Immobilised systems

The decline in interest in continuous fermentation during the early 1970s was followed by a period in which the focus of developments in fermentation technology was directed towards increasing volume output by the use of ever-larger batch sizes. These efforts culminated in the relatively sophisticated high-capacity vessels, described already in previous sections of this chapter. Fermenter productivity has been increased by the introduction of procedures such as high-gravity brewing (see Section 2.5). Some time savings have been made by the use of combined fermentation and maturation vessels (Section 5.4.3). However, these have been relatively modest and not of the order that was expected of continuous fermentation.

During the late 1970s through to the present, there has been a renaissance of interest in continuous brewing processes. This is because of the promise that many of the shortcomings of the early systems could be overcome by the use of relatively small vessels containing immobilised yeast cells. Such reactors have already found application at commercial scale for continuous flavour maturation and for the production of low-alcohol and alcohol-free beers by limited fermentation. As yet they have not been used for primary fermentation at production scale although several pilot scale systems have already been developed and commercial application in the near future would seem likely. Coupling of primary fermentation and warm maturation into a single continuous process catalysed entirely by immobilised yeast is of course feasible, providing the other upstream and downstream processes are capable of supporting the fermenters. In this case, it has been suggested that eventually it may be possible to produce beer within one day (Atkinson & Taidi, 1995)!

The major perceived advantages of the use of immobilised yeast reactors are:

(1) very rapid process times because of the high yeast concentration present in the reactor coupled with the ability to use rapid flow rates that would cause washout in open continuous systems;

(2) high efficiency of conversion of sugar to ethanol formation by restricting the extent of yeast growth (when used for primary fermentation);

(3) relatively small reactors giving large-volume continuous throughputs with a minimum of down-time;

(4) reduced risk of microbial failure because potential contaminants are unable to compete successfully with the existing yeast;

(5) rates of metabolism of immobilised yeast cells may be more rapid than free cells;

(6) no restriction on the choice of yeast strain;

(7) reduced requirement for yeast propagation;

(8) simplified yeast handling;

(9) simplified separation of product from yeast;

(10) improved product consistency;

(11) possibility of modular design allows improved flexibility compared to conventional continuous systems.

There are some disadvantages, particularly with respect to application to primary fermentation. It is advisable to use bright wort to avoid the possibility of clogging of support beds. In most breweries, this will require some form of pre-filtration. Brew-houses must be capable of a constant supply of wort, otherwise, as with conventional continuous fermentation, much of the volume productivity gain is lost. There is a requirement to store wort under sterile conditions. The large volumes of carbon dioxide evolved during primary fermentation may cause disruption of some mechanically fragile support media and excessive gas breakout can reduce the efficiency of mass transfer of solutes in packed beds. More significantly, as stated already, yeast growth is not a prerequisite of immobilised systems. This is an advantage with respect to efficiency of conversion of sugar to ethanol. However, the concomitant reduced uptake of wort amino acids results in beers which contain low levels of esters, and are therefore poorly matched to those produced by conventional batch fermentation. As will be described subsequently, one approach to circumventing this problem is the use of multiple tanks and a combination of free and immobilised yeast. In this instance, the supposed advantage of simplicity is perhaps brought into question.

Application of immobilised yeast reactors for primary fermentation within existing breweries at the expense of conventional fermenters and associated plant would be difficult to justify in terms of cost. However, it could be given serious consideration where the need arose for additional fermentation capacity, although obviously great care would need to be taken with respect to product matching. A method has been proposed whereby existing conventional fermenters could be modified for use as immobilised reactors by fitting retaining screens and using yeast entrapped in calcium alginate beads (Hsu & Bernstein, 1985). However, using existing brewery plant for duties for which it was not designed must at best be considered an unsatisfactory compromise.

Continuous warm maturation is an ideal application for immobilised yeast since it does not require yeast growth, involves little carbon dioxide evolution and can be fitted as an adjunct to conventional batch fermenters. At present, therefore, this application is a much more attractive proposition for fitting into existing breweries. For the same reasons production of low- or zero-alcohol beers by limited fermen tation is suited to an immobilised process. In the long term, use of immobilised reactors for combined continuous primary fermentation and maturation seems destined only to be considered seriously where entirely new breweries are being planned.

5.7.1 Theoretical aspects

Radovich (1985) defined immobilisation as 'physical confinement or localisation of microorganisms in a way that permits economic re-use'. McMurrough (1995) gave a more specific definition, 'cells physically confined or localised in a certain defined region of space with retention of their catalytic activity, if possible or even necessary their viability, which can be used repeatedly and continuously'. Thus, using the classification described in Section 5.6.2, immobilised yeast reactors would be viewed as totally closed systems where the process fluid, whether it be beer or wort, makes transient contact with actively metabolising but retained yeast cells. Unlike batch systems or conventional continuous fermenters, it is not essential that any cellular growth occurs within an immobilised reactor, although growth is not precluded. In this sense, the yeast may be viewed simply as a 'biocatalyst'.

In theory, compared to a conventional continuous fermenter much greater volumetric productivity may be achieved using an immobilised yeast bioreactor. With a completely open continuous fermenter, as in a chemostat, the maximum achievable flow rate (critical dilution rate) is governed by the maximum yeast growth rate which may be obtained under the particular selected operating conditions. Use of a dilution rate greater than the critical value results in washout of the yeast culture since the growth rate becomes less than the rate at which fresh medium is added (Section 5.6.1). In the case of an immobilised system, the yeast is retained and cannot wash out and therefore, dilution rates greater than the critical value may be used. In practice, the volumetric productivity is limited by the biomass loading in the reactor and the mass transfer of nutrients and products from the liquid phase to the yeast. These parameters are dependent on the method of immobilisation and the design and operation of the reactor, as will be discussed subsequently. Methods of immobilisation. Five methods for immobilising cells are recognised (Godia et al., 1987; Scott, 1987; McMurrough, 1995). These are floccu-lation, entrapment within a polymeric matrix, adhesion to a solid surface, colonisation of porous materials and retention behind membranes.

The simplest method of immobilisation is to take advantage of the natural floc-culent character of some yeast strains. This enables retention by sedimentation against an upward flow of process fluid. This is the operating principle of the continuous tower fermenter, described already (Section It has the advantage that there is no requirement for an immobilising support medium. Disadvantages are that the choice of yeast strain is restricted, flow rates must be modest to avoid washout and evolved carbon dioxide may disrupt floes.

Entrapment of cells in a porous matrix has been probably the most widely reported method of immobilisation. The matrix material must be non-toxic and retain entrapped cells but be sufficiently porous to allow passage of nutrients and metabolites. The materials which have been used include calcium alginate, K-carrageenan, polyacrylamide, agarose, pectin, gelatin, chitin and locust bean gum (Godia et al., 1987). Many of these, for example, calcium alginate, K-carrageenan and chitin, are approved for food use. The carriers may conveniently and inexpensively be formed into spherical beads, which can contain very high loadings of biomass. The low cost removes the need for the beads to be regenerable.

Mass transfer of substrates and metabolites between the beads and the process fluid is of critical importance to the efficiency of any immobilised cell reactor (Radovich, 1985). It is of particular importance with entrapped yeast supports since reactants and products must traverse through the matrix. Thus, mass transfer is dependent on diffusion gradients and these are influenced primarily by the bead size, the density of the matrix, the biomass loading, the bead concentration and the velocity of fluid flow over the bead surface. In this respect beads with a diameter of 0.2-3 mm offer a suitable compromise between level of biomass loading and diffusional path length. Unfortunately, most types of entrapment bead have low mechanical strength. They may be disrupted by evolution of gaseous carbon dioxide, and cell outgrowth may be considerable. Furthermore, they are compressible and, consequently, reduced flow rates due to compaction can be a problem with some reactor types. On the other hand they can be of roughly the same density as the suspending fluid which can make for easier mixing. The mechanical strength of the beads can be increased by using a greater degree of cross-linking in the gel or by the use of more robust surface layers. Of course, this is at the expense of diffusivity.

Cell immobilisation by attachment to surfaces is proving to be one of the most effective approaches for brewing applications at commercial scale since the available supports offer both mechanical strength and reasonable levels of biomass loading. In addition, diffusion between cells and process fluid is obviously less restricted than with entrapment systems. A wide range of supports have been used. These include wood chips, ceramics, glass, cotton fibres, diatomaceous earth, stainless steel, various resins, polyvinyl chloride and cellulose (Anselme & Tedder, 1987; Godia et al., 1987; Ryder et al., 1995).

The physiological basis of adhesion to inert surfaces is unclear but would appear to involve a combination of electrostatic and hydrophobic effects (Mozes et al., 1987). Yeast immobilised in this way may be by combinations of attachment to inert surfaces and cell-to-cell flocculation. These interactions are dependent on the nature of the surface, the state and type of cell and the conditions in the surrounding environment. Thus, the material and cell surfaces both have inherent charges and varying extents of hydrophobicity. These will also be influenced by environmental parameters such as pH, osmolarity, ionic composition and the presence of other solutes, which may either promote or block adhesion. Surfaces may be subject to chemical modification to promote adhesion, for example, addition of positively charged ligands. With regard to yeast, parameters such as starvation, mean cell age, cell size and cell concentration have effects on cellular morphology. These might be expected to influence adhesion and flocculation (Wood et al., 1992; Smart et al., 1995; Barker & Smart, 1996; Rhymes & Smart, 1996).

Attachment of cells to surfaces via covalent bonding is possible but has not been used widely since many of the procedures are cytotoxic. There have been some reports, for example, of use of glutaraldehyde cross-linked gels that contain covalently bound yeast cells coated onto glass beads (Godia et al., 1987). These have been tested for the production of fuel alcohol but not for the production of alcoholic beverages.

For brewing applications, DEAE-cellulose has proven to be the favoured support. It is available in commercial quantities supplied under the trade name Spezyme® (Cultur, Finland) and described as granulated derivatised cellulose (GDC). The material has a positive charge and binds negatively charged yeast cells in a monolayer. The material is robust and has a rough surface, which provides some protection against accidental dislodgement of cells; however, excessive shear forces can result in some shedding (Norton & D'Amore, 1994). The material is FDA approved, 0.40.8 mm granular size and regenerable by treatment with a 2% w/v sodium hydroxide solution at 80°C (Pajunen, 1995). In the same report, the yeast loading was given as being up to 500 x 106 cells per gram wet weight of carrier.

The fourth method of immobilisation, colonisation of porous materials, is another form of entrapment. The materials are usually provided with comparatively small surface pores and larger internal cavities. Single cells may bind to the surface of the carrier but the bulk enters the central matrix via the pores and grows in the internal cavities forming retained aggregates. Immobilisation is via a combination of surface adsorption and cell-to-cell aggregation. Most supports of this type are robust, regenerable, non-compressible, and, since most of the immobilised cells are protected within the internal matrix, losses due to shearing are minimal. For these reasons they may be used in reactors where there is a relatively high degree of agitation, to promote good rates of mass transfer. With some supports care must be taken to avoid excessive abrasion damage.

Biomass loadings are high and an open pore structure allows egress of carbon dioxide with minimal floe disruption. As with gel entrapment methods, solute mass transfer is restricted compared to surface adhesion. Norton and D'Amore (1994) concluded that with porous beads only a relatively shallow band of biomass, extending to a depth of 140 (im, took an active role in immobilised reactors. Similarly, Inoue (1988) argued, in this case with alginate beads although the point is the same, that cells in the core region exhibited limited metabolic activity because of oxygen starvation and high local ethanol concentrations. For this reason, it is important to use beads of an appropriate diameter to maximise the available surface area.

Porous supports which have been recommended for brewing applications include artificial sponges (Scott & O'Reilly, 1995), glass beads (Breitenbucher & Mistier, 1995; Hyttinen et al., 1995; Yamauchi & Kashihara, 1995) and ceramic rods with silicon carbide matrices (Krikilion et al., 1995). Of these, the latter two have been applied at commercial scale, and therefore received the most attention. Porous glass beads are sold under the trade name SIRAN®, produced by the Schott Engineering Company (Mainz, Germany). They are prepared from a mixture of glass powder and salt such that during the process of bead formation the salt is dissolved and removed leaving pores in the glass of 60-300 (im diameter. Beads of 2-3 mm diameter are recommended for fixed bed reactors and 1-2 mm diameter for fluidised bed types. Each bead contains approximately 55-60% pore volume. Yeast loading levels are reported to be in the region of 15 x 106 cells/g bead (Breitenbucher & Mistier, 1995). The beads are regenerable by treatment with hydrogen peroxide and steam sterilisation. Cleaning is more difficult than surface carriers because of the need to remove all organic matter from the matrix.

Reactors employing yeast entrapped within silicon carbide have been described in several recent publications (for example, van de Winkel et al. (1991, 1993, 1995), Krikilion et al. (1995)). The support consists of a ceramic cylinder, which contains a number of channels through which the process fluid traverses. The channels are embedded in a matrix of silicon carbide which has surface pores of roughly 8-30 (im diameter and internal pores of 100-150 (im diameter, depending on the grade used. Void volumes of up to 60% are achievable. This arrangement allows entrance of single cells via the small surface pores and provides space for extensive internal colonisation and yeast retention. The ceramic elements can be arranged in groups through which process fluid is circulated. Unlike fixed or fluidised reactors, there is little restriction to flow with these channelled elements, and therefore mass transfer efficiencies are high and there is no possibility of bed compression or abrasion damage. The elements are robust and the arrangement of channels allows easy egress of carbon dioxide. The elements may be regenerated and cleaned by successive forward and backward flushing with hot sodium hydroxide solution, detergent and hot water. This is followed by sterilisation with either steam or cold temperature treatment with a mixture of peracetic acid and hydrogen peroxide.

The fifth method of immobilisation is the use of a semi-permeable membrane that retains cells but allows exchange of soluble metabolites and gases. They may be used in many configurations. Cells may be attached to the membrane or freely suspended behind it. The former approach is commonly used in the special application of biosensors, the latter in membrane bioreactors (Scott, 1987). Many configurations of membrane reactor have been suggested, for example, membrane re-cycle, hollow fibre, dialysis membrane (Cheryan & Mehaia, 1984; Park & Kim, 1985; Godia et al., 1987). Few have progressed beyond the laboratory scale. They have inherent disadvantages of restricted mass transfer, a problem that may be exacerbated by a tendency to clog. Application at commercial scale for brewing seems most improbable.

A comparison of the methods of yeast immobilisation, together with examples pertinent to brewing applications is given in Table 5.7. No attempt has been made to quantify the differences in biomass loading achievable by each type of support since no single unit is used in the literature and comparisons are difficult. In any case, direct comparisons are of little value since they do not entirely reflect the volumetric productivity of which each type of support is capable. Thus, this parameter is influenced by not only the type of carrier but also the design of the bioreactor and the conditions under which it is operated. From a practical standpoint, the gel encapsulation supports seem unlikely to be applied at commercial scale for brewing applications. Of the other supports described in Table 5.7, both the DEAE-cellulose and glass bead supports have advantages over gel entrapment methods and both are clearly suitable for application at commercial scale. The silicon carbide approach looks particularly promising, having a good combination of high biomass loading, reasonable rates of mass transfer and unrestricted flow. Effects of immobilisation on yeast physiology. There is now a considerable body of evidence suggesting that the physiology of immobilised yeast differs from that

Table 5.7 Methods of yeast immobilisation used in brewing applications, advantages and disadvantages.










No requirement for support medium

Inexpensive supports High biomass loading Usable with any yeast strain

High mass transfer rates Robust



Not affected by COz

Restricted use of yeast strain Floe disruption by carbon dioxide evolution Wash-out at high flow rates

Restricted mass transfer Restricted yeast growth Potential cell outgrowth Fragile

Disruption by C02 Compression of reactor beds Non-regenerable


Cells may shear off at high flow rates Reduced biomass loading cf. other methods


Chitosan beads K-carrageenan

K-carrageenan Calcium alginate

Calcium alginate Calcium alginate

Diatomaceous earth DEAE-cellulose



DEAE-cellulose DEAE-cellulose

Primary fermentation

Primary fermentation (lab. scale)

Diacetyl removal using immobilised Bacillus polymixa (laboratory scale) Primary fermentation (pilot scale) Primary fermentation

(laboratory scale) Continuous alcohol-free beer process (pilot scale) Primary fermentation (laboratory scale)

Primary fermentation

(production scale) Continuous maturation

(production scale) Continuous wort acidification using immobilised lactic acid bacteria Continuous maturation and alcohol-free beer production Continuous alcohol-free beer production Continuous maturation

Seddon (1975)

Shindo et al. (1994) Willetts (1988)

Mensour et al. (1995)

White & Portno (1978)

Dziondziak & Seiffert

Narziss & Hellich

(1972) Pajunen (1995)

Pittner et al. (1993)

Mieth (1995)

van Dieren (1995) Nothaft (1995)

tu JO

Brew Your Own Beer

Brew Your Own Beer

Discover How To Become Your Own Brew Master, With Brew Your Own Beer. It takes more than a recipe to make a great beer. Just using the right ingredients doesn't mean your beer will taste like it was meant to. Most of the time it’s the way a beer is made and served that makes it either an exceptional beer or one that gets dumped into the nearest flower pot.

Get My Free Ebook

Post a comment