A stable platform for control of

fermentation

Chris Boulton

University of Nottingham

Mikkel Nordkvist

Alfa Laval

Current trends in fermentation

Current Trends in fermentation

• Bigger batch sizes

• High or very high

gravity worts

• Use of higher

temperatures

• Limited number of

serial fermentations

Current fermentation control

• Generally adequate

control of primary

parameters

– Initial extract

– Pitching rate

– Oxygenation

– Temperature

Problems to resolve

Some current problems

• Management of yeast stress – Sustainable serial re-pitching

– Management of yeast dispersion

• How to ensure short and consistent cycle times

• Predictable manipulation of yeast-derived beer flavour compounds

• Reliable markers of progress – Lack of suitable on-line sensors

Yeast dispersion and fermentation

performance

• Natural mixing of vessel contents is poor

– Spatial heterogeneity throughout most of

fermentation

– May hinder transport of yeast metabolites

– Off-line sample analysis may not accurately

reflect actual conditions in vessel

– Uncertainty of when crop forms

Provision of mechanical mixing

remedies man of these problems Rotary jet head (ISO-MIX A/S)

Flow meter

Variable speed

pump

Rotary mixing

head suspended

in wort

Initiation of yeast growth

(single pitching)

Yeast budding index

(% budding cells) for

stirred and unstirred

fermentations

More cells initiate

budding at same time

Synchronicity

maintained for longer

Ensures consistent

start point

Iso-Mix

Un-stirred

End of fermentation

• Essential to manage

yeast stress

• Remove crop as soon as

possible

• Very rapid yeast

sedimentation when loop

switched off

• Predictable and rapid

crop formation

IsoMix fermentation

Control

Loop off at 85h

Effect on process time in different breweries

26%

48%

17% 16%

20%

40%

20.20%

15% 17%

Brewery A, brand 1, 16.3 °P

Brewery A, brand 2, 18.5 °P

Brewery B, 13.7 °P

Brewery C, 15.0 °P

Brewery D Brewery E, 15.2 °P

Brewery F, 15.0 °P

Brewery G, 13.9 °P

Brewery H

Time reduction

Fermentation consistency

• Lager fermentation (18.5 P).

• 5000hl ccvs

• Mixing by Iso-Mix system ca. 250 hL/h

0

5

10

15

20

25

30

1 2

Tim

e t

o c

oo

l (d

ay

s)

Unmixed

controls

Mixed n = 18

02

46

810

1214

1618

0 4 8 12 16 20 24 28

Time (h)

Unmixed

Mixed

Tem

p (

oC

) Improved attemperation

• Crash cooling in 1800 hl conical with or without mixing

• Mixing by a single IM 20 RJH operated at ca. 250 hL/h

Ca.12h time saving

Influence of vessel filling

• Multi-filling of large vessels

• Prolonged fill times (up to 24h)

• Requires more decisions

– When to pitch

– When to oxygenate

• When does fermentation actually commence?

Options for pitching and

oxygenation during collection

3

4

5

1

2

6

Multi-filling with 6

brewlengths

Effect of prolonged pitching time

• Early pitched yeast out-competes late

pitched yeast for nutrients

• Produces heterogeneous population with

differing physiological condition

Rate of oxygen uptake by pitching yeast

0

2

4

6

8

10

12

14

16

18

0 0.5 1 1.5 2 3

Time after pitching (h)

ΔDOT

mg/L/min

Late pitched yeast has low

uptake rate

Effect of prolonged fill times on VDK

• Long fermenter fill with multiple batches of

wort

– Fed-batch system where fresh supply of group A

amino acids prevents uptake of Group B amino

acids

– Pushes VDK peak towards right

VDK profiles of 1500 vs 3000 hl high gravity lager

fermentations

All yeast pitched with

1st brewlength

Collection times 10

and 18h, respectively

Identical worts,

pitching rates and

oxygenation regimes

Effect of collection on beer volatiles

• Exposure time of yeast to oxygen during

vessel fill can be used to modulate ester

synthesis yeast

• Acts via repression of ATF genes by

oxygen

Experimental conditions

• 3 x 10 litre wort

fermentations

• Identical conditions

• All yeast pitched at start

Control – all wort added at

zero time

Trial 1 – 3 batches of wort

added over 12h

Trial 2 – 5 batches of wort

added over 24h

0

2.5

5

7.5

10

12.5

15

17.5

20

22.5

-25 0 25 50 75 100 125 150

Co

nce

ntr

atio

n (

pp

m)

Fermentation time (hours)

Effect of filling on ethyl acetate

model

3 batches 12 hours

5 batches 24h

0

0.25

0.5

0.75

1

1.25

1.5

1.75

2

2.25

2.5

-25 0 25 50 75 100 125 150

Co

ncen

trati

on

(p

pm

)

Fermentation time (hours)

Effect of filling on isoamyl acetate

model

3 batches 12 h

5 batches 24h

Bilverstone, et al., WBC Oregon, 2013

Improved monitoring of

fermentation

• Identification of key

stages in fermentation

still reliant on sampling

and off-line analysis

• Can be a cause of

prolonged cycle times

• Pumped loop system an

ideal location for suitable

in-line probes

Automatic in-line measurements

• VitalSensors Technologies (Denver, USA)

• Based on attenuated total reflection

sampling technique using mid-infra red

(MIR)

• 3 channels which can be calibrated for

ethanol, extract and CO2

Principle of sensor • Attenuated total reflection MIR sensor

• Ethanol channel

• Sugar channel

• CO2 channel

Installation

• Working volume ca.

5000 hL

• Sensor at ca. 2030 hL

close to sample port

• Communication with

Sensor Management

Station via Ethernet

• Integrated into brewery

control system

Sensor

Sample port

Sensor Management

Station

Results – extract and ethanol

Ethanol

Extract

0

1

2

3

4

5

6

7

8

9

0

2

4

6

8

10

12

14

16

18

20

0 24 48 72 96 120 144 168 192 216 240 264

Alc

oh

ol (

w/w

%),

fill

volu

me

(10

00 h

L)

Re

al e

xtra

ct (

P)

Time (h)

10T fermentation 1Extract sensor

Extract lab

Ethanol sensor

Ethanol lab

Volume

5000Hl

18.5oP

Pumped loop system

Opportunities for vessel design

and operation

In-line

monitoring

Pitching

Oxygen

(ester fine

tuning)

In-line

monitoring

Early warm cropping

reduces need for

in-tank crash cooling

In-line

cooling

Stabilising

agents

In-line

monitoring

Suitable for dual or

uni-tanking operation

In-line

cooling

Stabilising

agents

Thank you