Flow and Filtration: The Physics of Brewing

1
Flow and Filtration The Physics of Brewing

• Dr. Alex Speers
• Department of Food Science and Technology
• ltAlex.Speers_at_dal.cagt

2
Outline

• Introduction
• Brewing gums
• shearing
• Methods
• Rheometry
• Filtration
• Summary

3
Why study ?-glucans?

• Cause processing problems in brewing
• Under-modification of barley endosperm
• High viscosity of wort and beer
• Slow runoff of wort and beer
• Haze formation in packaged beer
• Clogging of membranes
• Increased production cost

4
Localization of barley ?-glucans
Structure of a barley kernel
5
Beta-Glucan and Arabinoxylan Content of Selected
Beers (ug / ml)
6
Chemical structure of barley ?-glucans
Unbranched chains of ?-D-glucopyranose residues
O
O
O
O
O
O
?-(1?4)- linkage ?-(1?3)- linkage
7
Chemical structure of arabinoxylans
8
Localization of gums

• Deposited mainly in in endosperm cell walls
• Barley endosperm cell walls contain
• 20 arabinoxylans
• 70 ?-glucans
• Barley aleurone cell walls contain
• 65-67 arabinoxylans
• 26-29 ?-glucans
• Beta-glucan content
• barley 0.14 - 8.9
• wort/beer 12 - 940 mg/L

9
Non-Fermentable Brewing Gums

• Defined as Non Starch Polysaccharides Gums - warm
  water extractable
• Tend to viscosify wort and beer
• Thus, add body/foam stability
• In the distant past - not a problem
• With advent of membrane filters, tight production
  schedules lighter beer
• Pose problems in some breweries some times

10
Beta-Glucan fringed micelles
11
Micelle-like Aggregation
12
Methods
13
Rheological Definitions

• Science of deformation and flow
• Three important terms are shear rate (?), shear
  stress (?) and viscosity (?) - note different
  symbols used.

?
V, F
h
??? V/h, ?? F/A ?????
14
Calculation Example

• Shear rate if dV 1 cm/s and h 1 cm?
• Shear rate 1cm/s 1 cm 1 /s
• Shear rate units /s or s-1
• Shear stress if F 0.001 N and A 1 m2 ?
• Shear stress 0.001 N/ m2 1 mPa
• Viscosity 1 mPa s

15
Shear stress/shear rate measurement rotational

• RPM -gt shear rate
• Torque -gt shear stress
• Viscosity shear stress/shear rate

16
Rheometry

• Cone and plate and coaxial fixtures

17
Shear stress/shear rate measurement pipe flow

• Flow rate -gt shear rate
• Pressure loss -gt shear stress
• Viscosity shear stress/shear rate
• Best suited for measuring Newtonian flow
  behaviour.

18
Rheometry

• Capillary viscometer

19
Rheometry

• Viscomat

20
Viscosity Dependence

• Temperature h A e DE/RT
• Concentration (gums,oP, Etoh)
• Shear rate
• Shear history

21
Shear effects
22
Shear effects
23
Non-Newtonian Flow

• Found at high gum concentrations

24
Rheological Notes

• Normally viscosity properly defined as apparent
  viscosity - mPa s ( cP),
• Kinematic viscosity is apparent viscosity divided
  by density (Stokes)
• (Misleading terms in literature),
• 1 mPa s is 1 cP viscosity of water at 20oC,
• Apparent viscosty depends on density,
  temperature, shear rate and shear history.

25
Rheological Notes

• Intrinsic Viscosity h
• Based on extrapolated Specific viscosity (h/ h s
  -1)/c -gt0
• Can be used to determine shape of polymer based
  on molecular weight
• h K M a

26
Effect of Concentration
Determination of C with 327 kDa b-glucan in a
control buffer
27
Early Results

• Using 327 kDa b-glucan at 50 g/L,
• ethanol (0-7), maltose (0-15) and
• pH (3.6-5.2)
• Viscosities were significantly
• different (Plt0.05).

28
Variation of h and C of b-glucan solutions
Treatment pH maltose ethanol h C
() () (mL/g) (g/L)
High ethanol 4.1 0.5 6.0 464 6.47 Low
ethanol 4.1 0.5 4.0 812 2.72 Control 4.1
0.5 5.0 815 3.11 High maltose 4.1
0.8 5.0 806 2.13 Low maltose 4.1
0.1 5.0 862 3.05 Low pH 3.6 0.5
5.0 741 3.95 High pH 4.5 0.5 5.0 827 3.05
29
Why Sporadic?

• Depends on crop year
• Stressed plant tends to more b-glucan (Kendall)

30
Why Some Breweries?

• Depends plant equipment
• Depends on process
• Possibly due to differences in shearing of wort
  beer

31
Brewing Shear Rates?

• Turbulent or laminar?
• NRE ???V L/ ?
• ??? density, V velocity L diameter ?
  viscosity
• Average shear rate in turbulence
• ? (?/?)3 / ?1/4
• ? average power dissipation per unit mass

32
Brewing Shear Rates?

• Turbulent or laminar?
• Turbulent flow cascades to laminar flow at small
  distance scales

33
Brewing Shear Rates

• Defined by Reynolds number of 2000-3000
• Note Re DVr/h
• Also note V is the average pipe velocity
• Generally get turbulent flow

34
Brewing Shear Rates

• Shear in Kettle 8600 s-1
• (Speers et al. 2002)
• Shear in Fermenter 20-60 s-1
• (Speers Ritcey, 1995)
• Shear in Yeast brink tank lt15 s-1
• (Kawamura et al. 1999)
• Average shear rate in pipe flow
• High 915 s-1
• Mean 500 s -1
• Low 175 s -1

35
Membrane filtration

• Theory developed in 30s
• Based on capillary plugging due to gradual
  restriction in diameter
• Surdarmana et al. 1996 Tech Quarterly
• t/V t/Vmax 1/Qinit
• Vmax maximum filtrate volume
• Qinit intial flow rate

36
Membrane filtration

• Theory developed in 30s
• Based on capillary plugging due to gradual
  restriction in diameter
• Surdarmana et al. 1996 Tech Quarterly
• t/V t/Vmax 1/Qinit
• Vmax maximum filtrate volume
• Qinit intial flow rate

37
Filtration Apparatus
38
Example Sudarmana Transform

• Medium viscosity arabinoxlyan in model beer

39
Relation of Intrinsic Viscosity and Filtration

• 1/Vmax a h for membrane test
• Filterability negatively correlated with h for
  commercial (DE) filtration
• Membrane filtration more suited for detection of
  b-glucan problems

40
Conclusions

• Ethanol, pH and maltose effect viscosity
• Shear strong effect on filtration
• Shear within brewery typically turbulent average
  40-1250 s-1
• Sudarmana fit works (Tech. Quart 3363)

41
Acknowledgments

• Students !
• NSERC
• Labatt Brewing RD
• NSDAM
• Westcan Malting
• Canada Malting
• Pfeuffer GmbH and Profamo Inc (Viscomat automated
  capillary rheometer)