Continuos Fermentation PDF
Continuos Fermentation PDF
Continuos Fermentation PDF
Storgrds, Erna. Process hygiene control in beer production and dispensing. 2000. 105 p. +
app. 66 p.
411
Kivist-Rahnasto, Jouni. Machine safety design. An approach fulfilling European safety requirements. 2000. 99 p. + app. 9 p.
412
414
Valmari, Tuomas. Potassium behaviour during combustion of wood incirculating fluidised bedpower plants. 2000. 88 p. + app. 75 p.
415
Mkel, Kari. Development of techniques for electrochemical studies in power plant environments. 2000. 46 p. + app. 128 p.
416
417
418
Seppnen, Veikko. Competence change in contract R&D. Analysis of project nets. 2000.
226 p. + app. 29 p.
419
Vaari, Jukka & Hietaniemi, Jukka. Smoke ventilation in operational fire fighting. Part 2. Multistorey buildings. 2000. 45 p. + app. 18 p.
420
Forsn, Holger & Tarvainen, Veikko. Accuracy and functionality of hand held wood moisture
content meters. 2000. 79 p. + app. 17 p.
421
Mikkanen, Pirita. Fly ash particle formation in kraft recovery boilers. 2000. 69 p. + app. 116 p.
422
Pyy, Pekka. Human reliability analysis methods for probabilistic safety assessment. 2000. 63 p.
+ app. 64 p.
423
Saloniemi, Heini. Electrodeposition of PbS, PbSe and PbTe thin films. 2000. 82 p. + app. 53 p.
424
Alanen, Raili. Analysis of electrical energy consumption and neural network estimation and forecasting of loads in a paper mill. 2000. 119 p. + app. 17 p.
425
Hepola, Jouko. Sulfur transformations in catalytic hot-gas cleaning of gasification gas. 2000.
54 p. + app. 80 p.
426
427
Dobrica, Liliana & Niemel, Eila. A strategy for analysing product line software architectures.
2000. 124 p.
428
Safety management systems Audit tools and reliability of auditing. 2000. 175 p. + app. 48 p.
429
Olin, Markus, Pasanen, Antti, Ojaniemi, Ulla, Mroueh, Ulla-Maija, Walavaara, Marko, Larjava,
Kari, Vasara, Petri, Lobbas, Pia & Lillandt, Katja. Implementation of IPPC-directive. Development of a new generation methodology and a case study in pulp industry. 2000. 81 p.
430
Virkajrvi, Ilkka. Feasibility of continuous main fermentation of beer using immobilized yeast.
2001. 87 p. + app. 50 p.
VTT INFORMATIONSTJNST
PB 2000
02044 VTT
Tel. (09) 456 4404
Fax (09) 456 4374
Ilkka Virkajrvi
Tt julkaisua myy
VTT TIETOPALVELU
PL 2000
02044 VTT
Puh. (09) 456 4404
Faksi (09) 456 4374
410
VTT PUBLICATIONS
V T T
P U B L I C A T I O N S
Ilkka Virkajrvi
P
P
ESPOO 2001
Cover picture: SEM micrograph by Paula Raivio, VTT Building and Transport
Virkajrvi, Ilkka. Feasibility of continuous main fermentation of beer using immobilized yeast.
Espoo 2001. Technical Research Centre of Finland, VTT Publications 430. 87 p. + app. 50 p.
Keywords
Abstract
Fermentation is the most time consuming step in the production of beer and
therefore the effective use of fermentation vessels is a crucial element in
brewing economy. One means of increasing the productivity of a batch process
is to convert it to a continuous one. Experiments in continuous fermentation
emerged during the 1950s and 1960s, but by the end of 1970s most of them had
been closed down. Immobilization technique revitalised continuous fermentation
research in the 1980s and led to industrial applications in the secondary
fermentation and in the production of low-alcohol beers.
This work demonstrated that an immobilized, continuous main fermentation is a
feasible process for production of lager beer. The immobilized main fermentation was stable for more than 14 months both in fermentation efficiency
and in aroma compound formation. The formation of aroma compounds could
be controlled by varying the composition and amount of gas feed into the first
fermentation stage. The division of immobilized main fermentation into an
aerobic and an anaerobic stage appeared to solve problems related to yeast
growth and viability.
The carrier material affected the formation of flavour compounds in small-scale
fermentations. Moreover the effect varied with the yeast strain used. The carrier
affected the economy of immobilized fermentation: the carrier cost could be as
high as one third of the investment. When a cheap carrier is used the investment
cost for a continuous, immobilized process was estimated to be only about 70%
of the investment cost of a batch process.
Preface
This work was carried out at VTT Biotechnology during the years 19952000.
The work formed a part of a wider project of the Finnish malting and brewing
industry aiming at continuous production of beer using immobilized yeast.
Financial support was provided by the Finnish malting and brewing industry and
Tekes, the National Technology Agency, which is gratefully acknowledged.
I am very grateful to Prof. Matti Linko, the former Laboratory Director, for his
encouragement to write papers and this thesis. I also thank the present Research
Director, Prof. Juha Ahvenainen, for providing excellent working facilities and
the possibility to finalise this work. Prof. Katrina Nordstrm provided many
valuable comments during the writing phase. Prof. Timo Korpela and Doc.
Pekka Reinikainen I thank for the critical reading of the manuscript.
My special thanks go to my co-authors. Jukka Kronlf, PhD and Esko Pajunen,
MSc provided valuable viewpoints from practical brewing. Nana Lahtinen, MSc
(ne Pohjala), Katri Lindborg, MSc and Terhi Vauhkonen, MSc I thank for their
pleasant co-operation, keen attitude and interesting discussions. Erna Storgrds,
PhD I thank for continuously reminding that there exist more microorganisms
than brewer's yeast even in a brewery. To Silja Home, Dr. Sci. (Tech.) go my
special thanks for her encouragement, continuous pressure to improve my
writing and valuable discussions.
This work would not have been possible without the pleasant and occasionally
humorous environment generated by all my colleagues at VTT Biotechnology,
especially those of the former Brewery group. Hannele Virtanen MSc, Airi
Hyrks, Marita Ikonen, Kari Lepist, Arvi Vilpola and especially Eero Mattila
have all helped throughout the years.
Lastly, I express my warmest thanks to Seija, Jussi and Juuso for letting me pile
papers at home and for understanding my occasional absent-mindedness.
II
III
IV
Virkajrvi, I. and Pohjala, N. 2000. Primary fermentation with immobilized yeast: some effects of carrier materials on the flavour of the
beer. J. Inst. Brew., Vol. 106, No. 5, pp. 311318.
VI
Contents
Abstract................................................................................................................. 3
Preface .................................................................................................................. 5
List of the original publications ............................................................................ 6
Abbreviations and terms ....................................................................................... 9
1. Introduction................................................................................................... 11
2. Beer fermentation and the brewing industry................................................. 12
2.1 Beer fermentation ................................................................................ 12
2.2 Beer flavour ......................................................................................... 13
2.3 Brewing industry ................................................................................. 14
2.4 Productivity of beer fermentation........................................................ 16
3. Continuous beer fermentation....................................................................... 18
3.1 Continuous, non-immobilized processes............................................. 18
3.1.1 Early attempts ......................................................................... 19
3.1.2 Large scale attempts................................................................ 20
3.1.3 Reasons for failure of continuous, non-immobilized
fermentations........................................................................... 24
3.2 Continuous, immobilized processes .................................................... 28
3.2.1 The Bio-Brew Bioreactor........................................................ 31
3.2.2 The continuous main fermentation system developed by Baker
and Kirsop ............................................................................... 33
3.2.3 Alginate as carrier for brewing................................................ 34
3.2.4 Use of immobilized yeast in batch fermentation..................... 34
3.2.5 Process development at Kirin Brewery Company Ltd., Japan 35
3.2.6 Process development at Labatt Breweries of Canada ............. 37
3.2.7 Process development at Meura Delta, Belgium ...................... 38
3.2.8 Process development at Sapporo Breweries Ltd., Japan ......... 40
3.2.9 Semi-industrial main fermentation at Hartwall Plc, Finland... 40
3.2.10 Some other relevant experiments in immobilized main
fermentation ............................................................................ 41
3.2.11 Flavour and economy.............................................................. 43
DEAE
diamino diethyl
DMS
dimethyl sulphide
FAN
GDC
HFCS
hl
PBR
VDK
vicinal diketones
Green beer
beer after primary fermentation, which usually has high concentrations
of vicinal diketones, also called young beer
Main fermentation
the first fermentation step in the production of lager beer, also called
primary fermentation; most of the flavour compounds are formed in
main fermentation
Primary fermentation
main fermentation
Secondary fermentation
the second fermentation step in the production of lager beer, also called
lagering and maturation; the main purpose is to remove buttery
off-flavour and its precursors to an acceptable level
Vicinal diketones
diacetyl and 2,3-pentanedione, which are responsible for buttery
off-flavour. This flavour may be essential in some ales, as it is in some
red vines.
10
1. Introduction
Brewing has changed from home brewing into very large-scale manufacturing.
Most beer is brewed by large companies, but on the other hand, the number of
small or very small breweries has increased in recent years. The competitive
factor for these small breweries is not price, but beers that differ from
mainstream beers. In recent years the tendency for globalisation has been
evident, with brewing companies acquiring breweries all over the beer drinking
world. This will make the market very competitive. Beer is a consumer product
for which the image is very important. Another important factor in consumer
products is the price. Only with a very efficient production and distribution chain
can the brewery make this factor work its advantage. However, every step taken
in cost reduction must preserve the flavour of the product.
Christensen (1997) showed that sometimes the market-leading, profitable and
well-managed companies fail to see and react to a change that will eventually
lead to loss of market dominance. Typically this change originally has a lower
performance than existing technologies, but its faster rate of improvement
rapidly changes the situation. Products of new technologies have features that
customers value: cheaper, simpler, smaller and more convenient to use.
Christensen (1997) called this event disruptive technological change. Examples
in his book include companies in hard disk manufacturing, steel making, and
excavator manufacturing, and thus the idea of disruptive technological change
appears to be rather universal and may be extendable to the brewing industry.
Process biotechnology has advanced enormously by science-oriented and
innovative brewmasters. They have published a great number of papers and
patents and established even large-scale continuous fermentation units.
However, by the late 1970s almost all of the continuous fermentation units had
disappeared, leaving a bitter aftertaste of disappointment.
The present study deals with the challenges of continuous processes for brewing.
It attacks the problems encountered in continuous main fermentation of beer and
uses immobilization technology to solve these problems.
11
12
The lager beer fermentation is divided into two phases: the main fermentation
and the secondary fermentation. The main fermentation lasts from 6 to 10 days
and temperatures used are between 7 and 15oC. Temperature profiles may also
be used. During the main fermentation most of the flavour compounds are
formed. At the end of the main fermentation the beer is cooled down to about
4oC and most of the yeast is separated from the beer. The secondary
fermentation can be performed in the same vessel as the main fermentation or
the beer can be transferred into a second vessel. The main objective of the
secondary fermentation is to remove diacetyl, which causes an off-flavour in
lager beer. The secondary fermentation normally lasts between one and two
weeks, but even six-week times have been reported by brewing companies. The
latter time was common before the lagering temperature was increased from 0oC
to 1015oC (Linko and Enari 1967).
Finally, the beer is stabilised by cooling it down to 0oC or even lower and by
maintaing this temperature for up to 3 days. Different stabilising agents (i.e.,
silica gels, polyvinylpolypyrrolines, tannins) may be used. Yeast and proteinpolyphenol complexes precipitate and these are then filtered off. Carbonating
and packaging ends the production.
13
14
1936
1947
1958
1967
1975
1982
1989
Number of
brewing
companies
Number of
brewing
plants
Million litres
sold
Million
litres/plant
750
404
211
124
57
35
26
750
465
252
170
120
87
65
5 300
10 300
10 400
13 700
18 500
22 800
23 200
7
22
41
81
154
262
357
The same tendency towards larger units/companies is valid today all over the
world. In the UK, consolidation has been very strong and already in 1991 six
companies produced 75% of the total amount of beer (Boulton 1991). Since then
consolidation has continued in the U.K., elsewhere in Europe and in South
America (Anon. 1999, Anon 2000a, Anon. 2000b, Anon. 2000c). Changes have
also taken place in the Nordic countries. But brewing is and will be from now on
15
maximum
productivity
average
productivity
Filling
Fermentation
ethanol
16
17
times.
Very similar advantages of continuous fermentation were reported by Steward
(1974) and Smith (1991). Additional costs of stirred continuous fermentation
included electrical stirring and cooling costs (Ricketts 1971). A 30% reduction
in hop rates, reduction of beer losses from 1.5% to 0.7% and marginal savings in
labour and detergent cleaning in the production of ale were reported (Seddon
18
1976). The fermentation time was reduced from three days to 4 hours and warm
conditioning from three days to zero (Seddon 1976).
The economic advantages of continuous systems were claimed to be such that it
is "inevitable that the future of the British fermentation systems lies in vertical
conical-bottom vessels and continuous processes" (Ricketts 1971). Two major
factors influencing the decision of New Zealand Breweries Limited to employ a
continuous fermentation system in early the 1950s were the restrictive
Government-imposed building regulations and excise paid by pitched wort (so
the brewing company paid taxes on the amount of beer in processing) (Davies
1988). Other factors included higher vessel utilisation and lower costs.
Fortuitous coincidences occurred: the need for brewery expansion at the moment
when the new technology was there and the ability to fabricate steel (Kennedy
1996).
In the following section early attempts at continuous non-immobilized
fermentation will be described. The next section will describe a number of larger
scale experiments, through which the advantages and disadvantages of
continuous non-immobilized fermentation are elucidated.
19
vessel and corresponding amount of beer was removed from the last vessel. The
residence time was 18 days in the first three vessels and 9 days in the last three
vessels. The beer was reported to be of normal quality. The amount of yeast in
the system was doubled in 28 days and the yeast had become rather granular
after 28 days of operation (Hlavacek et al. 1959). The experiments above were
not fully continuous and did not succeed. Because of their short operative life
they did not offer any advantages over batch processes.
A fully continuous stirred system was patented in 1906 (van Rijn). The process
had six vessels in series, such that each subsequent vessel was situated lower
than the previous one. Wort flowed into the first vessel and fermenting wort
overflowed into the following vessel. van Rijn must be credited for two reasons.
Firstly, the stirrer was equipped with rubber strips to remove precipitates and
dead cells from the walls and base of the vessel. Secondly, temperature control
was achieved by circulating water through hollow shafts of the stirrers.
20
Oxygen
column
Wort
CO2 outlet
Beer out
Coolant in
Steriliser
1st fermentation
vessel
2nd fermentation
vessel
Yeast outlet
Yeast
separator
21
production in May 1966 and had a designed output capacity of 36 million litres
per year. It had five A.P.V. Tower fermenters followed by four conditioning
vessels, two yeast settling vessels and beer cellar facilities. The investment costs
are reported to have been 60% of those of a comparable batch brewery, the
extract losses were decreased by 50%, fuel and power costs were said to be 50%
of those of the batch brewery and an additional financial advantage came from
billing practice (Anon. 1967).
Tower
fermenter
CO2
CO2
surplus
Wort
receiver
Conditioning
vessel
Pump
Pasteuriser
Final beer
Air
inlet
22
production (den Blanken 1974). The A.P.V. Tower fermenter could also be
slowed down or even shut down for up to four days (Seddon 1976).
Morton Coutts of Dominion Breweries of New Zealand patented a continuous
fermentation system (Dominion Breweries Ltd. 1956). The following description
of the Coutts' process at Dominion Breweries Ltd. is taken from Dunbar et al.
(1988). The process consists of three continuously stirred tank reactors
(Figure 4) in cascade and employs a flocculent lager yeast strain. A beer with ca
5.5 % alcohol (1.054 original gravity) was produced with a 45 h residence time.
After boiling, the wort is rapidly cooled to 0oC and trub is removed by
sedimentation. Dilution to fermentation gravity is carried out on line to the
fermentation systems. The Hold Up Vessel (HUV) offers some form of
microbial control providing an environment of low pH (< 4.5) and an alcohol
content of >2.0% w/v. The HUV comprises approximately 6% of the total
volume of the system. The flow into HUV consists of wort and recycled flow
from the second vessel (CF1) in the ratio of 1:1. Additionally, yeast is recycled
from the yeast separator (YS) to achieve control of the fermentation rate through
the amount of yeast in suspension. The HUV is continuously aerated. This
aeration is very important for control of growth and ester formation.
Fermentation vessels CF1 and CF2 comprise 66% and 22% of the system
volume, respectively. The flow from CF1 to CF2 is by gravity via a balance line.
Fermentation is at a maximum in CF1 and yeast growth is continued in CF1. The
Yeast Separator (YS) and the Yeast Washer (YW) are conical in shape and the
highly flocculent yeast is separated from the beer by gravity. Surplus yeast is
washed in counter current flow and the mixture of beer/deaerated water is used
to adjust the original gravity of the green beer, thus minimising extract losses.
The amount of yeast produced is similar to that produced in batch fermentation.
The rate of production could be altered by changing the wort flow into the
system. The total residence time can be varied between 36 and 97 hours.
The market in New Zealand was without competition, so the market for
continuously fermented beer was assured (Hough and Button 1972). The
Palmerston North Brewery of New Zealand Breweries Limited was the world's
first brewery to rely totally on continuous fermentation (Anon. 1987). The
Palmerston North brewery (12 million litres per year) was "very cost effective,
producing a good single brand of beer successfully and efficiently" (Anon.
1987), but in 1985 it was closed down, because instead of upgrading it was
23
decided for economical reasons to move the production to other breweries of the
company.
HUV
CF2
CF1
YS
YW
Deaerated
water
Dilution
water
Yeast
to sale
Wort
infeed
Yeast recycle
Beer
24
25
Coutts system no confirmed mutations in the Coutts system (Dunbar et al. 1988,
Davies 1988) were reported.
Contamination was another threat that was feared to affect continuous brewing
more than batch brewing. Gram-negative, anaerobic or facultative aerobes could
grow in stored wort and produce celery-like off-flavours (Ault 1965). The
principal strains were Aerobacter aerogenes and A. cloacae, with some
intermediate strains. Lactic acid bacteria could grow in the A.P.V. Tower and
were impossible to be removed (Ault 1965). Deliberate contamination of stirred
continuous fermentation led to a constant number of Obesumbacterium proteus,
Acetobacter, wild Saccharomyces or Torulopsis yeast in the vessel (Hough and
Rudin 1959). In all these cases the rate of beer production deceased because the
flow rate had to be adjusted in order to obtain fully fermented beer. Wild yeast
contamination was the most dangerous contaminant in the Coutts system
(Davies 1988).
In some designs there was no substrate gradient in a one vessel system, i.e.
glucose was continuously present in the fermenting beer, which may have
caused some problems. Maltose utilisation may be limited by the presence of
glucose. Too high ester concentrations may be caused by limited yeast growth
during fermentation. The flocculation capacity can be lost more easily.
Although the Coutts system can be operated with different outputs (Dunbar et al.
1988) Inflexibility and a greater degree of brand segmentation affected the
decision of New Zealand Breweries Limited to abandon continuous fermentation
(Davies 1988).
The low number of suitable yeast strains limited the applicability of the A.P.V.
Tower Fermenter (Hudson 1986). Even the Coutts systems uses a very
flocculent yeast strain, although with the use of centrifuges in yeast separation
and recycling should also facilitate the use of a non-flocculent yeast strain
(Davies 1988). In the A.P.V. Tower the yeast growth was severely restricted
(Hudson 1986), which was advantageous for yield, but made flavour matching
difficult. The stirred fermentation (Watney Mann process) was deliberately
designed so that the yeast growth was the same per unit carbohydrate fermented
as in the batch process. Slightly higher yeast growth in a stirred continuous
system compared to batch fermentation was reported (Maule 1973). 40% of dead
26
cells in a pilot scale A.P.V. Tower fermentation were found (Woodward 1967).
In stirred fermentations the proportion of living cells was satisfactory (Bishop
1970), although variations in proportions of dead cells in stirred continuous
fermentations were reported, leading to a lower productivity (Portno 1968a, b).
An economic analysis showed that the Coutts system had 20 to 42 % higher
capital costs over batch fermentation in cylindro-conical vessels (Davies 1988).
At a discussion panel of continuous fermentation at the same Convention
Warren (1988) reported that in "economic terms ... there was benefit from
neither one nor the other", i.e. continuous nor batch fermentation. Continuous
fermentation over batch fermentation was favoured in economic terms when the
brewery output was about 60 million litres per year (7000 brl/week) or more
(Royston 1970), although this was challenged by MacDonald et al. (1984). They
claimed that continuous fermentation failed to reach the designed output due to
the inflexibility of continuous fermentation in the face of changes in demand and
long start-up periods (MacDonald et al. 1984). The anticipated savings were not
made in practice. Finished beer storage in practice was large in order to meet
peak demands, labour was needed on a round-the-clock basis, which increased
the payroll costs, CO2 was not oxygen-free in the A.P.V. Tower fermentation,
the energy consumption was not reduced due to batchwise wort production and
microbial contaminations eliminated the savings in cleaning costs (MacDonald
et al. 1984).
It is clear that the problems encountered in continuous fermentation are very
many and varied, but there was no single reason to abandon it. Some of the
problems were met in one system, but not in an other. Moreover, in New
Zealand beer has been produced and still is produced by continuous fermentation
(Dominion Breweries). Although continuous wort production processes were
developed, only a very few totally continuous breweries were built. The Fort
Worth (Williamson and Brady 1965), La Cervecera del Norte brewery (Anon.
1967) and the Centri Brew Process (Schffel and Deublein 1980) should be
mentioned as examples of totally continuous processes. Inflexibility with regard
to the production rate and the changes between beers was surely one decisive
factor at least with the A.P.V. Tower, and changes in yeast flocculation capacity
may have caused problems in some stirred tank systems.
27
Cleaning minimised
28
29
30
Influence
Volumetric productivity
increased
Operation
continuous
Biomass separation
eased
Gas-liquid transfer
enhanced
Process design
simple
Substrate utilisation
enhanced
Yeast strains
indifferent
Hygiene
manageable
Scale-up
predictable
Product
consistent
31
yeast viability at the exit of the reactor was decreased. However, probably the
most serious problem was the high amount of -acetolactate in the green beer
(Narziss 1997). Decreased foam stability was noted by Dembowski (1992).
Preliminary
wort tank
Plate heat
exhanger
Filtered aid
wort filtration
Filtered
wort tank
Yeast
vessel
Filter
pump
Yeast
pump
Beer to
stabilization
Maturing
tanks
Plate heat
exhanger
BIOREACTOR
Figure 5. The Bio-Brew process (redrawn from Narziss and Hellich 1972).
Dembowski et al. (1993) developed the Bio-Brew further by optimising flow
through the kieselguhr yeast bed and adding a cooling plate inside the reactor.
This led to a slower decrease in yeast viability. Installing an aerobic phase in
front of the filter led to improved sensory quality of beer and to better stability of
the Bio-Brew (Dembowski et al. 1993), but the concentrations of low molecular
weight nitrogenous substances in the green beer still remained too high. Overall
the Bio-Brew experiments were not successful.
The very high VDK values of the Bio-Brew experiments still have an effect.
Although in many of the later experiments the high VDK-concentrations are
absent, the question of very high VDK-concentrations is still raised. The fact
that in the Coutts process, yeast does not produce more diacetyl than in batch
fermentation provided that the yeast and wort are kept the same (Dunbar et al.
1988), has not removed the doubt.
32
Beer out
Beer + CO2
Support plate
Filter
Yeast plug
Heating
coil
Cooling
coil
Wort in
33
34
First reactor
Cooling pipes
Third reactor
Filter
Off gas
Off gas
Beer
out
Wort
Air
Centrifuge
Filter
Heat treatment
P
Yeast
35
regenerating properties for repeated use (Yamauchi et al. 1994a). Aseptic filling
of the reactors was also impossible with alginate beads (Inoue 1995). Therefore,
alginate was replaced by porous glass beads developed by Kirin.
The reason for dividing the process into different units was to separate the
different phases of fermentation. The steriliser assured asepticity of the wort. In
the first bioreactor yeast growth, pH reduction and fusel alcohol formation
occurred. Out-flowing yeast in the fermenting beer was reduced below a certain
level by centrifugation. In the second reactors the yeast was in stationary (nongrowing) state for formation of ethanol and esters. The heat treatment converted
-acetolactate into acetoin and diacetyl. The third bioreactor contained yeast in a
stationary state for VDK reduction.
Kirin investigated primary fermentation in bioreactors containing 0.5 m3 and
10 m3 of carrier. In the larger bioreactors an additional cooling device was
necessary to maintain an even radial temperature distribution within the carrier.
Other problems encountered in scaling up were decreased fermentation capacity
per volume with increasing bioreactor size, and channelling (Yamauchi et al.
1994b). The reduced fermentation efficiency was attributed to higher flow rates
with increasing reactor size (Inoue 1995). Continuous operation of the system at
3.6 million litres per year for more than 6 months was possible (Yamauchi and
Kashihara 1995).
The product was sensorily acceptable lager beer, which however differed in
some characteristics from conventional batch beer. It had a higher alcohol
content and less fusel alcohols than the batch beer. It contained the same total
amount of organic acids as the batch beer, but the pattern was different: it had
less acetic acid and more succinic acid. The beer also contained more sulphite
and it had higher bitterness (Yamauchi et al. 1995a). Restricted yeast growth
was claimed to be the main reason for some of the differences. A series of
papers described the formation and control of different aroma compounds in this
process (Yamauchi et al. 1995a, Yamauchi et al. 1995b, and Yamauchi et al.
1995c).
The Kirin Brewing Company set up a small commercial scale production unit
using the described process on the island Saipan, producing 185 000 litres per
year. The brewing proved to be short lived. Reasons for the economic failure
36
were low demand and the limited number of products (Inoue 1995). The output
could not be run at one fifth of the designed output without deterioration of yeast
fermentative activity (Inoue 1995). Other operational disadvantages included
slow start up of the system (2 weeks), high energy costs because of the heat
treatment before the third bioreactor and beer losses with the centrifuged yeast
(Inoue 1995).
37
Thermal jacket
Green beer
Air/CO2
Wort
38
residence time was 22 h per bioreactor. The aeration was arranged by diffusion
through plastic tubing in the circulation loop. The productivity reported for one
matrix at 15oC was 6.6 hl beer per year at 12oPlato and 9.1 hl per year at
16oPlato. The productivity per matrix means that for an annual output of 100
million litres over 100 000 matrices would be needed.
Stage 1
Stage 2
Wort vessel
Wort supply
pump
Beer vessel
Immobilized
yeast reactors
Figure 9. The Meura Delta system for lager beer production (redrawn from
Andries et al. 1995).
The beer quality was said to resemble that of conventional batch fermented beer,
although the amount of higher alcohols was somewhat lower and the amounts of
esters were higher. The latter phenomenon can be controlled by aeration. The
utilisation of amino acids was similar to that in conventional batch fermentation.
The system was adjusted for production of top fermenting beer in semi-industrial
scale (Andries et al. 1997a). For this application the second immobilized yeast
loop bioreactor was replaced by a cylindro-conical tank with free cells. This tank
was equipped with an external circulation loop. Thus the system was a
combination of immobilized and free cell stages. The immobilized bioreactor
supplied the second stage continuously with free, viable cells. The beer was
similar to that produced traditionally. Improved productivity and decreased
investment costs compared with the totally immobilized system were the
claimed advantages. At least the Aubel brewery is using the Meura Delta system
39
with 500 matrices in the production of top fermented beer (Mensour et al. 1997,
Anon. 1997). In Canada at least one pub brewery has been using the Meura
Delta system (Rajotte 1998). A group in Versuhs- und Lehranstalt fr Brauerei
in Berlin (VLB) found the use of (Wackerbauer et al. 1996b) the loop reactor of
Meura Delta to be relatively easy, whereas Tata et al. (1999) reported somewhat
contradictory results.
40
day) (Kronlf and Virkajrvi 1999). Later the unit was extended to include a
continuous secondary fermentation unit with the same capacity (Kronlf et al.
2000). The yeast is removed from the green beer before the heat treatment
section of the secondary fermentation by cross-flow filtration. Flavour formation
was typical for lager beer (Table 4). The authors stated that it was fully possible
to produce good beer in less than two days with this system in normal
fermentation temperatures.
Table 4. Flavour compounds in beer produced by continuous, immobilized
fermentation in both the main and secondary fermentation (adapted from
Kronlf et al. 2000).
Beer analyses
Continuously
produced
Conventionally
produced
23.2
17.2
1.0
1.3
41.0
41.9
0.02
0.01
41
Wort
treatment
Wort feed
Beer cooling
CO2
Immobilized
yeast
reactor
CO2 removal
Optional yeast
separation
Beer out
Excess yeast
Figure 10. The fixed bed with a circulation loop, described by of Pajunen et al.
(2000a) (adapted).
The system was able to produce stout with close enough quality match over a
period of 8 weeks.
Pectate-immobilized yeast in a gas lift reactor was used by Smogrovicova et al.
(1997). The VDK in this beer was at the same level as in traditionally produced
beer. In a subsequent study (Smogrovicova and Domey 1999) the beer produced
using Ca-pectate immobilized yeast was found by taste testers to be comparable
with traditionally produced beer.
Two different systems for immobilized main fermentation was evaluated (Tata
et al. 1999). A fluidised bed reactor with yeast immobilized on a porous glass
bead carrier produced beer with quite acceptable flavour, albeit different from
the conventionally produced beer. The sintered silicon carbide loop reactor
system did not produce a completely fermented product with the same residence
time as the fluidised bed reactor system. Tata et al. (1999) ended their paper by
stating that immobilized yeast systems have advantages such as "stable and
42
43
% of batch
40
20
TOTAL
TOTAL HIGHER
0
-20
-40
-60
-80
-100
Sapporo
Labatt
Meura
VTT
Kirin
Figure 11. The differences (%) in total esters and total higher alcohol
concentrations between immobilized yeast fermented and conventionally
fermented beers in w/w per cent (data from Shindo et al. 1994, Mensour et al. 1997,
Andries et al. 1995, Linko and Kronlf 1991, Yamauchi and Kashihara 1995).
44
45
46
47
48
49
6. Results
6.1 Long term stability (II)
6.1.1 The apparent degree of attenuation
The apparent degree of attenuation remained high and very near the measured
attenuation limit of the wort throughout the whole operation of 442 days
(Figure 12). The changes after the prereactor can be explained partly by the
movement of the glass beads in the prereactor. From time to time in the
prereactor some of the glass beads rose and then fell back. This movement
removed non-bound yeast from the prereactor, causing changes in the total
amount of yeast in the prereactor. A similar kind of movement of the glass beads
was seen in smaller scale (0.8 dm3) and larger scale (100 dm3) experiments. The
changes in attenuation after the buffer tank can be partly explained by removal
of excess yeast from the buffer tank. The removal was carried out ca every
fortnight.
90
80
70
60
50
40
30
20
10
0
0
28
56
84
112
140
168
196
P rereactor
224 252
Tim e [d ays]
B uffer tank
280
308
336
364
392
420
448
476
M ain reactor
Figure 12. The apparent degree of attenuation in the outflows from the
prereactor, the buffer tank and the main reactor (II).
50
6.1.2 Asepticity
No harmful microbial contamination and no wild yeast contamination were
detected during the operation. The microbial counts for aerobic bacteria detected
are shown in Figure 13. Although a low number of contaminants was detected in
the samples taken from the system no colonisation occurred. This means that
although main fermentation is more sensitive to contamination than secondary
fermentation (Haikara et al. 1997) normal aseptic procedures are sufficient for
main fermentation.
100
90
80
70
60
50
Wort
40
Prereactor
Main reactor
30
20
10
0
0
28
56
84
112
140
168 196
224
252
280 308
336
364
392 420
448
Time [days]
Figure 13. The aerobe counts detected in the immobilized yeast reactor system (II).
51
15
3
10
2
5
0
0
28
56
PROPANOL
84
ACETTIME [days]
ALDEHYDE
3-methylbutylacetate [mg/l]
20
0
140
112
3-METHYLBUTYLACETATE
52
45
40
35
mg/dm
30
25
20
15
10
5
0
ETHY LA CETA TE
PROPA NOL
S1 AVE
S2 AVE
A CETA LDEHY DE
Figure 15. Comparison of the two sets of aroma compound analyses. S1 is the
average value for the first 137 days and S2 is the average value for days 378 to
442 (II).
Table 5. Concentrations of the aroma compounds at two different flow rates
[mg dm3] (II).
500 ml h1
750 ml h1
22.3
15.0
1.1
0.5
Propanol
16.6
15.6
2-Methyl propanol
12.5
10.7
3-Methyl butanol
45.3
38.7
3-Methyl butanol
18.2
14.6
4.3
5.3
Flow rate
Ethyl acetate
3-Methyl butyl acetate
Acetaldehyde
The aroma profile of beer from the immobilized fermentation was compared to a
commercial Finnish lager produced with the same yeast and wort. This beer was
matured, whereas beer from immobilized fermentation was not. A Finnish
brewery can have fermentation vessels of hundreds of thousands of litres and the
production of exactly the same beer in regard to aroma compounds in 20 litre
scale reactors is challenging. High gravity brewing was used in the industry,
53
whereas the wort was diluted to 11oP and autoclaved for the continuous
fermentation. For comparison, the results have been adjusted to the same
original gravity and the commercial lager has been adjusted to 100%
(Figure 16). The statistically calculated 95% confidence limits (n=16) for the
commercial lager are included. The most serious deviation is in the
concentration of 3-methyl butyl acetate. The beer produced by immobilized
yeast had less than 50% of the 3-methyl butyl acetate of the conventionally
produced beer. The differences in acetaldehyde or in particular in propanol were
not so serious in regard to their taste thresholds.
Ethylacetate
150
Acetaldehyde
3-Methylbutylacetate
100
50
0
3-Methylbutanol
Propanol
2-Methylbutanol
Immobeer
2-Methylpropanol
Confidence limits (95%)
54
90
80
70
60
50
1.5
2.0
2.5
3.0
3.5
4.0
Air
[ml/
4.5
min
]
120
140
5.0
160
80
100
60
40
20
in]
/m
l
[m
2
CO
N = 12 Q 2 = 0.5518
DF = 6 R 2 = 0.9161
Figure 17. The second order response surface for higher alcohols (III).
55
The confidence intervals for the second order coefficients for the terms air*air,
CO2*CO2 and air*CO2 were so large that the simplification of the model to a
linear one was justified (Figure 18). The goodness of fit was reduced from 0.92
to 0.90, but the goodness of prediction increased from 0.55 to 0.82.
90
80
70
60
50
1.5
2.0
2.5
3.0
3.5
4.0
Air
[ml/
4.5
min
]
120
140
5.0
160
80
100
60
40
20
0 40
in]
l/m
m
[
2
CO
N = 11 Q 2 = 0.8165
DF = 8 R 2 = 0.9033
Figure 18. The linear response surface for higher alcohols leaving the
prereactor (III).
Both models show that with a low CO2 feed rate and high air feed rate the
maximum amount of higher alcohols is formed in the prereactor. In a batch
fermentation high aeration has a similar effect. Kahler et al. (1965) also noticed
that higher aeration led to increased levels of higher alcohols and acetaldehyde
in a continuous fermentation. It should to be noticed that the high value for the
sum of higher alcohols is almost double the low value in the studied range.
56
3-Methyl
butanol
2-Methyl
butanol
Significance
Higher values
Siran
Siran
Siran
GDC
GDC
GDC
57
strongly flocculent strain 2-methyl butanol was also higher when using GDC
than when using Siran. The beers produced using Siran carrier had higher
concentrations of 3-methyl butyl acetate than beers produced with other carrier
materials. Ethyl acetate was higher in the beer produced using Celite carrier than
in the beer produced using Siran.
pH
Both the yeast strain and the carrier affected concentrations of total diacetyl, but
only the carrier had an effect on the concentrations of total pentanedione. With
Siran both total diacetyl and total pentanedione concentration were lower than
with GDC carrier. With Siran, lower total diacetyl concentrations were found
compared with Celite. No difference was found in total pentanedione
concentrations between Siran and Celite carriers. An interesting phenomenon is
higher pH of the young beers produced using Siran with many strains, although
the consumption of FAN was higher using Siran (see Figure 19).
4,5
4,4
4,3
4,2
4,1
4,0
3,9
3,8
3,7
3,6
0
10
20
30
40
50
Time [days]
Siran A24
GDC A24
Figure 19. pH of beer from the outflow from the reactor. A24 = VTT A-66024 (V).
58
cfu/ml
1,0E+05
Siran *
Beech wood *
Aspen *
Beech wood *
1,0E+04
1,0E+03
1,0E+02
1,0E+01
1,0E+00
0
20
40
60
80
Time [days]
Figure 20. Amount of K. terrigena in the outflowing green beer from the
contaminated reactors (VI).
The pH values of the green beers from the contaminated reactors were higher
than those of the beers from non-contaminated reactors (VI). The differences
were small but consistent with all the carrier materials used. The detection time
for this difference was ca two weeks.
The experiments were run in small-scale reactors in which aeration of the wort
was through the tubing only. This possibly contributed to the decrease in
viability of the outflowing yeast, but the decrease appeared to be more
pronounced in beers from contaminated reactors. At the end of the experiments,
samples from the carrier were taken from the top part and from the bottom part
59
of the reactor and these were analysed for the contaminant and for yeast
viability. In 7 of the 8 points analysed, the viability of the yeast was lower in the
contaminated reactor.
K. terrigena increased the concentration of dimethyl sulphide (DMS) in the
green beers. Three weeks after the contamination the taste threshold (50 g/l)
was exceeded. The carrier material appeared to affect the maximum
concentration of DMS in the outflowing beer.
5
Siran *
Siran
Beech wood *
Beech wood
Aspen *
Aspen
Beech wood *
Beech wood
Ratio [-]
4
3
2
1
0
0
20
40
60
80
Time [days]
Figure 21. The ratio of total diacetyl to total pentanedione in the green beer.
* indicates a contaminated reactor; no star indicates a control reactor (VI).
The fastest detection method for the contaminant was plate counting, because
the changes in the chemical composition of the beer were only detected later.
The slight elevation in the pH of the green beer or the change in the ratio
between total diacetyl and total pentanedione (Figure 21) may be used as alarms,
but both require a good knowledge of the process. This contaminant did not
produce nitrosoamines and the concentration of biogenic amines was also low.
The concentration of cadaverine was increased in the beer outflowing from the
contaminated bioreactors.
60
Wort tanks
Secondary
fermentation
Beer tanks
61
Operating costs for the immobilized process are 66 Euro per m3 of beer (6 385
000 Euro per year) and for the batch process 63 Euro per m3 of beer. The main
differences were the renewals of the carrier and the higher consumption of
electricity in the immobilized process.
The cost of the carrier material forms a very significant part of the investment
(Figure 23). The price used for the material was a quotation from a
manufacturer. The price of a cheap carrier may be below 0.3 Euro l1.
% of batch investment
110
100
90
80
70
60
50
0
Figure 23. The effect of carrier price on the total investment cost as a
percentage of the investment cost of a batch process (IV).
The electricity consumption of pumps was calculated from the rated power
multiplied by the running time per year. This will be an overestimate, because at
normal workload electric motors use less than the rated amount of electricity.
CIP costs of water and chemicals lead to savings for the continuous process.
Extra income was estimated to originate from lower losses of beer: there are less
tank bottoms. Each transfer from a tank is estimated to cause a 1% loss of beer
and each change of carrier results in a 1.5% loss of the reactor volume of beer. It
was estimated that in a batch process 2% of the fermentable extract is used for
yeast growth, compared to only 1.5% in an immobilized process.
62
When these extras are taken into account the total manufacturing cost for
processes are: batch 63 Euro m3 and immobilized 64 Euro m3 when the carrier
price is 4 Euro l1 (Table 7). If the carrier price is 0.3 Euro l1 the manufacturing
costs for the immobilized process are 1% less than those of the batch process.
With the annual world production of ca 1.3108 m3 (1 300 million hectolitres)
this indicates savings in the order of 40 million Euros per year in the
manufacturing costs.
Table 7. Manufacturing costs when savings are included (carrier price
4 Euro/litre) (IV).
Extra income
decreased losses
in transfers
less growth of yeast
Immobilized
1000 Euro/a
Euro/m3
155
1.6
67
0.7
Batch
%1)
Euro/m3
63.3
100
Manufacturing
cost
63.8
101
Carrier price
0.3 Euro/litre
63.0
99
1)
The feasibility of two cheap carriers was investigated (Linko et al. 1997) and
found that either aspen or beech wood chips can be used as a carrier material.
The use of these cheap carriers material will favour the economics of the
immobilized process. The use of wood chips as a carrier in secondary
fermentation is patented (Oy Panimolaboratorio Bryggerielaboratorium Ab
1998).
63
7. Discussion
Continuous, non-immobilized fermentation failed to fulfil the requirements
specified by Portno (1978):
The system must maintain a high concentration of yeast to allow
rapid fermentation.
The system must be capable of using a wide range of yeasts.
The system must be capable of producing a range of beers.
The system must maintain a fermentation gradient
The system must allow controlled growth of yeast at levels up to
that of a batch system.
The system must have a facility for control of growth rate to
maintain a young and vigorous culture.
The system must provide oxygen in controlled amounts in zones of
high substrate and low product concentration.
64
cell stage was reported (Andries et al. 1997b). The long operation time is
essential for economical reasons, although it is not sufficient in itself, as the
reported long operation times of the A.P.V. Tower demonstrated (Shore 1986).
The formation of flavour compounds was stable for long operative times (II).
The first period lasted 137 days and the second period 123 days. The
experimental phase between these constant gas feed periods did not cause
significant permanent changes in the formation of aroma compounds. In the
literature, aroma compounds from such long runs have seldom been reported. 3methyl butyl acetate decreased slowly during six weeks of operation in a 100
litre scale PBR system (Kronlf et al. 1995). The mixing caused by gas feed
may increase the stability of the system used in this study. Oxygen transfer from
the gas phase into the liquid phase and finally to yeast may occur in a large part
of the prereactor.
65
66
carrier will affect the flavour of beer produced. The origin of found differences
might well lie in the changes of the yeast cell wall when cell reacts with the
surface. These changes can and probably are strain specific. The final
optimisation for a particular flavour may include changes in wort quality (e.g.,
FAN), aeration, temperature etc.
7.4 Contamination
The primary fermentation systems are much more sensitive to contamination
than the less nutritious secondary fermentation systems (Kronlf and Haikara
1991). Moreover, the contaminating organisms of primary fermentation and
secondary fermentation reactors differ distinctively (Haikara et al. 1997, Kronlf
et al. 2000). Gram negative bacteria, such as Obesumbacterium proteus, other
enterobacteria or acetic acid bacteria are the most common contaminants in
immobilized yeast reactors used for primary fermentation. Besides bacteria, wild
yeasts may contaminate both primary and secondary fermentation systems
(Kronlf and Haikara 1991, Haikara and Kronlf 1995, Haikara et al. 1997,
Kronlf et al. 2000).
K. terrigena was able to colonise the reactors and survive in the process
conditions. The use of beech wood chips resulted in the highest concentration of
K. terrigena in the outflowing beer. Aspen chips and Siran supported lower
concentrations. When comparing the figures for pH, DMS and K. terrigena in
the outflowing green beers it was observed that the fastest detection method for
K. terrigena was plate counting. Increased concentrations of DMS were detected
in the green beer only on day 20, whereas K. terrigena was detected by plate
counting after approximately one week. The change in pH of the green beer was
seen after approximately two weeks. Another possibility is to follow the
concentrations of vicinal diketones and especially the ratio between total
pentanedione and total diacetyl. If this ratio starts to increase over a pre-set value
or shows a stable increasing trend, this could be taken as an indicator of
contamination. The reason for the increase in the ratio is unclear, but it could be
a consequence of the reduced viability of the yeast.
Biogenic amines were not produced during the experiments in amounts that
would be harmful to health. K. terrigena caused elevated concentrations of
67
7.5 Economics
The main economic advantages of continuous, immobilized fermentation are
based on the possibilities to use very short fermentation times and to diminish
the non-productive time. Using a very cheap carrier material, the investment cost
was estimated to be ca. 70 % of that of a batch fermentation at an annual output
of 100 million litres (IV). The smaller floor area requirement of the immobilized
process implied 15% savings in the building cost. The running costs of the
immobilized process were estimated to be 5% higher than those of the batch
process. The Kirin Brewing Company Ltd (Table 8) has estimated rather similar
costs (Inoue 1995) for the investment using their own carrier (Bioceramics,
porous glass beads). Contrary to results in (IV), Inoue (1995) calculated the
running costs to be 1.8 fold and the yield of beer to be lower. This may be a
reflection of their experiences from the small scale Boca Boca brewery on
Saipan.
Table 8. Estimated profit gained by immobilized yeast systems in a 100 million
litre brewery (adapted from Inoue 1995). Both main fermentation and secondary
fermentation were continuous, immobilized processes.
Batch
[%]
Immobilized
[%]
Investment
100
58
Running cost
100
176
Space required
100
44
Yield of beer
lower
68
7.6 Compendium
Using the immobilization technique a high concentration of yeast can be
maintained in the reactor. A change in the flow rate will not affect the amount of
yeast in the system, so the output can be varied. The high concentration of yeast
facilitates short fermentation times without an increase in temperature. With the
systems presented in this study the main fermentation lasted about 50 hours (II,
III and IV). In the early attempts at continuous fermentation very short
fermentation times were reported (down to 4 hours), but in many of these
experiments the fermentation temperature was elevated. Yamauchi et al. (1995a)
reported a 1 day fermentation time for their system of CSTR and PBR. Meura
Delta system with an inverted structure a loop reactor followed by a CSTR
was also able to ferment in 1 day (Andries et al. 1997a). Residence times
between 20 and 24 hours gave full attenuation in a fluidised bed reactor
(Mensour et al. 1996). The slow start-up encountered in the A.P.V. Tower
fermenter is abolished by using immobilization technology.
With immobilization technique the yeast strain is not limited to those with
suitable flocculation characteristics, but any production strain can be used. In the
papers II, III, V and VI five different strains were used. None of them posed any
difficulties, although their flocculation characteristics were different. The use of
a normal production strain in an immobilized fermentation process is reported by
many authors, e.g., van de Winkel et al. (1993), Smogrovicova et al. (1997)
whereas others simply describe the strain as brewer's yeast, e.g., Yamauchi et al.
(1995a), Maeba et al. (2000).
Maintaining a fermentation gradient is possible either by the use of multiple
vessels or by using packed bed type reactors. By controlling the amount of
oxygen entering the system the yeast growth together with formation of aroma
compounds can be controlled. When introducing air into wort just before the
reactor, oxygen is introduced into high substrate and low product concentration.
Although, all the requirements of Portno (1978) can be fulfilled by using
immobilization technique, there exist hindrances for the application of
immobilized main fermentation. The total production of beer in the world is
increasing slowly, but the increase is limited to very few countries: China,
Russia, Vietnam, India. Elsewhere the production capacity is in excess (Western
69
Europe, USA). The amount of money tied up in the existing fermentation vessels
alone could be in the order of 3 000 000 000 Euro (unit cost 0.16 million Euro,
unit size 4000 hl, cycle time 18 days, the world production 1 300 million
hectolitres). The lifetime of a fermenter can be 25 to 30 years. The capital tied
up in the conventional fermenters is a factor against the immobilized main
fermentation. Batchwise wort production diminishes the economic advantages of
continuous main fermentation, as more tanks for wort storage are needed in the
process. Different beers can be produced by changing the type of wort (Andries
et al. 1997b), without any change in the yeast.
The image of the product is a factor that cannot be ignored when considering
any major process changes in the beer production. There will always be
individuals and organisations that prefer to stay with traditional methods and
materials. The beer tasting guru Michael Jackson wrote (1992): "immobile yeast
might be the technique of the future and I probably won't like that, either",
although he did not find any faults in continuously fermented Dominion Kiwi
Lager. Fortunately, Meilgaard (2000) provided an example that new types of
beer may be successful on the market.
70
71
72
References
Andersen, K., Bergin, J., Ranta, B. and Viljava, T. 1999. New process for
continuous fermentation of beer. Proc 27th Congr. Eur. Brew. Conv., Cannes.
IRL Press, Oxford. Pp. 771778.
Andries, M., van Beveren, P., Goffin, O. and Masschelein, C. 1995. Desing of a
multi-purpose immobilized yeast bioreactor system for application in the
brewing process. European Brewery Convention Monograph XXIV. EBC
Symposium Immobilized Yeast Applications in the Brewing Industry Espoo,
Finland. Verlag Hans Carl Getrnke-Fachverlag, Nrnberg. Pp. 134144.
Andries, M., van Beveren, P., Goffin, O., Rajatte, P. and Masschelein, C. 1997a.
First results on semi-industrial continuous top fermentation with Meura-Delta
immobilized yeast fermentor. MBAA Techn. Quater., 34, 2, pp. 119122.
Andries, M., van Beveren, P.C., Goffin, O., Rajotte, P. and Masschelein, C.A.
1997b. Practical resulys using Meura-Delta immobilized yeast fermentation
system. Brewers' Guardian, February, pp. 2628.
Angelino, S.A.G.F. 1991. Beer. Food Sci. Technol., 44, pp. 581616.
Anon. 1967. Spain Continuous Brewing in Practise. Brewers' Journal, May,
pp. 5562.
Anon. 1987. Focus on...New Zealand Breweries Ltd. The Brewer, 78(867),
January, pp. 1417.
Anon. 1995. European Brewery Convention Monograph XXIV. E.B.C.
Symposium Immobilized Yeast Applications in the Brewing Industry, Espoo,
Finland, October 1995. Verlag Hans Carl Getrnke-Fachverlag, Nrnberg.
259 p.
Anon. 1997. Meura-Delta yeast system at Achouffe. Brewers' Guardian, April,
p. 10.
73
Anon. 1999. Beer Market overview. Consumer Goods UK, No 494, pp. 76
154.
Anon. 2000a. Die belgische Interbrew rckt weltweit auf Platz drei. Brauerei
Forum, 15(7), pp. 201203.
Anon. 2000b. Konzentration auf dem internationalen Biermarkt schreitet voran.
Brauerei Forum, 15(6), pp. 157.
Anon. 2000c. Biermarkt Frankreich fest in britischer Hand ... Brauerei Forum,
15(6), pp. 174175.
Ault, R.G. 1965. Spoilage bacteria in brewing. J. Inst. Brew., 71, p. 376.
Ault, R.G., Hampton, A.N., Newton, R. and Roberts, R.H. 1969. Biological and
biochemical aspects of tower fermentation. J. Inst. Brew., 75, pp. 260277.
Baker, D.A. and Kirsop, B.H. 1973. Rapid Beer Production and Conditioning
Using a Plug Fermentor. J. Inst. Brew., 79, pp. 487494.
Barbet, E. 1899. French Patent.
Barbet, E. 1905. British Patent 16 233.
Berdelle-Hilge, Ph. 1966. Verfahren und Vorrichtung zur kontinuerlichen
Behandlung von Flssigkeiten mit Enzymtrgern. German patent 1517 814,
September 2nd, 1966.
Bishop, L.R. 1970a. A System of Continuous Fermentation. J. Inst. Brew., 76,
pp. 172181.
Bishop, L.R. 1970b. Brewing Present and Future. Proceedings of the 11th
Convention of The Institute of Brewing (Australia and New Zealand Section).
Hobart, Australia, 8th13th March, 1970. Pp. 113116.
Borremans, E. 1997. Secondary fermentation of beer with immobilized yeast
(SFIY). Cerevisiae, 22(1), pp. 3337.
74
den Blanken, J.G. 1974. The tower fermenter for lager beer production. Brewers'
Guardian, October, pp. 3539.
Boulton, C.A. 1991. Developments in Brewery Fermentation. Biotechnology
and Genetic Engineering Reviews, 9, pp. 127181.
Christensen, C.M. 1997. The Innovators Dilemma when New Technologies
Cause Great Firms to Fail. Harvard Business School Press, Boston,
Massachusetts. 225 p.
Cop, J., Dyon, D., Iserentant, D. and Masschelein, C.A. 1989. Reactor design
optimization with a view to the improvement of amino acid utilization and
flavour development in calsium alginate entrapped brewing yeast fermentations.
Proc. 22th Congr. Eur. Brew. Conv., Lisboa. IRL Press, Oxford. Pp. 315322.
Davies, A.W. 1988. Continuous Fermentation 30 Years On. Proceedings of the
20th Convention of The Institute of Brewing, Australia and New Zealand
Section, Brisbane, Australia, 8th13th May, 1988. Pp. 165168.
Delbrck, M. 1892. Vortrag ber Schnellgrung und das Arbeiten mit gefesselter
Hefe. Wochschr. Brau., 9, pp. 695699.
Dembowski, K. 1992. Untersuchungen zur kontinuerlichen Biergrung am
Beispiel des Grfilterverfahrens. PhD thesis. Technische Universitt Munchen.
Pp. 255256.
Dembowski, K., Narziss, L. and Miedaner, H. 1993. Technologisch optimierte
Bierherstellung im Festbettfermenter bei sehr kurzer Produktionszeit. Proc 24th
Congr. Eur. Brew. Conv., Oslo. IRL Press, Oxford. Pp. 299306.
Dervakos, G.A. and Webb, C. 1991. On the merits of viable-cell immobilization.
Biotech. Adv., 9, pp. 559612.
Dirksen, J. 1998. Single phase cleaning for brewing operations. Presentation at
111th anniversary convention of Master Brewers Association of the Americas,
Minneapolis, September 2023, 1998.
75
76
77
78
Kronlf, J., Linko, M. and Pajunen E. 1995. Primary fermentation with twostage packed bed system pilot scale experiences. European Brewery Convention
Monograph XXIV. EBC Symposium Immobilized Yeast Applications in the
Brewing Industry, Espoo, Finland. Verlag Hans Carl Getrnke-Fachverlag,
Nrnberg. Pp. 118124
Kronlf, J. and Virkajrvi, I. 1999. Primary fermentation with immobilized
yeast. Proc 27th Congr. Eur. Brew. Conv., Cannes. IRL Press, Oxford. Pp. 761
770.
Kronlf, J.H.M., Virkajrvi, I., Storgrds, E.L., Londesborough, J. and Dymond,
G. 2000. Combined primary and secondary fermentation with immobilized
yeast. World Brewing Congress, Orlando. (To be published.)
Levin Y., Pecht, M., Goldstein, l. and Katchalski E. 1964. A Water-insoluble
Polyanionic Derivate of Trypsin. I Preparation and Properties. Biochemistry, pp.
19051913.
Linko, M. and Enari, T.-M. 1967. Lagering of beer I. Shortening of the lagering
period with the aid of raised temperature. Mallasjuomat, pp. 199218.
Linko, M. and Kronlf, J. 1991. Main fermentation with immobilized yeast.
Proc. 23rd Congr. Eur. Brew. Conv., Lisbon. IRL Press, Oxford. Pp. 353359.
Linko, M., Virkajrvi, I., Pohjala, N., Lindborg, K., Kronlf, J. and Pajunen, E.
1997. Main fermentation with immobilized yeast a breakthrough? Proc 26th
Congr. Eur. Brew. Conv., Maastricht. IRL Press, Oxford. Pp. 385394.
MacDonald, J., Reeve, P.T.V., Ruddlesden, J.D. and White, F.H. 1984. Current
appoaches to brewery fermentations. In: Progress in industrial microbiology,
Bushell, M.E. (ed.). Elsevier Amsterdam, Vol 19. Pp. 48198.
Maeba, H., Unemoto, S., Sato, M. and Shinotsuka, K. 2000. Primary
Fermentation with Immobilized Yeast in Porous Chitosan Beads. Pilot Scale
Trial. Proc. 26th Convention Inst. of Brew. Australia and New Zealand Section,
Singapore 1924 March, 2000. Pp. 8286.
79
80
Mensour, N.A., Margaritis, A., Briens, C.L., Pilkington, H., Russell, I. 1997.
New Developments in The Brewing Industry Using Immobilized Yeast Cell
Bioreactors. J. Inst. Brewing, 103, pp. 363370.
Mitchell, C.A. 1926. Vinegar: Its Manufacture and Examination. 2nd ed. Griffin,
London.
Moll, M. and Duteurtre, B. 1996. Fermentation and maturation of beer with
immobilized yeast. Brauwelt International, pp. 248525.
Mkinen, V. 1966. The continuous fermentation in the production of beer.
Mallasjuomat, 10, pp. 353357.
Nakanishi, K., Onaka, T., Inoue, T. and Kubo, S. 1985. A new immobilized
yeast reactor system for rapid production of beer. Proc. 20th Congr. Eur. Brew.
Conv., Helsinki. IRL Press, Oxford. Pp. 152157.
Narziss, L. 1997. Global brewing technology a look over the fence. Brauwelt
Int., 1, pp. 1621.
Narziss, L. and Hellich, P. 1971. Ein Beitrag zur wesentlichen Beschleunigung
der Grung und Reifung des Bieres. Brauwelt, 111, pp. 14911500.
Narziss, L. and Hellich, P. 1972. Rapid Fermentation and Maturing of Beer by
Means of Bio-Reactor. Brewers Digest, Sept, pp. 106118.
Norton, S. and D'Amore, T. 1994. Physiological effets of yeast immobilization:
Applications for brewing. Enzyme Microb. Technol. 16. Pp. 365375.
Norton, S., Neufield, R.J. and Poncelet, D.J.C.M. 1994. Canadian Patent
Application 2 1333 789.
Okabe, M, Oda, A., Park, Y.S., Noguchi, K., Okamoto, T. and Mitsui, S. 1994.
Continuous Beer Fermentation by High Cell-Density Culture of Bottom
Brewer's Yeast. J. Ferment. Bioeng., 77, pp. 4145.
81
Onaka, T., Nakanishi, K., Inoue, T. and Kubo, S. 1985. Beer brewing with
immobilized yeast. Bio/Technol., 3, pp. 467470.
Pajunen, E., Grnqvist, A. and Ranta, B. 1991. Immobilized yeast reactor
application in continuous secondary fermentation in industrial scale operation.,
Proc 27th Congr. Eur. Brew. Conv., Lisboa. IRL Press, Oxford. Pp. 361368.
Pajunen, E. and Grnqvist, A. 1994. Immobilized Yeast fermenters for
Continuous Lager Beer Maturation. Proc. 23rd Convention Inst. of Brew.,
Australia and New Zealand Section, 1015 April, 1994, Sydney, Australia. Pp.
101103.
Pajunen, E., Ranta, B., Andersen, K., Lommi, H., Viljava, T., Bergin, J. and
Guercia, H. 2000a. Novel Process for Beer Fermentation with Immobilized
Yeast. Proc. 26th Convention Inst. of Brew., Australia and New Zealand
Section, 1924 March, 2000, Singapore. Pp. 9194.
Pajunen, E.J., Viljava, T.T. and Lommi, H.O. 2000b. Novel Primary
Fermentation with Immobilized Yeast System. World Brewing Congress,
Orlando. In press.
Oy Panimolaboratorio Bryggerielaboratorium Ab 1998. Finnish Patent
102291, 13.11.1998.
Pedersen, S. 1993. Industrial Aspects of Immobilized Glucose Isomerase. In:
Industrial Application of Immobilized Biocatalysts, Tanaka, A., Tosa, T. and
Kobayashi, T. (eds). Marcel Dekker, Inc. New York. Pp. 185208.
Pilkington, P.H., Margaritis, A., Mensour, N.A. and Russell, I. 1997.
Fundamentals of immobilized yeast cells for continuous beer fermentation: a
review. J. Inst. Brew., 104, pp. 1931.
Pilkington, P.H., Margaritis, A., Mensour, N.A. Sobczak, J., Hancock, I. and
Russell, I. 1999. Kappa-Carrageenan Gel Immobilization of Lager Brewing
Yeast. J. Inst. Brew., 105(6), pp. 398404.
82
83
84
85
86
87
Published by
Author(s)
Virkajrvi, Ilkka
Title
Fermentation is the most time consuming step in the production of beer and therefore
the effective use of fermentation vessels is a crucial element in brewing economy.
One means of increasing the productivity of a batch process is to convert it to a
continuous one. Experiments in continuous fermentation emerged during the 1950s
and 1960s, but by the end of 1970s most of them had been closed down.
Immobilization technique revitalised continuous fermentation research in the 1980s
and led to industrial applications in the secondary fermentation and in the production
of low-alcohol beers.
This work demonstrated that an immobilized, continuous main fermentation is a
feasible process for production of lager beer. The immobilized main fermentation was
stable for more than 14 months both in fermentation efficiency and in aroma
compound formation. The formation of aroma compounds could be controlled by
varying the composition and amount of gas feed into the first fermentation stage. The
division of immobilized main fermentation into an aerobic and an anaerobic stage
appeared to solve problems related to yeast growth and viability.
The carrier material affected the formation of flavour compounds in small-scale
fermentations. Moreover the effect varied with the yeast strain used. The carrier
affected the economy of immobilized fermentation: the carrier cost could be as high
as one third of the investment. When a cheap carrier is used the investment cost for a
continuous, immobilized process was estimated to be only about 70% of the
investment cost of a batch process.
Keywords
beverages, beer, brewing, primary fermentation, immobilized yeasts, carriers, stability, flavours,
microbes, contamination
Activity unit
VTT Biotechnology, Process Technology, Tietotie 2, P.O.Box 1500, FIN02044 VTT, Finland
ISBN
Project number
February 2001
Language
English
Name of project
B1SU00097
Pages
87 p. + app. 50 p.
Price
Commissioned by
VTT Publications
12350621 (soft back ed.)
14550849 (URL: http://www.inf.vtt.fi/pdf/)
Sold by