohydrates to alcohol without this step limiting the invention).

The present invention relates to the production of beer (in particular, including light beer, light beer and North American beer). See, for example, Essentials of Beer Style -F. Eckhardt.

The method of claim 1 is that the continuous step is carried out in a gas-lifting bioreactor.

In carrying out a continuous step that is effective in various processes according to the invention, it is preferable to use immobilized cells (as opposed to free cells only) and this can be accomplished using a selected immobilizer or flocculent yeast. However it does

it is preferable to use flocculent yeast instead of immobilized cells, in particular preferred over-flocculating yeast for this purpose.

More details regarding preferred technologies and advantages associated with continuous processes are provided in the course of the detailed description of the present invention. It involves the use of adjustable gas mixers and the use of nitrogen, carbon dioxide and oxygen as well as air.

In addition, more details are given regarding the technological stages of periodic detention. It should be noted that in some embodiments of the present invention, attention to the periodic retention process goes beyond the question of "completing" the conversion of fermentable carbohydrates to alcohol (which in any case can actually be completed in the continuous processing step) . In these embodiments of the technological stage of periodic retention, the main focus is on the attainment (or improvement) of flavoring, in particular with regard to diacetyl and acetaldehyde.

Preferred embodiments of the present invention include the distribution of the fermented to a certain extent and / or at least partially fermented wort through a branch line (fixed branch line or by selectively connecting and disconnecting pipes) to multiple reservoirs over a continuous stage detention. During the distribution process, one tank is filled, followed by another, etc. In a particularly preferred embodiment, the throughput of the continuous reactor and the periodic holding capacity are matched in terms of dimensions and number of reactors / tanks for periodic holding so that the flow rate of the product obtained is consistent with the capacity. The ideal case is when the final product is drained from the batch tank just in time to be cleaned, reconnected and then refilled with a product which is released from the continuous fermentation step.

According to another aspect of the present invention, certain embodiments of the invention specifically target the oxygen content of the must / beer. This applies both to the continuous fermentation and to the periodic retention stages of the process. With respect to the continuous step, the concentration of oxygen has different effects, but it is clearly desirable to minimize its content in order to optimize the conversion of higher alcohols into active esters that impart taste. In this respect, it should be noted that the periodic retention stage may largely not affect the concentrations of higher alcohols if strict control of the O ₂ content is applied in order to manage the balance of the taste-active fusel content. esters. In this connection, it may be useful to pre-purge the CO ₂ mustard prior to continuous fermentation.

In one embodiment of the invention, the main purpose of the continuous processing steps is to ensure that a high degree of fermentation is achieved during the subsequent periodic fermentation that takes place during the periodic retention process.

For the sake of security, the full contents of the priority documents as part of this description are attached, although they are reproduced in full in this description.

Detailed description

The following is a detailed description in two parts of aspects of the present invention.

The description contains and / or references diagrams, formulas, figures and the like, each of which is denoted by the term "Figure" followed by a specific identification number, and each of the attached figures is described and / or labeled "FIG. . "And a specific ID. In these figures (FIG.) The elements shown may not be in the exact scale.

Explanations of the attached figures:

FIG. 1 is a flow chart of an experimental continuous fermentation system; FIG. the individual components of the technological equipment shown are given in Table 5.1.

FIG. 2 is a schematic diagram of an experimental gas lift lift bioreactor equipped with a suction tube (GPST type bioreactor) with a capacity of 50 L.

FIG. 3 is a cross-sectional view of the reactor of FIG. 2, including the location of the internal suction pipe and the internal separator.

FIG. 4 is a detailed diagram of the upper portion of the bioreactor of FIG. 2.

FIG. 5 is a detailed diagram of the bioreactor housing of FIG. 2.

FIG. 6 is a detailed diagram of the tapered bottom of the bioreactor of FIG. 2.

FIG. 7 is a detailed diagram of the gas tube dispenser of the bioreactor of FIG. 2. A total of 160 openings (0.16 cm in diameter) were drilled into a 1.27 cm diameter tubular sprayer, 0.8 cm in length from center to center and 0.6 cm wide.

FIG. 8 is a scheme for the continuous production of granules using known static mixers (patent application No. 2,133,789, Labatt).

FIG. 9 is an image of a glass pellet strain supplied by Schott Engineering.

FIG. 10 - an image of Celite kernel pellets supplied by World Minerals.

FIG. 11 is an image of k-carrageenan gel granules manufactured in the laboratory of Labatt Brewing Company Limited (Labatt).

FIG. 12 shows images of non-flocculating yeast; FIG. yeast forming chains and flocculating yeast. These images were taken with a microscope camera at 100 X magnification.

FIG. 13 is a microscope image of a medium of flocculating yeast strain LCC3021 at 100 X magnification. FIG.

FIG. 14 - Microscope image of LCC290 superfluid yeast strain at 100 X magnification.

FIG. 15 is a schematic diagram of the process using a static mixer for the preparation of κ-carrageenan gel granules. In a static mixer, fluid moves through the mixer, ensuring that the fluids pumped from the pipeline are mixed.

FIG. 16 is another diagram of a gas lift bioreactor equipped with a suction pipe used for main beer fermentation.

FIG. 17 is a photograph of an active gas lift bioreactor equipped with a suction pipe shown in FIG. 15.

FIG. 18 is a detailed diagram of a gas lift reactor provided with a suction pipe of FIG. 16.

FIG. 19 is a detailed diagram of a front cover of a bioreactor, wherein: 1 is an orifice for drawing in the liquid for the oxygen sensor; 2 48

a temperature sensor sleeve connected to a thermostatic control unit; 3 - temperature detector; fluid return port for oxygen sensor; 5 - inoculum opening; diaphragm sample hole fitted with a stainless steel cap.

FIG. 20 is a diagram of a fluid withdrawal port for an oxygen sensor, which port is provided with a filling device immersed in the liquid phase of the bioreactor.

FIG. 21 - devices and flow chart of continuous main fermentation in a system using a bioreactor with gas lift (see Table 5.1, which describes the devices in detail).

FIG. 22 is a schematic representation of the mechanism of carrageenan gel formation (adapted by Rees, 1972).

FIG. 23 is a schematic representation of the use of kappa-carrageenan gel granules containing immobilized light yeast yeast used in the main fermentation.

FIG. 24 is an image of kappa carrageenan gel granules containing immobilized light yeast brewer's yeast at zero fermentation time.

FIG. 25 is an image of light beer brewer's yeast captured in kappa-carrageenan gel pellets after two days of fermentation, showing signs of budding on individual yeast cells.

FIG. 26 is an image of the outer surface of a kappa-carrageenan gel pellet showing yeast cells for light beer after two months of continuous fermentation.

FIG. 27 shows an image of yeast cells for light beer in an outer portion of a kappa-carrageenan gel pellet after two months of continuous fermentation.

FIG. 28 is an image of light beer yeast cells in the center of a kappa-carrageenan gel pellet after two months of continuous fermentation.

FIG. 29 - image of a whole gel pellet of kappa-carrageenan after six months of continuous fermentation.

Detailed Description - Part 1:

WE ARE BREADING AND INOCULATE PREPARED

In the fermentation process, a polyploid strain of Saccharomyces cerevisiae (also referred to as Saccharomycesuval-uni and / or Saccharotnyces carlsbergensis) was used in this thesis. The brewing community usually refers these yeasts to the lower-fermenting yeasts in the production of light beer. This characteristic is explained by the ability of the yeast to precipitate from the liquid medium at the end of fermentation. Beer yeast for the English beer type, as opposed to light beer yeast, rises to the top of the fermentation vessel, which is why it is known as the top fermenting strain. The ability of the yeast to precipitate or

it is not necessarily dependent on whether the yeast is for light beer or English beer, but is a species strain. Yeast for light beer does not usually ferment at temperatures above 34 ° C, while English beer does not ferment melibiosis. Researchers use these characteristics to distinguish light beer yeasts from those of English beer (McCabe, 1999). The Saccharonyces cerevisiae yeast flocculant medium registered under number 3021 of the Labatt Culture Collection is used for both self-aggregation and quarragenan-immobilized ferments. To conduct experiments involving the use of

immobilized over-flocculating yeast, a variant of strain LCC3021 named CC290 was used.

Pure yeast cultures were stored at -80 ° C cryogenic temperatures in a freezer located at the Labatt Technology Development Department. When sterile yeast culture is required, it is pre-cultured aerobically at 21 ° C on an arap PYG support (3.5 g peptone, 3.0 g yeast extract, 2.0 g CN ₂ PO ₄ , 1.0 g MgSO ₄ .7H ₂ 0.1, 0 g (NH ₄ ) S0 ₄ , 20.0 g of glucose and 20.0 g of arap dissolved in distilled water to one liter volume).

The isolated colonies of yeast were then transferred to a tube containing 10mL pasteurized beer must and incubated with stirring at 21 ° C for 24 hours. The volume of this inoculum was gradually increased to 5 L by adding the previous culture to a suitable volume of beer must (10 mL in 190 mL, 200 mL in 800 mL and 1 L in 4 L). The yeast inoculum was then centrifuged and centrifuged at 10,000 rpm and 4 ° C for 10 minutes. The resulting wet yeast residues (30% water) after drying gave the desired yeast mass for all subsequent fermentation processes.

THE FERMENTATION ENVIRONMENT

Technically pure beer must produced by Labatt London brewhouse is used as a natural environment for all fermentation processes. These theses provide explanations for the relative weight of the must, expressed as Plateau degrees (° P). Formula 4.1 describes the relationship between relative weight and ° P.

° P = 135.997 · SG ³ - 630.272 · SG ² + 1111.14 · SG - 616.868 (4.1)

The wort used in these theses is 17.5 ° P, which corresponds to a relative weight of 1,072.

Table 4.1 describes the carbohydrates in this must as measured by the liquid chromatography (HPLC) method described in section 4.7.2. Approximately 73% of the carbohydrates in this beer must are fermentable, while the yeast used in this study cannot easily shorten 27% of the longer carbohydrate chains.

Table 4.1. Composition of carbohydrates in the casting must used in the fermentation process. The wort is made by Labatt brewery in London and has a relative weight of 17.5 ° P.

Average (g / L) The degree of unevenness (%) fermentable (%) non-fermentable (%) fructose 3.3 18,0 1.9 Glucose 16.5 4.3 9.3 Maltose 87,7 10,1 47,8 Maltotriose 25,2 10.8 14.2 Maltotetrosis 6.4 18.3 3.6 Polysaccharides 41,1 9,1 23,2 Total 177,2 73,2 26,8

The non-uniformity coefficient for most of the analyzed substances ranges from 10 to 20%. This unevenness is due in large part to the method used to obtain industrial production, as well as to the unevenness of the raw material in the various brews.

IMMOBILIZATION TYPES

In this study, three types of immobilization capture, adsorption and self-aggregation were tested. For industrial media, data provided by the supplier must first be presented and then supplemented by laboratory analysis. Images and size distributions of test media (when available) are presented elsewhere in the description.

Two types of adsorption matrices were tested in a gas-lifted experimental bioreactor equipped with a suction tube (abbreviated as GPST type bioreactor). Images of these media are presented for description. Schott Engineering provides Siran® sintered glass pellet carrier. The selected particles are 1-2 mm in diameter, have open pores for immobilization of brewer's yeast, with a pore volume of 55-60% and a pore size distribution between 60 and 300 pm, which is a suitable size for yeast cells. This type of carrier is reported to be biologically and chemically stable, easy to clean, reusable, steam sterilizable, non-dense and taste-neutral, making it one of the approved foods in the food industry .

California-based World Minerals supplies a granular carrier composed of kieselguhr soil. The carrier has advantages such as thermal and chemical stability, mechanical strength and stability. Kieselguhr, the basic raw material, is commonly used in the brewing industry to filter beer. The Celiez R-632 carrier is specifically designed to immobilize cells.

Supplier specifications are as follows: Fractional composition: 0,595 mm to 1,41 mm (14/30 mesh) Average pore diameter: 7,0 pm

Total pore volume: 1.19 cm ³ / g

Compact layer density: 0.334 kg / m ³ .

Grip-based carrier kappa carrageenan gel pellets are manufactured in the laboratories of Labatt Brewing Company

Ltd. The manufacturing process is described in Section 5.2, and the results of the manufacturing process are presented in Section 6.2.1.

The simplest method of self-aggregation immobilization is possible by selection of flocculant yeast strains. The industrial yeast LCC3021 for the production of light beer has a natural flocculation capacity and is considered to be of medium flocculation capacity. As fermentation progresses, small yeast colonies of 0.5 mm to 1.0 mm in size are formed. The yeast LCC290, a variant of the LCC3021 light beer yeast, forms much larger flocculates (1.0 to 5.0 5.0 wt depending on the degree of agitation) and is therefore classified as over-flocculating yeast. Images of different yeast flocculates are presented.

SAMPLING PROTOCOL

During fermentation, it is necessary to take samples of the fermentation fluid at multiple intervals. To perform this task, sterile sampling valves were provided by Scandi-brew®. These valves are made of stainless steel and are equipped with a chamber in which ethanol can be stored to maintain an aseptic environment. Before sampling, ethanol is discharged from the chamber by removing a protective cap from the outlet at the lower end. Fresh ethanol (75% vol) was introduced into the chamber and then the cap was inserted into the opening at the upper end of the tap. The valve lever is then withdrawn and approximately 50mL of liquid sample is collected in a sterile container. A second sample containing excess flocculent yeast is taken so that appropriate deflocculation can be carried out before cell counting. After sampling is completed, the valve chamber is flushed with hot water, peracetic acid and finally with ethanol. The safety cap is placed on the outlet and the chamber is filled with ethanol to prepare for further sampling.

MICROBIOLOGICAL CONTINUOUS CONTROL

Free yeast cell count and viability by methylene blue method

Liquid samples containing freely suspended yeast cells are first taken from the fermentation medium by the sampling procedure described above. For the enumeration of

cells using a Hauser Scientific Company hemocytometer with a volume of 10 ⁴ mL in combination with an optical microscope. Liquid samples are diluted with distilled water to reach a total amount of yeast cells in a counting box of 150 to 200. Heggart et al. (1999) describes all the factors that affect the viability characteristics of the yeast.

To assess the degree of viability in

sample, using the methylene blue staining method described by the American Society of Brewing Chemists (Technical Committee and Editorial Committee of the ASBC, 1992). Living cells can decolorize methylene blue dye by oxidizing it. On the other hand, dead cells turn blue. The following reagents are used to prepare methylene blue used for viability assessment:

Solution A: 0.1 g of methylene blue in 500 mL of distilled water

Solution B: 13.6 g of KH ₂ PO ₄ in 500 mL of distilled water

Solution C: 2.4 g Na ₂ HPO4.12H ₂ O in 100mL distilled water

A methylene blue buffer solution was then prepared by mixing 500 mL of solution A with 498.75 mL of solution B and 1.25 mL of solution C to give a final mixture of pH 4.6.

A mixture of the diluted cell suspension and methylene blue is prepared in a tube by thorough mixing. After allowing the mixture to stand for a few minutes (contact between the cells and the dye is provided), a drop of liquid is placed between the hemocytometer counting glass and the cover glass (with a specified volume). The viable cell count is determined by counting both live and dead cells in the count box, after which the live cell count is subtracted from the total cell count.

4.5.2. Immobilized yeast cell count Self-aggregation

When using yeast cells with a tendency to form flocculates, it becomes difficult to accurately estimate the number of cells present in the liquid sample as the cells tend to precipitate in the beaker. A deflocculating agent is used to obtain a representative sample. In these experiments a 0.5% vol. Sulfuric acid solution was used to destabilize the flocculent yeast cells to provide a representative count of the yeast cells. The same procedure for counting and determining viability described in Section 4.5.1 is used with sulfuric acid to replace distilled water as a dilution agent.

Counting of immobilized yeast cells - gel pellets

It is necessary to destroy the gel matrix prior to the counting of the gel cells, using a Politron® apparatus (Brinkmann Instruments). First, the pellet samples were passed through a sterile sieve (500 pm mesh) and then washed with sterile water. One milliliter of the pellets containing gel cells and 19 mL of distilled water were added to a 50 ml vessel. The Politron® apparatus is then used to physically break the gel, thus dissolving in the liquid. The methods for counting and determining the viability of the destroyed gel samples are then carried out as described in Section 5.5.1.

Continuous pollution control

All fermentation processes carried out during this study are subject to continuous pollution control. The continuous control program shall include at least one check per week of the liquid in 50 liter fermenters for continuous fermentation, as well as the must in the tanks. Liquid samples are taken in accordance with sterility requirements and then spread on a culture glass in Universal Beer Agar (UBA, Difco Laboratories) and 10 mg / L cycloheximide. These test samples are then incubated at 28 ° C for up to 10 days under aerobic and anaerobic conditions. The selected cups are placed in containers containing a tablet of AnaeroGen®, which removes any oxygen remaining in the vessel and creates the desired environment for anaerobic development. Using an indicator strip (changes pink if oxygen is present) allows us to check that the environment is anaerobic. Bacterial contamination, if present in the liquid sample, should be detected by this method.

Establishing wild or non-brewer's yeast requires a separate growth medium that should not favor bacterial and / or brewer's yeast growth. Porous plates made with yeast medium (YM, Difco Laboratories), to which 0.4 g / L CuS0 ₄ was added, were used for selective growth of all possible wild yeasts (cultivation at 25 ° C for 7 days). Cultivation of the liquid samples applied to arap PYN (peptone yeast extract medium, Difco laboratories) for 7 days at 37 ° C ensures the establishment of non-light beer brewer's yeast. The growth of brewer's yeast for light beer is inhibited at temperatures above 34 ° C, and thus the growth on these plates would detect yeast contamination for English beer.

ANALYTICAL METHODS

Appropriate measurements shall be made of all technical equipment related to the experiments as required by standard operating procedures.

Ethanol

Analyzes were made for the concentration of ethanol in beer and fermentation samples, using the gas chromatography method described by the Technical Committee and Editorial Committee of the American Society of Brewing Chemists (1992). isopropanol vol.%, followed by injection of 0.2 μΙ_ of this mixture into a Perkin Elmer 8500 gas chromatograph. The following description shows the details regarding fine tuning of the gas chromatograph:

Flame Ionization Detector (FID)

Automatic sampling unit ------ ------- Purified kieselguhr for chromatographic purposes 102, 80-100 mesh, airtight.

Helium gas carrier flowing at a flow rate of 20 mL / min Injector temperature - 175 ° C, detector temperature - 250 ° C and column temperature 185 ° C isothermal.

Carbohydrates

Concentrations of glucose, fructose, maltose, maltotriose, maltotetrose, polysaccharides and glycerol were measured using high performance liquid chromatography (SpectraPhysicsSP8100XR HPLC). A cation exchange column (Bio-Rad

Aminex HPX-87K) with potassium phosphate (dibasic) as a mobile phase to separate these carbohydrates as they elute through the system. The quantity of the substances is then determined using a refractive index detector to determine the corresponding peak of the substance. HPLC operates at a counter pressure of 800 psi, the column temperature is 85 ° C, and the detector temperature is 40 ° C. Samples are degassed and diluted to the required level. Then, 10 μΐ_ was injected into the system with a flow rate of 0.6 mL / min.

In the brewing industry, another measurement is usually used to determine the level of liquid carbohydrates. The relative weight of the fluid, expressed by plateau degrees, is measured using an Anton Paar DMA-58 densimeter.

the filtered and degassed samples are transferred to a glass tube, which is then subjected to electronic vibrations. The frequency of oscillations through the liquid is measured and then correlation is made with a fluid of a certain density (d / 100 g or ° P). It should be noted that this measurement is an approximation with respect to the total concentration (or relative weight) of carbohydrates in the sample, since the measurements were made on aqueous sucrose solutions at a temperature of 20 ° C, the relative weight of which is the same as the flow rate. a must.

Vicinap diketones

The total concentration of diacetyl (2,3-butanedione) and 2,3 pentanedione was determined using a Perking Elmer 8310 gas chromatograph equipped with an electron-capture detector. Argon gas containing 5% methane was flowed at 1.0 mL / min and samples were passed through a J&W DB-Wax column. The injection temperature was maintained at 105 ° C and the detector temperature set at 120 ° C. The Hewlett Packard 7694E Automatic Sampler facilitates analysis. The quantitative content is

calculated based on the determination of the peak area of the selected component in the sample and then compared with a standard calibrated value.

In order to determine the "total" concentration of these substances, it is first necessary to equilibrate these samples at a temperature of 65 ° C and then hold them at that temperature for 30 minutes. Treatment of this sample for preliminary analysis ensures the conversion of α-acetolactate and hydroxybutarate into their respective diketones, diacetyl and 2,3butanedione.

Esters and higher alcohols

Some of the most important aromatic compounds found in beer are determined using a chromatographic method. Acetaldehyde, ethyl acetate, isobutanol, 1-propanol, isoamyl acetate, isoamyl alcohol, ethylhexanoate and ethyl octanoate were determined using n-butanol as an internal standard. A Hewlett Packard 5890 gas chromatograph equipped with a flame ionization detector, automatic sampling device and a J&W DB-Wax capillary column is used. The injection temperature was set at 200 ° C and the detector temperature was 220 ° C. The temperature profile of the heating chamber is as follows: 40 ° C for 5 minutes, raising the temperature from 40 ° C to 200 ° C at a speed of 10 ° C / min, raising the temperature from 200 ° C to 220 ° C at a speed of 50 ° C / min and finally held at 220 ° C for 5 min. Helium gas at 30 mL / min (28 psig), hydrogen flow at 50mL / min (25 psig), and airflow at 300 mL / min (35 psig) are added to the 6.0 helium gas carrier flow mL / min. The duration of the entire 1 mL sample cycle is 40 minutes.

Other analyzes

Several other analytical measurements have been made of the fermentation fluid to be aged and packed. The final product is subjected to analysis conducted by Labatt's quality control department, which standards the end product for beer. The standards described by the Technical Committee and the Editorial Committee of the American Society of Brewing Chemists (1992) formed the basis for these measurements. A list of the analyzes and a brief description of the relevant measurements are given in Table 4.2.

Table 4.2 Quality control analyzes and standards

Indicators Description The air Total amount of air carried during filling; the standard is less than 1 mL. Carbon dioxide Carbon dioxide saturation level introduced into the product; it is expressed in%, the standard being 2.75%. Sulfur dioxide The amount of sulfur dioxide in beer is measured by a gas chromatograph; the standard is <10 mg / L. Dimethyl sulfide The amount of dimethyl sulfide in beer (aroma of boiled cereal mass) measured by a gas chromatograph; the standard is <70 µg / 1_. Bitterness The degree of bitterness due to the hops in the beer; measured on the basis of alpha acids in beer; 1 unit of bitterness (BU | is equivalent to ~ 1 mg / L alpha-acid. Color Beer color was determined spectrophotometrically; absorbance of the sample at 430 nm with a light path of 0.5 inch. pH Measured with a calibrated pH meter; pH = -Jud ₁₀ [H ⁺ ] Phantom extract The amount of soluble mass present in the liquid without offsetting the effect of alcohol on the relative density of the liquid; measured with an air meter and presented as ° P; PE

Valid The same as the ghostly extract, except for extract the alcohol taken into account in this measurement; YES Calculated output Based on a Balling experiment at which of 2.0665 g extract extract, 1,000 g of alcohol was obtained; IEE = 100 * [(2,0665 * (% w / w alcohol) + PE) / (100 + 1.0665 * (% w / w alcohol))] Turbidity Beer turbidity at room temperature (21 ° C) without sludge distribution; is determined by nephelometric scattering analysis of light and is measured in units (FTU); the standard is < 200 FTU Initial clarity The clarity of the beer when cooled from the room on cooling temperature (21 ° C) to 0 ° C, without sediment distribution; the measurement method is as above; the standard is <100 FTU Sparkling Measurement of the foaminess of the beer using the NIBEM; the foaminess is created in the controlled site and the degree of foam drop is taken into account; the standard is> 170 seconds

LCC290 Yeast Deposition Protocol

Preparation of the yeast test sample The over-flocculating yeast (LCC290) is grown in wort as described in Section 4.1. This crop was then centrifuged at 4 ° C and 10,000 rpm for 15 minutes to obtain a yeast precipitate for subsequent inoculation. The wort described in Section 4.2 was pasteurized at 100 ° C for 60 minutes and then one liter was sterile transferred into 6 x 2L flasks. Each flask was inoculated with 4 g of centrifuged yeast. The flasks were placed in a vibrator operating at 135 rpm (21 ° C ambient temperature) and allowed to ferment. Remove the flask at the following intervals: 24 h, 40 h, 48 h, 64 h, 71 h and 192 h. A small sample of carbohydrate analysis fluid, yeast concentration and viability is taken at each interval (the methods are described in Chapter 4). The remaining liquid / yeast mixture is subjected to the yeast precipitation test described in section 4.7.2.

Yeast precipitation protocol

The rate of precipitation of over-flocculating yeast strain

The LCC290 is measured according to the following method: Each sample is fermented for a specified period of time as described in Section 4.7.1. At the specified time, remove the corresponding flask. Samples are stirred to ensure that all particles are suspended. The contents of the flasks were then immediately transferred to 1000 mL graduated cylinders. When flocculates are deposited, the distance between the liquid surface and the fluid flocculation boundary is measured at 30-second intervals. The sedimentation rate is calculated using the following formula:

Sedimentation rate = AH = Hn-Ht (5.1)

At (tt _f )

Using the standard method of Kynch (1952), the sedimentation rate curve can be determined depending on the concentration in the diagram in which the curves are plotted for each fermentation interval.

METHODS FOR CALCULATING PARAMETERS OF

CIRCULATION AND MIXTURE

An acid injection system coupled to a data acquisition system is used to determine the mixing time and the circulation rate in a three-phase gas-lift gas-powered bioreactor equipped with a suction tube (GPST bioreactor).

By using strong acid in impulse mode it is possible to

Calculate both the circulation rate and the mixing time by continuously monitoring the change in pH throughout and then using the equation presented in section 3.2.2. The data acquisition system consists of:

PH meter detector (Cole-Parmer, cat. # P-05990-90) connected to a microprocessor based on a pH transmitter (model 2300) DT2805 Data Transmitter 386DX Personal Computer and

Fast Data Acquisition Program (written by C. Hudson and J. Beltrano in 1994 and modified by N. Mensour in 1998).

Ten milliliters of 10 N hydrochloric acid are injected into the annular portion of the GPST bioreactor (diagram is given in Figure 5.5) just below the location of the pH detector. This distance corresponds to a height of 26 cm below the main slab. The pH detector was calibrated in two locations with standard buffers (Beckman pH 7.0 green buffer and Beckman pH 4.0 red buffer) checked according to standard requirements, and the calibration was performed before the mixing experiments. The 4-20 mA current obtained from the pH meter is connected to a terminal board, where the current data is transformed into voltage, which is then measured using a data acquisition circuit located near the computer. The data acquisition program starts at the same time as the acid injection. The data collection time is 5 minutes at 50 Hz. The system in the program is set to 3750 points (selected from a total of 15000 points collected).

The collected data is transmitted from the laboratory to a more powerful computer (Pentium II microprocessor) for further analysis. The 2D Curve Table (Jandel Scientific Software, Labtronics, Canada) is widely used for data analysis because of its capacity to distribute large numbers of data groups, and because of its many features in terms of data input (data smoothness, curve drawing) given points, etc.) The Savitzky-Golay algorithm is used to process the actual data to eliminate noise, using the time domain equalization based on polynomial equalization using the least squares method moving window. The equilibrated data is then adjusted to reflect a change in pH compared to the pH initially measured. The decreasing sinusoidal function adjusts to the regulated data. The mixing time and speed of

circulation is then calculated from the leveled parameters.

These blending experiments were performed in existing fermentation processes using 50-L gas-lifted bioreactors with one of the three immobilized carriers (or LCC290 superflocculent yeast or LCC3021 yeast medium or κ-carrageenan gel granules). The liquid phase fermented beer with a relative weight of 2.5 ° P and the gas phase represented carbon dioxide gas. The fermentation temperature was controlled at 15 ° C. The surface velocities of the dispersed gas vary between 2.0 and 6.0 mm / s. At a given gas flow, the system is left to equilibrate for 10 minutes. The acid injection test can then be started and repeated 3 times. The next gas flow is selected and the pH of the mixture in the reactor is adjusted again to the starting level.

CONSTRUCTION OF AN EXPERIMENTAL SYSTEM BY

BIOREACTOR WITH GAS LOWER POWER SUPPLIED WITH SUCK PIPE

Fermentation system including bioreactor with

GAS LAND POWER FITTED WITH A TUBE PIPE

Fluidized bed systems and the use of suction tubes have shown their advantages when used in three-phase systems.

Two experimental bioreactors, equipped with suction tubes using gas lift (BPR bioreactors), were constructed and installed in the experimental brewery of Labatt Brewing Company Ltd. to carry out the experimental work for this study. Several existing vessels have been reconstructed to store the must and to collect the beer. The flowchart presented in FIG. 1 shows the whole process used in the continuous fermentation experiments, and Table 5.1 describes the equipment presented in FIG. 1.

The wort is supplied by the London Brewing installation through a 5.08 cm stainless steel pipeline and transported to 1600 L. beer must (WT1 & WT2) storage tanks with two tanks. It is possible to ensure continuous availability feeding the nutrient medium to the experimental fermenters (RI and R2) with continuous action. Each tank is equipped with an oxygen spray system and homogenization control, as well as a glycol cooling jacket for temperature control. This central nutrient source feeds 3 independent fermenters through a distribution chamber equipped with valves (V7, V8 and V9). Masterflex peristaltic pumps (PI and P2) are used to transfer the flow from the wort to the experimental bioreactors (RI and R2).

Table 5.1. Description of the individual devices and process line details shown in Figure 5.1

Item Description

Air 100 psig air supplied from a London installation CO ₂ 100 psig carbon dioxide supplied by a London installation

F1 Sterile inlet of carbon dioxide inlet

F2 Carbon dioxide inlet sterile filter

F3 Sterile exhaust filter. vent opening. channel on WT1 F4 Sterile filter on the exhaust outlet of the ventilation duct WT2 F5 Sterile inlet gas filter F6 Sterile inlet gas filter F7 Sterile WBT1 outlet filter GLR Glycol return line; 25 psig GLS Glycol delivery line; 45 psig NV1 Needle valve for carbon dioxide NV2 Needle valve for carbon dioxide NV3 NV4 Air inlet needle valve Needle valve for carbon dioxide inlet NV5 Air inlet needle valve NV6 Needle valve for carbon dioxide inlet P1 Masterflex peristaltic pump for P1 P2 Masterflex peristaltic pump for P2 PR1 Carbon dioxide pressure regulator PR2 Carbon dioxide pressure regulator PR3 Air pressure regulator PR4 Carbon dioxide pressure regulator ф PR5 Air pressure regulator PR6 Carbon dioxide pressure regulator R1 Experimental bioreactor; 50 L working volume R2 Experimental bioreactor; 50 L working volume RM1 Rotameter for measuring the flow rate of CO _2; 0 to 20 scfh rock RM2 Rotameter for measuring the flow rate of CO _2; 0 to 20 scfh rock RM3 Rotometer for measuring air flow; 0 to 2.5 scfh rock RM4 Rotameter for measuring the flow rate of CO _2; 0 to 10 scfh rock RM5 Rotometer for measuring air flow; 0 to 2.5 scfh rock RM6 Rotameter for measuring the flow rate of CO _2; 0 to 10 scfh rock

V1, V2, V3 Throttle Valves on WT1 Inlet / Outlet

V4.V5.V6 Throttle valves of WT2 inlet / outlet

V7, V8, V9 Beer dispenser ball valves

V10, V11 Ball valves for beer inlet inlet

V12 High-speed shut-off valve for inlet manifold gas

V13 Exhaust outlet valve of R1

V14.V15 Ball valve of the inlet opening for the must of R1

V16 High-speed shut-off valve for inlet manifold gas

V17 Exhaust outlet valve of R2

V18 WBT1 Exhaust Throttle Valve

WBT1 Young Beer Tank; 200 L working volume

WT1 Casting storage tank equipped with glycol shirts and spreaders; working volume 1600 L

WT2 Casting storage tank equipped with glycol shirts and spreaders; working volume 1600 L

Carbon dioxide and air are sprayed into the gas lift system through a sprayer constituting a stainless steel pipe (FIG. 7). Rotometers (RM3, RM4, RM5 & RM6) are used to control the flow rates of the streams spread in the system. Sterile filters (0.2 μτη mesh) are mounted on the gas pipes to prevent contamination of the bioreactor. The product flows from the reactor through an overflow system (FIG. 4). The power pump controls the residence time. Both reactors (R1 and R2) are connected to the Young Beer Tank (WBT1), which collects the overflowing liquid. Special 50 L collection tanks are used to collect and process the product as needed.

A more detailed diagram and exact dimensions of the 50-L experimental bioreactor are given in Section 5.1.1. The data for the processing and storage of the must must be presented in Section 5.1.2, while the purification and sterilization of the continuous fermentation system is discussed in Section 5.1.3. Section 5.1.4 describes the fermentation process to which this study relates.

Reactor design and performance

The 50-L displacement bioreactor designed for this study is made entirely of 304L stainless steel with 4 plexiglass viewing windows located in the reactor vessel so that particle and fluid motion can be monitored. The construction material was chosen because of its resistance to sanitary chemicals (caustic and acid), as well as its durability in sanitary steam treatment. Another important aspect of the construction is the minimized direct contact of the threaded devices with the fermentation medium. The holes are welded and fittings are used when necessary. The reactor is designed with an extended main section to increase the rate of gas evolution and thus to provide better transfer of the liquid-solid mass (Chisti & Moo-Young, 1993). The bottom of the reactor is conical with a cone angle of 90 ° to minimize solids collection at the bottom.

In FIG. 2 schematically shows experimental 50-L systems installed in Labatt's experimental brewery. This diagram shows the location of the gas inlet, the inlet for the liquid, the glycol cooling jacket, the exhaust outlet for the product, the temperature sensor and control system, and the location of the sampling holes for sanitary purposes.

In FIG. 3 shows the same GPST bioreactor with dimensions in cm.

FIG. 4 to FIG. 6 is a detailed schematic diagram of a 50 L bioreactor GPST with FIG. 7 is a diagram of the gas dispenser used in these experiments.

The internal suction pipe and the particulate separator are shown in FIG. 3. The ratio of the diameter of the suction pipe to the diameter of the reactor is 2/3 and was selected on the basis of literature (Chisti, 1991). The particulate separator is sized to provide better gas separation from the solid-liquid mixture. By increasing the diameter of the particulate separator (20.32 cm compared to the suction tube diameter 10.16 cm), it is possible to achieve a greater difference between the rate of bubble rise and the rate of lowering of the fluid. The intake of gas in the ring between the cylindrical and conical part of the suction pipe of the system will be reduced, resulting in better transfer of the solid-liquid mass.

A tubular sprayer (FIG. 7) was constructed for a carbon dioxide gas mixture in the suction pipe section. A total of 160 holes with a diameter of 0.16 cm were drilled into a tubular sprayer with a diameter of 1.27 cm. The openings are located 0.8 cm longitudinally from the center to the center, and in the transverse direction from the center to the center - 0.6 cm (8 rows of 20 holes each). Since mixing is a primary function of the spray gas, aperture diameter of 0.16 cm is selected.

Protocol for processing and storing beer must

In conventional fermentation methods, wort is not held for a long period of time without yeast. Oxygen-rich beer must is an excellent growth medium for many organisms, including yeast. Since the study of continuous fermentation in gas lift systems requires large quantities of yeast to be kept, it was necessary to investigate the transfer of must and its storage. The researchers believe that oxygen-free cold must must

ΊΟ

it should be stored for up to two weeks without contamination if the rainwater is transferred to storage tanks appropriately.

It is also necessary to provide adequate control of the pouring must temperature in the storage tank. In addition to the must tanks, suitable reservoirs for the experimental brewery have been constructed for use in fermentation processes. A test has been carried out on the ability of these tanks to maintain the required temperature. In FIG. 2 clearly illustrates that these tanks cannot be used without stirring unless a constant temperature of 4 ° C is maintained. Initially, the must must be introduced into these vessels at 4 ° C. Temperature measurements are made at several points in the tank to give a better picture of the actual temperature. When the liquid is left in the vessel for 24 hours without stirring, the temperature of the liquid near the top rises to approximately 24 ° C. In the middle of the tank, the temperature of the liquid also slightly increases (DT ~ 3 ° C), while in the cone of the vessel it remains close to the original temperature of 4 ° C.

By applying gentle stirring during the injection of carbon dioxide at a flow rate of 0.133 cm ³ / h in the collecting tank, it is possible to maintain a pouring must temperature of 4 ° C. For this purpose, pipes with a diameter of 2.54 cm are installed in the conical bottoms of the two storage tanks for the must, for stirring the must.

Non-aerated beer must from the Labatt installation, London, is transferred to a buffer vessel through a stainless steel pipeline of 5.08 cm in diameter. From this vessel, the must must pass through a pasteurizer and be fed into the WT1 or WT2 tanks, where the must must be stored at 4 ° C for no more than 2 weeks. The pasteurization step is carried out on site as a precautionary measure,

ensuring the elimination of unwanted microorganisms in the wort throughout the storage period. By using oxygen-free wort, the decomposition of warm oxygen mustard (formed by impure aldehydes) is minimized. In addition, oxygen intake through continuous gas fermentation can be closely monitored by the introduction of the spray gas.

Measurements of dissolved oxygen in the wort in the storage tank are made once. In FIG. 5.9 shows the dependence of the concentration of oxygen dissolved in the must in relation to the retention time. In the first example, the wort has been transferred to the tank and carbon dioxide (0.085m3 / h) has begun to spread to provide the necessary temperature control. The concentration of oxygen in the wort increases to the first day and reaches a value of approximately 1.3 mg / L and subsequently decreases to approximately 0.1 mg / L. In the second case, before filling with must, the tank is flushed with 0.85 m ³ / h of carbon dioxide in advance for a period of 3 hours. The initial high oxygen concentration sharply decreases and the desired oxygen content in the must (<0.1 mg / L) reaches within 2 days.

On the last attempt, the tank was pre-purged as described above, and continuous purging (0.085 m ³ / h) was carried out during the filling of the vessel as well as during the storage period of the wort. Dissolved oxygen is maintained at a minimum level (<0.1 mg / L) throughout the storage period of beer must. This method is adopted in all studies related to storage of beer must.

£ 15 &

I 10 s f h5

......... - • with agitation ▲ without agitation

A

--------- A —------- • · · · 1 1 *

-J ----------- 1 ----------- 1 ----------- 1 ----------- 1 ----------- 1 —'------- Mr

012345678

Measurement location Figure 5.8 - Effect of stirring by gas spraying as a function of temperature in a cylindrical conical tank for storing beer must. The wort is fed into the vessel at 4 ° C. Measurements were made 24 hours after filling the vessel. Carbon dioxide is sprayed into the vessel at a flow rate of 0.113 cm3 / h. Ambient temperature is 19.8 ° C. Location of measurements: (1) Top of tank near wall; (2) Top of tank in center; (3) The middle of the tank at the wall; (4) The center of the tank in the center;

c (5) Top of cone at wall; (6) Upper cone in center; (7) Lower part of cone.

Beer must in which 0,085 m 3 / h of carbon dioxide is sprayed after filling the tank

Beer tank pre-purged for 3 h with 0.85 m 3 / h gas

Beer tank pre-purged and refilled Figure 5.9 Control of dissolved oxygen in the must during its transfer from a London installation. Three filling methods were evaluated, the third being a combination of the first two methods. The pouring must temperature is maintained at 4¾.

5.1.2 Purification and sterilization

The must storage tanks are subjected to cleaning consisting of prewash with hot water (85 ° C), wash with caustic soda (40% caustic soda at 60 ° C, followed by hot water wash (85 ° C) The cleaning of these vessels is accompanied by the contact of the walls with a solution of peracetic acid (2% (w / v)) .The piping system of the wort is subjected to the same cleaning regime.

50-L capacity bioreactors are subjected to various purification and sterilization processes. The systems were washed with hot water (60 ° C) and then filled to the top with warm water (40 ° C). Diversol CX / A (DiverseyLever, Canada) cleaning agent was then added to this water to form a 2% solution (w / v). Air at a surface velocity of 5 mm / s is sprayed from the bottom of the reactor to ensure good dissolution and necessary contact within the reactor volume. After one hour of contact, the reactor was emptied and washed with a strong stream of fresh water. These cleaning procedures are repeated six times, ending with the last two

cold water-emptying cycles.

The young beer tank was cleaned using a 2% (w / v) solution of Diversol CX / A. Unlike bioreactors, no spray is used, as WBT is not a sprayer. Mechanical agitation by rolling the vessel is applied. The cycle is repeated twice, followed by water-emptying twice.

Prior to steam sterilization, 50-L bioreactors are connected to the young beer tank and the beer must feed tube is interrupted, as is the gas spray pipe. The valves, respectively valves V10, V11, V12, V13, V14, V15, V16, V17 and V18 are open and filters F5, F6 and F7 are removed. The gas pipe and these filters get out

subjected to sterilization under pressure individually for 15 minutes at 21 ° C. The steam is supplied via taps V12 and V16. They open slowly to minimize the possibility of equipment damage, and the reactor temperature is monitored closely. One-hour sterilization is performed, reaching a temperature of 100 ° C inside the reactor. The valves, respectively the valves V10, VII, V14, V15 and V18, are closed first. The steam supply is then stopped and the sterilized filter F7 connected. The sterilized filters F5 and F6 are immediately connected to the gas supply pipe and carbon dioxide is introduced at a surface velocity of 3 mm / s. This gas stream not only prevents the reactor from collapsing during cooling, but also displaces the air present in 50-L gas-lift bioreactors.

The wort feed tube is connected to the bioreactor after an internal temperature of 20 ° C is reached. With the valves V2, V5, V10 and V14 in the closed position and the valves V6, V7, V8, V9, V11 and V15 in the open position, steam is supplied by either V3 or V6, depending on from where the rainwater is coming from. The steam treatment cycle lasts for 1 hour, after which the valves V9, V11 and V15 close simultaneously with the steam supply. After the pipes reach room temperature (20 ° C), valves V3 and V6 are closed and steam supply is stopped. At this point, the entire continuous fermentation system including the yeast tank, 50 L bioreactors and the young beer tank are sterilized and ready for fermentation.

5.1.3 Fermentation Protocol

Experimental gas lift bioreactors equipped with a suction tube (GPST type bioreactors) are 50-L in volume and used for fermentation of the must. Glycolic

a thermal jacket provides temperature control while maintaining a liquid temperature of 15 ° C in all fermentation experiments. Each reactor is equipped with temperature detectors for measurement purposes and thermocouples, as well as an automatic solenoid actuator valve to regulate the glycol feed to the reactor, gas lifting fermenters are also provided with means for supplying the main mixing gas (carbon dioxide) or hydrogen) as well as with air-supply means for dispensing oxygen. The desired gas mixture is selected by adjusting the rotameter / needle combination, after which the gas mixture is passed through a sterile filter (Millipore, Millex®, -FG ₅₀ , 0, 2 gm) and then into the suction tube of the bioreactor. Air was injected into the reactor at a surface velocity of 0.39 mm / s (0.4 scfh) at all fermentations, while adjusting the flow rate of the main mixing gas to suit the type of concrete and mobilization.

Before starting continuous operation, the 50-L gas-fired bioreactor is put into operation in the established batch mode. After cleansing and sterilization as described in section 5.1.2, the bioreactor is filled with 50 liters of wort from the wort storage tanks (WT1 or WT2) and then through the sterile Scandi-BreW® liquefaction orifice. inject 200 g of yeast (4g / L). When k-carrageenan gel pellets are used, 20 L gel pellets are injected into the reactor to provide 4 g / L initial concentration of flocculent yeast medium LCC3021, which is considered system-regulated for continuous operation mode.

The fermentation medium (beer must) is continuously charged through the bottom of the reactor, while the "green beer is poured through a funnel opening located at the top of the reactor.

reactor. Since the reactor working volume is a constant value, the average residence time of the liquid in the reactor is adjusted by determining the flow rate of fresh beer must delivered to the reactor. Samples of liquid for chemical and microbiological analyzes are drained daily from the reactor through a vent and with a sterile valve (Scandi-Brew®) (methods described in Chapter 4 are used).

At certain times, the product resulting from continuous fermentation is removed from the 50 L bioreactor, where the main fermentation takes place, is collected in 40 L sterile stainless steel cans and subjected to post-fermentation treatment to obtain the final product - marketable beer, which is evaluated and compared to the control of industrially produced beer.

The selected 50-L bioreactor is disconnected from the young beer tank and immediately connected to the beer collection container. Once the desired amount of beer has been collected, the bioreactor reconnects with the young beer tank. The harvested green beer is subjected to a post-fermentation retention period to reduce the level of liquid diacetyl below 30 µg / L. The yeast transferred with the liquid precipitates and the liquid (cell concentration 1-5 million cells / mL) is transferred to a cold aging vessel (7 days at 2 ° C). After the aging period, the liquid was filtered, diluted to an alcohol content of 5% vol and carbonized before being poured into 341 mL bottles. The bottled liquid is pasteurized in a Labatt installation.

5.2. CONTINUOUS METHOD OF PREPARATION OF GRANULATED GEL

The object of this section is experimental work on the basis of which to evaluate a continuous process for the preparation of granulated gel with inoculated brewer's yeast in order to submit

immobilized yeast cells LCC3021 to tubular bioreactors with a gas lift capacity and volume of 50 L as described in section 5.1 and operating in continuous mode.

The manufacturing process (FIG. 8) involves the formation of an emulsion of a non-aqueous homogeneous phase (vegetable oil) and a water-dispersed phase (κ-carrageenan gel solution mixed with yeast cells), using static mixers. After this step, a sharp cooling is applied to induce gel formation from the polymer. The formed granules are then placed in a solution of potassium chloride, which helps to harden the granules and separate them from the oil phase.

The formation of a granular gel emulsion of κ-carrageenan is carried out at a temperature of 37 ° C, which is maintained in a water bath to prevent premature gelation of the carrageenan gel. The sterilized polymer was kept at 37 ° C using an adjustable temperature water bath, and the yeast inoculum was kept at 20 ° C before immobilization. Using Masterflex peristaltic pumps (Cole Parmer Company, USA), the gel and brewer's yeast are pumped through 24 6.4 mm diameter elements into a static mixer where the cells are dispersed uniformly in the gel. Using a Masterflex peristaltic pump, the sterilized oil stored at room temperature is transferred to a hot water tank to reach a temperature of 37 ° C.

The inoculated polymer (aqueous phase) is then mixed with the oil (homogeneous phase) through another group of static mixers to produce the desired emulsion. This emulsion was cooled sharply to 5 ° C in a water / ice bath, causing the polymer droplets to gel in granules. The granules are then placed in a solution of potassium chloride at a concentration of 22 g / L, which increases their strength and ensures their separation from the oil phase. This oil is recycled back to the oil tank, and the aqueous phase (granules and potassium chloride solution) is transferred to a separate container for size classification and then fed to 50 L bioreactors.

5.2.1 Static mixers of the Kenics type

The basis of this new granulation process is static Kenics type mixers (Cole Parmer Instrument Company, Niles, IL, USA). They are constructed of a series of tube-mounted stationary elements with an inside diameter equivalent to the diameter of the static mixer. These elements form transverse channels that help to separate and re-mix the liquid flow through the static mixer. This blending system achieves the transverse rupture of these fine jets and turns them into an increasingly homogeneous emulsion. Table 5.2 lists three types of static mixers used in this study.

Table 5.2 Indicators of static mixers used (supplied by Cole Parmer)

Model Diameter D (mm) of static mixer Number of elements (N _r ) G-04667-04 6.4 12 G-04667-06 9.5 12 G-04667-08 12.7 12

5.2.2 Gel production materials

The two main materials in the production of granulated gel are oil and polymer, k-carrageenan (type X-0909, lot 330360, Copenhagen Pectin, Denmark), which is a polysaccharide polymer extracted from red algae, was provided to us by Copenhagen Pectin A / С. This polymer has unique thermo-gelling properties, whereby its gelling temperature depends on the concentration of κ-carrageenan and potassium chloride [KCl]. The polymer was dissolved at 80 ° C in distilled water containing 2.0 g / L of KCI to a concentration of 30 g / L. The resulting gel solution has a gelation temperature of 28 ° C. The gel was autoclaved for 1 hour at 121 ° C and then placed in a 40 ° C water bath to prevent it from solidifying. Corn oil, industrial production (Pasquale Bros. Inc., Canada) is also sterilized for 1 hour at 121 ° C and then stored at room temperature 20 ° C until it is put into use. Brewer's yeast is prepared as described in section 4.1.

5.2.3 Measurement of pellet diameter

Pellet samples were placed in flasks containing 100 mL of a 22 g / L KCI solution. The granules remain immersed in this solution for 2 hours to increase their hardness. The oil is removed from the aqueous phase by washing with potassium chloride solution. The samples were then stored at 4 ° C to prevent microbial contamination prior to the assays.

The diameter of the pellets is measured using Optimas imaging software (Version 4.02, BioScan, Inc, USA) equipped with an Optimas camcorder (Version 4.02, BioScan, Inc, USA). A pellet sample is placed in a Petri dish coated with a fine water film (used to separate the pellets from each other), and then the dish is placed under the chamber. A total of 300 to 400 pellets are measured using this system. Optimas software capabilities range from 100 pm to several mm with a maximum absolute error of 30 pm. Optimas' data is analyzed using Microsoft Excel. The resulting sample size distribution is characterized by the mean sample diameter (D _B ) and the non-uniformity coefficient (COV).

5.2.4 Evaluation of granules - experimental design

Static mixers of three diameters (D _s = 6.4 mm, 9.5 mm and 12.7 mm) were compared to determine what type of granule should be obtained by measuring the average diameter of the sample granules and the coefficient of irregularity in terms of size distribution. The number of elements (N _d ) of static mixers varies from 12 to 120, while the volume of polymer fraction (Axis) is between 8.3% (v / v) and 50% (v / v) of gel in the oily solution. At 8 _C, more than 50% of the dispersed phase (gel) and the homogeneous phase (oil) become inversion, that is, inclusions of oil droplets into the polymer matrix instead of gel droplets in the oil matrix. The surface velocity of the oil / gel emulsion fluid through the emulsion region is adjusted from 3.6 cm / s to 17.8 cm / s. The surface velocity of the fluid (V _S l) through the static mixer is calculated by the following equation:

Vsl = (Qoil + Qcar) / S, where S is the cross-sectional area of the tubes located in the static mixer, Q _O ii is the volumetric flow rate of the oil phase and Q _car is the volumetric velocity of the gel carrageenan solution.

RESULTS AND DISCUSSION: FERMENTATIONS AND DYNAMICS OF

MIXING IN BIOREAKERS WITH GAS GROUND AND 50 L CAPACITY

The use of immobilized cells for ethanol production has been published. Over the past two decades, research has focused on optimizing ethanol production by combining the technology of immobilized cells with continuous production (Kuu, 1982; Gil, 1991; Maiorella, 1983). Many of them have great success with the use of continuous systems with the use of immobilized cells in the production of

ethanol became an industrial process. But implementing such a continuous process in the main fermentation of beer in the brewing industry is not so easy. Beer consists not only of ethanol, but also of a large number of flavoring compounds that create complications. The following section describes the results obtained by the author in the production of beer in an experimental fermenter of the GPST type.

6.1 PERIODIC FERMENTATIONS IN THE EXPERIMENTAL

GPST TYPE SYSTEM

Periodic fermentations using freely suspended yeast cells are carried out in an experimental tube bioreactor with a gas lift capacity of 50 L. These experiments provide a favorable opportunity to determine the feasibility of using such a brewing yeast fermentation system to obtain beer. . In addition, these experiments serve as a basis for comparison with the continuous fermentation processes of liquids. Two periodic fermentations were carried out in a 50 L bioreactor, which used a strain to produce a beer of the type of light beer from the Labatt Cultural Collection (LCC3021). During the fermentations, the rate of growth of the yeast was observed as well as the consumption of the nutrient medium.

Figures 6.1 and 6.2 show the yeast concentration and fermentation viability of lot 1 and lot 2. In both cases, the yeast growth corresponds to the known growth rates indicated in the literature. The viability, measured with methylene blue, remains high all the time and is close to 90 percent. The carbohydrate concentrations for Lot 1 and Lot 2 are presented in Figures 6.3 and 6.4. Monosaccharides, glucose and fructose were taken up first by the yeast, followed by the consumption of maltose and maltotriose. The level of maltotetrose as well as polysaccharides remains unchanged during fermentation.

Ethanol is one of the most important by-products of yeast metabolism. Anaerobic fermentation produces, under optimal conditions, about 48 g of ethanol and 47 g of carbon dioxide per 100 g of glucose metabolized. Small amounts of glycerol (3.3 g per 100 g glucose) will also be obtained, with this by-product involved in maintaining the redox balance in the fermenting yeast, as well as maintaining in the cells the osmotic balance, especially in the hypertonic environment, 6.5 and Figure 6.6 illustrate the change in the concentration of ethanol and glycerol depending on the duration of fermentation. Ethanol content levels rise very slowly at the start of the fermentation process due to the presence of oxygen in the fermentation medium, as the yeast cells are in the aerobic growth phase. Next

as oxygen is depleted, ethanol levels rise exponentially as fermentable sugars are consumed, and at that point concentration levels begin to fall.

Higher diketones are also very important by-products of yeast metabolism. The total diacetyl and pentanedione concentrations for batches 1 and 2 are given in Figures 6.7 and 6.8. These substances increase to approximately 40 hours in the fermentation medium, corresponding to the peak levels of yeast concentrations, and are the result of the synthesis of amino acids that the yeast produces during the growth phase. During the last fermentation portion, diacetyl, pentanedione and their precursors aacetolactate and α-ketobutarate are converted into their less aromatic active diols respectively.

From the results presented in this section, it is clear that the two periodic ferments have proceeded normally in a gas-lift bioreactor equipped with a suction tube. Yeast growth and carbohydrate growth are expected to occur as well as by-products, ethanol, diacetyl and pentanedione. Comparison of the batch parameters shows that both ferments have proceeded in a similar manner. Figure 6.9 compares the ethanol concentration for batches 1 and 2. The specific curves for the batches concerned have the same profile at many points, partially overlapping each other, thus showing the degree of repeatability. Although the sequence of substrate consumption and subsequent product extraction does not change in the gas lift system, the fermentation rate is different. Completion of the main fermentation, represented by the peak ethanol concentration, is reached in both batches in about 80-85 hours.

Reduction of diacetyl below 30 µg / l is achieved for an additional 20 hours. These results suggest that the main fermentation of high quality beer must can be completed in about 100 hours, compared to the classical periodic fermentation of light beer for 120-168 hours. The agitation performed in the GPST type bioreactor contributes to reducing the fermentation time due to the increase in mass transfer provided by such systems. With the help of this information, future fermentation experiments were conducted with the belief that the GPST type bioreactor did not significantly alter the metabolism of the yeast fermentation and that fermentation with free-suspended yeast in the system could reduce the periodic fermentation by at least 20 hours.

Viability

Figure 6.1 Total cell concentration and viability depending on the duration of LCC3021 yeast fermentation in lot 1 conducted in a GPST-type experimental bioreactor.

Cell concentration

Duration of fermentation (h)

Viability

Cell concentration

Figure 6.2 Total cell concentration and viability depending on the duration of LCC3021 yeast fermentation in lot 2 conducted in a GPST-type experimental bioreactor.

Polysaccharides

Maltose

Maltotetrosis

Fructose

Maltotriose

Glucose

Figure 6.3 Concentration of carbohydrates according to the duration of fermentation with yeast LCC3021 in lot 1, carried out in a GPST-type experimental bioreactor.

—O— Polysaccharides —B — Maltotetrosis — Maltotriose

,. f Maltose-D-Fructose-F-Glucose

Figure 6.4 Carbohydrate concentration depending on the duration of LCC3021 yeast fermentation in lot 2, conducted in a GPST-type experimental bioreactor.

Ethanol-O-Glycerin

Figure 6.5 Ethanol and glycerol concentration depending on the duration of lot 1 fermentation with yeast LCC3021 conducted in a GPST-type experimental bioreactor.

Ethanol Glycerin

Figure 6.6 Concentration of ethanol and glycerol depending on the duration of fermentation with yeast LCC3021 of lot 2, carried out in a GPST-type experimental bioreactor.

Diacetyl

Pentandion

Figure 6.7 Concentration of diacetyl and pentanedione according to the duration of LCC3021 yeast fermentation in lot 1 conducted in a GPST type experimental bioreactor.

Diacetyl

Pentandion

Figure 6.8 Diacetyl and pentanedione concentration depending on the duration of lot 2 fermentation with yeast LCC3021 conducted in a GPST-type experimental bioreactor.

♦ Lot 1Q Lot 2

Figure 6.9 Comparison between lot 1 and lot 2 with respect to ethanol concentration depending on the duration of fermentation. Yeast LCC3021 was used in an experimental bioreactor of the GPST type.

6.2 IMMOBILIZED CARRIERS

Several media were examined during the study to determine the best alternatives for future research. Three different ways of immobilization in a 50 L type GPST bioreactor were investigated. Two adsorbent carriers, which are commercially available and have sizes between 1 and 2 mm, have been evaluated. Siran®, a glass granular carrier supplied by Schott Engineering (FIG. 9), and Celite®, diatomite (kieselguhr) granules supplied by World Minerals (FIG. 10), were examined for their ease of storage and availability as commercial products. Adsorption-based carriers provide a suitable opportunity for more aseptic operations, since the reactor can first be loaded with the carrier, then sterilized on site and finally the yeast inoculated directly into the reactor. From an industrial use point of view, this option is very attractive as the carrier does not require special storage and the installation will not have a significant change in the inoculation modes.

Initial fermentation results in a 50 L LDPE bioreactor using these two carriers were unfavorable. The problems encountered were mainly due to the high particle densities of Siran® and Celiez® compared to the liquid medium. Systems involving three phases in the suction pipe and using gas lift work better when the ratio of carrier densities to liquid is maintained close to one. In the case of Siran®, this ratio is 1.34, and when using Celiez® this value is 1.31. The result of such high differences in the densities of the solid and liquid phases is a significant increase in the minimum value of the gas fluidization rate that is required for the operation of a 50 L type GPST bioreactor. For 4 liters of Siran® carrier (8

% (v / v) particulate charge) A gas velocity of 21.5 mm / s (based on suction pipe diameter) is required to achieve circulation. This higher gas velocity is not a significant problem when the test is carried out in aqueous solution, but since the liquid medium is a must of wort, damage with catastrophic effects occurs in the GSTP system. The required gas velocity causes excessive foaming in the reactor, which ultimately reduces the level of liquid under the suction pipe, thus stopping the circulation of the liquid and solid phases. A malfunction similar to the one described above when using Siran® is also observed when using Celiez® as immobilizer material. Due to these results, in the subsequent fermentation experiments using gas lift, we abandoned these two carriers on the basis of adsorption. The use of a carrier-based carrier such as κ-carrageenan allows the system to be charged with solids at 40% (v / v), requiring about 0.17 standard cubic meters per hour of gas (5.8 mm / s surface velocity of the gas) to effect fluidization and subsequent circulation. The positive results are explained by the operation of the system with carrageenan-encapsulated yeast cells, in which fluidization is easier due to the lower carrier density (about 1100 kg / m ³ ). The desired feeding of solids from the LCC 3021 (medium flocculant) and LCC290 (super flocculant) solids is also achieved, and gas velocities of approximately 3 mm / s are required to provide circulation. In the section

6.2.1 describes in more detail the g carrier of k-carrageenan and in section

6.2.2 - self-aggregating beer yeast flocculates LCC3021 and LCC290, which are evaluated as immobilized continuous main fermentation matrices in a GPST-type fermenter with volume

50L.

6.2.1 K-carrageenan gel pellets

Capture-based immobilization methods require the incorporation of yeast cells into the matrix before being introduced into the fermentation vessel. Because in-situ inoculation in the reactor was not feasible, it was necessary to obtain the gel pellets prior to the start of fermentation. It is not yet clear what effect the long storage period of the gel pellets had on the inoculation. In order to minimize potential negative effects due to the storage of the granules, it was decided to produce large quantities of gel granules within a short period of time (8 hours). For this purpose, the granulation process using the static mixer described in section 5.2 was used. Ideal granules would have a particle diameter ((D _B ) of between 0.8 mm and 1.4 mm and a minimum size distribution non-uniformity factor (COV). Several parameters need to be adjusted when preparing the granules in order to The following section provides a brief description of the choice of parameters for the preparation of the granules, and Section 6.2.1.2 describes the granules used in the continuous fermentation experiments.

6.2.1.1 Granule production method: choice of parameters

Characterized by the granulation method, it is undertaken jointly with other studies, giving particular importance to the following process parameters: stationary mixer diameter (D _s ), number of stationary mixer elements (Ns), surface flow rate of fluid (Vsl) and volume

of the polymer fraction (s _c ). Figures 6.12 to 6.21 show the results obtained from the experiments.

In FIG. 6.12 shows the typical size distribution obtained using a static mixer to immobilize yeast in a carrageenan gel. In this example, the following parameters are used: stationary mixer diameter 12.7 mm, number of elements of the stationary mixer - 60, surface flow rate of liquid 10.5 cm / s and volume of polymer fraction - 0.25, average diameter of the granules is 701 pm with an uneven coefficient of 45%. The cumulative size distribution is illustrated in Figure 6.13, which corresponds to the normal cumulative size distribution calculated by the mean of the samples and the standard deviation. The Smirnov method (Scheaffer et McClave, 1990) was used to test the normal state. The maximum difference between the experimental data and the relevant data was calculated to be 0.0274. The modified value for D corresponds to the data following a normal distribution, which should be below 0.895 at 95% confidence level. In our case, the value of the modified D is calculated at 0.174, which is below the limit of 0.895, and it can be concluded that our indicators correspond to a normal distribution. All the data collected from this granulation method shows a one-size-fits-all distribution. Ponceletet al. (1992), however, show the occurrence of concomitant peaks and / or secondary peaks corresponding to granules with diameters smaller than 200 µm for abginate granules obtained by dispersing in a vessel by stirring. The smaller granules resulting from our method may be lost during the washing of the granules and therefore not be present in the size distribution data.

The effects of the surface velocity of the liquid and the diameter of the static mixer on the average diameter of the granules and on the coefficient of irregularity are shown in Figures 6.14 and 6.15 respectively. The average diameter of the granules decreases as the surface velocity of the liquid increases with all three sizes of static mixer diameters with a more pronounced effect on a static mixer with a diameter of 12.7 mm. Granules with a median diameter greater than 700 pm cannot be obtained in static mixers with a smaller diameter (6.4 mm and 9.5mm) at all velocities tested, whereas granules with a diameter of 12.7 mm are obtained granules greater than 700 pm at a fluid speed of less than 11 cm / s. All three static mixers produce granules with an irregularity ratio between 38% and 58%. It is also seen that as the velocity increases, the coefficient of irregularity decreases in all three cases. The coefficient of irregularity changes with the change in the diameter of the static mixers, the lowest values being obtained with the static mixer with the smallest diameter.

At a superficial fluid velocity of 3.5 cm / s, the volume of the polymer fraction influences the average diameter of the granules, whereas at velocities above 7 cm / s there are no differences in the experimental values of s _c and they vary between 0.083 and 0.5 (Figure 6.16 ). It can be seen that there is little or no effect on the coefficient of irregularity depending on the volume fraction of the polymer as surface velocities increase (Figure 6.17).

Figures 6.18 and 6.19 illustrate the effect of the surface velocity of the liquid and the number of elements of the static mixer on the average diameter of the granules. As the fluid velocity increases, the average diameter of the granules decreases for all static mixers with different number of elements (Figure 6.18). The average diameter of the granules at a given fluid velocity is similar in number of elements from 24 to 120 elements, while in the configuration of 12 elements the obtained granule diameters are larger than the other five configurations, Figure 6.19 also shows that the average The diameter of the granules is achieved with a static mixer of at least 24 elements.

Figure 6.20 shows the effect of the surface velocity of the fluid on the coefficient of irregularity for several static mixers with different number of elements. The velocity of the fluid does not appear to affect the non-uniformity coefficient for all tested configurations. The influence of the number of static mixer elements on the coefficient of irregularity is more pronounced (Figure 6.21). The coefficient of irregularity decreases with the increase of the elements of the static mixer and reaches a minimum of 45% for 60 or more elements. These results are consistent for surface velocities ranging from 3.6 cm / s to 17.8 cm / s.

It can be assumed that an increase in the diameter (Dj) of the static mixer will create a heterogeneity of shear force in the mixer, leading to an increase in dimensional dispersion, which is measured by the coefficient of non-uniformity. At the same time, an increase in D _s would decrease

the intensity of the shear force, thereby increasing the average diameter of the granules. These two effects are observed when conducting experiments with the smallest size static mixer to produce the smallest medium diameter (400 pm - 500 pm) granules and the smallest non-uniformity ratio (approximately 40%) or the most the smaller dimensional dispersion.

The energy required to produce an emulsion is proportional to the area of the separation surface between the polymer and the oil phase. With smaller granule sizes, more energy is required to form an emulsion. Berkman and

Calabrese (1988) stated that an increase in the average surface velocity (V _s ) of the fluid leads to an increase in the scattering energy per unit mass of the fluid, thus favoring a reduction in the size of the granules. An increase in the average surface velocity of the liquid (tested between 3.6 cm / s and 17.8 cm / s) results in a decrease in the average size of the granules. Such speed increases the pressure difference between the inlet and outlet openings of the static mixer. The pressure difference is proportional to the dissipation energy per unit mass of the liquid. Therefore, as the velocity increases, the scattering energy in the system increases, which favors the reduction of the granule size. The decrease in the diameter of the granules D _B is observed with the increase of the surface velocities of the fluid. Thus, Walker (1983) showed that a dynamic equilibrium is formed between the formation of granules and the coalescence between the granules at high velocities corresponding to significant turbulent levels. For constant values of the diameter (D _s ) of static mixers and the number of elements (N _s ), the surface velocity has little effect on the coefficient of irregularity. Therefore, velocity is a parameter that allows the management and selection of the average diameter of the granules without substantially modifying the size dispersion.

The scope of this study found that the bulk gel fraction of carrageenan (e _c ) had little effect on the average diameter of the granules or on the coefficient of irregularity, except for the lowest velocity tested at 3.6 cm / s, at which the average diameter of the granules decreases with decreasing (s _c ). Audet and Lacroix (1989) studied this parameter for the production of granules of carrageenan in a two-phase dispersion system (vessel with stirring and periodic action, do not use a static mixer with continuous operation) and they concluded that _has not proved

influence on the average diameter of the granules in a polymer solution with a carrageenan concentration of 3% (w / v). The specific effect of the concentration of the κ-carrageenan gel on the size distribution of the granules was tested by Audet and Lacroix (1989), who indicated that this parameter had a strong influence on the size distribution. Increasing the gel concentration leads to an increase in the average diameter of the granules (D _B ) and the coefficient of irregularity (COV). This effect is explained by increasing the viscosity of the gel at higher concentrations, which results in smaller shear forces in the emulsion, and thus larger granules. Although the effect of gel concentration on the size of the granules has not been investigated in these abstracts, other means of adjusting the granule size could be used if necessary.

As the number of elements (Ns) increases, the average residence time of the fluid in the static mixer increases, resulting in a more homogeneous mixture and thus the formation of smaller and more dispersed granules. In the experimental work, the equilibrium state in the dispersion (measured by the coefficient of non-uniformity) is reached at about 60 to 72 elements. Middleman (1974) points out that 10 elements are sufficient to achieve equilibrium at low viscosity emulsions (0.6 to 1.0 cP). The carrageenan solution used in these experiments [3% (w / v)] has an average viscosity of 200 cP and the oil viscosity is 25 cP. As a result, this higher viscosity requires a longer residence time in the mixer to achieve pseudo-homogeneity.

Figure 6.12 Size distribution of granules obtained using a static mixer. The technological parameters in this experiment are: = 12.7 mm, N _y = 60, ν _ςι = 10.5 cm / s, and

CC = 0.25.

♦ Experimental data O Standard distribution

Figure 6.13 Cumulative size distribution of granules obtained using a static mixer. The technological parameters in this experiment are: Dj = 12.7 mm, N; = 60, =

10.5 cm / s and £ _c = 0.25.

• Ds = 12.7 bDs = 9.5 ± Ds = 6.4

Figure 6.14 Surface fluid velocity (cm / s) depending on the average diameter of the granules (pgp). Three different static mixers (D 12.7 mm, 9.5 mm and 6.4 mm) were evaluated. The number of elements of the static mixers (M ^) is constant 48 and the polymer fraction (£ _e ) is maintained at 0.25.

(0

• Ds = 12.7 Ds = 9.5 ▲ Ds = 8.4

Figure 6.15 Surface velocity of the fluid (cm / s) depending on the coefficient of irregularity of the diameter of the granules (%). Three different static mixers (D _s - 12.7 mm, 9.5 mm and 6.4 mm) were evaluated. The number of elements of the static mixers (N _s ) is constant - 48, and the polymer fraction (E _s ) is maintained at 0.25.

100

♦ ε = 0.5 ε = 0.25 1 ε = 0.125 · 8 = 0.083

Figure 6.16 Surface velocity of the liquid (cm / s) depending on the average diameter of the granules (pm) Four polymer fractions (£ _c = 0.5, 0.25, 0.125 and 0.083) were estimated. The number of elements of static mixers (Ν <) is constant - 48, and the diameter of static mixers (Dr) is 12.7 mm.

(8

♦ 8 = 0.5 ε = 0.25 1ε = ύ.125 · 8 = 0.083 Figure 6.17 Surface velocity of the liquid (cm / s) depending on the coefficient of irregularity of the diameter of the granules (%) An estimate of four polymer fractions (= 0.5, 0.25, 0.125 and 0.083).

The number of elements of the static mixers (Ν _? ) Is constant 48 and the diameter of the static mixers (D>) is 12.7 mm.

101

♦ Ns = 12 hls = 24 A Ns = 48 Δ Ns-60 □ Ns = 72 · Ns = 120

Figure 6.18 Surface fluid velocity (cm / s) depending on the diameter of the granules (pm). The number of static mixer elements (N5) is changed to 12, 24, 48, 60, 72 and 120. The polymer fraction was kept constant - 0.25 and the diameter (D _s ) of static mixers was 12.7 mm.

♦ No. = 12 Ns = 24 A No. = 48 Δ No. = 60 □ No. = 72 · No. = 120

Figure 6.19 Surface velocity of the fluid (cm / s) depending on the coefficient of irregularity of the diameter of the granules (pm). The number of static mixer elements (N $) changes to 12, 24, 48, 60, 72 and 120. The polymer fraction (£ .-) is kept constant 0.25 and the diameter (D _s ) of static mixers is 12.7 mm.

102

♦ VS 1. = 3.6 hVSL-7.0 aVSL = 10.6> V6L = 13.2 dVSL = 17.8

Figure 6.20 Average diameter of the granules (pm) depending on the number of static mixer elements (Ν $). The surface velocity of the fluid (cm / s) changed to 3.6, 7.0, 10.6, 13.2 and 17.8. The polymer fraction (£ _β ) is kept constant - 0.25 and the diameter (D $) of static mixers is 12.7 mm.

Figure 6.21 Coefficient of irregularity of the diameter of the granules (%) depending on the number of elements of the static mixers (N $). The surface velocity of the fluid (cm / s) changes to 3.6 and 17.8. The polymer fraction (f _c ) is kept constant - 0.25 and the diameter (D ^) of static mixers is 12.7 mm.

103

6.2.1.2. Granule production method: characteristics of k-carrageenan granules

From the data presented in the previous section, it is possible to select the parameters of the granule production process in order to obtain granules with desirable characteristics for fermentation experiments. To reduce the unevenness factor to the minimum value for a specific diameter of the static mixer, 60 mixing elements were selected to obtain an oil / gel dispersion.

Granules with an average diameter of approximately one mm were selected to minimize the external mass transfer and to facilitate separation of the fermenting liquid by mechanical means. To achieve this, the largest stationary mixer (12.7 mm) was tested and the lowest test speed (3.6 cm / s) was selected. Because the polymer fraction had little effect on the coefficient of irregularity, a gel / oil ratio of 50/50 (ε = 0.5) was used to maximize productivity to the maximum possible.

Several batches of LCC3021 clustered yeast gel pellets were produced in the laboratory using the method described in section 5.2 with the following parameters: D = 12.7 mm, Ns = 60, V _SL = 3.9 cm / s and ε ₀ = 0.5. The resulting granules (FIG. 11) were passed through a series of sieves to remove granules with a diameter greater than 2.0 mm and less than 0.5 mm. The particle size distribution of the obtained particles is presented in Figure 6.23. Figure 6.24 illustrates the cumulative size distribution of these granules. This is a typical distribution that is used in all fermentation trials in 50 L gas-filled reactors.

104

Figure 6.23 Size distribution of k-carrageenan gel pellets in continuous fermentation experiments in a 50 L gas-lift bioreactor equipped with a suction tube.

105 f

Figure 6.24 Cumulative size distribution of k-carrageenan gel pellets in continuous fermentation experiments in a 50-L gas-lift bioreactor equipped with a suction tube.

106

6.2.1.3 Optimizing the granule production method for industrial use

A method for producing 10 L granules per hour using a static mixer has been investigated and experimentally scaled. Several aspects of this process require further research and / or optimization before being considered commercially viable. It is necessary to increase the volume capacity in order for the experimental bioreactor to be loaded with the specified volume of immobilized cells. For example, for a 2000 hL GPST type bioreactor, approximately 800 hL granules are required. In order to provide these volumes, it is necessary to increase the flow of gel and oil. The data suggest that the increased speed when using static mixers with diameters from 6.4 mm to 12.7 mm will lead to the formation of too small granules to be used in the fermentation system. It follows that it is necessary to increase the diameter of the static mixer, thereby increasing the average diameter of the granules. However, the use of a larger diameter static mixer will also increase the granularity of the granules, thereby increasing the percentage of granules beyond the desired range.

Another alternative is to implement the system using medium-sized static mixers (12.7 mm) connected in parallel. With a system of ten static mixers, the output will eventually reach 100 L / h. For 2000 hL for industrial implementation, the process must be carried out continuously for 800 hours, or approximately 34 days, to produce the required volume of granules. Several systems may be used to reduce production time, but this would lead to other complications. Production time could have been shorter if it had been

107 granules may be stored for a longer period of time, while maintaining viability. This is possible if the process involved in drying the granules or storing the granules in a vacuum container could be investigated.

Poncelet et al. (1993) have published a paper which should take into account the type of static mixer used to obtain the dispersion. With this proposed system, using a different type of static mixer, unlike the Kenics used in this study, it would be possible to use a larger diameter mixer without compromising the size distribution of the granules (maintaining a low uneven coefficient).

Additional problems with the existing experimental process are related to the operation of the system at 40 ° C and the use of vegetable oil and potassium chloride solution for the production of granules. Due to the high temperature of the production process, both heating and cooling systems are required. The possible thermal shock to which the yeast cells are subjected requires further studies to evaluate possible adverse effects. In this study, granules with immobilized cells with high yeast viability (over 90%) were obtained, but no other effects on the yeast population that could be due to the process were investigated.

As oil is used to produce the desired emulsion and subsequent granule formation, and the oil acts as a surfactant and thus retains the foaming, there is oil residue on the surface of the granules. Although these residues would be of help during the fermentation stage, any residues carried into the finished beer product would be

108 harmful as foam is desired in the final product. Large volumes of potassium chloride solution used to separate the solid phase (granules) from the oil and the method of separating this saline solution from the granules prior to their introduction into the bioreactor requires attention. Otherwise, it may be necessary to clean this solution from the reactor and then load the granules into it.

As the granules with immobilized cells are produced outside the reactor, sterilizations are applied during the formation of the granules and sterility is maintained until the granules are introduced into the bioreactor. At different points in the path of transfer of the granules between the tanks, the possibility of their contamination is created, which requires constant monitoring, since the presence of pollutants can result in their co-immobilization in the granules. As a result of diligence in the laboratory, it was possible to produce sterile granules. However, the environment in which the installation is located may not be appropriate in comparison to the laboratory conditions, which requires more strict controls.

6.2.2 flocculating yeast cells

One of the most natural forms of immobilization is the self-assembly of microorganisms into aggregates of cells. Calleja and Johnson (1977) have suggested that the cells cause contact with each other and form aggregates with all their binding properties. The first involves cells of different sexes that are attracted to each other by releasing pheromones (□ and α-factors). This type of binding is temporary and is a protein-protein bond between □ and α-agglutinins attached to the cell walls.

The cells can also be grouped, not separating from the mother cells during budding. This inseparability may be

109 specific to a particular yeast strain or may be caused by depletion of the culture medium or mutation of a number of genes. This phenomenon is about chain formation, not flocculation. The connections between these cells can be irreversibly broken by mechanical shearing (Stratford, 1996). Stewart and Russell (1981) defined flocculation as a reversible "phenomenon whereby the yeast cells adhere and form larger particles and either sediment rapidly from the medium in which they are suspended or rise to the surface of the medium." Many facts prove that flocculation is genetically controlled and that the mechanism of flocculation is based on selected molecules that act as bridges between walls of adjacent cells. Specifically, individual lectins are thought to bind to the D-mannans of the contacting cells in the presence of Ca ²⁺ ions (Calleja and Johnson, 1977). It has been found that this protein / carbohydrate bond can be reversibly inhibited by gelling agents or by specific sugars.

In FIG. 12 shows three possible configurations of yeast cells, namely non-flocculating yeast forming yeast chains and flocculating yeast. In the case of yeast-forming yeasts

Q chains, although the cells clustered, are not considered to be a type of flocculation, since these cells are, above all, never single, and flocculation suggests single cells that together form multiple due to a favorable environment (Ca ²⁺ ions and low levels of inhibiting sugars). For flocculating cells, the specific size of the blocks may depend on the genetics of the cells as well as on the hydrodynamic conditions to which the cell is exposed (shear medium).

In FIG. 13 and FIG. 14 show two types of Labatt yeast strain for light beer with different flocculation rates. Yeast strain

110

Medium flocculation LCC3021 is presented in FIG. 13. In the presence of calcium ions, this strain forms groups of 0.5 mm to 1.0 mm in size if glucose has been depleted from the liquid medium. In FIG. 14 shows the image of the LCC290 superflocculent yeast strain, which forms blocks larger than 1 mm in size and aggregates to groups up to 5 mm in diameter with low shear. Under moderate agitation conditions, the particle diameter of the LCC290 flocculates is between 1 and 2 mm. Several methods for determining the size of yeast flocculation have been proposed (Speers & Ritcey, 1995; Akiyama-Jibiki et al., 1997; Teixera et al, 1991; Stewart & Russell, 2000). B Brewer's Yeast (Stewart & Russell, 2000) suggested that methods for determining yeast flocculation be reduced to three categories, namely sedimentation methods, stationary fermentation methods, and direct monitoring of flocculation formation in the growth medium. .

The sedimentation methods were first described by Burns in 1937 and modified by Helm et al in 1953. They are currently part of the standard methods of analysis and are recognized by the Technical Committee and the Editorial Committee of the American Society of Brewing Chemists ( 1992). This method refers to in vitro methods because the precipitation characteristics of the yeast are determined in calcium sulfate buffer and not in the actual fermentation medium. The stationary fermentation method involves the growth of yeast in hops and the flocculation characteristics are measured in vivo. Both methods involve measuring the absorption of sedimented yeast samples compared to deflocculated yeast samples using a UV spectrophotometer in the visible part of the spectrum. Stewart and Russell (2000) present a method for measuring yeast flocculation by visually describing the degree of flocculation

Ill observed in yeast growth specimens in glass bottles closed with screw caps. They use subjective measures to express the degree of flocculation. For example: 5 - very strong flocculant, 4 - strong flocculant, 3 - moderately strong flocculant, 2 - weak flocculant, 1 - poor flocculant and 0 - non-flocculant. The over-flocculating yeast strain LCC290 received the characteristic 4-strong flocculant, while the yeast strain LCC3021 was rated 3-moderate flocculant.

flocculation ability is an important feature in the brewing industry, as the tendency to use natural yeast that either settles or floats is often used as a method of separating these yeasts from the fermentation fluid. But the yeast strain that floats before fermentation is complete is undesirable since the liquid will not reach the ideal alcohol content and residual sugar level. In continuous fermentation, and in particular continuous fermentation in a bioreactor equipped with a suction tube, and using gas lift, the flocculating yeast acts as an immobilized matrix. Their desire to settle down! compensated by injecting the spray gas which

O maintained in suspended state. In such a system, the risk of under-fermentation of the liquid medium is eliminated, since the solids are constantly circulated and maintained in intimate contact with the fermenting liquid.

Section 6.2.2.1 describes the precipitation parameters and fermentation indices of the LCC290 superfluous yeast. The interest is in identifying the onset of j flocculation for this particular yeast strain. In addition, the rate of sedimentation of the yeast is determined to provide valuable j

112 information that could be used in the future to construct sedimentation tanks.

6.2.2.1 Characterization of LCC290 over-flocculent yeast

Prior to the continuous fermentation of LCC290 super flocculent yeast in a GPST type bioreactor, it was decided to characterize the yeast on a laboratory scale - fermentation in a shake flask. Figure 6.28 shows the evolution of the yeast population as a function of time. As expected, the concentration increases sharply during the first 48 hours, when there is sufficient nutrient medium and the presence of oxygen in the beer must to allow the yeast to grow. But when the yeast continues to consume carbohydrates in the absence of air, they do not reproduce and enter the anaerobic fermentation phase. Once the carbohydrate supply is over, small populations of yeast begin to die off. This phenomenon is shown in Figure 6.29, where cell viability decreases from approximately 97% to just over 90%.

Figure 6.30 shows the consumption of carbohydrates depending on the duration of fermentation. Yeast cells first consume monosaccharides, glucose and fructose, followed by maltose and maltotriose. However, yeast cannot metabolize both maltotetrosis and long-chain polysaccharides (poly-1 and 2). As the concentration of carbohydrates decreases (represented by the curve of relative weight in Figure 6.31), the concentration of ethanol increases proportionally. After approximately 37 hours of fermentation, the concentrations of ethanol and carbohydrates are equal.

Taking into account the pattern of growth and metabolism of carbohydrates, it can be assumed that the over-flocculating strain LCC290 manifests itself as an industrial strain

113 yeast LCC3021. Fermentation reaches the final stage of completion of the process when the relative weight of the liquid reaches approximately 2.7 ° P. It is customary for flocculating strains of yeast to form large groups (blocks) and precipitate from solution before the end of fermentation; this phenomenon is known in the brewing industry as the "hold-up" of fermentation. In our attempts at fermentation, we were able to complete the fermentation because the flasks were agitated and thus the yeast was maintained in suspended state and in close contact with the culture medium.

Another important feature is the study of the ability of the yeast to flocculate. In particular, we were interested in establishing the rate at which the yeast would settle, as well as obtaining an indication of when that particular yeast strain began to flocculate. These two characteristics are important for continuous fermentation experiments as they play a role in maintaining the yeast population in the fermenter using gas lift. In FIG. 6.32 shows the yeast precipitation curves in samples tested for 24 hours during fermentation, the flocculation being inhibited by the presence of certain sugars; glucose is a known inhibitor, which means that flocculation will only start if that inhibitor is depleted. In a sample at 40 h periodic fermentation, cells begin to flocculate and precipitate from the solution when tested by the method described in section 4.7. Sedimentation is very fast for all test intervals except for the 24-hour period during which no precipitation is observed. During the slowest settling experiment for 40 hours, the yeast completely precipitated in the test apparatus for 90 s. Less than 71 hours are required for sedimentation

60s.

114

Figure 6.28 Total concentration of yeast cells from LCC290 in periodic fermentation with stirring depending on the duration of fermentation, maintaining the fermentation temperature at 15 ° C.

115

Figure 6.29 Viability of yeast cells measured using methylene blue, depending on the duration of fermentation.

—6 Polysaccharides

-B- Maltotetrosis - · - Glucose

Fructose

Maltotriose

Maltose

Figure 6.30 Concentration of carbohydrates versus duration of periodic fermentation with LCC290

116

Glitz »#

Ethanol

Relative weight

Figure 6.31 Ethanol and glycerol concentrations and relative weight of liquid depending on the duration of periodic fermentation with LCC290.

w

Figure 6.32 Height of the separation surface of a yeast suspension depending on the duration of precipitation. The values for this graph are determined based on several fermentation intervals.

117 1.2E-03

E ~ 1.0E-03 f

I B.OE-04 to £ 6.0E-04 x 4.0E-04

2.0E-04

0.0E + 00

40 hrs A48 hrs »64 hrs d71 hrs 192 hrs

Figure 6.33 Yeast suspension sedimentation rate depending on cell concentration. The values for this graph are determined based on several fermentation intervals.

118

Researchers have suggested that sedimentation rate is a function of the concentration of solids (Soe and Clevenger, 1916). Figure 6.33 shows the rate of solid phase precipitation for a given concentration of yeast cells. The data shown in this graph were obtained by a method proposed by Lynch (1952), with precipitation data obtained at each fermentation interval. The points are on approximately one line, confirming the same phenomenon observed by Soe and Clevenger (1916).

The results obtained during the sedimentation test indicate that the LFC290 over-flocculent yeast flocculates at a relative weight of the liquid 6 ° P and below. This value could be used as a benchmark for continuous fermentation to indicate when the relative weight of the fluid in the pseudo-steady state should be maintained if it is desirable to allow the cells to flocculate. Operation of the reactor above 6 ° P carries the risk of destabilization of the flocculating cells and may lead to failure in the population of immobilized cells. Overflooding yeast deposition characteristics indicate that the yeast population will stop too quickly if left stagnant. In a three-phase GPST bioreactor, it is possible to keep these cells in circulation, but in the event of a failure in the gas supply system, the cell population will stop quickly and additional spreading may be necessary. solid resuspension gas. For post-fermentation treatment, these sedimentation characteristics are important because solids separators such as gravity separators can be used to remove solids periodically. In the brewing industry this would reduce the workload of

119

solid particulate centrifuges, which will allow longer operation of the centrifuges. Smaller losses of beer could be expected and the degree of deviation from taste due to centrifugation (although minimal) would be minimized due to the lower levels of yeast biomass passing through the centrifuge.

6.3. EVALUATION OF TECHNOLOGY FOR THE CONTINUOUS FERMENTATION OF BEER USING GAS LOW POWER

The first and most important goal of these theses is to evaluate the feasibility of operating a continuous continuous-state experimental PCR type bioreactor using k-carrageenan gel granules to capture cells of Saccharomyces carlsbergensis (described in Section 6.2.1). In addition, our desire was to investigate whether a good-tasting North American light beer can be produced in such a system. We also set ourselves the goal of determining the minimum residence time required for the complete dilution of high relative gravity mustache (17.5 ° P), as well as establishing the operating limits for the oxygen content of the continuous fermentation system .

The minimum residence time for consuming all the sugars in the must must be 24 hours. These data can be compared to the duration of the classical periodic fermentation, which is five to seven days. The dissolved oxygen concentration, measured by an Ingold oxygen detector at a specific location in the bioreactor, tends to zero regardless of the oxygen added to the spray gas (ranging from 0 to 20% (v / v)). This indicates that the oxygen supplied to the must is either rapidly consumed by the yeast cells, or

120 just came out with the exhaust gas. The free cell level in the beer overflow is in the order of 108 cells per 1 mL of green beer. The levels of higher diketones, diacetyl and 2,3-pentanedione, as well as the level of acetaldehyde, decrease in proportion with the decrease in oxygen in the dispersed gas. (Figures 6.34 and 6.35). The measured esters (ethyl acetate and isoamyl acetate) and higher alcohols (propanol, isobutanol, isomyl alcohol) were not found to be affected by the change in oxygen supply (Figure 6.36).

Figure 6.37 compares different aromatic compounds in two

I type of tested beer products obtained in a continuous system using immobilized cells, the comparison being made with a control beer obtained industrially (periodic fermentation without using cells). With respect to the filing of the

Oxygen several differences were observed in esters (ethyl acetate, isoamyl acetate) and higher alcohols (propanol) when comparing beer obtained by continuous fermentation and control beer obtained under industrial conditions. Tests on the finished product from beer produced with 2% oxygen were evaluated using an established taste set to be relatively close to the control beer (industrial product). But beer obtained with 20% oxygen was evaluated in an unacceptable way by the signs of oxidation of aromatic compounds and paper and wine testing.

In a pseudo-steady state, the experimental bioreactor is operated for a duration of 24 hours over a period of 6 .1 weeks. Green beer has good taste and no major defects are observed (sulfur is not observed). The amount of oxygen in the spray gas is an important element in these experiments. Beer produced with 2 to 5% oxygen in the spray gas gives the best test results. This important control requires attention and attention should be paid

121 of more accurate oxygen measuring apparatus and measurements to be carried out on a larger group of samples for analysis before and after fermentation. In traditional intermittent fermentation, the wort is dosed with oxygen before being transferred to the fermenter. After inoculation of the medium, the dissolved oxygen concentration decreases sharply as the yeast cells consume it (the yeast grows during the first 24 hours of fermentation). Therefore, the rest of the fermentation occurs mainly under anaerobic conditions. The use of a continuous homogeneous system for the main fermentation does not allow the concentration of oxygen to change throughout the process. For this reason, it is probably very difficult to achieve full compatibility with regard to the taste of beer produced using a continuous and intermittent fermentation process.

Due to these differences, the bioreactor used in these initial evaluations produces good taste beer and an assay profile. By using a bioreactor with gas lift and relatively small pellet size (~ 1 mm), it is possible to increase the volume capacity of the bioreactor by reducing the time for main fermentation by several days. The level of biomass in the outlet beer indicates that the rate of yeast growth in the bioreactor using immobilized cells is equivalent to that in periodic fermentation using free cells, with the processes being carried out under similar conditions. These observations confirm the relationship between taste and yeast growth rate. This may explain the failure of previous attempts to produce good beer in systems with limited cell growth. Controlled delivery of the gas mixture can be a powerful tool in precision regulation

122 of organoleptic properties in continuous fermentations using immobilized cells.

High levels of diacetyl in the final fluid have also been observed by other researchers (Virkajarvi & Pohjala, 1999; Kronlof et al., 2000). In North American light beer, the measured level of diacetyl is 30 pg / L compared to the level of 400-800 pg / L in the final fluid obtained by continuous fermentation. The use of traditional aging technology (cold aging at 2 ° C for 14 days) reduces diacetyls to the desired limits, but is detrimental to the overall productivity of the process. The use of rapid auxiliary fermentation described in Chapter 2 would help to reduce diacetyls without significantly reducing productivity (a two-hour process). However, the additional costs may be high and all brewers may be reluctant to subject the beer to high temperatures (80 to 90 ° C).

123

Diacetyl * · Pentandion

Figure 6.34 Concentration of higher diketones depending on the percentage of oxygen in the spray gas. The fermentation was carried out in a 50 L gas lift apparatus filled with 40% (v / v) k-carrageenan gel pellets. The total velocity of the spray gas is kept constant at 6.4 scfh. The stay is 24 hours.

Figure 6.35 Acetaldehyde concentration depending on the percentage of oxygen in the spray gas. The fermentation was carried out in a 50 L gas lift apparatus filled with 40% (v / v) k-carrageenan gel pellets. The total velocity of the spray gas is kept constant at 6.4 scfh. The stay is 24 hours.

124

Ethyl acetate- · - Propanol Isobutanol-ώ— Isoamylacetag ⁹ ** Isoamyl alcohol

Figure 6.36 Concentration of esters and fusel alcohols depending on the percentage of oxygen in the spray gas. The fermentation was carried out in a 50 L gas lift apparatus filled with 40% (v / v) k-carrageenan gel pellets. The total velocity of the spray gas is kept constant at 6.4 scfh. The stay is 24 hours.

Figure 6.37 Comparison between the finished product produced with 2% oxygen and 20% oxygen in the spray gas, compared to industrially produced beer and taste limits. For products obtained by continuous fermentation, the process was carried out in a 50 L gas lift apparatus filled with 40% (v / v) k-carrageenan gel pellets. The total velocity of the spray gas is kept constant at 6.4 scfh. The stay is 24 hours.

125

6.4 SPEED OF MIXING AND CIRCULATION IN THREE-PHASE FERMENTATION PARTICIPATED IN A BREAKER WITH A GAS LAND FORCE WITH A SUCKING TUBE

Mixing experiments were performed using the acid injection method described in section 4.8. The goals of this experimental phase were two. First, we wanted to evaluate whether the gas lift system was well mixed, calculating the fluid circulation rate and mixing time for several immobilized carriers at different surface velocities of the fluidized gas. The difference between these experiments and those published in the literature is that the external experimentation was carried out in a real fermentation medium with active yeast cells and not with samples (solutions of water-solids-gas). These blending experiments also include the calculation of the surface velocity of the fluid, which could be used to scale the experimental system.

6.4.1 Calibration of the instrument.

The pH meter used in the mixing studies shall be calibrated using the method described in section 4.8. A 4-20 mA signal emitted by the pH meter passes through a 250 Ohm resistor and transforms into a 1-5 volt signal that could be registered on a data card. The pH meter is then immersed in several buffer solutions as follows: buffer solution with pH 4.6, buffer solution with pH 4.0, buffer solution with pH 5.0, buffer solution with pH 6.0 and finally in buffer solution with pH 7.0 (Beckman certified standards, Cole Parmer). Figure 6.38 presents the data from the Ingold pH meter system that corresponds to 5 standard pH solutions. Figure 6.39 is a pH calibration

126 meters, with the diagram showing the value of the actual pH depending on the measured signal voltage. The most appropriate calibration line for this system (correlation coefficient 0,9996) is determined by the following equation:

pH = 3.68 * voltage - 3.53 (6.1)

The mixing data collected was transformed using Equation 6.1 to obtain the actual pH values as measured in the gas lift bioreactor.

The corresponding time for which the pH meter reflects the change in pH values is also measured. Figures 6.40 through 6.43 show the typical pH response time for four different steps of pH change, namely 0.6, 1.2, 2.3 and 3.4 units. The actual data is adjusted to the impulse function, using data from the Curve Table 2D (Jandel Scientific) and analysis software entered into the measurement system. The obtained curve alignment has a correlation coefficient of 0.994. These graphs calculate the response time of the instrument when the pH values change, corresponding to 98% of the change, and then plot the corresponding points in the diagram, depending on the step of pH change (Figure 6.44). Response time declines with a change in pH within the test range. For the smallest change values of 0.6, the response time of the pH meter is 6.4 seconds. The response time is reduced to about 4.2 seconds with a pH change of 3.4.

127

Figure 6.38 Characteristic data recorded through the Ingold pH data acquisition system for different buffers as a function of time. The frequency is 50 Hz for a total of 15000 points.

Figure 6.39 Calibration curve of pH meter for pH values according to the measured voltage (V). The line of best fit or the calibration curve corresponds to the following equation: pH = 3.68 * voltage - 3.53, at a correlation coefficient of 0.9996. (n = 2500)

128

[PtieeCumJ y "= w-" b <1- "xp <-0" - <yap (1-2 ^o - ⁵ L2> -c) / <1)> ² hh) L8483804 Fth * -882683,46

Response time (s)

Figure 6.40 Characteristic data for the response time of a pH meter as a function of pH change in step ~ 0.6. The sensitivity is adjusted using software containing a curve table. The best fit of the impulse function has a correlation coefficient of 0.995.

[PulseCum] y = e + b <1-® <p (- (JHSn (1-2 ⁰ - ^a / 2> 4syd)) ²

Figure 6.41 Characteristic data for the pH response time versus pH change in step ~ 1.2. Sensitivity is controlled by software containing a curve table. The best fit of the impulse function has a correlation coefficient of 0.999.

129 [PufceCum] y ^ + bc-exp ^ hcZpa - ^^ - sUf) ² ^ 0.86871074 F «tat ° 3872162.7

Vol • _β · * · 1 • h '··' · · ' Ί I • •

Response time (s)

Figure 6.42 Characteristic data for the pH response time depending on the pH change in step ~ 2.3. Sensitivity is controlled by software containing a curve table. The best fit of the impulse function has a correlation coefficient of 0.999.

[PUwCum] y'l + iXI-exjX-Cx-dltXI - ^^ Hyd)) ²

Figure 6.43 Typical response time for a pH meter as a function of pH change in step ~ 3.4. Sensitivity is controlled by software containing a curve table. The best fit of the impulse function has a correlation coefficient of 0.999.

130

Figure 6.44 Response time of Ingold pH meter as the step changes. (N = 3). The instruments were immersed in higher pH buffer solutions and then immediately immersed in pH 4.0 buffer solution. The best fit line for each curve is used to calculate the time it takes the instrument to reach 98% of the step change.

131

The characteristics of a type of pH meter are very important, especially when the system is used to estimate mixing times and circulation rates. The instrument must have low response times to reflect changes in the environment it measures. In particular, the pH response time should be lower than the reactor circulation rate if used for accurate measurement. However, quasi-instantaneous response is not desirable, since a slight delay in the response will simply reflect on the measurement of the circulation rate and reduce it to zero.

6.4.2 Mixing time and speed calculation

The experiments on the duration of mixing and the calculation of velocities were performed in three types of immobilized cell fermentations. The experiments were carried out in an experimental bioreactor with a volume of 50 L equipped with a suction tube using an aqueous solution containing no solid phase for fermentation and then a fermentation liquid medium with a relative weight of 2.7 ° P containing or k-gel pellets. carrageenan or LFC290 yeast overgrowing or medium flocculating yeast LCC3021. Figures 6.45 and 6.46 show exemplary data images that were collected using a pH meter after a pulse injection of acid into the aqueous solution (described in Section 4.8 | Method). Figure 6.45 shows the sensitivity of a pH meter to! injection of acid into an aqueous solution that does not contain

I j solid phase. Figure 6.46 shows the sensitivity to

I injection of acid into a fermentation fluid medium containing flocculating yeast LCC290 with high flocculation | ability. The signals are adjusted to the decremental sinusoidal

I curve, the correlation coefficient being 0.96 and 0.90 respectively and j

132 calculated for the corresponding equation. The parameters "b" and "c" of the diagram correspond to the values of "a" and the decremental sine wave equation (3.1). These numerical values are used in Equations 3.2 and 3.3 to calculate the circulation rate and mixing time for a system. Appendix C contains the rest of the mixing data and the curves plotted for all experiments.

Figures 6.47 and 6.48 show diagrams of mixing duration depending on the rate of circulation of the injection of acid into an aqueous solution containing no solid phase. Figures 6.49 and 6.50 show the corresponding diagrams showing mixing experiments using high-flocculation yeast LCC290, and Figures 6.51 and 6.52 show the results of the mixing process test involving k-carrageenan. The mixing time and circulation time of LCC3021 flocculant yeast medium is shown in Figures 6.53 and 6.54. Depending on the type of solid phase, the mixing time and the circulation rate decrease with the corresponding increase in the gas surface velocity. The relationship between the circulation velocity and the surface velocity of the fluid corresponds to equation 3.11 proposed by Kennard and Janekah (1991) for all four systems tested. The mixing time corresponds to the dependence according to equation 3.11 for the solid solution aqueous solution system and for the k-carrageenan gel pellet system. The two systems with flocculent yeast in the immobilized matrix do not show a clear agreement with the theoretical model of Kennard and Janekah. Systems with LCC290 and LCC3021 have an initial mixing time (until the surface gas velocity exceeds 4 mm / s) which is lower than predicted in the model. But the rate of decline

133 of the mixing time decreases at a gas surface velocity greater than 4 mm / s, while in the model the values are lower. Table 6.1 gives the derived equations for the duration of circulation and the duration of mixing depending on the surface velocity of the gas.

Figure 6.55 shows the mixing time depending on the surface velocity of the gas for all four systems. For a solid-phase aqueous solution, the longest duration of 98% mixing was observed (~ 220 seconds at V _sg 3mm / s), and the system with LCC290 showed a better ability to minimize the effect of acid injection in system (~ 110 seconds at V _sg 3mm / s). The values for LCC3021 and the k-carrageenan system are between those of the solid-phase aqueous solution and the LCC290. The solid phase in the bioreactor with gas lift helps to disperse the fluid elements in the liquid phase by stimulating the formation of turbulent motions, providing co-axial mixing. However, LCC290 superflocculent yeast at the same solid phase loading (16% (w / v)) as medium medium LCC3021

flocculation ability, provide a shorter mixing time at all surface gas velocities tested.

Figure 6.56 shows the duration of circulation depending on the surface velocity of the gas for all four test systems. At a gas surface velocity of 2 mm / s, the circulation time varies between 28 seconds and 35 seconds, respectively, for the system with no solid phase in aqueous solution having the shortest mixing time and the system with LCC290 showing the lowest circulation. At higher gas speeds, the difference between the four systems is reduced by approximately 3 seconds. However, at all speeds tested

134 the system with the LCC290 shows slightly lower circulation rates, while the non-solid phase system has the highest circulation rates.

Figure 6.57 shows the relationship between the surface velocity of the gas and the surface velocity of the liquid. Equation 3.7, proposed by Livingston and Zhang (1993), was used to calculate the surface velocity of a fluid for a given circulation duration and the type of solid phase. The surface velocity of the liquid increases with the corresponding increase in the surface velocity of the gas. The LCC3021 system and the non-solid phase have similar behaviors, while the LCC290 and xanthate systems show some similarities. The equation proposed by Kennard and Jenekah was used to construct the curves in Figure 6.57 showing the dependence of the surface velocity of the liquid on the surface velocity of the gas. Figure 6.58 shows a diagram of the experimentally calculated surface velocity of the liquid, depending on the theoretically calculated surface velocity of the liquid, using equation 3.10 for this purpose. For all four systems, the proposed equation is appropriate for use and is expressed by a linear function with a slope of i and an exit point at y = 0. Table 6.1 lists the correlations between the systems tested in the course of this research.

135

Time (s)

Figure 6.45 Inertia time per pH meter in a 50 L gas lift bioreactor equipped with a suction tube. Sections of acid were injected into the aqueous solution containing no solid phase. The surface velocity of the gas is set at 1.94 mm / s. On the basis of the data, a damping sine curve is constructed. The correlation coefficient is 0.96.

136

Rankl Eqn8001 [UDF1] y »Slne decay (a, b, c, d) ^ 0.89998615 DFAcJ ^« 0.89983725 FitSMErr-0.016230014 Fttat> 24545221 «« 028431867 b = 0.040116063

C «0.17963988 d« 2.4564802

Figure 6.46 Inertia time per pH meter in a 50 L gas lift bioreactor equipped with a suction tube. Sections of acid were injected into a fermentation medium containing LCC290 with high flocculating ability. The surface velocity of the gas was adjusted to 1.94 mm / s. On the basis of the data, a damping sine curve is constructed. The correlation coefficient is 0.90.

137

♦ Experimental ~~ * Theoretical

Figure 6.47 Mixing time depending on the gas surface velocity in a 50 L gas-lift bioreactor equipped with a suction tube. Sections of acid were injected into the aqueous solution containing no solid phase. The mixing time corresponds to the time required for 98% of the step change, which must be reduced to zero (N = 3).

u Experimental - Theoretical Figure 6.48 Duration of circulation depending on the surface velocity of the gas in a 50 L gas-lift bioreactor equipped with a suction tube. Sections of acid were injected into the aqueous solution containing no solid phase. (N = 3).

138

♦ Experimental **** Theoretical |

Figure 6.49 Duration of mixing, depending on the surface velocity of the gas in a 50 L gas-lift bioreactor equipped with a suction tube. Sections of acid were injected into a fermentation medium containing LCC290 with a high flocculation capacity (flocculus size> 1.0 mm), the mixing time corresponding to the time required for 98% of the step change, which must be reduced to zero (N = 3).

Surface gas velocity (mm / s) лна Experimental Те Theoretical figure 6.50 Duration of circulation depending on the surface velocity of the gas in a 50 L gas-lift bioreactor equipped with a suction tube. Sections of acid were injected into a fermentation medium containing LCC290 with high flocculating ability (flocculation size> 1.0 mm).

139

♦ Experimental- “Theoretical

Figure c.51 Mixing time, depending on the gas velocity at 50 L, with a gas lift equipped with a suction tube. Sections of acid were injected into fermentation medium containing LCC3021 yeast immobilized with k-carrageenan gel pellets (granule size 1-2 mm). The mixing time corresponds to the time required for 98% of the step change, which should be

♦ Experimental · “· Theoretical Figure 6.52 Duration of circulation depending on the surface velocity of the gas in a 50 L gas-lift bioreactor equipped with a suction tube. Sections of acid were injected into fermentation medium containing LCC3021 yeast immobilized with k-carrageenan gel pellets (granule size 1-2 mm). (N = 3).

140

♦ Experimental —- Theoretical

Figure 6.53 Duration of mixing, depending on the surface velocity of the gas in a 50 L gas-lift bioreactor equipped with a suction tube. Sections of acid were injected into fermentation medium containing yeast LCC3021 (granule size 1-2 mm). The mixing time corresponds to the time required for 98% of the step change, which must be reduced to zero (N = 3).

♦ Experimental Theoretical

Figure 6.54 Duration of circulation depending on the surface velocity of the gas in a 50 L gas-lift bioreactor equipped with a suction tube. Sections of acid were injected into fermentation medium containing yeast LCC3021 (flocculation size <0.5 mm). (N = 3).

141

♦ LCC3021 □ LCC290 Δ ′ Carrageenan

Figure 6.55 Duration of mixing, depending on the surface velocity of the gas in a 50 L gas-lift bioreactor equipped with a suction tube. Sections of acid were injected into a fermentation medium containing a solid phase of different particles. The mixing time corresponds to the time required for 98% of the step change, which must be reduced to zero (N = 3).

♦ LCC3021 о LCC290 Δ Carrageenan * No solid phase Figure 6.56 Duration of circulation depending on the surface velocity of the gas in a 50 L gas-lift bioreactor equipped with a suction tube. Sections of acid were injected into a fermentation medium containing a solid phase of different particles. (N = 3).

142 <o

Gas surface velocity (mm / s) ♦ LCC3021 O LCC290 Δ Carrageenan * No solid phase

Figure 6.57 Surface fluid velocity (mm / s) versus gas velocity (mm / s) for the four test systems - yeast LCC3021, yeast LCC290, carrageenan gel granules and solid phase aqueous solution. carried out in an experimental gas lift bioreactor equipped with a suction tube.

Figure 6.58 Theoretical surface velocity of the fluid (mm / s) compared to the experimental surface velocity of the fluid (mm / s) for the four test systems. Theoretical values were calculated using the following relationship proposed by Kennard and Jenekah (1991): VSL axis VSGM. The line has a slope of 1 and the Y-segment is 0.

143

Table 6.1 Description of the calculated correlations for mixing time, circulation rate and surface fluid velocity for the four test systems

Mixing time Speed of circulation Surface velocity of the fluid Water solution without t _m = 336, 04 V _SG ° ' ⁴ tc = 37.94 V _S g ^{0 4} V _SL = 189.06 V _SG ⁰²⁸³ solid phase Yeast from LCC290 t _m = 181.55 V _SG ° ' ⁴ t _c = 44.67 V _SG - ° ' ⁴ In _SL = 134.75 In _SG ⁰⁴¹⁹ K-carrageenan gel t _m = 254.68 V _SG - ° · ⁴ tc = 41.73 All ^-0.4 V _sl = 171, 92 Vsg ^{0 283} Yeast from LCC3021 t _m = 322.07 Vsq · ⁰ ' ⁴ tc = 37,907 ^ ⁴ In _SL = 158.12 In _SG ⁰⁴²⁷

For the correlation of the surface velocity of the liquid, Kennard and Janekah (1991) suggest a power of 0.41 in distilled water and 0.64 when the solution contains carboxymethylcellulose and ethanol. Systems with LCC290 and LCC3021 have power ratios of 0.419 and 0.427, respectively, while the k-carrageenan system and the aqueous solution without solid phase have a power factor of 0.283.

An advantage of the technology using gas lift and suction pipe is that the system provides mixing in such a way that the fluid leaving the reactor is completely homogeneous. When operating the 50 L experimental systems as continuous mode fermenters from the bottom of the 50 L total volume reactor, fresh nutrient medium is injected at a rate of 36 mL per minute. This represents an approximate 1000-fold dilution of the feed components. Using the mixing indices calculated for the LCC290 yeast system, the fluid elements are mixed in a circulating loop of about 3 reactors, whereas a circulating loop with 10 reactors is required for a solid solution aqueous solution system. In addition, the residence time (24 hours) is approximately 100 times longer than the mixing time (180 seconds). The fast

144 mixing, combined with the dilution of the nutrient medium in the system, and the large difference in mixing time depending on the residence time clearly indicate a good mixing system. The real prerequisite for using a gas lift bioreactor was the fact that it provided an ideal mixing environment for beer fermentation. The results of the blending process study were obtained for all three fermentation carriers.

6.5 EVALUATION OF MULTIPLE IMMOBILIZATION METHODS TO BE USED FOR CONTINUOUS MAIN FERMENTATION IN SYSTEM β

With gas lift

Continuous fermentation is carried out in experimental

L gas lift bioreactor equipped with a suction tube using three types of immobilizing media: k-carrageenan gel pellets, LCC290 superflocculent yeast and medium flocculation yeast LCC3021. All fermentations are carried out with the same level of yeast inoculum (4 g / L) and industrial beer must is used for the production of light beer. Fermentations are started in batch mode to ensure rapid reduction of sugars in the beer must and to support the yeast growth in the fermenter. In flocculent yeast, the periodic phase provides for the formation of yeast flocculates that could remain in the bioreactor after the onset of continuous fermentation. Once the level of diacetyl in the fermentation liquid drops below 30 µg / L, a continuous supply of beer must begins. The following sections describe in more detail the results of fermentation analyzes for these three types of immobilized matrices.

145

6.5.1 Use of k-carrageenan gel pellets: Grip

The use of k-carrageenan gel pellets as an immobilization matrix helps to evaluate the feasibility of continuous main fermentation of beer using gas lift (Section 6.3). It is also necessary to evaluate whether such a system would be able to operate for a longer period (up to 2 months) without major technological difficulties, including fermentation instability and pollution. The operating parameters for these fermentation experiments are presented in Figure 6.59. The surface velocity of the carbon dioxide gas is set at 5.5 mm / s and air is injected into the reactor at a velocity of 0.9 mm / s. It follows that the oxidation rate is adjusted to 3% of the total amount of gas sprayed, which corresponds to the results given in section 6.3, which indicates the production of the preferred beer when 2-5% oxygen is introduced into the system. The fermentation temperature was adjusted to 15 ° C. Some instability can be noticed in the data and it is due to the type of process control circuit used.

Figure 6.60 presents the quantification of the population of free yeast cells and their viability depending on the duration of fermentation. Viability remains relatively high during the two-month fermentation with a temporary decrease, measured around two hundred hours. This corresponds to the point just before the continuous supply of beer must begins. In periodic fermentation, the viability is usually reduced at the end of the fermentation because there is a lack of nutrient medium for the cells. After the continuous supply of beer must begins, the viability returns to 90%. The population of free yeast cells is low

146 during the first 400 hours of fermentation, and then in the next 300 hours it increases about 10 times - from ~ 100 million cells per 1 mL to ~ 1.5 billion cells per 1 ml_. Once this maximum concentration is reached, the free-cell yeast population maintains a pseudo-stable state for the remainder of the fermentation period.

The sudden increase in the concentration of free cells probably correlates with the population of immobilized yeast cells. At the beginning of the fermentation, the yeast entrapped in the gel grows in the gel as it takes up all the space available. Once the gel pellets are filled with yeast, the increasing population will pass into the liquid medium. Our results show that during the first 400 hours, immobilized yeast grows inside the gel and after 700 hours the yeast no longer has room to grow in the gel, which is why more cells are released into the fermentation fluid.

This instability in the yeast population is affected by ethanol and relative weight indicators (Figure 6.61). During the initial 200 hours of fermentation, it can be expected that the ethanol content will increase excessively and the relative weight will decrease, following the traditional kinetics of periodic fermentation. Between 200 and 600 hours, ethanol reaches that part of the curve that is parallel to the abscissa and corresponds to 45 g / L and a relative weight of about 6 ° P. This result is not expected as the fermented liquid at the end of the fermentation process has a relative weight of 2.5 to 2.7 ° P. At about 600 hours, when the free yeast cell population has just reached its maximum, the ethanol level rises to about 70 g / L and the relative weight drops to about 2.2 ° P. A closer look at the content of specific carbohydrates depending on

147

from the duration of the fermentation (Figure 6.62) shows that the concentration of maltose does not reach equilibrium until approximately 600 hours of the process. As expected, other sugars are down. The other two key assays are continuously monitored throughout the two-month continuous process, diacetyl and 2,3-pentanedione (Figure 6.63). The low points at approximately 180 hours correspond to the end of the periodic phase. As the material was fed continuously, the content of diacetyl and 2,3-pentanedione increased sharply to about 500 pg / L and 400 pg / L, respectively. This initial increase is expected because the supply of fresh nutrient medium stimulates the growth of yeast, which consequently increases the levels of excess metabolites leading to the production of diacetyl and 2,3-pentanedione. During fermentation, levels of 2,3-pentanedione remain above 400 pg / L, while the diacetyl concentration decreases from 500 pg / L to 275 pg / L in the middle of the fermentation process. This point also coincides with the level of diacetyl, falling below the level of 2,3 pentanedione, which was observed in the feasibility assessment referred to in section 6.3.

148

200 400 600 800 1000 1200

Duration of fermentation (h)

Temperature

Carbon dioxide

Air Figure 6.59 Performance parameters of fermentation with carrageenan immobilized yeast depending on the duration of the fermentation.

149

Viability

Cells Figure 6.60 Total yeast cell concentration and viability of immobilized yeast in k-carrageenan depending on the duration of fermentation.

Ethanol

Viability Figure 6.61 Ethanol concentration and relative weight in continuous fermentation with immobilized yeast in k-carrageenan depending on the duration of fermentation.

150

-0- Polysaccharides

Maltose • 43 ”Maltotetrosis Q · Glycerin

Maltotriose

Figure 6.62 Carbohydrate content of continuous fermentation with immobilized yeast in k-carrageenan depending on the duration of fermentation.

- • - Diacetyl

Pentadion

Figure 6.63 Concentration of higher diketones in continuous fermentation with immobilized yeast in k-carrageenan depending on the duration of fermentation.

151

6.5.2 Use of a superfluidizing strain of brewer's yeast:

Samothle smoked a wound

Continuous fermentation in an experimental 50 L GPST-type bioreactor was carried out with LCC290 super-flocculent yeast for 3 months. Spraying CO ₂ gas is injected at a rate of ~ 2,5 mm / s, and the air is supplied at a rate of ~ 0,4 mm / s, in order to ensure the growth of brewer yeast (this corresponds to a total of 1,5 L gas per minute , of which 3% is oxygen), the fermentation temperature in the reactor is adjusted to ~ 15 ° C throughout the process. Stopping the power supply forced us to lower the temperature in the fermenter to 4 ° C over a period of three days (about 1700 hours) (Figure 6.64). The reactor was cooled to slow the metabolism of the yeast and to maintain its viability during power outages. This unexpected case gave us the appropriate opportunity to assess the flexibility of the system with respect to situations that would normally occur in industrial production. After the power was restored, the reactor temperature was again adjusted to 15 ° C and the fermentation continued for another 30 days.

Upon completion of the initial batch phase (~ 180 hours), a continuous supply of wort to the system at a flow rate of 2.16 L per hour was started, thus providing a residence time of 24 hours, based on the reactor working volume of 50 L After the initial batch phase, the cell concentration increased, reaching 3 trillion cells per ml per 750 hours of fermentation. (Figure 6.65). The yeast mass is then reduced to approximately 1 trillion cells per mL for about 1000 hours and remains at this level until the end of fermentation. The yeast viability is 90% throughout the fermentation process (Figure 6.65).

152

Ethanol concentration and the relative weight of the liquid medium reach a pseudo-stable value shortly after continuous loading begins (Figure 6.66). The ethanol concentration rises to about 70 g / L and the relative weight of the liquid is ~ 2.3 ° P during the remaining fermentation time. The carbohydrate concentration data shown in Figure 6.67 confirm this pseudostable condition for approximately 270 hours during the continuous fermentation process. The polysaccharide concentration drops from ~ 42 g / L to ~ 33 g / L about 1400 hours of fermentation. This result is explained by the differences in the batches of the must. Since light yeast breweries cannot consume these polysaccharides, this anomaly in the nutrient medium of the wort does not have a noticeable effect on the performance of the main fermentation apparatus. This difference in the proportion of non-fermentable sugar should be detected by the members of Expert Committee on Experimental Tests, which should note that the product has a "poor" consistency.

Diacetyl and 2,3-pentanedione concentrations throughout the process are presented in Figure 6.68. As with continuous fermentation using k-carrageenan, diacetyl and 2,3 pentanedione levels increase further with the onset of continuous wort feeding. Diacetyl reaches about 375 pg / L, while the concentration of 2,3-pentanedione is about 600 pg / L. These concentration levels are maintained throughout the fermentation until electricity is interrupted at ~ 1,700 hours. As the liquid remains for 3 days without metabolism of the yeast (no nutrient medium is supplied), the levels of higher diketones decrease. After the wort is restored, the diacetyl and 2,3 pentanedione return to their pseudo-steady state levels.

153

Figure 6.64 Performance parameters of fermentation with LC290 depending on the duration of fermentation.

6.00E + 09 &

5.00E + 09

100%

85%

4.00E + 09

3.00E + 09

2.00E + 09

1.00E + 09

OOOE + OO

70%

55%

40%

25% about about X VO about about about about X

S *

I ¹ r * I h

1000 1500 2000 2500 3000

Duration of fermentation (h)

10%

-A- Cells- · - Viability

Figure 6.65 Total yeast cell concentration and viability of LC290 yeast depending on the duration of fermentation.

154

I -A- Ethanol ~ E ~ Relative weight |

Figure 6.66 Ethanol concentration and relative weight of LC290 yeast in continuous fermentation depending on the duration of fermentation.

—E- Polysaccharides · And- Maltotetrosis -Y ^- Maltotriose Maltose -Θ- Glycerin

Figure 6.67 Carbohydrate content of continuous yeast fermentation LC290 depending on the duration of fermentation.

155

Diacetyl - · - Pentadione Figure 6.68 Concentration of higher diketones in continuous fermentation with LC290 yeast depending on the duration of fermentation.

156

6.5.3 Use of medium flocculant yeast:

self-flooding

Fermentation was performed under several operating modes using LLC3021 medium flocculating yeast strain as an immobilization matrix. the fermentation was carried out in an experimental 50 L gas-lift bioreactor equipped with a suction tube. As with the two earlier immobilization methods, the initial concentration of yeast was set at 4 g / L. These brewer's yeasts ferment beer must for the industrial production of light beer (described in section 4.2) and ferment the batch until all fermentable sugars have been consumed and diacetyl levels fall below 30 pg / L. The fermentation temperature was adjusted to 15 ° C and the rate of the spray gas was adjusted to the same level as that of fermentation with LCC290 (carbon dioxide surface velocity of –2.5 mm / s and –0.4 mm / s of air providing approximately 3% oxygen in the total amount of gas sprayed.) (Figure 6.69).

This initial periodic stage enables the flocculation of the yeast cells and is therefore easier to retain in a system using gas lift. At the end of the periodic fermentation, the wort feeding rate is adjusted to 2.16 L per hour, which corresponds to a residence time of 24 hours, taking into account the reactor working volume of 50 L. The yeast population (Figure 6.70) increases to about 1 trillion cells per milliliter and stays at this level for 1000 hours (between 500 and 1500 hours during continuous fermentation. The brewer's yeast population suddenly increases about ~ 1500 hours of fermentation and then stabilized at 2 trillion cells / mL the brewer's yeast population

157 was unexpected. The viability of the yeast throughout the fermentation process was maintained at 90% (Figure 6.70).

Figure 6.71 shows the ethanol concentration and relative weight of the fermentation fluid during the three-month continuous fermentation. Shortly after the start of the batch stage (180 hours), the ethanol concentration stabilizes at 70 g / L and the relative weight reaches a minimum of ~ 2.2 ° P. It is anticipated that a sudden increase in the yeast population would reflect a decrease in ethanol concentration. The most logical explanation for the increase in brewer's yeast population is that the greater part of the population enters the growth phase, resulting in a doubling of their concentration. A decrease in ethanol concentration could be expected in connection with an increase in the concentration of brewer's yeast, but this is not the case since ethanol 'remains in a pseudo-stable state and its value is 70 g / L in | the whole course of the process. Concentration of carbohydrates as a function of the fermentation duration (Figure 6.72) shows the same results, both ethanol characteristics and relative weight.

[j With this course of fermentation a pseudo-steady state is reached j approximately 250 hours of continuous fermentation.

!

| Figure 6.73 shows the concentration characteristics of {diacetyl and 2,3-pentanedione depending on the duration of | fermentation. As with the results for higher diketones using k-carrageenan gel and LCC290, the concentrations of j

diacetyl and 2,3-pentanedione increase after the start of the fermentation phase and reach a pseudo-stable state

at a value of -225 µg / L and 400 gg / L, respectively.

158

ο ο

Duration of fermentation (h)

Carbon dioxide "*" Air

Temperature

Figure 6.69 Fermentation performance parameters with LCC3021 depending on fermentation duration.

159

6.00E + 09

-5.00E + 09 i E * = 4.00E + 09

I 53.OOO + O9 h § · * 2.00E + 09 w S * ⁵ 1.00E + 09

OOOE + OO

Duration of fermentation —Jr · Cells - # - Viability

Figure 6.70 Total concentration of yeast cells and viability of LCC3021 depending on the duration of fermentation.

Figure 6.71 Ethanol concentration and relative weight in fermentation with LCC3021 depending on the duration of fermentation.

160

· ·· Polysaccharides -Ο Maltotetrosis - Maltotriose

Maltose —G— Glycerin

Figure 6.72 Concentration of carbohydrates in continuous fermentation with brewer's yeast LCC3O21 depending on the duration of fermentation.

Pentadion

Diacetyl- - Figure 6.73 Concentration of higher diketones upon fermentation with LCC3021 depending on the duration of fermentation.

161

6.5.4 Comparison of different media

Sections 6.5.1 to 6.5.3 present the fermentation characteristics of k-carrageenan gel pellets, LCC290 superflocculent yeast and LCC3021 medium flocculent yeast as immobilization matrices. All three carriers are considered suitable for continuous main fermentation in a gas-lifted experimental bioreactor equipped with a suction tube and a volume of 50 L.

The residence time for all three cases is 24 hours, fermentation using the LFC290 superflocculent yeast reaches a pseudo-steady state much faster than the use of the LCC3021 medium flocculent yeast systems and immobilized cells in k-carrageenan. Fermentation with LCC290 achieves a maximum ethanol concentration of 70 g / L in about 250 hours. Using LCC3021, a steady state ethanol concentration of 70 g / L is achieved for about 600 hours. During continuous fermentation with k-carrageenan, ethanol is equilibrated at two separate points during the process. First, ethanol reaches 45 g / L between 200 and 500 hours and then rises to 70 g / L after about 757 hours and remains at this concentration until the end of the experiment.

The three fermentation systems appear to reach a maximum free cell concentration of -1 trillion cells per milliliter. Variability in beer yeast concentration affects ethanol production in the k-carrageenan system (lower initial pseudo-stable ethanol concentration compared to the LCC290 yeast system). The concentration of brewer's yeast peaks at different time intervals in each system. For continuous fermentation with LCC290, ethanol peaks between 500 and 1000 hours, while fermentation with LCC3021

162

the calculated maximum value is between 1500 and 2000 hours. The k-carrageenan immobilization system reaches a maximum cell concentration between 700 and 1000 hours of continuous process.

The pseudo-stable values of diacetyl and 2,3-pentanedione concentrations in the three fermentations with immobilized LCC290 superfluous yeast cells, medium flocculating yeast LCC3021 and immobilized in k-carrageenan yeast are different. For fermentation with LCC290, stabilization is achieved at about 375 µg / L, while for fermentation with LCC3021 the stabilization level is at about 225 µg / l. In fermentation with k-carrageenan, the concentration of diacetyl reaches 500 gg / L and, in the middle of the fermentation process, the level gradually decreases to about 200 µg / L within 500 hours. The concentration of 2,3-pentanedione reflects the concentration of diacetyl at all three fermentation runs, with concentrations of 2,3-pentanedione being higher than those of diacetyl during all fermentations with LCC290 and LCC3021. Fermentation with k-carrageenan shows a different trend - with diacetyl levels higher than those of 2,3-pentanedione during its first pseudo-steady state, after which the concentration of diacetyl falls below that of 2,3-pentanedione. The data on the concentration of brewer's yeast and the resulting ethanol also indicate that two separate and unique pseudo-stable conditions are observed during fermentation with k-carrageenan.

The work of comparing different fermentation systems and evaluating which one works better can be complex when the qualities of the systems are evaluated on the basis of more than one criterion. For example, if only total ethanol productivity were used to measure success, all three test systems would be equivalent, since obtaining 70

163 g / L of ethanol for a reactor volume of 50 L over a 24-hour period was achieved in all systems.

Towards the production of marketable beer is

have more requirements beyond ethanol production. The proposed fermentation system could be evaluated in terms of its ability to produce quality beer (among other indicators and ethanol productivity and diacetyl levels), taking into account the possible gradual increase in carrier costs, carrier availability, ease of use. working with the system, protecting the environment in the storage of media, the stability of the media, and the flexibility of the work provided by the media. In order to evaluate a multidimensional problem, business environments use a dimensionless analysis method called the “Comparison Chart for Reporting Results” (Kaplan and Norton, 1996). The first stage involves identifying the criteria by which the system is to be evaluated. Each of the criteria is then rated on a scale of 1 to 5, with 1 indicating the least approval and 5 indicating the highest approval. At the end of the analysis, the result of each option is calculated and the alternative with the highest score is the best choice.

Table 6.2 presents the results of the analysis according to the “Comparison Map for Results” conducted for the immobilization media, which are considered as possible alternatives for use in the experimental gas lift powered bioreactor equipped with a suction tube and a volume of 50 L for fermentation. A total of 6 carriers were evaluated and their main objective was the feasibility of producing marketable beer. These carriers are as follows: Chitopearl, chitosan Celite granules, kieselguhr, Siran® glass granules, k-carrageenan gel granules, medium flocculent yeast

164

LCC3021 and over-flocculating yeast LCC290. Each media was evaluated according to the above scale. LCC290 super flocculent yeast performs best, followed closely by medium flocculent yeast LCC3021.

The remaining four carriers received results between 16 and 20. A third preference was given to the k-carrageenan system, since beer was commercially available in the experimental unit. The evaluation of the carriers clearly indicates that future work on the research of continuous fermentation systems should focus on self-aggregation as a way of immobilization. Affordability (easily accessible), costs (low costs due to no need for additional equipment for work), convenience of work (coherence within existing operations of the installation) should be provided by such an alternative that outweighs the possible instability of yeast flocculates in a stirring system. It is possible to use the tendency to cut self-aggregated particles to adjust the size of flocculates during the fermentation process and it is possible to

achieves greater mass transfer, leading to a further increase in volume capacity.

Table 6.2 Comparison of several carriers for immobilization during the main fermentation of beer in a bioreactor system with gas lift

Chitopearl® Celite® Siran® Carrageenan LCC3021 LCC290

Good-quality beers

Costs 2

Accessibility 1

Convenience with 3 work

14

13

52

13

165

Deposits- 4 3 4 2 3 3 not / surrounding environment Sustainability 4 1 1 3 2 3 Flexibility 3 3 3 1 2 5 Total 20 17 16 18 26 30

6.6. Beer production of North American light beer by technology using gas lift and suction pipe

The production of North American (CA) light beer poses many obstacles for brewers. Light beer of the CA type is characterized by a light color and low bitterness, low residual sugar, no dominant aroma, resulting in no residual taste. Because of these intrinsic properties, brewers can mask very few taste deficiencies. High levels of diacetyl (bitterness), acetaldehyde (green apple), as well as a lack of taste of sulfur compounds (burnt rubber, odor of skunk, spoiled eggs, boiled vegetables) are the most common taste problems that hinder brewers . Although the bacterial contamination of the fermentation medium may also cause the aroma of the beer to be at the wrong level, improper control of the fermentation process more often results in higher than expected levels of non-compliance with the flavor requirements.

During continuous fermentation experiments conducted as part of these theses, the levels of contamination in the casting must and in the fermentation vessels were controlled by strictly established aseptic methods, fermentation experiments with all three types of carriers, which lasted several months, showed there are no detectable levels of pollutants (observations were made using the methods described in Chapter 4). Higher than desired levels of diacetyl (aiming to be below 30 pg / L) and acetaldehyde (aiming to be below 10 mg / L) create a problem for the product obtained by continuous main fermentation, but these levels are not due to bacterial contamination. These findings are not dissimilar to those reported in the literature (Pajunen et al., 2000; Kronlof et al.

166 al., 2000). The Belgian brewer explains the high levels of acetaldehyde produced in the process of continuous fermentation with sales-related indicators and labels the product as apple flavored beer (Andries et al., 1996).

High levels of diacetyl accompanying main fermentation are also a normal occurrence in the brewing industry. Some brewers have adopted the practice called "free temperature increase" after the completion of main fermentation to help reduce diacetyl content. Others simply choose to hold the products for a longer period during maturation to reduce the higher diketones (diacetyl and 2,3-pentanedione) to the desired levels. In another approach, several research groups are investigating the speed-maturing technology discussed in Chapter 2 to reduce diacetyl levels. While this approach is effective, it adds another degree of complexity to the whole brewing process, which some find difficult to accept.

The savings from the rapid maturation process are very clear, but it may be advisable to minimize technological difficulties in the early use of continuous brewing in the brewing industry to facilitate the transition from traditional batch fermentation to continuous production. It was therefore decided to investigate periodic retention (batch retention) after continuous main fermentation in experimental gas lift and 50 L reactors to control the high levels of diacetyl in the finished beer product. This additional technological step was not foreseen at the beginning of this Ph. D. program, but it was necessary to take such a measure to compare beer,

167 obtained in continuous mode, with beer controls obtained in a batch process.

6.6.1 Periodic retention of the product after continuous main fermentation

An important parameter for determining the completion of major fermentation is the level of diacetyl in the final fermented liquid. The conversion of the diacetyl precursor, α-acetolactate, into diacetyl is a limiting step with respect to diacetyl (Figure 3.5). This first reaction is chemical and depends largely on temperature. If submitted 'green beer to the cold aging process before

the chemical conversion of α-acetolactate to diacetyl, the resulting beer may have levels of diacetyl above the taste limit of 20 µg / L, unless long periods of cold aging are used to ensure a slow conversion of the precursor. For all three continuous fermentations described in section 6.5, the diacetyl level at the reactor outlet was above the desired values of 30 µg / l in the undiluted beer. If the liquid is filtered at this stage to remove brewer's yeast, the diacetyl will remain high, which means that a periodic period at higher temperatures should be used to reduce the diacetyl content to the required level.

The beer obtained by continuous fermentation is collected and held in 40 L stainless steel vessels at 21 ° C. Small samples of the liquid (100 mL) were drained regularly and analyzed for diacetyl, ethanol, relative weight, esters and fusel alcohols. Figure 6.74 shows the decrease in diacetyl as a function of the retention time of the product in the warm state for one batch of beer fermented in continuous brewing yeast LCC290 as an immobilized matrix. The period of retention in a warm state results

168 regarding the reduction of diacetyls from ~ 600 pg / L to less than 30 pg / L, which is considered in the brewing industry to be a "pre-decline.

In another experiment, the effect of stirring on the reduction of diacetyl content during the retention period was investigated. Carbon dioxide gas (0.14 m ³ / h) was sprayed from the bottom of the collecting vessel through a 1.27 cm stainless steel pipe system to stir the liquid during retention. Figure 6.75 shows the results of the experiment. Stirring by CO ₂ spraying has been shown to have very little effect on the rate of decrease of diacetyl content in the retention vessel. This result may be indicative of the unsatisfactory results of carbon dioxide mixing and therefore no increase in the reaction rate of the first chemical reaction (conversion of α-acetolactate to diacetyl) is observed or no increase in mass transfer rate for the second reaction, to be done quickly (converting diacetyls to acetone using brewer's yeast). It may also be possible for the suspension in the non-agitated vessel to contain sufficient cells to reduce the diacetyl content by converting it to acetone, which is not aromatic when inactive. with regard to the rate of chemical conversion (first stage).

The effects of periodically holding batches in the heated state after continuous main fermentation on the concentration of esters and fusel alcohols, as well as on the concentration of ethanol and relative weight, are presented in Figures 6.76 and 6.77, respectively. These results indicate that retention in the warm state has little effect on

169 concentrations of acetaldehyde, ethyl acetate, propanol, isobutanol, isoamyl alcohol and isoamyl acetate (Figure 6.76) The relative weight of the fluid in the holding vessel decreases from 2.7 ° P to 2.0 ° P during the first 12 hours of holding and then stabilizes to this lower value. Ethanol concentration was stabilized at 70 mg / L throughout the 65-hour test period. These results indicate that the retention period influences mainly the concentration of diacetyl and 2,3-pentanedione, while its effect on the content of esters, fusel alcohols and ethanol is minimal.

The periodic retention was carried out with a liquid obtained by continuous fermentation in a 50 L gas-lifted bioreactor using LCC290 superfluidic yeast, medium flocculent yeast LCC3021 or immobilized yeast yeast in k-carrageenan. Data to reduce

the diacetyl content of these three test systems are presented in Figure 6.78. The diacetyl content is reduced to a target value of 30 pg / L in all systems. But the time required to reach this reduction varies in all three cases. In the case of LCC290, a reduction of 600 pg / L to 30 pg / L is achieved in approximately 48 hours, and in fermentations with LCC3021 and k-carrageenan, only ~ 24 hours and ~ 40 hours are required to reach the set point. It was accepted that this discrepancy was due to the diacetyl baseline values and not to the type of immobilized matrix used.

Figure 6.79 illustrates the same results for the diacetyls of Figure 6.78, adjusting the retention time based on the fermentation results with LCC3021 and k-carrageenan. The actual diacetyl depletion schedules for the last two fermentations have been offset so that their initial values fall on the diacetyl depletion graph obtained at

170 the use of LCC290 super flocculant yeast. With this transformation, the decrease in diacetyl content for the three systems falls on the same line. Using Table Curve 2D software, these results, expressed graphically, represent a curve corresponding to the kinetic equation (Levenspiel, 1972) of the first line of the graph shown in Figure 6.80. The experimental data in Figure 6.79 were found to fit the following equality: [Diacetyl] = 648.54 is ⁽ ' ⁰ ' ^{0426 4} (6.1) with a correlation coefficient of 0.96. These results strongly confirm the theory that all three immobilized systems show the same diacetyl reduction potential, and the results are not surprising since diacetyl reduction is often associated with a yeast strain (Nakatani et al., 1984) .The k-carrageenan system immobilizes LCC3021 in its yeast gel structure, and brewer's yeast LCC290 is the chosen variant MA LCC3021.

After the diacetyl level was below the setpoint of 30 pg / L, the resulting beer was again kept cool (2 ° C) for 7 days before proceeding to the final stages of production (filtration, dilution, carbonation and bottling). packaging). Table 6.3 gives a brief analysis of the beer obtained in an experimental system using LCC290 brewer's yeast, LCC3021 brewer's yeast or immobilized brewer's yeast as immobilized matrices, Figure 6.81 presents a radar graph of esters and fusel alcohols in three types of continuous brewed beer and one industrial control brew obtained using the batch mode. Compared to the industrial brew in the batch (control), the liquids produced in the continuous mode have a lower ester content (ethyl acetate, isoamyl acetate), higher

171 propanol content and lower isobutanol content, primary amyl alcohol and isoamyl alcohol. The levels of acetaldehyde in the products obtained by continuous fermentation are higher than the control fluid. The foaming levels, the initial clouding of cold, the turbidity of heat, dimethyl sulfide, sulfur dioxide, carbon dioxide and air are all within the limits of the standards.

Several other parameters (apparent extract, actual extract, calculated initial extract, color, bitterness), which are closely influenced by the dilution of the product and the initial concentration of ethanol to the final values of this indicator - 5,0% (v / v), are different from the control, since the products obtained in continuous mode require a greater degree of dilution with water to reach the desired ethanol content due to the higher initial ethanol concentration values (70 g / L vs. 60 g / L in a batch process).

Figure 6.82 shows the radar graph for alcohol, pH, color, bitterness of the same liquids described above. Alcohol, diacetyl and pH levels are good and are within the set range, while color and bitterness are out of the norm. The lighter color is related to the higher dilution required by the fluids produced in continuous mode, and this can be adjusted by increasing the degree of coloration of the nutrient medium from the wort. The bitterness values are also dependent on the dilution and could also be adjusted by the wort.

172

Warm-up retention period (h)

Diacetyl is Pentadione The diacetyl set point

Figure 6.74 Concentration of higher diketones as a function of the retention time of batches in the warm state after continuous main fermentation with brewer's yeast LCC290 in a gas lift reactor.

[♦ Without stirring · With stirring · Diacetyl set point figure 6.75 Diacetyl concentration depending on the duration of warming-up of batches after continuous main fermentation with brewer's yeast LCC290 in a gas lift reactor. One sample was stirred by carbon dioxide spray and the other sample was left in a static position.

173

Warm-up retention period (h) • Acetaldehyde Ethyl acetate Propanol A isobutanol oIsoamylalcohol Isoamyl acetate

Figure 6.76 Esters and Fusel Alcohol Concentrations as a function of the duration of warmed-up batch retention after continuous main fermentation with brewer's yeast LCC290 in a gas lift reactor.

Warm-up retention period (h)

F about Z δ

A Ethanol Relative weight Figure 6.76 Esters and fusel alcohol concentrations depending on the duration of the batch retention in the warm state after continuous main fermentation with brewer's yeast LCC290 in a gas lift reactor.

174

Periodic retention time (h) ♦ LCC290 * LCC3021 · Carrageenan Diacetyl set point

Figure 6.78 Concentration of diacetyl as a function of warming-up time of batches after continuous main fermentation with LCC290 brewer's yeast, LC3021 brewer's yeast and immobilized brewer's yeast in a k-carrageenan gel in a gas system

Duration of Periodic Hold (b) ♦ LCC290 LCC3021 Adjustment • Carrageenan Adjustment Prev. reduction limit

Figure 6.79 Concentration of diacetyl as a function of warming-up time of batches after continuous main fermentation with LCC290 brewer's yeast, LC3021 brewer's yeast and immobilized brewer's yeast in k-carrageenan gel in a gas lift system. The fermentation values with LC3021 and k-carrageenan are adjusted to have the same starting point as the fermentation with LCC290.

175

Figure 6.80 Concentration of diacetyl as a function of the retention time of batches in the warm state after continuous main fermentation with LCC290 brewer's yeast, LC3021 brewer's yeast and immobilized brewer's yeast in a k-carrageenan gel in a gas lift system. The fermentation values with LC3021 and k-carrageenan are adjusted to have the same starting point as the fermentation with LCC290.

176

Table 6.3 Brief description of the analyzes of beer produced in a gas lift system.

Indicators Liquid obtained by industrial batch fermentation Liquid obtained by continuous fermentation with LCC290 Liquid obtained by continuous fermentation with LCC3021 Liquid obtained by continuous fermentation with carrageenan Air (mL) <1 0.40 0.35 0.35 Carbon dioxide (%) 2.75 2.90 2,81 2.73 Sulfur dioxide (mg / L) <10 0 0 0 Dimethyl sulfide (cdd) <70 59 30 30 Bitterness (BU) 12,00 11.30 13,74 13,18 Color (SRM) 3.20 2.20 2.10 2,30 pH 4.10 4:15 4.10 4.09 Diacetyl (pg / L) <20 12 12 9 Alcohol%, (v / v) 5,00 4.99 5.03 5.04 Alcohol%, (w / w) 3.93 3.93 3.96 3.96 Phantom extract (° P) 1.55 1.02 1,40 1.44 Valid 3,36 2.84 3.24 3.27

extract '(° P)

Initial extract (° P) 11,00 10.5 11,0 11,0 FTU <200 45 50 47 Initially <100 43 51 54 clouding of cold (FTU) Sparkling > 170 167 187 175 Acetaldehyde (mg / L) 4.4 +1, 3 10,0 21.9 11.6 Propanol 12.8 ± 6.8 57.6 51,7 84,3

(mg / L)

177

Ethyl acetate (mg / L) 32.4 ± 4.3 11,0 10.6 8.7 Isobutanol (mg / L) 21.6 ± 3.4 10.6 8.3 8.9 Source amyl alcohol (mg / L) 20.00 ± 2.3 15.2 9,4 6.4 Isoamyl alcohol (mg / L) 60.9 + 8.6 48,0 40,9 46.5 Isoamyl acetate 2.5 ± 0.7 0.32 0.25 0.18

(mg / L)

178

Figure 6.81 Radar graph of beer obtained by continuous fermentation in a gas lift system with brewer's yeast LCC290, brewer's yeast LCC3021 or brewer's yeast LCC3021 immobilized in k-carrageenan gel pellets. The esters and fusel alcohols of the finished beer product are shown in this graph. After continuous main fermentation, all products are subjected to periodic warming of batches.

179

Color (SRM)

I 'Control- El - LCC290' A · LCC3021 Carrageenan • j j j

I

Figure 6.82 Radar graph of beer obtained by continuous fermentation in a gas lift system with brewer's yeast LCC290, brewer's yeast LCC3021 or brewer's yeast LCC3021 immobilized in k-carrageenan gel pellets. The esters and fusel alcohols of the finished beer product are shown in this graph. After continuous main fermentation, all products are subjected to periodic warming of batches.

180

6.6.2 Choosing the Right Spray Gas Carbon dioxide is suitable for most breweries as it is a natural by-product of yeast fermentation. The released carbon dioxide is collected in the breweries and then the gas stream is purified to remove the minor impurities that can be carried with the stream (usually sulfur compounds). This purified stream is then compressed and stored in liquid form for further use in the brewery (99.95% purity). The use of carbon dioxide as a gas for dispersion in continuous fermentation seems to be a logical choice in terms of performance. The installation could use collection system to extract _CO2 leaving continuous fermentation. The use of other gases would only add to other difficulties in the execution of existing operations.

But in order to make continuous fermentation a viable alternative to the current batch operations, it is necessary to produce a product similar to the existing trademarks. It is believed that by reducing the biological / biochemical effects to which the brewer's yeast is subjected during continuous fermentation, it would be possible to achieve a similar product. Carbon dioxide is known to adversely affect the metabolism of yeast during the main fermentation. This effect is enhanced by the use of high tsilindrichnokonichni apparatus where the pressure a pressure which is inherent in the system, impedes the free release of CO ₂ from the liquid medium. Under these conditions, the beer products produced have a lower ester content and higher fusel alcohols. Successful attempts have been made to remove some of the CO ₂ from

181

fermentation by periodically injecting an additional gas stream from the bottom of the cylindrical conical apparatus. It can be assumed that the effects of CO ₂ inhibition are reflected in the reduction of fusel alcohols and the increase of ester levels in the final product.

In view of these facts, the use of nitrogen instead of carbon dioxide as a gas to be dispersed in the gas lift system was examined. Several liquids fermented in 50 L reactors and gas lift were obtained from the Labatt Experimental Brewery, using the following steps: After 14 hours of fluid collection from the reactor (24 hours residence time) green Beer is separated from brewer's yeast. This fluid was then held for 48 hours at room temperature (2GC), which provided the diacetyl and acetaldehyde to reach Labatt standards (diacetyl <30 µg / L and acetaldehyde <10 tdD). This fluid was then left for aging. in cold weather for 7 days, after which it is processed according to industrial practice.

Table 6.4 compares the performance of finished beer products obtained by continuous fermentation with brewer's yeast LCC290 and CO ₂ and N ₂ gas spray compared to standard industrially produced liquid (control). An equivalent estimate of the liquid obtained from nitrogen spraying was compared with that produced by the liquid. Analyzes show that there is twice as much 1-propanol in the liquid, while the concentration of dimethyl sulfide is approximately three times lower. Color and bitterness indicators are higher than industrially produced liquid, as is the foaming potential measured by the NIBEM test. Beer obtained by CO ₂ spraying has a lower content of esters (ethyl acetate, isomyl acetate) and a higher content of 1,182 propanol compared to the liquid obtained with nitrogen gas sprayed. The ratios of esters (ethyl acetate, isoamyl acetate, ethyl hexanate, ethyl octanate, ethyl decanate) to fusel alcohols (1-propanol, isobutanol, isomyl alcohol) are calculated for the control fluid and for the fluids obtained with carbon dioxide dispersion of 0.30 , 0.27 and 0.15.

Table 6.4 Chemical analyzes for several products obtained by continuous fermentation in a gas lift system using LCC290 ultra-flocculent yeast.

Analysis Liquid from industrial batch fermentation Liquid from continuous fermentation and dispersion of nitrogen Liquid from continuous fermentation and spraying. of CO ₂ Acetaldehyde (mg / L) 2,53 3.68 4.80 Ethyl acetate (mg / L) 28.08 30,90 14.14 1-propanol (mg / L) 13.03 28,45 40.03 Isobutanol (mg / L) 17.15 17,89 7.01 Isoamyl acetate (mg / L) 2.57 2.06 0.69 Isoamyl alcohol (mg / L) 74.54 78.45 51.35 Ethylhexanoate (mg / L) 0,140 0,180 0,074 Ethyl octanoate (mg / L) 0,110 0.280 0,059 Ethyl decanoate (mg / L) 0,0079 0,0630 0,0081 Diacetyl (µg / L) 6 9 10 2,3-pentanedione (pg / L) 4 5 14 Sulfur dioxide (mg / L) 1.3 1 0 Dimethyl sulfide (mg / L) 79 24 64 Bitterness (BU) 11.5 20,6 15.5 Color (SRM) 3.1 4,1 3.7 Foam 176 210 195 CAA (mg / L) 92 84 pH 4.13 4.10 4:19

183

Act. extract (° P) 3,38 4.08 3,601 th most common Initial extract (° P) 10.97 12.77 11.50 Alcohol (v / v%) 4.93 5.74 5.17

Figure 6.83 is a radar graph showing the concentrations of esters, fusel alcohols and acetaldehyde for the three liquids. The profile of the liquid obtained by spraying nitrogen closely follows the profile of the control beer, except for the higher propanol content. CO ₂ sputter fermentation shows a much lower content of esters and fusel alcohol that are not similar to controls. For both liquids obtained by continuous fermentation, the levels of diacetyl and acetaldehyde are below Labatt's standards.

The data obtained show that the nitrogen dispersion increases the production of esters to levels similar to those for commercial beer products, while the carbon dioxide spray produces liquids with a relatively lower concentration of esters. These results show that the metabolism of yeast in altered CO ₂ media changes. Propanol levels, irrespective of the type of gas being sprayed, are much higher with the beer obtained by continuous fermentation compared to the measured values of the control beer obtained from industrial fermentation. Although propanol concentrations are below the 100 mg / L limit, the observed differences may be an indicator of a slight change in metabolism in continuous fermentation compared to periodic fermentation. It is also possible that the higher level of propanol is due to the continuous supply of the amino acid threonine, from which the decomposition of the oxyacid produces propanol.

184

Figure 6.83 Radar graph comparing the esters and fusel alcohols of liquids obtained by continuous fermentation with nitrogen gas and CO2, with those obtained by industrial periodic fermentation.

185

CHAPTER 7. CONCLUSIONS

As a result of the studies conducted within these theses, the following conclusions can be drawn. Quality beer without major flavors may be produced in an experimental gas-lift bioreactor equipped with a suction tube and a volume of 50 L, using continuous main fermentation, followed by a two-day periodic retention to adjust the diacetyl content. A minimum residence time of 24 hours or 1 revolution of the volume of the reactor per day is achieved during fermentation of beer with high relative weight (17.5 ° P), resulting in fermented liquid (2.5 ° P). . Excessive flocculating yeast LCC290, medium flocculating yeast LCC3021 and immobilized yeast in k-carrageenan are applicable carriers for the gas lift system. The use of dense preformed media such as Silan® glass granules and Celite® kettle granules is not a practical alternative for use with a suction tube and gas lift device. A minimum of 2 months of continuous fermentation has been achieved

mode and use of k-carrageenan gel pellets and a minimum of 3 months of continuous mode fermentation using LCC290 and LCC3021, without detecting microbial contamination or instability of the reactor performance. In addition, the continuous mode system is capable of operating with possible and acceptable differences in industrial beer must during a longer operating mode.

In continuous fermentation using LCC290 ultra-flocculent yeast and nitrogen as a spreading gas, followed by a two-day periodic retention, the product obtained in taste corresponds most closely to the industrial beer taken for control. Two-day periodic detention providing

186 a decrease in the concentration of diacetyl in the product leaving the reactor for continuous main fermentation is effective but not an optimal regulatory mechanism. The decrease in diacetyls in the three continuous fermentation systems tested is very similar and, as previously assumed, this feature could be due to the strain type. The duration of the periodic retention does not affect the concentrations of esters and fusel alcohols in beer during the retention stage.

The use of 3% oxygen in the spray gas provides adequate oxygen levels in the medium of the must, thereby maintaining a viable brewer's yeast population (over 90%) throughout the fermentation process to produce good-tasting beers. Continuous fermentations with the yeast LCC290 and LCC3021 as immobilized matrices reach a pseudostable state much faster than the system using k-carrageenan gel granules. The instability in fermentation with k-carrageenan for immobilization is probably the result of an increase in the population of immobilized brewer's yeast causing the product to ferment below the set levels. For ideally continuous production, this phenomenon is too undesirable and is associated with prolonged start-up time, increasing the time required to interact and start over again, followed by severe damage. A longer start-up phase also requires a longer continuous-mode phase to become attractive to the application.

The process of continuous preparation of gel pellets provides the necessary properties of the pellets for testing in experimental apparatus. However, further optimization of the granule production process is required to obtain granules with a narrow size distribution interval.

187

The process also requires further research to determine its suitability for use on an industrial scale. In addition to Kenics-type research, a static mixer can be provided on a larger scale by increasing the diameter, not just increasing the number of static mixers, and this option should be tested to make it acceptable for use.

The acid pulse indicator instrument used in these theses allowed us to determine the mixing time and the circulation rate in the experimental 50 L bioreactor during actual fermentation in LCC290 superfluidic yeast medium and circulating yeast21 in k-carrageenan. The mixing data corresponds to a damping sinusoidal function from which the mixing duration and the circulation rate are calculated.

Speed mixing in the apparatus with suction tube and gas lift is ensured, with mixing time estimated in less than 200 seconds for the three types of immobilized media. Mixing time slightly decreases with increasing surface gas velocity in all three cases. For all surface gas velocities tested (2mm / s to 6mm / s), the LCC290 system shows the fastest mixing time (between 100 s and 120 s). The duration of the circulation is similar for the three types of carriers regardless of the surface velocity of the gas. The duration of circulation also decreases linearly with the corresponding increase in gas velocity. Complete mixing in the liquid (98% pulse sensitivity) occurs within three to six reactor cycles in the reactor for all immobilization media tested. These results confirm that the tested 50 L experimental system provided the necessary

188

mixing for continuous fermentation. Inappropriate mixing durations are indicative of possible dead zones that are not desired during the fermentation of beer.

This research clearly demonstrates the feasibility of further scale-up of a production system designed, constructed and operating in Labatt's experimental brewery. The use of a suction tube bioreactor and gas lift using ultra-flocculent yeast LCC290 and nitrogen gas spraying followed by two-day batch retention at 21 ° C is recommended as a preferred system.

Detailed Description - Part 2:

CHAPTER 4. MATERIALS AND METHODS

4.1 Yeast strain and characteristics

In this experimental work, a strain of Saccharomyces cerevisiae is used for light beer from the Labatt Cultural Collection (LCC) 3021. Saccharomyces cerevisia is synonymous with Saccharomyces uvarum Beijerinck var. carlsbergensis Kudryavtsev, 1960 (Kurtzman, 1998). At 37 ° C, LCC 3021 does not develop. This helps to distinguish LCC 3021 light beer from most English beer. LCC 3021 is a strain belonging to lower-fermenting brewer's yeast, like most strains for light beer, but there are exceptions. Also, this strain ferments glucose, galactose, sucrose, maltose, raffinose and melibiosis, but not starch. The ability to ferment melibiosis is one tool used by taxonomists to distinguish this strain from yeast for English beer.

189

Like most yeast, LCC 3021 is a polyploid and reproduced by mitotic fission. Under normal conditions, light beer yeasts do not reproduce by meiosis. This is related to the advantage that brewer's yeast strain is genetically stable, since cross-breeding of genetic material is less likely (Kreger-van Rij, 1984).

4.2 Preparation of the yeast inoculum.

Yeast is taken from vials stored in a freezer at a cryogenic temperature of -80 ° C and seeded on a peptone nutrient medium from yeast extract (PYN) agar ((peptone 3.5 g / L; yeast extract - 3.0 g / L; KH ₂ PO ₄ - 2.0 g / L; MgSO ₄ .7H2O -1.0 g / L; (NH ₄ ) ₂ SO ₄ -1.0 g / L; glucose - 20.0 g / L ; arap - 20.0 g / L in dH ₂ 0) to obtain well separated colonies A sterile material consisting of several colonies is taken from developing yeast over 3-4 days and inoculates 10 mL of wort poured into a tube, left to develop at 21 ° C overnight and then material is added to a larger volume of must, usually 200 mL, to increase the yeast biomass In the following days, this mixture is added to another larger volume of must, etc. until the desired amount of yeast is obtained Typically, approximately 20 g per liter of wort is expected to be obtained The culture is centrifuged at 4 ° C and 1.0 X 10 ⁴ rpm (radius = 0.06 t) for 10 minutes. As a result of centrifugation, the liquid is separated and the resulting brewer's yeast sludge with appropriate humidity is used for fermentation.

4.3 Fermentation medium of wort

Beer musts weighing 17.7 ° P are fed to breweries in Labatt, Canada. The concentration of fermentable carbohydrates, relative weight and free amine nitrogen in the casting must used for fermentation throughout the experimental period

190

work are given in Annex A2.1. Further details on the composition of the wort are provided by Dale et al., 1986, Hoekstra, 1975, Hough et al., 1982, Klopper, 1974, and Taylor, 1989.

Periodic fermentations: The wort is heated in an autoclave to 100 ° C for 45 minutes and then cooled before being inoculated with granules of immobilized cells or free-suspended brewer's yeast.

Continuous fermentations: The wort used for continuous fermentation is pasteurized using water vapor (Fisher combined heat exchanger, Eurocal 5FH type is used) before being fed into the bioreactor with gas lift and this wort is continuously monitored for microbial contamination. as described in section 4.6. If dirt is detected in the beer must, it is immediately drained and the installation is charged with new beer must.

The steam pasteurizer operates at a volume velocity of 0.8 m ³ / hr. The apparatus is provided with a tubular retention section where the wort is kept at an average temperature of 85 ° C with a minimum temperature of 80 ° C. The holding area volume is 1.13 x 10 ' ² t ³ , which provides a residence time in this section for 51 seconds. After the heating step, the wort is quenched to 2 ° C at the outlet of the unit.

4.4 Method of immobilization

X-0909 kappa carrageenan gel was provided to us by Copenhagen Pectin A / S. Kapa-carrageenan gel pellets containing trapped yeast yeast cells were prepared in a static mixer, as described in detail by Neufeld et al, (1996), at an initial cell concentration of 10 ⁷ - 10 ⁸ cells per 1 mL of gel , which are determined for each experiment. As shown in FIG. 15, the process in a static mixer is based on emulsion formation

191 between non-aqueous continuous phase (homogeneous phase), vegetable oil (Mazola maize oil) and water-dispersed phase, kappa-carrageenan solution (3% (w / v)) in KCI inoculated with brewer's yeast, using in-line polyacetal static mixers (ColeParmer Instrument Co., USA). In the scheme in the heating section where the yeast is rapidly mixed with the carrageenan solution, resulting in an emulsion, the temperature is 37 ° C. The gelation of the droplets of kappa-carrageenan in the emulsion is caused by the rapid cooling in ice medium and subsequent solidification in a solution of potassium chloride (22 g / L). A static mixer with a diameter of 12.7 mm and 24 elements is used to obtain a mixture of yeast and carrageenan. A second 42-element mixer with a diameter of 12.7 mm is used to form the emulsion. The pellets used in the experiments were 0.5 mm <(pellet diameter) <2.0 mm.

4.5 Cumulative size distribution of kappa-carrageenan gel granules containing immobilized yeast cells

A sample of the granules is taken every 30 L of the production of gel pellets to calculate the particle size distribution on the basis of a moisture-weighted mass. Each sample has a wet weight of approximately 500 g. A mass screening is used to determine the particle size distribution of the granules. The granules pass through a series of sieves with mesh sizes of 2.0, 1.7, 1.4 1.18, 1.0 and 0.5 mm. A 4.5 L solution of KCI (22g / L) was used to facilitate the sieving of each pellet sample. Kappa-carrageenan gel pellets of regular spherical shape are accepted so that the sieve diameter is taken as the particle diameter. It is also assumed that the particle density is the same and does not depend on the particle size.

4.6 Yeast cell counts and viability.

192

Viability of freely suspended brewer's yeast and cell concentration: The method of the American Brewing Chemists Association, using methylene blue staining, was used to determine the viability of the yeast cells (Technical Committee and Editorial Committee of the ASBC, 1992). The strain is evaluated by determining whether the development of the yeast is viable or not, and this determination is made on the basis of the ability of the living cells to oxidize the colorant to a colorless form. Dead cells do not have the ability to oxidize the dye, including the dye in blue. Buffering a solution of methylene blue was prepared by mixing 500 mL of Solution A (0,1 g methylene blue / 500 mL dH ₂ 0) in 500 mL of Solution B (498.65 mL of 13.6 g CN ₂ P0 _4/500 mL dH ₂ 0 mixed with 1.25 mL of 2.5 g Na ₂ HPO _4. 12H ₂ O / 100 mL dH ₂ O) to give the final buffer solution of methylene blue at pH 4.6. The diluted yeast solution was mixed with the solution of methylene blue in a tube until approximately 100 yeast cells were suspended in a microscopic field. A small drop of the well-mixed suspension was placed on a microscope slide. After one to five minutes of exposure to the strain, the cells stained in blue and the cells that remained colorless are listed. The percentage determination of living cells is based on the total number of cells listed. Cell concentration was determined using an optical microscope and a hemacytometer (Hauser Scientific Company).

Immobilized Cell Viability and Cell Concentration: The gel pellets were separated from the enzyme liquid by passing the mixture through a sterile sieve (500 µm mesh size) and washing with 10 mL of distilled water. 1 mL gel pellets containing trapped brewer's yeast were added to a sterile 50 mL container containing 19 mL of distilled

193 water. The granules are then broken using a Poltron® apparatus (Brinkmann Instruments) to release the cells from the gel. The cell viability and their concentration were then determined as described above for the free-suspended cells.

4.6 Microbiological assays

Liquid Phase Analyzes: At least once a week, continuous fermentation samples are taken for microbiological analysis. The wort used in continuous fermentation is also tested for contamination before being fed to the bioreactor. For testing the presence of aerobic and anaerobic bacteria, the samples were applied to Universal Beer Mustard Agar (UBA, Difco Laboratories) by adding 10 mg / L cycloheximide and incubated at 28 ° C for 10 days. The plates tested for anaerobic bacterial contamination are placed in an anaerobic flask with AnaeroGen® (oxo compound), which absorbs the oxygen remaining in the flask, creating an anaerobic environment. Anaerobic indicator (oxo compound), which turns pink in the presence of pink, in the presence of used to check anaerobic conditions in the flask. Wild yeast contamination was tested by placing samples in yeast medium (arap YM, Difco Laboratories) plus CuS0 ₄ (0.4 g / L) incubated at 25 ° C for 7 days. Peptone, yeast extract, agar medium described above, is used to check for non-light yeast contamination by testing at 37 ° C for 7 days. The absence of yeast development in the peptone-agar medium at 37 ° C indicates that there are no brewer's yeast to produce English beer beer that develops at 37 ° C.

194

Gel Phase Analyzes: An analysis was performed in our laboratory to confirm that the immobilized cells in the granules to be used for fermentation did not contain bacterial contamination before being fed into the bioreactor. The primary concern was to avoid contamination with organisms that spoil the beer, such as Pediococcus sp. and Lactobacillus sp. or wild yeast. Carrageenan 3 mL gel pellets were inoculated into 100 mL of several different liquid media described below, placed in 250 mL flasks at 25 ° C and shaken at 100 rpm in an incubator shaker. NBB (Nachweis von Bierschadlichen Bacterien) fluid medium (BBL cat. # 98139, based on liquid medium οτΝΒΒ containing 0.02 g / L cyclohexamide) is a semi-selective medium used to detect bacteria that spoil beer, such as for example Pediococcus sp. and Lactobacillus sp. Liquid medium containing copper sulfate (16 g / L YM Difco fluid medium; 0.4 g / L CuSO ₄ ) is a semi-selective test medium for wild yeast contamination. Finally, Standard Methods (STA) + cyclohexamide liquid medium (16 g / L Standard Methods, Difco liquid medium; 0.02 g / L cycloheximide) is used to test for bacterial contamination in water, waste water, dairy products and nutrients ( Power and McCuen, 1988). The specific environment is chosen to identify and identify possible organisms that can damage the beer within three days. The contaminated samples shall be measured by determining the turbidity in the sample and a presumed identification of the contamination shall be made.

Methodology for establishing respiratory failure of yeast cells: Method of staining with triphenyltetrazolium chloride (TTX): This method is used to determine the respiratory failure of brewer's yeast from interruption of the population and is based on the principle that TTX is

195 a colorless salt that forms a red precipitate upon reduction. When TTX is deposited on yeast colonies developing on yeast-peptone-dextrose-agar (yeast extract - 10 10 g / L; peptone 20 g / L; dextrose - 20 g / L; agar - 20 g / L in dH ₂ 0), respiratory surface yeast will reduce TTX and these colonies will turn dark pink to red. But respiratory failure yeasts will not reduce the colorant and it will remain its original color.

The culture medium was periodically diluted to a suitable concentration of microorganisms -100 cells / 0.2 mL. The brewer's yeast-peptone-dextrose plate is then incubated for approximately 3 days at 21 ° C until the yeast colonies are visible in aerobic media. Each plate was then coated with 20 mL TTX at 50 ° C. After cooling the specific solutions at 50 ° C, the top layer of TTX agar was applied by mixing Solution A (12.6 g / L NaH ₂ P04; 11.6 g / L Na ₂ HP0 ₄ ; 30.0 g / L arap in dH20, autoclaved at 121 ° C, 15 min) with solution B (2.0 g / L 2,3,5-triphenyltetrazolium chloride in dH20, autoclaved at 121 ° C, 15 min). The plates were studied after 3 hours of incubation at room temperature. The percentage of respiratory failure is represented as the percentage of unstained colonies relative to the total observed amount.

4.8 Scanning electron microscopy of yeast immobilized in carrageenan gel pellets.

Kappa-carrageenan gel pellets containing immobilized brewer's yeast are removed from the bioreactor through a sampling port and placed in a glass vial with a screw cap, immersed in a small amount of fermentation liquid medium. The vial is immediately covered with ice and transported in an insulating container to the scanning optical microscope. Capacaragenan gel pellets containing immobilized brewer's yeast are fixed in 2

196

% (v / v) glutaraldehyde prepared from Sorensen phosphate solution, 0.07 M, pH 6.8 (Hayat, 1972). Thereafter, a further fixation of 1% (v / v) osmium tetroxide prepared in the same buffer is carried out and dehydration by a series of alcoholic solutions with different alcohol content - 50, 70, 80, 90, 95,100% (v / v) 15 minutes in each solution and then 3 shifts in 100%. Prior to the critical drying point (Ladd Research Industries, Burlington, VT), the carbon dioxide granules were frozen in liquid nitrogen, destroyed, and placed in 100% alcohol by carbon dioxide. Frozen destruction ensures the detection of the inner surface with minimal damage. After drying to the critical point, 30 nm gold / palladium-coated spray (Polaron SC500 powder coating machine, Fison Instruments, England) was applied to the samples. and then scanned with the Hitachi S-4500 auto-electron emission of the Hitachi S-4500 electron microscope Nissei Sangyo, Tokyo, Japan.

4.9 Bioreactor Sampling Protocol

The bioreactor sampling tank has an opening fitted with a diaphragm valve (Scandi-Brew Type T). The tank is filled with a solution of ethanol - 70% (v / v) to maintain aseptic conditions around the aperture between sampling. To sample, remove the plug located at the base of the ethanol tank, dry and rinse thoroughly with ethanol before opening the opening. The samples are placed in a vial or bottle with a screw cap, the volume varies from 5 to 60 mL, depending on the requirements of the assay. In order to test microbiological contamination, 10 mL of the fermentation liquid was passed through a membrane filter using a vacuum pump. The diaphragm with apertures of 0.45

197 μίτι shall be placed in a suitably selected environment as described in section 4.6.

For chemical analyzes, a sample of 60mL was drained through the septum of a 100 mL vial and filtered through a syringe-equipped filter FP-050, Schleicher and Schull, a two-layer syringe filtration system. The required sample volume is then poured into a 10 mL vial and covered with a septum of Teflon and an aluminum cap. The required sample volumes are given in Table 4.1.

Table 4.1 Sampling volume requirements for various chemical analyzes

Sample Volume (mL) ethanol 5 short-chain diols 10 volatile islands in beer 12 higher diketones 5 carbohydrates / relative weight / 12 free amine nitrogen / protein 12

4.10 Measurement of dissolved oxygen

Dr. Digox 5 Analyzer Thiedig measures the oxygen dissolved in the fermented must in the range 0.001 - 19.99 mg / L (Anon, 1998). Vilacha and Uhlig, (1985) have tested many dissolved oxygen measuring instruments in beer and have found that the Digox analyzer provides accurate and accurate measurements.

The electrochemical method used in Digox 5 is based on a three-electrode circuit with a potentiometer. The measuring cell consists of a measuring electrode (cathode) and an electrode counter (anode). These electrodes are immersed in the liquid to be measured

198 oxygen concentration. The reaction of the measuring electrode is observed after fixing of a certain measuring potential. On the large silver measuring electrode, the oxygen molecules are reduced to hydroxyl ions. Two water molecules react with one molecule of oxygen and four electrons are absorbed, resulting in four hydroxyl ions (reaction 4.1)

0 ₂ + 2H ₂ 0 + 4e -h · 40NG (4.1)

The stainless steel anode absorbs the four electrons released on the cathode to provide the flow of electrical current.

In Equation 4.2, the measured current I is directly proportional to the oxygen concentration C _L o:

I = KxC (_o (4.2) where the constant K is influenced by the faraday constant, the number of converted electrons per molecule, the surface area of the cathode surface and the width of the boundary layer on the surface of

the measuring electrode.

Constant measurement potential is important for selectivity (with respect to oxygen) and precision of measurement. The measured voltage is stabilized by a reference electrode that is not loaded with current. This, together with a voltage regulator that provides electronic feedback, also provides a constant measuring potential. The surface of the measuring electrode is connected electrolytically to the reference electrode by means of a diaphragm.

The error based on the measured range of the final value of dissolved oxygen is 3% (Anon, 1998).

The dissolved oxygen analyzer is calibrated using active calibration, whereby Digox 5 produces a certain amount of Faraday (0.500 mg / L) oxygen and then performs a double check with the measured values in the matrix.

199

This allows the apparatus to be calibrated under pressure, temperature and flow conditions corresponding to those of the measurement for one minute. Since the exchange of molecules in the sensor is a diffusion process, it is influenced by the temperature obtained as a result of the higher reaction rates and increases the current measured. The Digox 5 is therefore also equipped with a sensor that measures temperature and automatically compensates for fluctuations.

Digox 5 has some other advantages as well. Because Digox 5 does not use an electrolyte, the sensitivity loss is relatively small and only small deposits are observed on the measuring electrode. In addition, sensitivity can be determined at any time by performing active calibration. This is a simple procedure to clean the electrode and recalibrate the instrument. In most cathode membranes, silver chloride is deposited and the electrolyte solution is altered, resulting in a progressive decrease in values. For this reason, it is recommended that the membranes and electrolytes be replaced for a few weeks and then recalibrated, a long and slow operation. The calibration of membrane sensors is usually performed in the laboratory at oxygen saturation levels, which can cause significant variations, especially in beer must and beer matrix at very low oxygen levels. The temperature has a threefold effect on membrane-based oxygen sensors: the membrane permeability will change, the partial pressure of oxygen will change and the solubility of oxygen in the electrolyte will change. Temperature compensation for these three factors in the membrane sensors is difficult.

Measurement of dissolved oxygen in beer must during storage: Flexible Tygon® pipe (inner diameter

200 ¹ / inch) meeting the food quality requirements aseptically connects to the sampling hole located near the upper end of the conical bases of the T-1 and T-2 beer storage tanks (see section 4.2.1). A variable speed peristaltic pump provides a volumetric flow rate of 11 L / hr through the dissolved oxygen analyzer unit ((Masterflex® L / S ™ Digital Standard Drive, Cole-Parmer cat. # P-07523-50)). Measurement of dissolved oxygen in the casting must is then recorded after 4-5 minutes.

Measurement of dissolved oxygen in the bioreactor: Before the measurements of dissolved oxygen in the bioreactor are performed, the Digox 5 analyzer is cleaned. The sensor inlet is connected to a sterile Tygon® tube (inner diameter

% inch) meeting the food requirements. 70% (v / v) ethanol solution was pumped through the analyzer at a volume flow rate of approximately 10L / hr for 15 min. The dissolved oxygen analyzer is connected to a laboratory water tap and hot water (70 ° C) is passed through the sensor for a minimum of two hours. This method is used instead of steam sterilization since the materials from which the analyzer parts are made cannot be subjected to temperatures above 70 ° C. During the next two-hour sanitation period, the tube and the inlet and outlet openings of the apparatus are tightened to maintain sterility in the analyzer. A freshly sterilized laminar flow suction tube is connected to the analyzer inlet and outlet. The free ends of the tube are then aseptically connected to stainless steel openings with an internal diameter% at the bottom of the bioreactor and measurement is made. When the openings of the bioreactor are not used, they open

201

closed hermetically using a short sterilized Tygon® tube.

The dissolved oxygen in the bioreactor is measured online by draining the fermentation fluid through an opening located at the bottom of the bioreactor. The fluid is withdrawn from the bioreactor and passed through a stainless steel filter (see Section 4.1.2) connected to a 1/4 inch tubular element that passes through the bottom of the bioreactor. The fluid is then passed through a Tygon® flexible tube (internal diameter% inch) to meet the food requirements, which is connected to a variable speed peristaltic pump, providing a dissolved oxygen analyzer with a volumetric flow rate of 11 L / hr (Masterflex ® L / S ™ Digital Standard Drive, Cole-Parmer cat. # P-07523-50). The fermentation liquid is then recycled through a second stainless steel tube of% inch internal diameter that passes through the bottom of the bioreactor. A Tygon® tube meeting the food quality requirements (Cole-Parmer, 1999) is used to connect the sensor to the bioreactor due to its low oxygen permeability of 30 cm ³ mm / (s-cm ^2- cmHg) x1O · ¹⁰ . The measurement is made after 4-5 minutes of circulation.

4.11.2 Chemical analyzes

Calibration is carried out using appropriate standard reagents. All reagents used for analysis are> 99% pure. Where necessary, further purification is carried out.

4.11.1 Ethanol

Ethanol concentration was determined using gas chromatography using an internal standardized method of the Technical Committee and Editorial Committee of the American Society of Brewing Chemists (1992). The degassed samples are treated directly with a solution

202

of isopropanol, 5% (v / v) and injected into a Perkin Elmer 8500 gas chromatograph equipped with a flame ionization detector and a Dynatech automatic sampler. A purified kieselguhr column is used for chromatographic purposes, 80-100 mesh, using helium as the gas carrier. Chromatographic analysis conditions: flow rate 20mL / min, injector temperature 175 ° C, detector temperature 250 ° C and column temperature 185 ° C.

4.11.2 Brief description of carbohydrates

The glucose, fructose, maltose, maltotriose, maltotetrose, poly-1 (polysaccharide peak) and glycerol concentrations were determined using a spectro-physical (SP8100XR) liquid chromatograph equipped with a cation exchange column (Bio-Rad Aminex, HPX-87K) and a refractive index detector (Spectra-Physics, SP6040XR). The portable phase is 0.01 M potassium phosphate and the system is equipped with an automatic spectro-physical sampler (SP8110). The device operates at a counter pressure of 800 psi. The flow rate of the sample and the elution agent through the column was 0.6 mL / min, the temperature in the column was 85 ° C and the temperature in the detector was 40 ° C. The injected volume was μΙ_.

4.11.3 Relative weight

The relative weight of the must and fermentation medium in this study is described with respect to the actual extract (Plateau degree, ° P), which is the accepted unit used in the brewing industry. The fermentation samples are filtered as described in section 4.8, and analyzed in a density meter (Anton Paar DMA-58) to determine the relative weight of the must (Plateau degree). The fermentation samples are placed in a glass U-tube and subjected to electronic oscillation to determine the relative weight, thereby determining

203 plateau indirectly. The plateau degree is the numerical value, expressed as a percentage (w / v) of a sucrose solution in water at 20 ° C, whose relative weight is the same as that of the beer must in question. Since the scale of the Plateau scale and the resulting tables relating to the relative weight of the solution in which the substances are dissolved are based on an aqueous solution of sucrose, it is approximately the same as the extract. The extract is determined by the total amount of soluble mass in the beer material, "as is" and / or potentially during processing (Hardwick, 1995), such as carbohydrates, proteins, tannins. Currently in

the brewing industry the relative weight of the extract is expressed as a Plateau degree, since there is no more appropriate and better measure relating to changes in the composition of beer musts of different origins.

4.11.4 Total diacetyl content

Total diacetyl (2,3-butanedione) in beer and fermentation samples is measured using a capillary separation (Hewlett-Packard 5890) and detection of the presence of an element by electron capture using the method of the Technical Committee and Editorial Committee of the American Society of Brewing Chemists (1992). The method is used to determine the total amount of diacetyl, as it measures the amount of diacetyl and its precursor alpha-acetolactate. The gas carrier is made up of 5% methane in argon at 1.0 mL / min and a J&W DB-Wax column is used. The separation ratio is 2: 1 and the additional gas is helium at 60 mL / min. The injection temperature is 105 ° C and the temperature of the detector is 120 ° C.

The system is equipped with the Hewlett Packard 7694E automatic sampler and uses 2,3hexandione as the internal standard. The cycle duration is 40 min with time for

204 vial equilibration 30 min at 65 ° C, compression time - 2 min at 4.8 psig, contour fill time 0.2 min, equilibrium contour time - 0.1 min and duration of injection - 0.27 min. The pressure of the gas carrier is 18.8 psig, the temperature in the fluid conduit is 95 ° C, and the contour temperature is 65 ° C.

4.11.5 Volatile substances in beer

The volatile substances in beer are acetaldehyde, ethyl acetate, isobutanol, 1-propanol, isoamyl acetate, isoamyl alcohol, ethylhexanoate and ethyl octanoate and are measured using an internal standard using a gas chromatograph (Hewlett Packard 5890) and a flame detector. The gas carrier is helium at a flow rate of 6.0 mL / min and the gas chromatograph is equipped with a Hewlett Packard 7694 automatic sampler.

The injection temperature in the gas chromatograph is 200 ° C and the temperature in the detector is 220 ° C. Temperature profile of the heating chamber: 40 ° C (5 min), 40-200 ° C (10 ° C / min), 200-220 ° C (50 ° C / min), 220 ° C (5min). The gases for flame ionization detector j include: carrier with a flow rate of 6.0 mL / min, helium additive - 30 mL / min and ί 28 psig, H ₂ - 50 mL / min and 25 psig and air - 300 mL / min and 35 psig.

u The septum is purged with a flow rate of 0.8 mL / min. The discharge pressure is 4.0 psig. When the automatic sampler is connected via an indicator located in the injection port, the pressure in the vial is 15.9 psig, the pressure of the gas carrier is 7.1 psig, the discharge pressure in the column is 4 psig, the flow rate of the branch flow is 18 mL / min and the flow rate in the column was 6 mL / min. Temperature zone: vial at 70 ° C, contour at 80 ° C, portable tube - at 150 ° C.

The duration of the gas chromatography cycle is 40 min, with the equilibration time in the vial being 35 min; pressure build-up time - 0.25 min, contour filling time - 0.1 min, equilibrium contouring time - 0.1 min, injection duration - 3 min, and contour sample volume -1 mL.

4.11.6 Free amine nitrogen (CAA)

Determination of the concentration of free amine nitrogen in the fermentation sample was performed according to the method of the Technical Committee and Editorial Committee of the American Society of Brewing Chemists (1992), using a Perkin Elmer LS50B spectrophotometer. This spectrophotometric method shows a color reaction between ninhydrin and the nitrogen present in the sample. The amount absorbed depends directly on the presence of free amine nitrogen.

a) Color reagent: 19.83 g disodium hydrogen phosphate (Na ₂ HP0 ₄ )

30.00 g potassium dihydrogen phosphate (KN ₂ PO ₄ ) 2.78 g ninhydrin monohydrate

1.50 g fructose

b) Dilution reagent: 2,00 g potassium iodate (KYU ₃ )

596 mL of distilled deionized water

404 mL 95% (v / v) ethanol

They are refrigerated and used at room temperature.

c) Glycine stock solution: 0.1072 g / 100 mL distilled deionized water

d) Glycine standard solution: dilute the glycine stock solution 1: 100 (v / v) with distilled deionized water. This standard solution contains 2 mg / L of free amine nitrogen.

Samples were diluted at 100: 1 with distilled water and 2 mL of the diluted sample was introduced into three tubes. The control sample was prepared by placing 2 mL of distilled deionidized water in each of the three tubes. They are also preparing

206

three tubes, each containing 2 mL of standard glycine solution.

To each sample was added 1 mL of color reagent and then placed in a water bath at 100 ° C for exactly 16 min. The tubes were then cooled in a water bath at 20 ° C for 20 min. Five mL of the dilution reagent was added to each tube and mixed thoroughly. The samples were then allowed to stand for 10-15 min. Measure the absorbance at 570 nm using a spectrophotometer and calculate the free amine nitrogen by the formula 4.3:

CAA (mg / L) = (A $ - Av - Af) 2d / As (4.3), where CAA is the amount of free amine nitrogen in the sample in mg / L, Aβ is the average content of absorbed substance from the test solution, Av - the mean of the absorbed substance in the control; Af is the average value of the absorbents corrected for turbidity and beer opacity; 2 is the amount of CAA in standard glycine solution; d - sample dilution factor; As is the average value of the substance absorbed for the standard glycine solution.

CHAPTER 5. CONTINUOUS FERMENTATION IN THE SYSTEM OF

BRIEFACTOR WITH GAS POWER

For continuous fermentation of beer, a bioreactor equipped with a suction tube was selected, which utilizes a gas lift to effect continuous circulation of the medium (a GPST type bioreactor). This type of bioreactor was chosen because of its excellent mass transfer (liquid-solid phase) and high mixing process performance. Liquid-solid mass transfer is particularly important as it involves the transfer of the fluid from the liquid phase to the solid, which is a biocatalyst from immobilized cells, providing substrates for encapsulated brewer's yeast. These

207

bioreactors also provide good aeration, low energy consumption and are of simplified construction. This makes gas-lift bioreactor systems very attractive for large-scale operation such as those used on an industrial scale for wastewater treatment (Driessen et al., 1997; Heijnen, 1993).

5.1 Description of a bioreactor fitted with a suction tube using gas lift

This section describes in detail the gas lift bioreactor used in this research.

5.1.1 Bioreactor housing

A 13 L (8 L displacement) bioreactor equipped with a suction tube using a gas lift (GPST type bioreactor) designed for the purpose of this research works with a three phase fluidized bed (liquid (solid / gas), in which the immobilized cells are maintained in suspended state by carbon dioxide providing internal circulation of the liquid (Heijnen, 1996), as shown in FIG. 16. A photo of the bioreactor is shown in FIG. 17, and a detailed drawing is given in FIG. 18. Carbon dioxide and air are fed from the cone-bottom of the bioreactor through a sintered stainless steel spray device (CO ₂ Part # 9222 spray nozzle, Hagedorn & Gannon, USA) with a length of 0.11 th outer diameter of 0.013 t. Carbon dioxide is used as the fluidizing gas and air is used to deliver oxygen to the yeast cells.

In the columnar bioreactor is a concentric suction tube that functions as an upstream tube (lift) in this fluidized system, while the outer ring acts as a downstream tube. The inner suction pipe is attached to a cylindrical particulate separator and is mounted with

208 using 3 stainless steel suspension fixtures in the upper expansion section of the bioreactor. The placement of the suction tube and the particulate separator inside the bioreactor minimizes the risk of microbial contamination from the external environment.

Typically, the bioreactor is provided with a grid with openings to separate the immobilized cells from the liquid at the outlet of the bioreactor. But the grate tends to become clogged, which is why a stainless steel cylinder is used to separate the granules from immobilized cells from the liquid phase as they move above the upper end of the suction tube and move down into the ring. The particles could rather collide with the cylinder and fall back into the liquid phase volume than leave the bioreactor through the outlet. Therefore, there is a small area near the outlet of the bioreactor that lacks particles of immobilized cells. The upper expanded section of the bioreactor also enlarges the surface area for gas bubbles.

5.1.2 Front cover of the bioreactor

In FIG. 19 is a schematic view of the front cover of the bioreactor. The openings in the front cover of the bioreactor are designed to minimize the risk of contamination. The holes are either welded directly to the front cover or sealed fittings (Swagelok®) are used. An inoculum orifice, a temperature sensor sleeve, a thermometer, a gas sample septum and openings for the drainage of liquid samples to measure dissolved oxygen and return samples are installed in the front cover. A temperature sensor is mounted in the sleeve, which is inverse to a control system for measuring the temperature. This control system gives a feedback signal to a solenoid valve that opens and closes the glycol feed to the thermal jacket of the bioreactor. The temperature is up

209 monitors continuously using a thermometer (a watertight thermometer equipped with a Cole-Parmer thermocouple # 90610-20) and a T-type probe welded to the front cover of the bioreactor. Dissolved oxygen was measured using an analyzer (Dr. Theidig, Digox 5) requiring a flow rate of 9-11 L / hr liquid flow through the analyzer to accurately measure oxygen content. The fluid is discharged from the bioreactor through an inner diameter tube that passes through the housing of the front cover and reaches the fluid. As shown in FIG. 20, a filter is installed at the end of the tube to remove large particles from the liquid which is pumped into the dissolved oxygen analyzer. The liquid is then returned to the bioreactor through another tube with an internal diameter located in the front cover.

5.1.3 Sanitary valves for aseptic sampling

The bioreactor is equipped with a diaphragm sampling valve (Scandi-Brew®) welded to the wall of the bioreactor. The valve is designed for sampling under aseptic conditions. The membrane closes directly to the side of the fermentation fluid, which

allows the valve to be subjected to full steam and alcohol sterilization through two openings (FIG. 18). A small outer ethanol tank encloses the membrane to maintain sterility between sampling. This valve is used for all sampling and it is assumed that the composition of the fluid at the sampling site is not significantly different from that of the fluid flowing through the outlet of the bioreactor. As noted in the Materials and Methods chapter of this research, samples of the bioreactor are taken through a valve located on the outer wall of the bioreactor. To confirm the assumption that the composition of the liquid flowing through the outlet of the bioreactor is the same in

210 comparison with that of the fluid in the samples taken from inside the bioreactor, a mixing duration study was performed.

A pulse indicator was used to determine the mixing time in a gas lift bioreactor (Chistie, 1989). 1 mL of 10 N HCl was rapidly injected into the bioreactor ring and the change in pH was recorded. By injecting 10 N NaOH. The pH returns to its original value. The pH electrode (ColeParmer, cat. # P-05990-90) is 277 mm long and 3.5 mm in diameter. A pH Ingold sensor, model 2300, is used to monitor the pH values. Two-point calibration is performed with specific standard buffers - Beckman pH 7.0 green buffer, Part # 566002, and pH 4.0 red buffer. Beckman, Part # 566000. Data are logged at 3750 Hz in 300 seconds using a software program from Cheryl Hudson and John Beltrano, established in 1999 (University of Western Ontairo, London, Ontario).

The pH values were then processed using the Savitsky-Golay algorithm in the 2D Curve Table (Jandel Scientific Software, Labtronics, Guelph, Ontario). The Savitsky-Golay algorithm is a method of time domain equalization based on polynomial least squares alignment through a moving window in the pH meter (Anon, 1996). The leveled data is then normalized and a ΔρΗ plot is plotted against time. Mixing time is taken to the nearest minute when the pH reaches ~ 95% of equilibrium values. Mixing times are measured using three different carbon dioxide volume flows: 283 cm ³ / min, 472 cm ³ / min (volumetric flow rate is used throughout this research) and 661 cm ³ / min. In all three cases, the pH values in the bioreactor reach equilibrium (~ 95%) in less than 2 minutes,

211

Ί as shown in Appendix 1. The mixing time is assumed to be too short and proves our initial assumption that the system in the bioreactor is well mixed. This allows us to assume that the composition of the fluid samples taken from the bioreactor do not differ significantly from the composition of the fluid flowing from the outlet at an average residence time of 24 hours in the bioreactor. The accompanying figures show that mixing by dispersion is also compounded by a particular fluid circulation, which is typical of bioreactors using gas lift (Chisti, 1989).

5.2 Flow chart of the continuous fermentation system

A flow chart of the continuous fermentation system, located in the experimental small brewery of Labatt Brewing Company Limited, London, Ontario, is given in FIG. 21. Table 5.1 describes the individual parts of the system. In short, the pouring must is obtained from the Labatt installation in London, sterilized using a pasteurizer using instant evaporation (Fisher sheet heat exchanger, combined flow, Eurocal 5FH type), and stored in large collection tanks tanks (T-1 and T-2). During continuous fermentation, the must must be transferred at regulated flow to a bioreactor (BR-1) containing immobilized yeast cells. The fermented liquid leaves the bioreactor as an overflow and is collected in a T-3 collection vessel. The following sections describe in detail the operation of the continuous fermentation system given in FIG. 21.

5.2.1 Obtaining and storing must must

Non-oxygenated beer must used for continuous fermentation is obtained by pipeline from

212 installation of the Labatt in London and fed to a cylindrical conical storage tank of 1600 L capacity, in which carbon dioxide is pre-dispersed to minimize the increase in oxygen in the must. Prior to use, all vessels in this scheme, including T1 and T-2 collection tanks, are cleaned and sanitized in accordance with Labatt's established rules. The cast must then be pasteurized using water vapor and fed at 2 ° C to the T-1 or T-2 collection tank (also pre-treated with carbon dioxide spray). The wort is kept in these tanks for up to 2 weeks at 2 ° C, supplying liquid to the BR-1 bioreactor for continuous fermentation. At the end of the two-week period, the bioreactor power is changed so that the wort is fed from the second tank containing fresh wort. Two identical T-1 and T-2 beer must tanks are used to minimize downtime while switching to fresh beer must. In all cases, the wort must be tested for contamination for at least 2 days before being fed to the bioreactor (BR-1). If the beer must is contaminated, it is removed and fresh beer must be fed immediately and pasteurized.

Minimizing dissolved oxygen in beer must during storage: The purpose is to store the beer must with a minimum constant oxygen content and at a low temperature without cooling. This requirement is related to the prevention of side effects involving oxygen that break down the must (Narzip et al., 1993) to ensure consistent supply of the must to the bioreactor and minimize the risk of microbial contamination of the must. during storage. Large cylindrical conical vessels (T-1 and T-2) with volume

213

The 1600 L (net) used for storing must must for subsequent continuous fermentations was originally designed for periodic fermentations rather than as reservoirs for storing must. Therefore, the cooling of these vessels is not suitable for maintaining a temperature of 2 ° C. After three days of storage of the must, the temperature difference from one section to another varied by 15 ° C (Table 5.2).

These warm areas in the tank increase the risk of developing germs. Mixing in the tanks was therefore necessary to ensure a uniformly low temperature in all sections.

For this reason, a tubular sprayer was installed in the conical base of each of the T-1 and T-2 beer must tanks. Experiments were conducted to determine the best regime for filling the beer must tank and maintaining a constant low dissolved oxygen level. In the first experiment, the tank is filled with beer must, which is fed without oxygen enrichment and pasteurization. After the storage tank is filled with 1600 L of wort, 0,113 m ³ / hr of carbon dioxide is sprayed from its base. In the second experiment, the must was also fed without oxygen enrichment and pasteurization. Before filling in the tank is sprayed carbon dioxide (0,85 m ³ / hr) for 3 hours, and small amounts of CO ₂ (0,113 m ³ / hr) was sprayed into the tank during filling thereof. This low-flow carbon dioxide stream is bubbled continuously into the pouring must stored in the tank until it is fed to the continuous fermentation. I / Ι in both experiments during the weekly storage of oxygen concentration in the casting must is constantly observed.

214

Figure 5.7 shows the dissolved oxygen concentration as a function of storage time. When the storage tank is not pre-purged with carbon dioxide, the air in the vessel volume increases the oxygen in the must. Without preliminary spraying, a significantly long period of time is required to reach a minimum and constant level of dissolved oxygen in the must. When the tank has been purged, the concentration of dissolved oxygen in the must must remain at a constant minimum throughout the storage period. Therefore, continuous fermentation involves the pre-dispersion of carbon dioxide in the tanks for the storage of must (T-1 and T-2) and purging with a flow of low-carbon carbon dioxide during the storage of the must. maintain a minimum positive pressure in the tank.

The temperature profile of the storage vessels is also compared when spraying with 0,113 m ³ / hr carbon dioxide and without spraying. In addition to beer must, this is also done in water, using a thermometer-related probe for this purpose (a watertight thermometer equipped with a Cole-Parmer thermocouple # 90610-20). Water from the city water supply network (1600 L) is fed into the beer must storage tank, allowed to equilibrate for three days, and the water temperature is recorded in different sections of the tank. The water from the tank was then sprayed for 24 hours with 0.113 m ³ / hr of carbon dioxide and the temperature was read again. The ambient temperature was recorded in all cases and the temperature at a set point in the tank volume was 2.0 ° C.

As shown in Table 5.2, when carbon dioxide is sprayed, the temperature in the wort storage tank is

215

more even - in the range between 0.1 and 4.1 ° C in the measured area, and the contents in the tank are not cooled. This lower temperature helps prevent the development of germs in beer must during storage.

The gas extracted from the wort passes through a sterile gas filter located at the top of the tank. The wort is then transferred by a variable speed peristaltic pump (P-1) (Masterflex® L / S ™ Digital Standard Drive, Cole-Parmer cat. # P-07523-50) to the inlet of the bioreactor (BR- 1) with a volume of 8 L, using a flexible Norprene® L / S 16 pipeline that meets the food requirements.

5.2.2 Continuous fermentation in a suction tube bioreactor system and use of gas lift

The wort is fed into the bioreactor through a 1/4 opening located in the conical bottom of the BR-1 bioreactor. A mixture of filter sterilized air (Millipore, Millex®-FG ₅ o, 0.2 µm) and carbon dioxide (99.99% purity) was fed into the bioreactor via a sintered stainless steel sprayer. A rotometer (R-3) is used to control the flow of carbon dioxide at normal temperature and pressure and a regulator to recalibrate the mass flow (M-1) to regulate the air flow at normal temperature and pressure, the fermented liquid is removed from the The bioreactor, by pouring through the outlet and through a reinforced PVC pipe with an internal diameter of 1 ”is fed into a collecting vessel (T3), which is cooled by an external cooling coil through which glycol passes, thus maintaining a temperature of 4 ° C.

5.2.3 Collection of the product

The collection vessel (T-3) is provided with a long inlet (inner diameter 1 ”) that is designed so that the fermented liquid flows down the walls of the vessel to

216

minimize foam formation. This vessel is provided with a sterile gas filter (Millipore, Millex®-FG50, 0.2 µm) to separate gas from the bioreactor (BR-1). The collection vessel is periodically emptied using a 1/4 valve (V-12) located above the base of the vessel.

5.2.4 Glycol cooling circuit

The glycolate was transferred from the brewery in London to the experimental installation at -23 ° C and a pressure of 45 psig. It circulates through the cooling jackets of the beer must storage tanks (T-1 and T-2), the gas lift bioreactor (BR-1) and the product collecting vessel (T-3). Both the storage tanks and the bioreactor are equipped with temperature sensors that provide feedback to the thermostats, which in turn regulate the flow from the cooling glycol to the cooling jackets. The reservoirs store the must at 2 ° C, while the temperature in the bioreactor is adjusted from 12 to 22 ° C depending on the specifics of the experiment. The collection vessel does not have automatic temperature control, but the glycol flow is manually controlled to maintain a container temperature of approximately 4 ° C No precise temperature control is required in the collection vessel (T-3) since the fluid in this vessel is drained and not subjected to analysis or further procedures.

Glycol is also used to isolate and cool the pipes for transferring the must from the must tanks (T-1 and T-2) to the bioreactor (BR-1). After the glycol circulates through a shirt, it returns to the main tube located in the experimental small brewery and then returns to the London installation, usually at 15 ° C and 40 psig.

217

5.2.5 Sterilization of the bioreactor

The bioreactor (BR-1) is filled with 2% (v / v) Diversol® CX / A (Diversey Lever, Canada) solution, which is a detergent and gas is sprayed overnight. The liquid is then drained from the reactor and washed with cold water. This cycle of treatment with a purifier and a wash with water is repeated twice. In order to prepare the bioreactor for steam sterilization, the wort and gas supply pipes are switched off. The steam supply pipe is connected to the inlet of the bioreactor and the following valves are opened, respectively: the valves of the gas supply pipe for the spray gas (V-7, V-6), the valve (V-17) for the gas supply , product outlet pipe valves (V-9.V-11), diaphragm sampling valves (V-8, V-10) and product drain valve (V12). The steam vent of the installation is then slowly opened and the bioreactor valves adjusted so that a small stream of steam can be observed at the outlet of each external opening. After steam treatment for 60 minutes, all external valves of the bioreactor are closed (V-17, V-8, V10, V-12) with the exception of the valve of the diversion pipe for must (V-6). When the steam supply valve is closed, the divert valve of the wort (V-6) is closed and the sterile filter is connected to the collecting tank to prevent contamination by non-sterile air entering the system when cooled. . When the pipe of the steam installation is closed to establish excess atmospheric pressure while the system is cooling, the gas supply pipe to the bioreactor is also shut off via V-17.

5.4 Starting the fermentation system

The wort is fed from the installation to a pressure vessel made of stainless steel of 20 L capacity and heated in an autoclave

218 for 45 minutes at 100 ° C. The immobilized cells were aseptically transferred into the cooled beer must (40% (v / v)). The hermetically sealed container is transported to an experimental installation of the small brewery where the bioreactor system is located. The 20 L vessel is connected to a high-speed fixture (Cornelius Anoka, MN, USA), which is connected to a reinforced PVC pipeline 3/8 (Cole-Parmer, USA). The other end of the PVC pipeline is connected to the diaphragm valve (V-8) to sample the wall of the bioreactor. A 20 psig filter-sterilized carbon dioxide was fed to the 20 L tank and the membrane sampling port was opened so that the mixture of immobilized cells was transferred from the vessel into the bioreactor without inoculum being exposed to ambient air. The internal components of the 20 L fast-acting vessel device are removed to prevent clogging with immobilized cells during transfer to the bioreactor. The cumulative particle size distribution (smaller than the predetermined size) for kappa-carrageenan gel granules is shown in Figure 5.8.

The arithmetic mean particle size D _pn was calculated to be 1.252 mm and the mean Sauter Dp _{Sni diameter} was 1.17 mm. The average particle diameter is 1.255 mm. The experimental data and the average particle diameter calculations are given in Appendix 1.

After inoculation with immobilized cells, the bioreactor is in batch mode until sugar and diacetyl concentrations reach the setpoints - less than 3 ° Plateau relative to weight and less than 1000 µg / L diacetyl. The system is then prepared for continuous operation. To be flushed with hot water and sterilized with water vapor, the brew pipe, the valves, respectively the valves V-2 (or V-4 for T-2), V-5, and V6 are opened, and V -1 (or V-3 for T-2) and V-7 are closed by isolating

219 wort feed pipe. The wort feed pipe is flushed with hot water at approximately 80 ° C, which is supplied by V-2 (or V-4 for T-2). After the hot-water wash cycle, the water vapor pipe of the installation is connected to the same place and steam sterilization is performed with the wort to supply the wort for a minimum of 30 minutes. At the same time, when the steam pipe is off, the bypass valve (V-6) is also closed. After the system has cooled, the V-2 (or V-4 for T-2) is closed and the steam supply pipe is switched off. The valve V1 (or V-3 for T-2) of the beer must reservoir is opened and the supply of beer must begins via the pump (P-1). The wort is sent to the outlet through the bypass valve (V-6) until the condensate in the pipe is replaced with fresh cold wort. The bypass valve is then closed and the valve (V-6) of the inlet opening of the bioreactor is opened, giving rise to continuous fermentation.

Every two weeks, the tank from which the beer is fed

I must alternate with one of the beer must storage tanks (T-1 and T-2). After supplying T-1 beer must for two weeks, the feed pump P-1 is stopped and the valve (V-5) at the inlet of the bioreactor is closed. The wort transfer tube is then connected to the second wort storage tank (T-2) and the tube is flushed and sterilized as described in the previous paragraph. The continuous fermentation is then continued only after a short interruption of less than 1 hour.

220

ο. φ Σ co <0 0_

CO., LTD co V co CO., LTD I do X CL \ o s Φ Σ X X X ro CO CL CO CL co CL co CL X CO., LTD \ o \ o \ D h— Σ Σ σ s o Φ Φ Φ Φ co Σ Σ Σ Σ \ o W LU LU LU LU d e Mr Rücker e Θ Θ 1— 1— D “c Ι- 1— h- C6 c ΟΣ cr C CD co CO., LTD co x ' CO., LTD ¢ 0 ¢ 6 X Q. X & X CL CL X CL X Φ H Φ Φ Γ ^- Φ 1— Φ 1— that E Φ E Φ E Φ E Φ E Φ e x Σ x Σ X Σ X σ X Σ © Φ f tz Oh E Oh E a tz o E Oh h * CO., LTD o CL Oh CL o o CL Oh CL □ δ ο <5 CO o Φ Q. e X x CL tz X. X CL e * X CL E X. X CL E X X X X X X X Q. x Σ X Σ X S X S X s CD co co co E o L. o e o CM E o CM E o CM 5 CM E O CM Art CT CT CT Q. z e o ' tz o tz o ' E o E o '

cl

o E 1— Φ X § E 1— in. 1— L- —1 X co CO., LTD CO., LTD CO., LTD ¢ 0 C0 § _J o ο- J3 M- £ 1 CM n JO CM JO CM o co ν V V V V V

φ ο CO

X ο

H 4 h- 4 o x CJ t l · - fU σ Σ CO m C0 X CO., LTD co X X X o E λ co that that X Gj X X X Φ (X) L · E E E Σ 1— o ¢ 0 CO., LTD CO., LTD a E cc K X X X CD X Φ Φ Φ o 3 X X X X X X Oh S 5 Φ Φ Φ (Q r · o X X X X μ- X co co co 2Γ co CL 2 o CL X δ CL X JI O S ιέ; 2 5 δ X co X CO co Co 0 co <0 co o X CL Φ Σ co 2 CL CL CL ¢ 0 co CO., LTD CO., LTD CO., LTD E J. X co s co o co About that o co co E Q. CL CL CL £ φ co Φ CT Φ « s E E Φ CL Φ CL Φ CL o [= X 0 X Θ

CM (J- 1 C0 CM axis ¢ - X | J- co CO., LTD CL C0 CO., LTD X o X X CL Φ Q. CL Oh R o o tn co m Ι- K Ι- H Ο s Ο o K E <x X X § X X X X ci o a a 0 X o o X co X X co X co co CO., LTD co co CO., LTD X X X X co co co co CO., LTD ¢ 0 co CO., LTD L_ l_ 1— l_ co co ¢ 0 CO., LTD co co co co CL λ CL CL J3 CL F E F E F E F E X X X X Θ θ θ e

T-3 gas outlet filter <2 bar polypropylene, PTFE membrane,

0.2 micrometer

CD

221

M-1 Mass Air Flow Controller to BR-1 <500 seem 316 HCt, polyamide, Viton®, enclosed in a circle

R-1 Carbon dioxide rotometer to T-1 <10scfh 316 HCt, acrylic block

* Fr. ο CE p CD \ o X I f f p is: X X Q. Fr. l l ι- ι- k— 1— Ο Ο Ο o I X X X co co co co t- " V " T— CO., LTD co co co

LLL

Q. Q. her ff

SS ate

I- ΙΟо

0_θ-

α. Q.

О О Sat Sat

Ф Ф 0_ CL

CM CO I I ss axis

o o o o E e E E l_ 1— in L_ Φ Φ Φ Φ E E EC IE E 6 ® ® ®_ o o Ο Ο Φ c c c c m W CO., LTD co cz p p p o o o o o o > > X X X X > > o o o o o o o o o * X * X q e; e; s s s s o L-> in in. e; c; c; Φ Φ Φ Φ X X X X X CE e e E o o in o CQ b H k- I— I- Oh o Oh Oh o Oh o o Oh Oh X X X X X X X X X X co co co co co co co co co co T— y— T " T " r— co co co co co co co co co co

l £ f X E (X c; L I l X

CL

ф Q.

I axis m

L L

5

t— CMCO

OS OS OSt

O_ 0-0_ >>

co hJ co> ± co <c>>:>:>

V-10 Valve (diaphragm) for outlet BR-1 beer must opening 12 mm 316 HCt, silicone bed

V-11 Valve (throttle), spare, outlet for BR-1 1 ”316 NSt beer, viton bed

222

V-12 Valve (ball) for pouring liquid from T-3 1/4 316 NSt, silicone bed

o o o o o o q q q q q q l_ l_ in. L- L_ F. F. Φ Φ Φ Φ q q q q q q Oh Oh Oh o o o CO., LTD CO., LTD m co m m o o o o Oh o x X X X X I o o o o o o a kt * X x x x x x q q q q q q s s s x s x o o o o o υ Oh o Oh o o o x r X X X X co co co CO., LTD co co X— T— χ— co co co co co co

and p

X CO., LTD

Ф q

CO 2

<M μ- 1 s cc co 2 s -ι- Et J. ό: x o X q co * s 2 o o x s q o X £ 1 " cc E q q m CO., LTD X φ co £ 2 ra co q o <0 co £ a. Q. F. co £ o co E CO., LTD {= Ι- fi co Ο co \ O m κ CO., LTD s. co x_. CO., LTD. - μ- co. X CO., LTD cO co Q H I 1- Oh co co X m \ o 3 'S C0 CL co CL X H 03 1- ¢ 0 o ¢ 0 ΐ I • s X Φ Φ c; X q X φ F. o Φ o 2 co 2 o T CL zr Q. co ¢ 0 h5 o. co 13, X q q q co X X X c μ- H 1— CO., LTD ι I I q F. Φ Φ 2 CD CD CD

co 't U) to S oo

223

Symbols used in FIG. 21

Flexible clutch and nipple

Gas filter

Pressure regulator

Pump

X Valve

224

Cumulative percentage of the sieve product

100X% w / w

Particle diameter D mm

Figure 5.8 Cumulative size distribution of kappa-carrageenan gel pellets containing immobilized yeast cells (n = 5)

225

Table 5.2 Water temperature profile of the beer must storage tank (T-1 or T-2) after equilibration for 3 days without carbon dioxide spraying and after hourly carbon dioxide spraying with a flow rate of 0,113 m ³ / h

Location of measurement in cylindrical conical vessel Temperature (° C) Without spreading of CO ₂ With diff. of CO ₂ 10 cm below the surface of the liquid and 10 cm from the wall of the vessel 20,6 0.4 10 cm below the surface of the liquid and in the center of the pan 20,1 0.1 At the bottom of the cylindrical part and 10 cm from the wall of the vessel 3.8 3.7 At the bottom of the cylindrical part and in the center of the vessel 3.7 4,1 Experimental installation environment 21.4 19.8

226

Dissolved oxygen concentration (mg / L)

The duration of the holding of the must (d)

Figure 5.7 Concentration of dissolved oxygen in the wort depending on the retention time in the wort storage tanks (T-1 and T-2) and under different filling conditions of the tanks.

227

¹ CHAPTER 6. IMMOBILIZATION OF BEER Yeast IN CAPA-CARAGENAN GEL TO OBTAIN LIGHT BEER Beer Researchers have studied various matrices for the physical capture of healthy cells, including calcium alginate, 1992; 1987; Masschelein and RamosJeunehomme, 1985; Nedovic et al., 1996; Shindo et al., 1994; White and Portno, 1978), arapoca (Hooijmans et al., 1990; Lundberg and Kuchel, 1997) and carrageenan bodies (Norton et al., 1995; Wang et al., 1982). Carrageenan is a food grade material and is preferred for cell encapsulation because of its very good strength compared to other gels (Biiyukgiingor, 1992).

The first part of this chapter describes the observation of three cycles of repetitive fermentation with yeast cells immobilized in gel capsules of kappa-carrageenan. The viability of the immobilized cells and the release of cells in the liquid phase were investigated. The fermentation process parameters, including ethanol, maltose, maptotriose, fructose and glucose, were monitored throughout the periodic fermentation process and then compared to control fermentation parameters using only free-yeast cells under the same conditions under the same conditions. relation to the nutrient medium.

There is little published information on the physical effects on cells after a long period of immobilization (Virkajarvi and Kronlof, 1998) and prolonged exposure to external stresses and fermentation products. The second part of this chapter examines the viability, distribution of the cell population and ^{1 The} presentation of Chapter 6.0 is published (Pilkington at al., 1999).

228 the appearance of yeast cells immobilized in carrageenan gel pellets for a longer period of continuous fermentation in a gas lift bioreactor. The relative percentages of yeast with respiratory failure in the population of immobilized and freely suspended cells in the bioreactor have also been examined over a long period of time.

The carrageenan is prepared from repeating 3-6 anhydrogalactose units and selected carrageenans, different in number and position of the sulfate ester groups in the recurrent galactose units. In FIG. 22 shows schematically the mechanism of carrageenan gelation. When carrageenan is in the asyl state, its polysaccharide chains are helices of arbitrary configuration. When enough spirals are formed to provide cross-links for a continuous chain, gelling is observed. The more spirals form or when the spirals form aggregates, the gel becomes healthier and more resistant (Rees, 1972).

The three common types of carrageenan are lambda, jota and kappa.

As illustrated in Figure 6.2, they differ in the content of sulfate esters and the amount of sulfate esters influences the solubility of the polysaccharide chain. Lambda-carrageenan is high in sulfate content and does not tend to form gel (Marrs, 1998). Iota-carrageenan forms a highly elastic dilute gel in the presence of calcium ions and shows no significant syneresis. Syneresis is observed when the tendency of the gel to form spirals or aggregates is so strong that the chain tightens, resulting in drops of liquid coming out of the surface (Rees, 1972). Capacarragenan has a moderate content of sulfate groups due to

229 which forms a stronger and more stable gel in the presence of potassium ions and undergoes a certain syneresis. The increase in gel strength provided by kappa-carrageenan makes it suitable for immobilizing healthy yeast cells.

FIG. 6.2. Chemical structure of lambda, jota and kappa carrageenan

An important characteristic of carrageenan is its reversible thermogelling property. As the solution of carrageenan cools, the viscosity increases and gelling is observed. As the solution is heated, the viscosity decreases and the carrageenan returns to its ash state. By adjusting the composition of the gelling cations in the solution, the temperature at which the carrageenan is converted from sol to gel can be changed. The gelation temperature of kappa-carrageenan increases as the concentration of potassium chloride in the solution increases. This dependence is used to design the process of cell immobilization, since strict requirements regarding temperature fluctuations can be avoided (Neufeld et al., 1996). The gelling temperature of the carrageenan may be adjusted in such a way that it is sufficiently high for the carrageenan to gel in fermentation conditions,

230 and low enough that the yeast cells can be mixed with carrageenan when in ash state without having adverse effects on cell viability prior to gelation of the granules.

However, there are many factors that necessitate further study of the effects of immobilization in gel matrices on metabolism and physiology of yeast cells. The immobilized cells are not exposed to the same microenvironment as the free cells in the liquid phase, as there are additional barriers from the gel matrix and other trapped yeast cells that must be overcome before ©

substrates can be moved to their surfaces (FIG. 23). There are many studies on the rates of mass transfer in gel matrices (Estape et al., 1992; Hannoun and Stephanopoulos, 1986; Korgel et al., 1992; Kurosawa et al., 1989; Merchant et al., 1987; Qyaas et al., 1995 ; Venancio and Tiexiera, 1997) in order to clarify the possible negative effects that the restriction of the culture medium may have on immobilized cells during fermentation. The actual diffusion coefficients of small molecules in the carrageenan gel are comparable to the diffusion coefficients of the same molecules in water, which makes the gel possible for molecular diffusion of small molecules such as glucose and ethanol. However, in typical fermentation using immobilized cells, nutrients are rapidly transferred to the immobilized cell granules mainly by convective transport in addition to molecular diffusion (Hannoun and Stephanopoulos, 1986). Once the nutrients enter the granules, the movement is relatively slow as molecular diffusion dominates. This means that the yeast cells in the periphery of

Gel granules may have a distinct advantage in nutrient delivery over those at the center of the granules.

The aging of immobilized yeast should also be taken into account. The trapped cells age during continuous fermentation for months, and they ferment under defined pseudo-steady-state conditions. But during periodic fermentation, the yeast cells are exposed to the external environment, which changes over time, and the cells are reused for only a limited number of fermentations before being deposited. More ^w research is needed to study the long-term effect of continuous fermentation on the viability of yeast cells in terms of fermentation rates.

Part A of this chapter examines the kinetics of settlement

of yeast in kappa-carrageenan gel pellets during three cycles of repetitive fermentation. The viability and concentrations of cells in immobilized and freely suspended yeast were monitored, along with the concentration of ethanol and sugars, as well as plateau degrees. Part B investigates the effect of fermentation duration on the location of cells and their distribution in granules and the morphology of yeast cells. A scanning electron microscope was used to examine immobilized yeast in different portions of kappa-carrageenan granules under four different conditions: 1) immediately after granule production; 2) after two days of fermentation; 3) after three months of continuous fermentation in an experimental bioreactor using gas lift; 4) after six months of continuous fermentation in an experimental bioreactor with

232 use of gas lift. Live yeast and their concentration in both immobilized cells and free cells in the liquid phase were also measured. The relative percentages of respiratory failure yeast (immobilized and free cells in the liquid phase) after four months of continuous fermentation in a bioreactor using gas lift were also examined. The data obtained are compared with those of traditional periodic fermentation of beer. Yeast strain was used in all studies to produce light beer.

6.1 Experimental methodology

Preparation of kappa-carrageenan gel pellets: X-0909 kappa-carrageenan gel was provided to us by Copenhagen Pectin A / S. Kappa-carrageenan gel pellets containing immobilized yeast cells for light beer production were produced using a 2.6 x 10 ⁷ cell / mL gel initial static mixer (Patent Application 2133789 (Neufeld et al., 1996) and a granule diameter of 0.5 to 2.0 mm.

fermentation medium: Beer musts with a relative weight of 17.5 ° P were supplied from Labatt breweries in Canada, as described in detail in the Materials and methods section.

Part A: Process kinetics for repeated batches using immobilized yeast in kappa-carrageenan gel pellets

The ferments were carried out in 2 L Erlenmeyer flasks at 21 ° C with shaking at 150 rpm. The loading of the carrier is 40% (v / v) of granules with immobilized cells, and the total fermentation volume is 1 L. The duration of each fermentation is seven days. During fermentation, R1 must be fermented with fresh granules with immobilized cells and at the end of

233 fermentation these granules are separated from the fermentation liquid by passing the mixture through a sterile stainless steel sieve (mesh size 500 µm). The granules are then fermented with fresh sterile wort at the same proportions in a second (R2) and then in a third (R3) periodic fermentation. Samples are taken twice daily for the first three days, then once a day for the fourth and fifth days for each fermentation. The fermentations are carried out with two or three parallel samples. All fermentations were carried out in parallel control fermentations using free-suspended cells, the conditions being the same except for free-cell fermentation at 4 g / L. Samples are analyzed for viability of free and immobilized cells, their concentration as well as the concentration of carbohydrates and ethanol in the liquid phase.

The productivity ratios Y _p / _S of ethanol and total fermentable glucose were calculated using equation 3.20 for the three immobilized cell fermentation cycles and the three free cell control fermentations. For all fermentations, productivity factors are calculated from the start of fermentation until the time when maltose is consumed.

Ethanol Y _{eT anol productivity} , i. _E. the amount of ethanol produced in the total working volume of the bioreactor per unit fermentation time is calculated using equation 3.25 for R1, R2 and R3 and for the free cell controls for the period from the start of fermentation to the time when the maltose consumption is complete. For ethanol productivity and productivity ratios, the contribution of

234

immobilized and freely suspended yeast cells are no different.

The local maximum specific growth rate _max and doubling time of the cells is calculated for the control of free cells, using equations 3.3 and 3.4.

Part B. Viability and morphological characteristics of immobilized yeast as a function of prolongation of fermentation time

Periodic fermentation conditions: Periodic fermentation was carried out in 2 L Erlenmeyer flasks at 21 ° C with shaking at 150 rpm. The loading of the carrier is 40% (v / v) and the total fermentation volume is 1 L.

Continuous fermentation conditions: Experimental gas lift bioreactors are used to carry out continuous fermentation. All collected data are from a 8 L displacement bioreactor with the exception of scanning electron microscope images of bi-monthly fermentation samples taken from a 50 L bioreactor using the same fermentation medium and immobilization method. Immobilized cell granules (40% (v / v)) were fluidized in the reactor volume using a mixture of air and carbon dioxide. The reactors operate under different conditions - the fermentation temperatures are regulated at 12.17 and 22 ° C and the residence time is between 0.9 and 1.8 days. The gas lift reactor reached a maximum ethanol concentration of 73 kg / m ³ during the six-month experiment with an average concentration of 58 kg / m ³ .

Microbiological assays: Samples are taken from the liquid phase of the gas lift bioreactor at least once a week to be tested for contamination, including for wild

235

yeast, non-fermented yeast and yeast aerobic and anaerobic bacteria. After four months, the yeast cells from the liquid phase were subjected to analysis for change in respiratory failure.

Scanning Electron Microscope (SEM): Kappa-carrageenan gel pellets (1.0-1.5 mm in diameter) containing immobilized yeast cells for light beer fermentation are taken for analysis with SEM at four different times: 1) after receipt of granules with immobilized cells and before inoculation of the granules in the fermentation medium; 2) periodic fermentation after two days; 3) continuous fermentation after 2 months in an experimental bioreactor equipped with a suction tube and using gas lift; 4) continuous fermentation after 6 months in an experimental bioreactor equipped with a suction tube and using gas lift. The methodology for analysis with SEM and the preparation of the relevant samples is described in section 4.7. Using the methods described in section 4.6, the concentration of yeast cells and their viability (immobilized and freely suspended) were determined at the same time as the CEM analyzes.

6.2 Results and Discussion

Part A: Kinetics of the Immobilized Yeast Reprocessing Process In kappa-carrageenan, the fermentation time is significantly reduced each time freshly used wort is fed to the already used immobilized cells, as shown in Figures 6.4 (a), (b) and (c) showing the content of maltose, maltotriose, glucose, fructose and ethanol depending on the duration of the

236 fermentation for three repeated periodic fermentation cycles. From these figures it can be seen that the time for complete consumption of sugars is 64 hours for R1,44 hours for R2 and 26 hours for R3. Free-suspended cells in control fermentation containing no immobilized cell granules, shown in Figure 6.5, require 82 hours for complete consumption of sugars. From the graph presented in Figure 6.4, it can also be seen that the final ethanol concentrations are highest in the third of the three recurrent fermentations with immobilized cells. Since kappa-carrageenan is hydro-gel, certain amounts of ethanol are transferred to the granules when they are re-fed to fresh wort. Therefore, at times zero for R2 and R3, some ethanol and i are present in the fermentation liquid

the initial concentration of glucose, maltose, maltotriose and fructose is lower in fermentations with immobilized cells (Figure 6.4) compared to the control fermentation with free cells (Figure 6.5). Productivity ratios have been calculated for fermentations such that the yield of g of ethanol per g of sugar consumed can be established on the basis of comparison.

Figures 6.6 (a) and (b) compare the concentrations of maltose and ethanol, depending on the fermentation time for R1, R2 and R3. During R1, maltose is raised by yeast cells almost as soon as they are introduced into fresh brewer's yeast. Ethanol concentrations peaked in R1, and also reached higher concentrations than the first two periodic ferments. As shown in Figure 6.6 (b), the initial delay in ethanol production at R1 is dramatically reduced when these immobilized cells are reintroduced into R2, and further reduced by the use of the same cells in R3.

237

Figure 6.7 (a) shows the concentration of immobilized cells per unit of total bioreactor volume depending on the fermentation time for R1, R2, and R3. The free cells that separated from the immobilized cell matrix and passed into the liquid phase in these fermentations are shown in Figure 6.7 (b) as a function of duration. Figure 6.7 (c) shows the total amount of immobilized and free yeast cells per unit of total reactor volume for the three batch fermentations. The data in Figure 6.7 (a) show that the concentration of immobilized cells in the kappa-carrageenan gel continues to increase after its initial inoculation in the yeast at R1, When the same granules are introduced into fresh beer must for fermentation in R2, the development in the gel pellets continues. In the third duration, during which the encapsulated cells are reintroduced into fresh beer must, the rate of increase in the concentration of immobilized cells slows down. The concentration profile of the free cells that are separated from the gel matrix by capa-carrageenan and pass into the liquid phase, the immobilized cells and the total amount of cells in fermentation R1 is shown in Figure 6.8

At R1, the concentration of immobilized cells in kappa-carrageenan gel pellets increased at a rate similar to that of control fermentation, which contained cells in the liquid phase only. This is confirmed by comparing the average growth curve of free cells in the control fermentation of Figure 6.9 with the same growth curve of immobilized capsule carrageenan cells in gel granules at R1 in Figure 6.10. During the course of R1, the gel granules have not yet achieved complete colonization and the gel matrix has not yet reached

238

shows an inhibitory effect on the growth of yeast cells inside the granules. In R2, the matrix limits the cell growth in the volume of the granules, as evidenced by the smaller increase in the number of cells during this fermentation cycle. This could be due to the type of gel or the accumulation of yeast cells in the granules or the lack of nutrient supply to the cells.

Table 6.1 shows the ethanol yield of the fermentable sugars of the substrate and Yp / s for the three periodic ferments and the control fermentation. Table 6.2 shows the volumetric productivity of ethanol, which is calculated on the basis of the data in Table 6.1. Sugar ethanol yields during fermentations do not differ significantly from each other or from controls. All yields are above 90% of the theoretical yield of 0.51 predicted by the Guy-Lussac equation. As noted earlier, the production of biomass and other byproducts formed by yeast cells impedes productivity over 95% of theoretical (Hardwick, 1995). The volumetric productivity of ethanol in the bioreactor in the three recurrent fermentations differs significantly from each other. Ethanol productivity increases with each cycle of repetitive fermentation, and in R3 immobilized cells are more productive than control fermentation. The total amount of ethanol obtained in R2 is not much higher than that obtained in R1, but the duration of fermentation is less than half of R1 and the control fermentation. There are many factors that can contribute to this increase in the rate of fermentation with immobilized cells with each recurrence of their use, such as the adaptation of yeast cells to fermentation conditions and

239

the progressive increase in cell concentration. The total cell volume per unit volume of the bioreactor at R3 becomes significantly larger than that of the control. In Figure 6.7. (B) The concentration curve of the freely suspended cells (separated from the gel matrix) in the liquid phase depending on the duration of the fermentation indicates that the number of cells separated from the gel granules increases with each batch stage. As the granules are more fully filled with yeast cells, more cells are released in the liquid phase. Hiisken et al. (1996) conducted studies examining bacterial cell colonies expansion and eruption / cell separation from capo-carrageenan gel. Vives et al. (1993) reported that the maximum concentration of yeast cells they reached in kappa-carrageenan gel pellets was 10 ⁹ cells per gram of gel, which corresponds to the concentration attained in the gel particles at R2. A similar maximum cell concentration has been found in the continuous fermentations described in Part B. But the maximum cell load of the gel matrix will depend on the initial cell load, gel composition and other factors.

240

Figure 6.4. a) R1, concentration of maltose, maltotriose, glucose and ethanol depending on the duration of fermentation in repeated batch fermentations using yeast cells for light beer immobilized in cap-carrageenan.

241

Concentration of maltose, maltotriose, glucose, fructose, ethanol (kg / m ³ )

Duration of fermentation (h)

Maltose Glucose Fructose Ethanol -G- Maltotriose

Figure 6.4. b) R2, concentration of maltose, maltotriose, glucose and ethanol depending on the duration of fermentation in repeated batch fermentations using yeast light beer cells immobilized in cap-carrageenan.

242

Concentration of maltose, maltotriose, glucose, fructose, ethanol (kg / m ³ )

Duration of fermentation (h)

Figure 6.4. c) R3, concentration of maltose, maltotriose, glucose and ethanol depending on the duration of fermentation in repeated batch fermentations using yeast cells for light beer immobilized in cap-carrageenan.

243

Concentration of maltose, maltopsiosis, glucose, fructose, ethanol (kg / m ³ )

-e- Maltose Glucose Fructose

-G- Maltotriose Ethanol

Figure 6.5. Concentration of maltose, maltotriose, glucose and ethanol depending on the duration of fermentation in control fermentations using free-brewed yeast for light beer (no immobilized cells).

244

Maltose concentration (kg / m

Duration of fermentation (h)

-f-R1 - «- R2

-A-R3

Figure 6.6. a) Concentration of maltose, depending on the duration of fermentation in repeated batch fermentations R1, R2 and R3, using yeast cells immobilized in gel capsules of kappa-carrageenan.

245

-4-R1 -T-R2 - * - R3

Duration of fermentation (h)

Figure 6.6. b) Ethanol concentration depending on the duration of fermentation in repeated batch fermentations R1, R2 and R3, using yeast cells immobilized in gel capsules of kappa-carrageenan.

246

Concentration of immobilized cells per unit of total volume of the bioreactor (cells / mL)

-4-R1 - «- R2 - and - R3

Duration of fermentation (h)

Figure 6.7. a) Average concentration of immobilized yeast cells for light beer per unit of total reactor volume, depending on the duration of the fermentations at R1, R2 and R3. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

247

Concentration of cells in the liquid phase per unit volume of the bioreactor (cells / mL)

- ♦ -M - • -R2 -i-R3

Duration of fermentation (h)

Figure 6.7. B) Concentration of light beer yeast cells that have separated into the bulk of the liquid phase per unit of total reactor volume depending on the duration of the fermentations at R1, R2 and R3. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

248

Duration of fermentation (h)

Figure 6.7. c) Total concentration of light beer yeast cells (immobilized and in the liquid phase) per unit of total reactor volume depending on the duration of the fermentations at R1, R2 and R3. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

249

Yeast cell concentration profile per unit of total bioreactor volume (cells / m1)

Duration of fermentation (h)

Liquid Phase Cells Immobilized Cells Total Cells

Figure 6.8. Cell concentration - immobilized, in liquid phase and total, depending on the duration of fermentation for R1 - the first of the three recurrent fermentations using light beer yeast immobilized in capo-carrageenan gel pellets.

250

Concentration of cells in the liquid phase per unit volume of the bioreactor (cells / mL)

Figure 6.9 Average concentration of freely suspended yeast cells for light beer in control fermentation (n = 3) per unit total reactor volume depending on the duration of fermentation. No immobilized cells are present during these fermentations.

251

Concentration of immobilized cells of

Figure 6.10 Concentration of immobilized yeast light beer cells per 1 mL of gel pellets depending on the duration of the fermentations at R1, R2, and R3. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

252

Table 6.1. Y _{P / s} yield of product P, substrate ethanol, glucose (Gl), fructose (Fr), maltose (Mal) and maltotriose (MT3) for R1, R2, R3 and control fermentation with free-suspended cells

253

Table 6.2 Ethanol productivity in bioreactor [V _eTaHO n = (kg ethanol produced) / (t ³ volume of bioreactor. H)] for periodic fermentations with immobilized cells (R1, R2 and R3) compared to periodic fermentations with free suspended cells.

fermentation VeraHon (kg / PTG) * R1 0,470 th most common R2 0.668 R3 1,246 th most common Free cell control 0.805

* The value is calculated when the maltose is completely consumed.

Another factor that contributes to the increase in volume productivity in the bioreactor observed with each repetitive batch fermentation is the adaptation of the yeast cells. At the end of the first fermentation, the yeast cells have adapted their metabolic mechanism to the given fermentation conditions. This may result in a reduction in the delay period at the start of subsequent periodic fermentations, thereby increasing the fermentation rates. During these studies, all control fermentations were performed with freshly brewed yeast for light beer. It would be interesting for the freely suspended yeast from the control to be reintroduced together with reused immobilized cells to further investigate this effect in terms of effects on cell concentration.

Figure 6.11 shows that the viability of immobilized cells (using the methylene blue method as an indicator) is low (<50%) when initially immobilized cells are introduced into the pouring must at

254 fermentation R1, but the viability of the immobilized cells is over 90% after 48 hours of fermentation. Yeast cells rapidly settle in the granules and their viability remains high throughout the R3 fermentation period. But at R3, viability slightly declines towards the end of fermentation. During the three recurrent fermentations, the free cells that pass into the liquid medium have a higher viability than the immobilized ones. The immobilization matrix may have a negative effect on the viability of the yeast cells (mass transfer restrictions and / or space constraints) or the live yeast cells may preferentially separate from the immobilization matrix in the liquid medium over dead cells. Using the mean data for the three separate control ferments with free-suspended yeast contained in Appendix 1, a plot of Ln (X / Xo) was plotted against the duration of the fermentation presented in Figure 6.12. The slope is equal to the local maximum specific cell growth rate at 21 ° C in wort with shaking at 150 rpm. The local maximum specific yeast growth rate was found to be 0.096 hr ' ¹ , and the cell doubling time was 7.22 hours. w _max, which is found in this study, is defined as a local area since, as noted in the theoretical, actual mu _max used in equation Monod, is reached only when S is significantly greater value compared to the constant Monod K _s . More studies are needed to calculate the Monod K _s constant for a given substrate in these fermentations to confirm that the calculated local _max is the true maximum value, as defined by the Monod equation.

255

Immobilized cell viability,%

-4-M

Figure 6.11 Viability of immobilized yeast cells for light beer (methylene blue) depending on the duration of the fermentations R1, R2 and R3.

256

Figure 6.12 Ln (X / Xo) depending on the duration of the periodic fermentation during the exponential growth phase of the freely suspended yeast in control fermentations; X is the concentration of cells at time t, and X0 is the concentration of cells at t = 0 (n = 3).

257

Part C. Viability and morphological characteristics of immobilized yeasts depending on the prolonged duration of fermentation

After obtaining immobilized cell granules using static mixers, and before the gel granules are exposed to the fermentation medium, the cell concentration is 2.6 x 10 ⁷ cells / mL gel granules (Table 6.3, where values are averaged of two samples). The scanning electron microscope image shows that the cells are separate from each other and are evenly distributed in the volume of the gel pellets (FIG. 24).

Table 6.3. Viability (methylene blue) and concentration of freely suspended and immobilized yeast yeast cells trapped in capa-carrageenan gel pellets, depending on the duration of fermentation

Go on body Fermentation method Freely suspended yeast in the liquid phase Immobilized yeast in the gel phase Life. (%) Conc. on the kp. (cells / mL liquid) Life. (%) Conc. per cell (cells / mL gel) 0 - - - - 2.6E + 07 2 days Period. 98 5.5E + 07 92 2.35E + 08 2 months Indisputable. 93 2.35E + 08 76 8.60E + 08 6 months Indisputable. 92 2.11E + 08 <50 * 1.40E + 09 *

* Based on one sample.

Viability was> 90% after 2 days of periodic fermentation, and the concentration of cells in the volume of gel pellets increased tenfold (Table 6.3). Cells (> 90% viable) also begin to separate from the gel and move into the bulk volume of the liquid phase of the fermentation process, providing a concentration of 10 ⁷ cells / mL of

258 liquid. In the volume of the gel pellets small colonies of yeast are formed, with many marks on each cell, as shown in FIG. 25.

The viability of the immobilized cells decreased after 2 months of continuous fermentation in a bioreactor with gas lift (Table 6.3), but the cells in the bulk liquid phase remained high viability (> 90%) and this finding applies to several different continuous fermentations in experimental bioreactors with gas lift. The photo of FIG. 26 from a scanning electron microscope shows that after two months, large yeast colonies form near the peripheral part of the granules, confirming the results of other researchers (Bancel and Hu, 1996; Gdia et al., 1987; Wada et al., 1979 Wang et al., 1982). The morphology of the yeast located at the outer end of the granules with immobilized cells was compared with the morphology of the cells located in

the center of the gel pellets, this comparison was made for several samples and images from a scanning electron microscope were used. The cells located in the peripheral part of the granules are ovate and smooth with many pimple marks (FIG. 27), which is an indicator of yeast multiplication (Smart, 1995). The cells depicted in the center of the granule (FIG. 28) appear to be deformed and exhibit little tendency to form puncture marks. The lack of budding scars may indicate a possible restriction of nutrients at the center of the granules, such as oxygen. The surface irregularities observed on the yeast surface of FIG. 28, may also be an indicator of cell aging (Barker and Smart, 1996; Smart, 1999).

259

The viability of the yeast immobilized in the volume of carrageenan gel drops below 50% after six months of continuous fermentation in a bioreactor with gas lift (Table 6.3). It should be noted that for the sixth month, data on the concentration of immobilized cells and viability were collected at one point only, and the reported data for the fourth month were similar concentrations of immobilized cells 1.14 x 10 ⁹ cells / mL gel and viability <50 %. While the gradual decrease in the viability of immobilized cells has been observed for a long time, the viability of the cells in the basic liquid phase remains relatively high. In addition, even though the overall viability of the immobilized cells in the granules is low, a continuous fermentation of the brewery is made over six months of continuous operation in the bioreactor. Possible reasons for this are the significant contribution to the fermentation process of the freely suspended yeast cells, which are of high viability. Also viable immobilized cells located at the periphery of the gel pellets, where there are smaller barriers to mass transfer, than the cells located in the center of the pellets also have a potential contribution. It is unclear whether the immobilized cells tend to redistribute in the volume of the gel matrix or remain in the static position where they were originally located. During the six-month fermentation period, a maximum concentration of 10 ⁹ cells / mL gel pellets is reached in the volume of these granules.

On ΦΙ / ΙΓ. 29 is an image of an entire granule taken on a scanning electron microscope. This pellet has a hollow center and, with many pellets tested, approximately half show such a structure. This cavity may be the result of structural degradation, which was further aided by the preparation of

260

samples for SEM. This cavity has not been observed in the preparation of fresh granules. Previous studies by other researchers (Bancel et al., 1996) have shown that cell growth induces a weakening of the gel network. Audet et al. (1988) reported that the addition of hornbill adhesive to capacarragenan alters the mechanical strength of gel pellets to immobilize bacteria.

Throughout the six-month beer fermentation experiment, the gas-lift bioreactor was tested for contamination at least once a week. No bacterial contamination was detected throughout the experiment. During the last two months of the experiment, contaminating yeasts were detected at a concentration of between 1 and 5 colony forming units per 1 mL. These yeasts can grow in a medium of PYN at 37 ° C, but they cannot grow aerobically or anaerobically in a medium of DUBA, do not ferment dextrin and do not show growth in a medium of CuS0 ₄ (selective for wild yeast).

After a five-month beer fermentation experiment, the average percentage of yeast cell respiratory failure is 7%, which is higher than that typically found using a strain during industrial batch fermentation (average 2%). Other researchers have reported similar data (Norton and D'Amore, 1995). Yeast respiratory failure is the result of a change that causes the yeast to be unable to process glucose to carbon dioxide and water. These yeasts have mitochondria with a permanent decrease in activity, which usually results from a change in the mitochondria of DNA (Hardwick, 1995).

261

Artifacts (structures resulting from processing) in the preparation of SEM samples may lead to misunderstandings. Technologies such as nuclear magnetic resonance (NMR), spectroscopy (Fernandez, 1996) and confocal microscopy (Bancel and Hu, 1996) have been used to test for invasion. The MRI images allowed studies to be carried out to study the transfer, movement and spatial distribution of cells and biochemistry in biolayers. Studies (Bancel and Nee, 1996) also show that confocal laser scanning microscopy by serial optical sectioning can be used to monitor cells immobilized in porous gelatin microcarriers.

Although methylene blue is used in the brewing industry as a standard indicator of cell viability, this method has many drawbacks (Mochaba et al., 1998). It measures whether the yeast population is viable or not based on the ability of living cells to oxidize the dye and convert it to a colorless form. Dead cells lack the ability to oxidize the dye and it remains blue (O'Connor-Cox et al., 1997). Methods for determining the amount of micro-organisms by Petri dish and glass culture are based on the ability of cells to grow and produce macro-colonies on agar plates or micro-colonies on a microscope slide (Technical Committee and Editorial Committee of the ASBC, 1992). In the ongoing work on the viability of yeast immobilized in matrices for large periods of time, Labatt is currently using not only the methylene blue method but also the methods mentioned above, as well as the confocal study

262 microscopy using vital staining. In addition to the topic of determining cell viability, it should be noted that the problem of "viability" of immobilized cells should also be explored in future work. When viability is used to reveal the ability of cells to grow and reproduce, vitality determines the parameters of fermentation, activity or ability of the yeast to recover from stress (Smart et al., 1999).

CHAPTER 7. TASTE QUALITIES OF BEER OBTAINED THROUGH CONTINUOUS FERMENTATION IN A GAS LIFTING SYSTEM

POWER

7.1. Experimental procedure

The use of continuous fermentation for production

Beer is very different from other methods using immobilized cells, since the product obtained is not measured with respect to only one component of interest, such as ethanol content. Rather, there must be a balance of several compounds in order to produce a quality final product. The effect of oxygen on the yeast has been investigated to produce metabolites conferring taste during continuous fermentation and during the auxiliary periodic storage. The effect of residence time on taste metabolites was also examined at two levels. Finally, by continuous fermentation, industrially produced enzyme decarboxylase alpha-acetolactate was added to the casting must, and the total concentration of diacetyl in the liquid phase was observed.

7.1.1 Effect of the relative air content of the fluidised gas fed to the bioreactor on the bioreactor

263 yeast metabolites during major continuous fermentation

The amount of air, and therefore oxygen, in the fluidized gas supplied to the bioreactor changes, while the residence time, temperature and all other process control parameters remain constant. The total volume flow rate of the gas is kept constant at 472 mL / min at normal temperature and pressure, the temperature is 15 ° C, and the kappa-carrageenan gel granules contain immobilized yeast LCC 3021, which are used throughout the test period. with an initial cell concentration of 1 x 10 ⁸ cells / mL gel. Four different volumetric air flow rates were used in the system experiment (Table 7.1) and the average residence time R _t in the bioreactor was 1.18 days.

Table 7.1 Volumetric flow rate of air supplied to the bioreactor during gas spraying during continuous fermentation. The total volume flow rate supplied to the bioreactor is 472 mL / min at normal temperature and pressure and the rest of the gas is carbon dioxide.

Volume airflow (mL / min) Fluid gas percentage air content (% (v / v)) begining (DAYS) End (days) Total time (days) 94 19,9 10 26 17 354 75,0 27 40 14 34 7.2 41 58 18 0 0 59 66 8

264

The following assays were performed repeatedly during the experiment: free amine nitrogen, total amount of fermentable carbohydrate (such as glucose), ethanol, total amount of diacetyl, volatile substances in beer (selected esters and alcohols), concentration of yeast cells in the liquid phase and viability .The bioreactor is also tested for contamination at least once a week.

The dissolved oxygen concentration in the bulk volume of the liquid phase is measured when the continuous fermentation is assumed to be in a pseudo-steady state for each volumetric air flow rate (minimum three revolutions of the bioreactor).

7.1.2 Batch storage period after continuous fermentation: Effects of oxygen on yeast metabolites.

Even when the amount of oxygen in the fluidised gas flowing into the reactor is relatively low (34 ml / min at normal pressure and temperature), the concentrations of acetaldehyde and total diacetyl detected during the experiment and conducted as described in section 7.1. 2, are unacceptably high for the beer market in North American beer. Therefore, a new approach is applied in which, after continuous main fermentation in the bioreactor, the liquid is stored in batches for 48 hours at a slightly elevated temperature of 21 ° C to reduce the concentration of these two substances. Also, the results in the previous section 7.1.2 indicate a significant effect, in that the amount of air in the fluidised gas influences the measured quantities of compounds that impart taste. Therefore, the effect of yeast taste metabolites on the

265 aerobic compared to anaerobic conditions under additional periodic retention after the main fermentation.

Continuous main fermentation was carried out in a bioreactor with a gas lift and a volume of 50 L, using the highly flocculent variant of the yeast strain LCC3021 in the experiment, as the volume requirements for the test samples were too large compared to a bioreactor with a volume of 8 L. The operating conditions are as follows: 1180 mL / min CO ₂ and 189 mL / min air in the fluid

gas at normal temperature and pressure, average stay R in the bioreactor -1.0 day, temperature 15 ° C and beer must with high relative weight of 17.5 ° P.

A total of 4 samples (100 mL vials) were taken, two being held under anaerobic conditions and the other two exposed to an aerobic environment.

The procedure with anaerobic samples is as follows: two 100 mL vials and 6 25 mL vials are autoclaved and then placed in an anaerobic cabinet (Labmaster 100, mbraun, USA) with argon gas. The 100 mL vials were allowed to equilibrate for 45 minutes and then sealed with Teflon® aluminum caps and compartments. A 50 mL syringe equipped with a 3 inch needle, gauge 16, and sterilized with 70% (v / v) ethanol solution is used to sample the bioreactor by piercing the membrane of the sample valve membrane. The sample was then injected into a 100 mL pre-purged anaerobic vial. It is necessary to provide a groove in the vial with an additional sterile syringe needle to release pressure in the volume of the vial as it is being filled. After being taken from the bioreactor, aerobic samples are exposed to the atmosphere by fully opening the diaphragm sampling valve.

266 samples and pouring into an unopened 100 mL sample bottle without the use of a syringe and needle.

The sample fluid was left to stand for 2 hours at room temperature to precipitate the cells from the solution, leaving cells with a concentration of approximately 10 ⁶ cells / mL in the basic liquid phase. After precipitation, the liquid from each 100 mL vial was decanted into three 25 mL vials to allow analyzes to be made without altering the course of the fermentation process due to sampling. After aerobic samples were transferred to smaller vials, they were incubated at 21 ° C without caps. The anaerobic samples were transferred to three smaller vials and closed with Teflon® aluminum caps and compartments. To avoid creating pressure due to separation of

carbon dioxide in the volume of the vials, whilst preventing exposure of the samples to the aerobic environment, the barrier is punctured with a needle. The end of the needle exposed to the external environment is immersed in ethanol (less than 1 cm from the presser head), thus preventing air from entering the samples. Samples for analysis are taken at 2, 24 and 48 hours: A sample is also taken directly from the bioreactor and immediately analyzed to determine the state of the fermentation process in the volume of the bioreactor for the appropriate duration. Samples are analyzed for total fermentable carbohydrate content (such as glucose), ethanol, total diacetyl content and volatile substances in beer (selected esters and alcohols).

7.1.3 Effect of residence time of the fluid on the yeast metabolites during continuous main fermentation

267

In order to investigate the effect of the residence time of the liquid on the metabolic activity of the yeast, an experiment was conducted in which, during continuous main fermentation, a stepwise change in the volume flow of the must must be fed to the bioreactor. Fermentation utilizes LCC3021 yeast cells immobilized in kappa-carrageenan gel pellets. The temperature in the bioreactor was kept constant,! 7 ° C, throughout the test. The volume flow rate of the gas supplied to the bioreactor is also constant and is 472 mL / min at normal pressure and temperature. The gas is a mixture of air (11 mL / min at normal temperature and pressure) and carbon dioxide (461 mL / min at normal temperature and pressure). The initial concentration of yeast cells in the capo-carrageenan gel is 2.6 x 10 ⁷ cells / mL gel pellets and the bioreactor contains 40% (v / v) granules.

The following analyzes were performed repeatedly during the experiment: carbohydrates, free amine nitrogen, total amount of fermentable carbohydrates (such as glucose), ethanol, total diacetyl, volatile substances in beer (selected esters and alcohols) and concentration of yeast cells in the liquid phase and viability. The bioreactor was also tested for contamination at least once a week.

7.1.4 Use of industrially prepared decarboxylase alpha-acetolactate to reduce the total amount of diacetyl during continuous main fermentation of beer

Most breweries in North America believe that high concentrations of diacetyl are an undesirable taste defect in beer. In continuous main fermentations carried out so far, total concentrations of diacetyl are generally above the limit

268

levels for traditional batch fermentations (70-150 µg / L) in North American light beer. During the periodic fermentation, diacetylt decreases in the later stages of fermentation when oxygen is not in large quantities and no additional sugars are introduced. In the case of continuous fermentation, a constant low level of oxygen and fresh beer must be supplied in the bioreactor during continuous spraying. Therefore, a novel strategy is explored using an industrially prepared enzyme to regulate the concentration of diacetyl in a continuous bioreactor.

In fermentation of wort, diacetyl is formed when alpha-acetolactate, an intermediate of valine synthesis, oxidizes decarboxylates outside the yeast cell. The yeast cell then reabsorbs the diacetyl and turns it into a less flavourful acetoin. This oxidative decarboxylation of alpha-acetolactate to diacetyl is limited at periodic fermentations of beer must. In continuous fermentations, total diacetyl at the outlet of the bioreactor is at unacceptable concentrations (300-400 µg / L). The industrially prepared enzyme-decarboxylase alpha-acetolactate of Novo-Nordisk A / S can be converted directly to acetoin, thus avoiding the undesirable diacetyl intermediate (Figure 7.1) (Jepsen, 1993).

269

Alpha-acetolactate

COz + pH]

Figure 7.1 Effect of decarboxylase alpha-acetolactate

270

Decarboxylase alpha-acetolactate is added to the bioreactor to investigate its total effect (end effect) on the total concentration of diacetyl. Other strategies for diacetyl reduction have been explored, which include periodic retention for 48 hours after fermentation and second-stage immobilized cell fermentation systems using Alpha-Laval technologies (Anon, 1997). These two strategies have shown success in reducing diacetyl levels at the output of the bioreactor. By using decarboxylase alpha-acetolactate (DKAA) in the casting must to reduce the diacetyl content of the product discharged from the bioreactor, the post-fermentation treatment periods can be minimized or eliminated.

Decarboxylase activity of alpha-acetolactate is optimal at a pH of 6.0 at 10 ° C for light beer, which is typical of industrial beer must. Alpha-acetolactate activity is maximized at 35 ° C (Anon, 1994) Under typical fermentation conditions where the temperature and pH are lower, alpha-acetolactate activity is less than optimal.

In 1997, Health Canada amended the Food and Drug Administration (SOR / 97-81) of Canada to allow the use of alpha-acetolactate decarboxylase in alcoholic beverages, which is an open door for its use in Canadian breweries . Bacillus subtilis, carrying the decarboxylase alpha-acetolactate gene code from Bacillus brevis, produces this enzyme. Since alpha-acetolactate is an enzyme produced by a genetically modified organism, it is publicly available.

271 understanding that it was necessary to examine his behavior

before it is used as a commercial product.

For the experiments, yeast LCC3021 was used to prepare light beer. Beer musts with a relative weight of -17.5 ° P are delivered from Labatt in London. Ethanol concentrations, total fermentable carbohydrates (such as glucose), total diacetyl, and cell concentration in the liquid phase are measured. The yeast cells were immobilized in kappa-carrageenan gel pellets as described in Chapter 4. The bioreactor made three turns before being assumed to have reached a pseudo-steady state. As noted earlier, the diacetyl method used in this development refers to "total diacetyl" as this method measures the amount of diacetyl and its precursor alpha-acetolactate. Thus, the observed decrease in total diacetyl during the course of the experiment could be due to the combined effect of converting the alpha-acetolactate enzyme directly to acetoin and a subsequent decrease in the concentration of its diacetyl derivative.

Alpha-acetolactate decarboxylase was provided for laboratory use by Novo Nordisk A / S, Denmark as Matures® L. The enzyme activity is 1500 ADU / g, where ADU is the amount of enzyme from which, under standard conditions, decarboxylation of alpha-acetolactate is receive 1 μ mole of acetoin per minute as described in Novo Nordisk Method AF27 (Anon, 1994).

Continuous fermentation conditions: Continuous fermentation is carried out in an 8 L volume bioreactor equipped with a suction tube using gas lift. It supplies gel capsules of kappa-carrageenan, which immobilize yeast cells for the production of light

272 beers in the amount of 40% (v / v). A gas mixture of carbon dioxide (438 mL / min at normal temperature and pressure) and air (34 ml / min at normal temperature and pressure) is sprayed into the bioreactor. The fermentation temperature was adjusted to 15 ° C throughout the test period and the residence time of R _t in the bioreactor was 1.5 days. Under these conditions, the total concentration of diacetyl is monitored and an average value of the pseudo-steady state diacetyl concentration is reached. The decarboxylase alfaacetolactate 72 gg / L (108 ADU / L) was then added to the must and the concentration of diacetyl was observed in the bioreactor.

Experiment 1: The wort is taken from the brewery in 20 L stainless steel vessels and heated in an autoclave for 45 minutes at 100 ° C. The wort is kept at 2 ° C in a controlled water bath until fed to the bioreactor. When a pseudo-steady state concentration of diacetyl is reached in the bioreactor, 72 gg / L (108 ADU / L) of DKAA is added to the must in a 20 L container. The initial biomass concentration in the kappa carrageenan gel pellets was 3 x 10 ⁷ cells / mL gel.

Experiment 2: In order to minimize the risk of contamination, the system is closed in at the outlet and other safeguards are also provided to the system, as described in Chapter 4. As in Experiment 1, beer must is taken. from the brewery in 20 L stainless steel vessels and heated in an autoclave for 45 minutes at 100 ° C. The wort is kept at 2 ° C in a controlled water bath until fed to the bioreactor. The initial biomass concentration in the capacaragenan gel pellets was 3 x 10 ⁷ cells / mL gel. When in the bioreactor is reached

273

concentration of diacetyl under pseudo-steady state conditions, 72 gg / L (108 ADU / L) ODKAA was added to the must in a 20 L container.

Experiment 3; Oxygen-free beer must at 17.5 ° P (14 hL) was fed into a large beer must (T-1) storage vessel in the experimental installation. The wort must then be pasteurized using steam and stored, spraying carbon dioxide in it to maintain a constant dissolved oxygen concentration of <0.10 mg / L, as described in Chapter 5. From this tank, the wort must is fed into the bioreactor to reach a total concentration of diacetyl under pseudo-steady state conditions. Then, under aseptic conditions, alpha-acetolactate 72 gg / L was added to the wort to conduct the remainder of the experiment. Addition of alpha-acetolactate is accomplished by measuring the amount of wort remaining in the storage tank and by calculating the amount of alpha-acetolactate required to reach an enzyme concentration of 72 gg / L (108 ADU) / L). The appropriate amount of enzyme is then dissolved in 10 L of sterile beer must. This solution is transferred to a 20 L stainless steel pressure vessel, which is connected via a sterile tube to the sampling hole of the beer must storage tank (T-1). The alpha-acetolactate solution is then introduced under pressure into a beer must storage tank using sterile carbon dioxide for this purpose. To ensure the proper mixing of the alpha-acetolactate solution with the must in the tank, the flow rate of carbon dioxide sprayed into the vessel is raised to 4720 mL / min at normal temperature and pressure for 1 hour, after which normal flow is restored. The storage tank is contained

274 wort to which sufficient alpha-acetolactate was added to complete the experiment. The initial biomass in the kappa-carrageenan gel pellets was 10 ⁸ cells / mL gel.

Results and discussion

7.2.1 Effect of the relative amount of air in the fluidised gas supplied to the bioreactor on the yeast metabolites during the main continuous fermentation

Figures 7.2 - 7.11 graphically depict the viability of the yeast in the liquid phase and the cell concentration, as well as the concentrations of free amine nitrogen (CAA), total fermentable carbohydrates (such as glucose), ethanol, total diacetyl, acetaldehyde, ethyl acetate, 1-propanolate , isobutanol,

isoamyl acetate, isoamyl alcohol, ethylhexanoate and ethylchanoate depending on the duration of continuous fermentation. All operating parameters of the bioreactor are kept constant throughout the test except for the percentage of air in the gas sprayed into the bioreactor, which is indicated directly in the figures. Table 7.2 gives the mean values for each pseudo-steady state analysis (after a minimum of three reactor revolutions).

275

Table 7.2 Summary table showing the effect of air flow through the sprayer in the bioreactor on the concentrations of the yeast in the liquid phase and the key yeast metabolites in the bioreactor over a residence time of R _{t of} 1.18 days, giving the mean values at the pseudostable condition .

Average * concentration Volume airflow (mL / min) 94 354 34 Conc. per cell (cells / mL) 3.87E + 08 2.98E + 08 4.73E + 08 Total farms. glucose (g / 100 ml_) 1.36 1,25 2.07 CAA (mg / L) 196,9 171,7 162,8 Ethanol (g / 100 mL) 6.14 5.46 5.74 Total diacetyl (ug / L) 346 1417 389 Acetapdehyde (mg / L) 75.62 329,48 28,63 Ethyl acetate (mg / L) 22,38 21.13 18.01 1-propanol 44.74 50.89 53.04 Isobutanol (mg / L) 8.73 16.09 8.05 Isoamyl acetate (mg / L) 0.38 0.21 0.30 Isoamyl alcohol (mg / L) 58.62 61.64 59.16 Ethylhexanoate (mg / L) 0.060 0.030 0.053 Butyloctanoate (mg / L) 0.031 0.013 0.025

* Average for the last four days for each operating parameter

Figures 7.2 and 7.3 show that in this experiment the yeast population in the liquid phase did not reach zero. The flavoring compounds investigated in this development were prepared by combining free and immobilized yeast cells and the relative contribution of each source was not determined. There is more than one source available

276

suspended yeast cells: the growth of biomass and cells that have separated from the gel pellets and passed into the basic liquid medium. Studies with composite models of separated and cultured cells have shown that when cells are separated from the surface layers of biomass, even if the bioreactor operates at a high dilution rate, there is still a population of cells in the fluid (Karamanev, 1991).

Figure 7.4 shows the concentration of free amine nitrogen (CAA) in the liquid phase. It is interesting to note that the minimum concentration of free amine nitrogen is observed at an air flow of 34 mL / min at normal temperature and pressure. This does not coincide with the maximum concentration of ethanol or the minimum concentrations of total fermentable carbohydrate (such as glucose).

The concentration of ethanol in the liquid phase in the reactor decreases while the total fermentable carbohydrate (such as glucose) increases when the volume flow rate of air in the spray gas increases from 94 to 354 mL / min, as can be seen from Figure 7.5. This may indicate that there is more cell respiration rather than fermentation due to the increase in oxygen available. When the volume flow rate is again reduced from 354 mL / min to 34 mL / min at normal temperature and pressure, the ethanol concentration again increases but does not reach the concentration observed at a volume flow rate of 94 mL / min, as other factors influence on the system resulting in cell aging, the effects of prolonged exposure to a relatively high amount of oxygen at a time when the flow rate is 354 mL / min, and changes in the population of immobilized cells. Figure 2.4 White and Portno (1978) noted changes in concentrations of yeast metabolites associated with taste.

277

properties, depending on the duration of continuous fermentation carried out in their vertical fermenter.

Figure 7.6 shows the pronounced effect of oxygen on the production of total diacetyl, since diacetyl is generally considered an undesirable flavoring substance in beer, one of the main reasons for optimizing the amount of oxygen in the bioreactor is to regulate flavoring levels. After a purge period of 354 mL / min, the flow rate drops to 34 mL / min at normal temperature and pressure and the total diacetyl decreases. During periodic fermentation, it is known that an increase in oxygen leads to the formation of alpha-acetolactate, which is a precursor of diacetyl (Kunze, 1996).

Figure 7.7 shows a clear relationship between the amount of air in the spray gas and the concentration of acetaldehyde. As the percentage of air in the spray gas increases, the amount of acetaldehyde also increases. Acetaldehyde gives the beer a green apple flavor and is usually present in commercially produced beer at levels less than 20 mg / L.

278

1.00E + 10Concentration of cells in the liquid phase (cells / mL)

1.oooo + o0

1.00E + 08 ·

1.00E + 071.00E + 06

1.00E + 05 · ♦

1.oooo + o +

1.00Ε + ΟΉ

1.00E + 02

1.QOE + O1

1.00E + 00 *

I,; r *

Continuous fermentation duration (h)

Figure 7.2 Concentration of yeast cells in the liquid phase depending on the duration of continuous fermentation. The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

279

Figure 7.3 Viability of the yeast in the liquid phase depending on the duration of continuous fermentation. The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

•

280

Figure 7.4 Concentration of free amine nitrogen remaining in the must, depending on the duration of continuous fermentation. The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

281

7.00

Concentration of ethanol and fermentable glucose (g / 100mL)

6.00

5.00

4.00

3.00 ·

200

1.00

0.00> ♦

·. ♦

TTT

And,. f! 'u' ·· · -! k't-M jnL / min — r — ► '/ ·· V +.

! ' ', J. > · .'R

D 'W': ·. \ *

ί ^Λ ^ 3. · No.> * · »· '♦ ♦♦♦ • t'

G.,. ••• I * ·· /. 4? '

I t

Ί | <v; ¹ ' ^r

• '! I I i r i i -------------, I ......

5 10 15 20 25 30 36 40 h

♦ ♦♦ ♦ Ethanol

Farm.pucosa

Duration of continuous fermentation (days)

Figure 7.5 Concentration of ethanol and total fermentable carbohydrate (such as glucose) depending on the duration of continuous fermentation. The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

282

Figure 7.6 Total concentration of diacetyl depending on the duration of continuous fermentation. . The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

283

Duration of continuous fermentation (days)

Figure 7.7 Concentration of acetaldehyde in the liquid phase depending on the duration of continuous fermentation. The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

284

Table 7.2 and Figures 7.8 - 7.9 give the concentrations of ethyl acetate, isoamyl acetate, ethylhexanoate and ethyl octanoate in a pseudorandom state of the system depending on the duration of continuous fermentation. For all esters measured, a stepwise change of aeration flow from 94 to 354 ml / min at normal temperature and pressure results in a decrease in concentration. When the aeration flow is reduced from 354 mL / min to 34 mL / min at normal temperature and pressure, the concentrations of isoamyl acetate, ethyl hexanoate and ethyl octanoate increase. However, they do not increase to the values observed at aeration flow rate of 94 mL / min. The performance diagrams of these compounds are very close together, with ethylhexanoate and ethyl octanoate showing more

relative change compared to isoamyl acetate. The ethyl acetate concentration decreases when the volume flow rate of air is reduced to 34 mL / min at normal temperature and pressure.

For all these esters measured during the study, their concentration increases when the air is completely eliminated from the fluidizing gas. The concentration of each ester increases and then the curve of the diagram takes on a cone shape when the concentration of cells in the liquid phase of the bioreactor decreases sharply.

Figures 7.10 and 7.11 give the contents of higher alcohols, isoamylapacol, isobutanol and 1-propanol depending on the duration of continuous fermentation. For the alcohols measured, the concentration increases as a result of a stepwise aeration increase from 94 to 354mL / min at normal temperature and pressure. Isogoganol shows the largest relative change in change in aeration flow.

285

Concentrations of 1-propanol are well below the limit values for flavoring agents, which are 600-800 mg / L, but during the experimental continuous fermentation, the concentration was well above that found for conventional industry beer obtained by periodic fermentation at which the concentration is usually below 16 mg / L. This is not the case for isoamyl alcohol or isobutanol which are within the normal range. The compound 1 propanol is supposed to be produced by the reduction of acidic propionate. (Gee and Ramirez, 1994). Other researchers (Hough et al., 1982; Yamauchi et al., 1995) also reported the formation of 1-propanol of the amino acid metabolism of α-aminobutyric acid and threonine with the corresponding oxyacid and aldehyde, respectively α-butanoic acid and propionaldehyde .

Because excess diacetyl, acetaldehyde, and fusel alcohols are not desired in beer to limit their production, it is important to regulate oxygen. As discussed in the literature, when oxygen uptake into yeast cells increases, the anabolic formation of amino acid precursors increases and thus a higher content of alcohols, oxyacids and diacetyl is observed. It is known that the concentration of esters decreases with the increase of oxygen, since the formation of esters is catalyzed by the transfer of acetyl. This acetyl transfer is inhibited by unsaturated fatty acids and ergosterol, which in turn are increased by the presence of oxygen (Norton and D'Amore, 1994).

Under the conditions of the bioreactor used in this experiment under pseudo-steady state (after a minimum of three revolutions per

286 bioreactors) the measured concentrations of dissolved oxygen in the liquid phase in the bioreactor are closer to zero (less than 0.03 mg / L).

This experiment did not allow a direct comparison of data from 94mLymin and 34mL / min air flow rate in the fluidized gas at normal temperature and pressure, as they were separated from the highest flow rate (354 mL / min). This is because of the physiological state of the yeast as a result of exposure to the previous conditions in the bioreactor. The immobilizing matrix and the continuous one

fermentation can also lead to other changes in the taste of the product.

No contamination was detected at any point of the bioreactor during the experiment.

Other strategies, such as the addition of nutrients such as zinc, could be explored to balance the requirements for a certain amount of oxygen to maintain the viability of the yeast and the need to minimize the amount of oxygen to produce beer with the necessary flavors. , magnesium or the provision of other exogenous substances necessary for the yeast cells to maintain their viability. The addition of such substances makes it possible to reduce the oxygen content requirements of the yeast. Alternatively, very low oxygen concentrations can be used for most of the time, with pulses being oxygenated periodically at certain sites to maintain cell viability.

287

b 10 15 20 25 30 36 40 46 60 65 60 66 70

Duration of continuous fermentation (days)

Figure 7.8 Concentration of ethyl acetate as a function of the duration of continuous fermentation. . The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

288

Isoamyl acetate concentration (mg / L)

Figure 7.9 Concentration of isoamyl acetate, ethylhexanoate and ethyl octanoate in the liquid phase depending on the duration of continuous fermentation. . The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

z £ o

S 'oh

Isoamyl acetate Ethylhexanoate Ethyl octanoate

289

Isoamyl alcohol concentration (mg / L)

Duration of continuous fermentation (days)

Figure 7.10 Concentration of isoamyl alcohol and isobutanol in the liquid phase depending on the duration of continuous fermentation. The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

Isoamyl alcohol Isobutanol

290

Figure 7.11 Concentration of 1-propanol in the liquid phase as a function of the duration of continuous fermentation. The volumetric air flow rate at normal temperature and pressure supplied to the bioreactor via a spray device is shown in the graph. The remainder of the gas is carbon dioxide and during the experiment the total gas flow rate is a constant of 472 mL / min at normal temperature and pressure.

291

7.2.2 Post-fermentation period of product retention in batches: Effects of oxygen action on yeast metabolites.

Since total diacetyl is not within the required range for industrially produced beer at the end of periodic fermentation, several approaches are used to reduce the concentration of this compound to acceptable limits. One

this approach is to use a warmed-up retention period immediately after continuous main fermentation.

Figures 7.12 - 7.21 graphically depict the concentrations of total fermentable carbohydrate (such as glucose), ethanol, total diacetyl, acetaldehyde, ethyl acetate, 1-propanol, isobutanol, isoamyl acetate, isoamyl alcohol and ethylhexanoate in the liquid phase dependence. Samples taken from the fermenter for continuous fermentation in the pseudostable condition are kept in aerobic or anaerobic state, as shown in the legend of each of the figures.

In Figure 7.12, the concentration of total fermentable carbohydrate (such as glucose) decreases rapidly during the first two hours and then decreases at a slower rate during both the retention period at both aerobic and anaerobic samples. A possible reason for this observation is that in the first two hours before decantation more yeast is present and the sugar concentration is higher at the beginning of the retention period. There is no significant difference between aerobic and anaerobic samples with respect to the reduction of fermentable glucose, although some differences may be observed initially.

292

The ethanol concentration of Figure 7.13 increases rapidly at the start of the retention period and then aerobic and anaerobic samples increase the ethanol concentration over time in parallel. The initial increase in ethanol concentration for the anaerobic sample coincides with the period during which the sugars content is most reduced. At the end of the retention period, the ethanol concentration was higher in the anaerobically treated samples.

It can be seen from Figure 7.14 that aerobic samples show an increase in acetaldehyde at the outset when exposed to aerobic conditions outside the bioreactor. The combination of aerobic conditions with sugar consumption and ethanol production could explain this result. At the end of the 48-hour retention period, the acetaldehyde concentration drops from 17 mg / L to 9 mg / L in the anaerobic sample, providing a concentration within the range of quality North American light beer (less than 10 mg / L).

The concentration of total diacetyl depending on the retention time is given in Figure 7.15. The results show that the elimination of oxygen from the system during this retention period provides more favorable conditions for the reduction of diacetyl. The shape of the curve of total diacetyl can be explained by the depletion of free amine nitrogen and the subsequent intracellular production of valine to which diacetyl is a by-product (Nakatani et al., 1984a; Nakatani et al., 1984b). The total concentration of diacetyl at the end of the main continuous fermentation is 326 µg / L and at the end of the anaerobic retention period is 33 µg / L, which is well below the limit for industrially produced beer.

293

Figures 7.16 - 7.18 show diagrams of the concentrations of ethyl acetate, isoamyl acetate and ethyl hexanoate esters depending on the duration of post-fermentation retention. The same aerobic and anaerobic samples were observed for all esters. The concentration of esters in aerobic and anaerobic samples does not differ, whereas later during the retention period, the concentration of esters in aerobic samples decreases and anaerobic increases. Since the concentration of esters in continuous fermentation is somewhat low compared to the concentration of esters found in commercially produced beer, it is desirable to select conditions that favor the production of esters.

Figures 7.19 - 7.21 show the concentrations of isoamyl alcohol, 1-propanol and isobutanol depending on the duration of post-fermentation retention. At the end of the 48 hour retention period, there was no significant difference in these alcohols for aerobically and anaerobically treated samples. However, at 24 hours the aerobically treated samples in all cases showed a high concentration.

Figure 7.22 presents radar graphs that allow comparison of several flavoring substances after a 48 hour aerobic and anaerobic retention period with industrially produced beer. Radar graphics are commonly used in the brewing industry as they allow the comparison of different characteristics of beer, the data of which is plotted (Sharpe, 1988). In this figure, it can be seen that the beer obtained by continuous fermentation and then retained under anaerobic conditions is closest to ordinary market beer. It can be seen from Appendix 6 that the anaerobic fluid is within the normal range of

294

market beer with the exception of the content of 1-propanol, which is significantly higher than the periodically fermented beer. This higher than normal content of 1-propanol was observed in all products obtained by continuous fermentation in this study.

The formation of 1-propanol is observed during the main continuous fermentation stage and its content does not decrease significantly during the retention period, where the conditions are aerobic and anaerobic. Kunze (1996) found that during periodic fermentation, the following factors increase higher alcohols such as 1-propanol: mixing, intense aeration of beer must and re-adding fresh beer must to available yeast.

Ultimately, the ideal case would be to completely eliminate the secondary retention period by optimizing the conditions in the bioreactor in the main continuous fermentation. But another benefit could be gained by using the retention period by optimizing the retention temperature (the removal of diacetyls by the yeast is highly temperature dependent), the amount of fermentable sugars remaining in the liquid at the start of the retention period, optimizing the concentration of yeast present, the hydrodynamic characteristics of the retention vessel (diacetyl removal could be improved by improving the contact between the yeast and beer) and undertaking no measures to eliminate oxygen at this stage.

The calculations for the volumetric capacity of beer are given in Appendix 3. The process described in this section, which involves continuous operation of the bioreactor with a 24-hour standby time and subsequent periodic holding for 48 hours, is 1.8 times 295.

productive compared to the conventional industrial batch process. The relatively fast industrial batch process with a cycle of 7.5 days has a volume production of beer of 0.093 t ³ (the volume of the vessel in t ³ x day), while the continuous process described here has a productivity of beer of 0.165 t ³ (the volume of the vessel in t ³ x day). If further studies allow the retention period to be shortened to 24 hours, beer productivity will be 2.3 times higher than the industry benchmark. If the ideal case of a 24-hour continuous process without subsequent batch retention is achieved, the volumetric productivity of beer will be 7.5 times higher than the industry standard for recurring work. In addition to the increase in volume productivity, additional benefits are also gained from going from a batch to a continuous process, such as a shorter time to market, a reduction in the size of the brewery and a lower breeding frequency of yeast, which should be balanced with careful analyzes of the relative costs of labor.

Other researchers (Kronlof and Virkajarvi, 1996; Nakanishi et al., 1993; Yamauchi et al., 1995) focus their attention on the study of multi-stage continuous fermentations in which the first stage of continuous fermentation (aerobic) has as a result only partial consumption of fermentable sugars in beer must. While this approach has shown some success in obtaining flavors, these systems are complex. During the first aerobic stage of such a system, an environment that is more susceptible to microbial contamination (ie, high concentrations of sugars, temperature and oxygen with low ethanol concentration) is created. In the gas lift bioreactor system used in this study, in the bioreactor

296 has a low concentration of fermentable sugars, a low pH value, a high ethanol concentration and a low oxygen concentration, which makes the environment unsuitable for possible contamination.

297

Figure 7.12 Average concentration of fermentable gyukosis as a function of the retention time of aerobically and anaerobically treated samples after continuous main fermentation in a bioreactor with gas lift. The deviations along the lines intersecting the curves represent the upper and lower bounds of

Retention time after fermentation (h) “* ·· Aerobic · * - Anaerobic

Figure 7.13 Average ethanol concentration as a function of the retention time of aerobically and anaerobically treated samples after continuous main fermentation in a bioreactor with gas lift. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

298

Aerobic - * · Anaerobic

Retention time after fermentation (h)

Figure 7.14 Average concentration of acetaldehyde depending on the retention time of aerobically and anaerobically treated samples after continuous main fermentation in a bioreactor with gas lift. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

299

Retention time after fermentation (h)

Figure 7.15 Average concentration of diacetyl as a function of the retention time of aerobically and anaerobically treated samples after continuous main fermentation in a gas lift bioreactor. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

300

Retention time after fermentation (h)

Figure 7.16. Average concentration of ethyl acetate as a function of the retention time of aerobically and anaerobically treated samples after the main continuous fermentation in a bioreactor with gas lift. The deviations along the lines intersecting the curves represent the upper and lower bounds of AIPPAPYAANTYAPMITA PAMMI =

Retention time after fermentation (h)

Figure 7.17. Average concentration of isoamyl acetate as a function of the retention time of aerobically and anaerobically treated samples after the main continuous fermentation in a bioreactor with gas lift. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

301

Retention time after fermentation (h) Figure 7.18. Average concentration of ethylhexanoate depending on the retention time of aerobically and anaerobically treated samples after the main continuous fermentation in a bioreactor with gas lift. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

302

Retention time after fermentation (h)

Figure 7.19. Average concentration of isoamylalcohol depending on the retention time of aerobically and anaerobically treated samples after the main continuous fermentation in a bioreactor with gas lift. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

Figure 7.20. Average concentration of 1-propanol as a function of the retention time of aerobically and anaerobically treated samples after the main continuous fermentation in a gas lift bioreactor. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

303

Retention time after fermentation (h)

Figure 7.21. Average isobutanol concentration as a function of the retention time of aerobically and anaerobically treated samples after the main continuous fermentation in a bioreactor with gas lift. The deviations along the lines intersecting the curves represent the upper and lower bounds of the experimental data (n = 2).

304

Figure 7.22. Radar graph of normal concentration data obtained after 48 hours of aerobic and anaerobic retention following continuous main fermentation in a gas lift bioreactor. Data on normal concentrations are based on the averages of duplicate samples and data on industrial beer are taken as the average of the data given in Annex 6.

305

7.2.3 Effect of the residence time of the liquid on key yeast metabolites during continuous main fermentation

Figures 7.23 - 7.28 show the results of the liquid phase assays obtained in the bioreactor. Table 7.3 lists the mean concentrations and flow rates at the pseudorandom state (after a minimum of three turns of the bioreactor). Two downtime were used in the experiment. While the viability of the yeast in the liquid phase does not change significantly as the flow rate of the wort fed to the bioreactor increases, the concentration of yeast cells changes, as can be seen from Figure 7.23.

Table 7.3 (a) Effect of residence time in the bioreactor on the yeast in the liquid phase and concentrations of key yeast metabolites, presented as mean values at the pseudo-steady state; (B) Effect of residence time in the bioreactor on the yeast in the liquid phase and concentrations of key yeast metabolites at the bioreactor outlet, presented as mean values at the pseudo-steady state.

(a) Average concentration Duration of stay in the bioreactor

1.8 days 0.9 days

Conc. per cell (cells / mL) 2.38E + 08 1.32E + 08 Total farms. glucose (g / 100 mL) 0.29 6.09 CAA (mg / L) 106,3 246,4 Ethanol (d / 100 mL) 5.16 4.80 Total diacetyl (ug / L) 292 460 Acetaldehyde (mg / L) 19,47 37.07 Ethyl acetate (mg / L) 41,00 38,29

1-propanol (mg / L) 44.95 13.53 Isobutanol (mg / L) 22,78 9.13 Isoamyl acetate (mg / L) 0.90 1,28 Isoamyl alcohol (mg / L) 76.67 51.39 (b) Average flow Duration of stay in the bioreactor 1.8 days 0.9 days Conc. per cell (cells / min) 7.38E + 08 8.22E + 08 Total farms. glucose (g / 100 min) 8.93E-03 3.72E-01 CAA (g / min) 3.65E-04 1.50E-03 Ethanol (g / 100 min) 1.60E-01 2.93E-01 Total diacetyl (g / min) 9.06E-07 2.81 E-06 Acetaldehyde (g / min) 6.04E-05 2.26E-04 Ethyl acetate (g / min) 1.27F-04 2.34E-04 1-propanol (g / min) 1.39E-04 8.25E-05 Isobutanol (g / min) 7.06E-05 5.57E-05 Isomyl acetate (g / min) 2.80E-06 7.30E-06 Isos! Milapochol (g / min) 2.38E-04 3.13F-04

Figures 7.24 and 7.25 show the concentrations of free amine nitrogen (CAA) in beer must substrates and total fermentable carbohydrate (such as glucose), both of which increase when the residence time decreases from 1.8 to 0.9. days. From the mass balance presented in Table 7.4, it can be seen that the rate of depletion of total fermentable carbohydrate (such as glucose) increases while the rate of free amine nitrogen depletion decreases with decreasing residence time in the bioreactor. The productivity factor Y _p / _S of the fermentation product ethanol from the fermentable glucose substrate increases from 0.3 to 0.5 s

307

reducing downtime. Since drought and carbon dioxide are dispersed in the system, there are likely to be minimal losses of ethanol in the gas phase, which would have an impact on the Y _p / _S productivity factor by affecting the ethanol balance. A study conducted in collaboration with Budac and Margaritis (1999) demonstrates that volatile beer flavors, including ethanol, acetaldehyde, ethyl acetate and isoamyl acetate, are established in the main space of the bioreactor with gas lift during continuous fermentation. Chromatography-mass spectroscopy was used for this purpose.

Table 7.4. Mass balance of free amine nitrogen and total fermentable carbohydrate (such as glucose) using the mean of Table 7.3, effect of residence time.

Downtime 1.8 days 0.9 days 1.8 days 0.9 days

Free amine nitrogen Total farm glucose (g / min) (g / min)

Inlet * * 8.84E-04 1.74ΕΌ3 4.71 E-01 9.26E-01 Outlet 3.65E-04 1.50E-03 8,93E-03 3.72E-01 Consumption (AS) 5,18E-04 2.3E-04 4.62E-01 5.54E-01 Odds. of manuf. 0.3 0.5

(Yp / s) _____________________________________________________________________ * The concentration of the inlet is given in Appendix 1.

The concentration of the fermentation product ethanol in the liquid phase decreases as the residence time changes. But the system as a whole produces more ethanol based on the mass flow rate in the shorter residence time of the liquid. Because the purpose of this research is to produce not only ethanol, but

308

beer that has many components in balance while maximizing ethanol productivity in balance with other factors. At the end of the main fermentation of commercial beer, most of the fermentable glucose substrate must be consumed.

Figure 7.26 shows the change in the concentration of acetaldehyde and total diacetyl depending on the change in the flow of beer must. In both analyzes, the concentration at the corresponding flow rate increases as the residence time of the liquid decreases. Acetaldehyde is released from the yeast during the first fermentations of beer during the first days (Kunze, 1996).

Figures 7.25 and 7.27 show the effect of reducing the residence time on the concentrations of higher alcohols, 1-propanol, isobutanol and isoamyl alcohol in the liquid phase in the bioreactor. The concentrations of the three higher alcohols decrease as the residence time in the bioreactor decreases.

The mass flow rates of ethyl acetate and isomyl acetate given in Table 7.3 (b) increase in accordance with the decrease in the residence time of the liquid phase. It can be seen from Figure 7.28 that the concentration of ethyl acetate in the liquid phase decreases while the concentration of isomyl acetate increases. Because this experiment provides an increase in cell growth in the liquid phase without increasing oxygen supply to the system, conditions in the bioreactor contribute to the production of esters. Hough et al. (1982) argue that under conditions of increased cell growth and decreased oxygen, esters are formed.

309 ffl

3.00E + 08 OBE + OE

♦ ♦ ♦ ♦

2.00E + O8

1.60ECU <

R, “In days> and ♦

♦ ♦♦ ♦

100

K m ₽ January 5 «m ^Z A * £ ₂ = 1

1.00E + 086.00E-Y7O.OOE + OO

Rt-Old дни

T ~ r “τΐβ

- + o «m a • e · s

with T

Φ m s §

&

w o 0 z vo o o o

--T ---- 1— I I

2 4 8 8 10 12

Duration of continuous fermentation (days) * Concentration of cells

Cell viability Figure 7.23. Concentration of yeast cells in the liquid phase, depending on the relative duration of continuous fermentation, effect of the residence time of the liquid in the bioreactor. Rt is the residence time of the liquid phase in the bioreactor, expressed in days.

310

Та Ethanol ^a Ferm. glucose

Duration of continuous fermentation (days)

Figure 7.24. Concentration of ethanol and fermentable glucose in the liquid phase, depending on the relative duration of continuous fermentation, effect of the residence time of the liquid in the bioreactor. Rt is the residence time of the liquid phase in the bioreactor, expressed in days.

311

Duration of continuous fermentation (days)

Figure 7.25. Concentration of free amine nitrogen and 1-propanol in the liquid phase depending on the relative duration of continuous fermentation, effect of the residence time of the liquid in the bioreactor. Rt is the residence time of the liquid phase in the bioreactor, expressed in days.

♦ CAA

1-propanol

312

Duration of continuous fermentation (days)

Figure 7.26. Concentration of total diacetyl and acetaldehyde in the liquid phase, depending on the relative duration of continuous fermentation, effect of the residence time of the liquid in the bioreactor. Rt is the residence time of the liquid phase in the bioreactor, expressed in days.

♦ Diacetyl

Acetaldehyde

313 so

c 0 X £ \ 0 o w

L X

X S 3 · CO ο. IX S X ο * <

Β, -Ι.Ι days

-ΓΙΟ ♦♦♦

R, -0.9 DAYS

-r14 ιβ • • 10>

4o q Ο X ο χ q w q s E <o ο co (0 X

K s σ <8 a E § X o * ♦ Isobutanol ¹¹ Isoamyl alcohol

Duration of continuous fermentation (days)

Figure 7.27. Concentration of isobutanol and isoamyl alcohol in the liquid phase depending on the relative duration of continuous fermentation, effect on the residence time of the liquid in the bioreactor. Rt is the residence time of the liquid phase in the bioreactor, expressed in days.

314

Duration of continuous fermentation (days)

Figure 7.28. Concentration of ethyl acetate and isomyl acetate in the liquid phase, depending on the relative duration of continuous fermentation, effect of the residence time of the liquid in the bioreactor. Rt is the residence time of the liquid phase in the bioreactor, expressed in days.

♦ Ethyl acetate

Isoamyl acetate

315

7.2.4. Use of industrially produced decarboxylase alpha-acetolactate to reduce total diacetyl during continuous main beer fermentation

Experiment 1: The bioreactor was contaminated with aerobically developing gram positive cocci prior to completion of the experiment. The bioreactor itself was found to be contaminated because the microbiological testing of the wort applied showed no contamination. This demonstrates the need to improve precautions against contamination of the bioreactor. A decrease in the concentration of diacetyl is observed before the system stops operating when decarboxylase alpha-acetolactate (DKAA) is added to the feed. Unfortunately, it was not possible to draw any conclusions from these data due to the confounding effects due to contamination of the bioreactor.

Experiment 2: As a result of several precautionary measures with respect to the bioreactor, the system operated without contamination during the course of Experiment 2. Data for this experiment are given in Figures 7.29 - 7.31. Table 7.5 summarizes the pseudo-steady-state concentrations of total diacetyl before and after the addition of DKAA to beer must. The concentration of total diacetyl decreased by 47% with the addition of DKAA to beer must, which makes it promising to use this enzyme in the future (average values taken after three turns). As can be seen from Figures 7.30 and 7.31, the total fermentable carbonitrate (such as glucose) and the cell concentration had a slight deviation during this experiment, which could be due to small differences in the wort applied to the bioreactor before and after the addition of SCA.

316

600

600.

400300

200-

100.

Adding DKAA ---- ►

₀ 4 ------------------> ------------------ <---------- ------------------------- 1 ------------------> ----- -------------- 1 ----------------- 0 50 100 150 200 2S0 300 350

Continuous fermentation duration (h)

Figure 7.29. Concentration of total diacetyl in the liquid phase depending on the duration of continuous fermentation, the effect of adding DKAA to the fermentation soybean of casting must. Experiment 2.

7 ·!

And!

I I ή,

6 ·

5 '

43 ----- ►

Adding

SCA

----- ► * Ethanol ^A Farm. glucose, -I II | ·, _Ί | I ......, ............ || II II

100 150 200 250 300 350

Continuous fermentation duration (h)

Figure 7.30. Concentration of fermentable carbohydrate (such as glucose) and ethanol in the liquid phase depending on the duration of continuous fermentation, effect of adding DKAA to the fermentation medium of the must, Experiment 2.

317

1.00E + 10

Concentration of cells in the liquid phase (cells / mL)

1.00E + 09 ♦

1.00E + 08 ·

1.00E + 071.00E + 06 ·

1.00E + 051.00E + 041.00E + 03.

1.00E + 02

1.00E + 01

1.00E + 00 + 0

SO

100

Adding DKAA

------>

-! -------------- 1 ------------- r150 200 250

300

350

Continuous fermentation duration (h)

Figure 7.31. Concentration of cells in the liquid phase depending on the duration of continuous fermentation, effect of the addition of DKAA to the fermentation medium of the must, Experiment 2.

318

In order to avoid possible confusion related to differences in the must during the experiment 3, a large amount of beer must (14 hL) was taken from the brewery and DKAA was added directly to the must remaining in this storage vessel after the initial phase of the pseudo-steady state is reached. This avoids possible differences in the must must which may have an effect on the fermentation process according to Experiment 2. This canister is also provided with a carbon dioxide spray device to maintain it consistently. low dissolved oxygen in the wort. Figures 7.32 to 7.34 illustrate the effect of adding DKAA to the casting must on concentrations of total diacetyl, total fermentable carbohydrate (such as glucose), ethanol and free-suspended cells during continuous fermentation.

Table 7.6 also gives the average concentration of total diacetyl before and after the addition of DKAA to the fed wort (average values taken after three turns).

No contamination was detected throughout the experiment. The concentration of total diacetyl decreased by 45% with the addition of DKAA. No significant differences were observed in the concentrations of ethanol, fermentable carbohydrate (such as glucose) and free-suspended cells, which is consistent with the findings of Aschengreen and Jepsen (1992).

319

500

450

400

350

300

250

200

150

100

♦ ♦ ♦ * ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ • Adding DKAA ------ r '1 -----'!

100

150

200

250

300

350

Duration of continuous fermentation (h) Figure 7.32. Concentration of diacetyl in the liquid phase depending on the duration of continuous fermentation, effect of the addition of DKAA to the fermentation medium of the must, Experiment 3.

320

I do 3-] K. 2 7 · c p 1 £ 6 ' a • 8 · - C. X Oh B- 4-  Ethanol 2 δ i Adding DKAA same Farm. glucose i X to 3 2 me ф ΞΓ 1 ^A A X £ 0 * 4 4 A 4 A A * * ^F _k

60 100 160 200 260 300 350

Continuous fermentation duration (h) Figure 7.33. Concentration of ethanol and total fermentable sugar (such as glucose) in the liquid phase, depending on the duration of continuous fermentation, effect of adding DKAA to the fermentation medium of the must, Experiment 3.

1.00E + 11-D i am i 1.00E + SO 1.00E + 0 » I am X 1.00EY6 f 1.00E + 07 X X b * 1.00E + 06 FN 1.00E + O6 X 1.00E + 04 X X 1.00E + 03 · me Q. 1X f = r X about 1.00E + 02 1.00E + 01 1L0ENHN

Adding DKAA

----- ►

I I · · ''

50 100 160 200 250 300 350

Duration of continuous fermentation (h) Figure 7.34. Concentration of cells in the liquid phase depending on the duration of continuous fermentation, effect of adding DKAA to the fermentation medium of the must, Experiment 3.

321

The results of Experiments 2 and 3 show that DKAA has a significant effect on the concentration of total diacetyl during continuous fermentation in bioreactors with gas lift, helping to reduce the average diacetyl concentration by 46%. This makes it possible to reduce or

elimination of secondary treatment carried out to reduce diacetyl in continuous gas lift systems. In these initial experiments, a relatively high dosage of DKAA is used and this is necessary to optimize the amount, method and timing of dosing in the must, if this enzyme is to be used in the process. Additional savings can be made if an enzyme that is accessible and with higher levels of activity under the conditions of beer fermentation is used, or the enzyme itself is immobilized, thereby ensuring its reuse (Dulieu et al. , 1996). Another consideration will be related to the public acceptance of enzyme supplements that are produced using genetically modified organisms.

The dosage recommended by the supplier is 2 kg / 1OOO hL and at a cost of an industrially obtained enzyme of $ 131.05 / kg, $ 0.26 / hL will be added to the fermentation material costs. The dosage used in the experiments was 72 µg / L (108 ADU / L) or 7.2 kg / 1000 hL DKAA, which represents an additional cost of $ 0.94 / hL. The savings from the use of DKAA for reducing diacetyls during continuous fermentations using gas lift will depend on the optimal dosage of the enzyme under the conditions of the bioreactor and the time saved by its use.

322

Table 7.5 Effect of the addition of DKAA under pseudo-steady-state conditions of the fermentation medium from the must on the concentration of total diacetyl during the continuous fermentation of beer in a bioreactor with gas lift

Experiment Average overall concentration Percentage of diacetyl (pg / L) decrease in (DKAA-free) (DKAA, 60 pg / L) diacetyl Experiment II 495 260 47

Experiment III 445 245 45 * pseudo-steady-state averages after three revolutions.

The above supports the contention that continuous fermentation using immobilized yeast and associated free cells in a suction tube bioreactor and using gas lift is a possible alternative to periodic beer fermentation based on the following criteria:

- attainment of appropriate taste

- higher volume capacity of the bioreactor

- minimal complications

- demonstration of continuous work over a long period of time

- regulation of air (oxygen) in the fluidizing gas to regulate the taste

- addition of α-acetolactate decarboxylase enzyme to control diacetyl content

- no bacterial contamination

- economic advantages.

323

There are still issues that require further research, but the technology is ready for larger scale testing. Gas lift bioreactors are already being used on an industrial scale for wastewater treatment, which increases the prospects for the technical feasibility of continuous fermentation systems. Brewery Grolsch in Netherlands reported the use of a bioreactor with a gas lift and volume 230 t ³ for the treatment of wastewater from the brewery (Driessen et al., 1997). One of the biggest obstacles to utilizing continuous fermentation on an industrial scale in the brewing industry may be the adoption of the new process by brewers in an industry that is deeply attached to tradition.

This study gathers data on secondary yeast metabolites obtained during continuous fermentation, which was done to highlight the importance of regulating oxygen in the fluidizing gas to shape the taste of the beer. The findings show that, under certain operating conditions, an increase in the fluid in the fluidised gas in the bioreactor results in an increase in acetaldehyde, diacetyl and higher alcohols (isoamyl alcohol and isobutanol), while ester concentrations (isoamyl acetate, ethylhexanoate and ethyl hexanoate, ethyloctanoate, and ethylacetate. These data confirm that it is possible to control the taste of beer by the composition of the fluidizing gas in the bioreactor, which allows the production of unique products.

Unless the presence of air in the fluidizing gas is eliminated, the liquid phase of the bioreactor adjusts the concentration of free-suspended cells to be greater than 10 ⁸ cells / mL. This way the system has more than

324 one population of yeast cells coexisting in the bioreactor - immobilized yeast and suspended in the liquid phase of yeast. Due to the high growth rate of viable cells in the liquid phase of the bioreactor, it is possible to use the continuous process in the bioreactor as a yeast cultivator.

When the secondary 48-hour batch retention period is added to the continuous main fermentation, a flavor profile is provided within the commercial beer range. The temperature of this retention period is 21 ° C and the importance of minimally exposing the liquid to oxygen during the retention period on the taste formation has been demonstrated experimentally. The retention period adds two days to the process, as well as additional complications, but it is significantly faster than the commercially used periodic ferments, which last between seven and fourteen days.

Ultimately, the ideal case is to eliminate the secondary retention period by optimizing the conditions during the first continuous process in the bioreactor. However, further reduction of the secondary retention time can be achieved in a short period by optimizing the retention temperature (diacetyl removal by yeast is highly temperature dependent), the amount of fermentable sugars remaining in the liquid at the beginning of the retention period, the concentration of yeast present, the hydrodynamic characteristics of the retention vessel (diacetyl removal can be improved by improving the contact between yeast and beer) and by taking other measures for eliminating oxygen from this stage.

325

Other researchers (Kronlof and Virkajarvi, 1996; Nakanishi et al., 1993; Yamauchi et al., 1995) have focused on the study of multi-stage continuous fermentations, in which the first stage of continuous fermentation (aerobic) results in only partial consumption of of fermentable sugars present in the must. While this approach has shown some success in taste, these systems are complex. The first aerobic step in such systems creates an environment that is more susceptible to microbial contamination (ie, high concentrations of sugars, temperature and oxygen at low ethanol concentrations). In the gas lift bioreactor used in this research, there is a low concentration of fermentable sugars in a pseudo-steady state, a low pH value, a high ethanol concentration, and a low oxygen concentration, making the environment unfavorable to possible contamination. For smaller breweries, minimal complications and the development of a stable and resilient pollution process is an important success factor.

The addition of industrially produced decarboxylase alfaacetolactate (DKAA) to the wort fed for continuous fermentation shows an average diacetyl reduction of 46%. But since DKAA is an enzyme derived from a genetically modified organism, there is a public perception that research is needed before using such an enzyme in a commercial product. In addition, the commercially available diacetyl regulatory enzymes do not have optimal fermentation activity.

In more than six months of continuous fermentation using cap-carrageenan to immobilize cells, more than 90% of the free-suspended cells in the liquid phase

326

remain viable while the viability of the immobilized cells decreases to less than 60%. Scanning electron microscope images show that cells located near the periphery of the gel pellets have multiple pimple marks and correct morphology, while those close to the pellet core have irregular shape and smaller pimple marks, demonstrating a decrease in growth. The micrographs also show that the yeast, which is located near the core of the granules, shows signs of aging. As discussed in Chapter 5, kappa carrageenan has many characteristics that make it desirable for use as an immobilization matrix. However, there is currently no industrial method for granule production, as yeast retention in the matrix is part of the granule production process and the handling of the granules in an industrial installation increases complications and costs. Other methods of immobilization such as self-aggregation or flocculation could be explored in the future. This would eliminate the complications of handling the granules under plant conditions and if the yeast flocculates

disrupted regularly, regular removal of adult cells from the bioreactor may be ensured.

Claims (15)

1. An alcoholic beverage production method using an in-house bioreactor to carry out a continuous fermentation step in which beer must containing fermentable sugars is initially fermented using self-assembled flocculent yeast cells and fed a limited amount oxygen, after which at least the partially fermented product, which is released from the continuous process, is fed to a periodic processing step. Method according to claim 1, characterized in that the alcoholic beverage is beer. Method according to claim 2, characterized in that the beer is a light beer. The method of claim 2, wherein the beer is a light beer. 5. A method according to claim 2, wherein the beer is a North American beer. The method according to claim 1, characterized in that the yeast is high-flocculating or super-flocculating yeast. A method according to claim 1, characterized in that said fermentation is the main fermentation. A method according to claim 1, characterized in that, before fermentation, the wort is purged with carbon dioxide. 328 9. The method of claim 1, wherein carbon dioxide containing about 3% oxygen is sprayed into the reactor. A method according to claim 1, characterized in that the completion of the main fermentation is carried out during the periodic retention step. 11. The method of claim 10, wherein the concentration of diacetyl is reduced. 12. The method of claim 10, wherein the concentration of acetaldehyde is reduced. 13. The method of claim 10, wherein the concentration of higher fusel alcohols remains substantially unchanged during the periodic process of retention. A method according to claim 1, characterized in that the product, which is discharged from the continuous process, is distributed through a pipeline with branches to one of the many vessels for holding the product in batches in which the periodic retention process is carried out. A method according to claim 1, characterized in that the continuous process is a process for achieving a certain degree of fermentation for carrying out subsequent periodic fermentations in the periodic stage of retention.