Foaming of milk Hilton Deeth

Dairy Webinar 20 November 2013

Outline Significance of foaming in the dairy industry Foams – general concepts Causes of reduced foaming  Lipolysis

 Other

Ways to improve foaming Ways to reduce foaming Some recent research on foaming

Significance of foaming

Significance of foaming/frothing to the dairy industry Beneficial Hot frothing for making cappuccino coffee Cold frothing for making freddoccinos? 

Poor frothing milk is a real problem

Detrimental Cold frothing during pumping, filling vats, etc 

too much froth causes product losses

Excess froth during reconstitution of powders, e.g. infant formulae

Foams – General concepts

Foams: General Concepts A foam is a two phase system in which a distinct gas bubble phase is surrounded by a continuous liquid lamellar phase

Kamath et al., 2011

Foams: General Concepts 2 Foam formation involves the creation and stabilisation of gas bubbles in a liquid.

Measured by foamability and foam stability • •

Foamability is the amount of foam formed Foam stability is how long the foam lasts

Foams: General Concepts 3 Milk foams are stabilised by milk proteins Surface active agents reduce foaming by displacing the proteins Foaming requires air incorporation through: • Steam injection (as in cappuccino machines) • Air injection • Agitation (as in home frothers, e.g. Caffitaly, Nespresso Aeroccino • Other, e.g. pouring from one container to another

Causes of reduced foaming

Common myths about reduced foaming Added water in the milk. Adding water has little effect on frothing.

Too much or too little fat. The fat content has some effect on frothing: skim milk generally gives more froth

Due to additives in the milk. No The milk is too fresh. Refrigerated storage of pasteurised milk for up to three days has no effect on foaming.

Causes of reduced foaming Not completely known but… Lipolysis is the major cause Other reported causes include: Cow factors – stage of lactation, breed, etc The milk from some individual cows foams poorly

Free fat (ruptured milk fat globule) Mastitis

Lipolysis – what is it?  breakdown of fats (triglycerides, TG) to produce:

- free fatty acids (FFA) and - diglycerides/monoglygerides (DG/MG) (or glycerol) LIPASE

TG FFA + DG DG  FFA + MG [MG  FFA + glycerol]

Why is it a problem for foaming? FFAs and di- and monoglycerides

are surface active and reduce foaming Can replace proteins at the bubble surface

Lipolysis and steam foaming

Foamability %....

The amount of foam produced (foamability) as % of original milk volume decreases with free fatty acid level 40 35 30 25 20 15 10 5 0 0

1

2

3

4

Free Fatty Acids (mequiv./L)

5

6

Milk foaming problems LIPOLYSIS  Two major causes : Spontaneous lipolysis on farm: Initiated by cooling to < 100C

Induced lipolysis: on farm from air incorporation at teat cup cluster causing foaming of warm raw milk (due to inadequate maintenance of milking machines) Can also occur in the factory

Spontaneous lipolysis  After cooling and refrigeration for ~16 h:  “normal” milk has a FFA content of ~0.5-1.0 (mmoles per litre)  “spontaneous” milk has a FFA of > ~1.5 but can be as high as 10.  Major factors:

 cows in late lactation  cows on poor feed  certain cows/certain bulls’ progeny  all of the above

Induced lipolysis – in raw milk  Agitation - with air [causing foaming]

 Pumping – particularly with air intake  Homogenisation – very effective

In practice, is always combined with pasteurisation (~72°C/15 sec) which destroys milk lipase  Mixing homogenised (pasteurised) milk and raw milk

Ways to improve foaming

Ways to improve foaming Selection of good farm milk Heating – pasteurisation, UHT Homogenisation  Foaming

increases with pressure of homogenisation

Addition of milk solids, particularly proteins Addition of gums Adding calcium

Selection of milk  Lipolysis, the major cause of foaming

problems, occurs mostly at farm A small amount can be induced during transport

 Problem in bulk milk usually due to small

number of suppliers  Can often be narrowed down to milk from certain tanker runs, then to individual suppliers

Steam frothing values (foamability) of 12 tanker milks over 12 months Month

Tanker A

Tanker B

Tanker C

Tanker D

Tanker E

Tanker F

Tanker G

Tanker H

1 2 3 4 5 6 7 8 9 10 11 12

106 120 104 102 103 95 90 92 83 72 54 57

19 22 33 36 55 130 86 66 70 39 64 57

78 78 100 85 94 90 98 93 88 96 170

32 27 41 42 48 90 92 92 59 44 42 61

58 43 47 49 56 63 77 78 71 56 55

90 90 61 52 82

10 10 20 10 20

Aver age

90

56

97

56

59

84

17

Tanker I

Tanker J

Tanker K

Tanker L

120 70 107

11 17 19 23 28

56 47 69 72 113

43 70 108 90 107

104 89 85 83 71

115 78 80 83 71 76

66 37 70 79 93 100 102 70 72 78 84 54

110 98 95 53 72 71

83 73 108 90 107 80 110 98 95 53 72 71

79

84

76

83

87

FFA and steam frothing values of individual farm supplies in tanker G Farm supplier

FFA

Steam frothing value

G1 G2 G3 G4 G5 G6 G7

1.56 1.16 0.84 1.40 6.47 2.76 3.95

30 80 100 60 0 1 0

Effect of heating and homogenisation on SFV Milk

Raw Pasteurised Pasteurised and homogenised

Steam frothing values (average of 5 replications) 43 86 125

Homogenisation pressure Steam frothing values (MPa) (average of 10 replications)

Deeth and Smith, 1983

6.9

92

13.8

112

20.6

124

Steam frothing values of bulk raw and corresponding pasteurised milks 140

120

Pasteurised

100

80

SFV

Raw

60

40

20

0 1

2

3

4

5

6

7

MONTH

8

9

10

11

12

Effect of heating and homogenisation on SFV Milk

Raw Pasteurised Pasteurised and homogenised

Steam frothing values (average of 5 replications) 43 86 125

Homogenisation pressure Steam frothing values (MPa) (average of 10 replications)

Deeth and Smith, 1983

6.9

92

13.8

112

20.6

124

Steam frothing values of bulk raw and corresponding pasteurised milks 140

120

Pasteurised

100

80

SFV

Raw

60

40

20

0 1

2

3

4

5

6

7

MONTH

8

9

10

11

12

Proteins and foaming of lipolysed milk  Milk foams are stabilised by proteins at the air-liquid interface  Lipolysis produces surface-active lipids  Suface-active lipids compete with proteins and reduce foaming  Addition of milk proteins improves foaming

Effect of adding SMP on steam frothing value 160 140 120 100

SFV

80

milk a milk b

60 40 20 0 0

1

2

3

% SMP added

Deeth and Smith, 1983

4

5

Effect of adding κ-carrageenan on foam stability 600

Half life (mins)

500

400

300

200

100

0

0

10

20

30

40

50

60

70

80

90

Temperature (°C) Control

0.0001

0.0003

Adding 0, 0.01% and 0.03% κ-carrageenan to reconstituted low-heat skim milk powder Kamath and Deeth, 2011

Effect of adding calcium chloride on foam stability 800 700

Half-life (mins)

600

500 400 300

200 100

0 0

10

20

30

40

50

60

70

80

90

Temperature (°C)

Control

10 mM

15 mM

20 mM

Adding 0, 10, 15 and 20 mM calcium chloride to reconstituted low-heat skim milk powder 20 mM addition not recommended – risk of coagulation Kamath et al., 2011 (JDS)

Ways to reduce foaming

Possible ways to reduce foaming of dry formulations Avoid using milk powders forming very stable foams at and below the reconstitution temperature Add milk fat? Add calcium-binding agents, e.g. citrate, polyphosphate

Foamability of reconstituted low, medium & high heat skim milk powder 110

Initial foam volume (mL)

100

n=3

90 80 70 60 50 40 30 20 10 0 0

20

40

60

80

100

Temperature (oC) low heat (Mfd: Dec 2003)

low heat (Mfd: Sep 2004)

medium heat (Mfd: Sep 2004)

high heat (Mfd: Jan 2005)

So low-, medium- and high heat SMP have same foamability

Foam stability of low, medium and high heat skim milk powders 600 Half Life (mins)

500

Possible mixing temp

n=3

400 300 200 100 0 0

10

20

30

40

50

60

70

80

90

Temperature (oC)

low heat (Mfd: Dec 2003) medium heat (Mfd: Sep 2004)

low heat (Mfd: Sep 2004) high heat (Mfd: Jan 2005)

BUT low-, medium- and high-heat SMP have different foam stabibilities

Practical implication of foam stability of reconstituted powders  Different skim milk powders have similar

foamability but vary considerably in foam stability  Mixing often done at 40-50°C corresponds to high foam stability of some powders

 Medium-heat powders should be avoided but individual batches should be tested

Effect of adding 1.4% soft milk fat and oils to reconstituted SMP- foam stability

Is result of one trial; may not be same for all batches of milk fat

Foam stability depressing effect of calcium-binding agents a

c

b

300 200 100 0 0

20

40

60

80

100

400

350 300 250 200 150 100 50 0

Half-life (mins)

Half-life (mins)

Half-life (mins)

400

5mM

200 100

0

0

20

40

60

80

100

0

10mM

Control

5mM

20

40

60

80

100

Temperature (°C)

Temperature (°C)

Temperature (°C) Control

300

10mM

Control

5mM

10mM

20mM

a. Addition of EDTA b. Addition of sodium citrate c. Addition of sodium hexametaphosphate [Note: Some UHT milks have added citrate or SHMP ]

Kamath, 2007

Some recent foaming research

Recent UQ Research on foaming  Done by Dr Sapna Kamath  Set up foaming apparatus in lab which uses

compressed air  Used for foaming at different temperatures  Determined : Foamability – amount of foam produced measured immediately after foaming Foam stability – time before amount of foam remaining decreases to 50% of original volume

Apparatus for Measurement of Foamability and Foam Stability Room temperature: 22oC Air pressure: 5-6 psi Air flow rate: 2.4 mL/s

air inlet

glass tube pressure regulator sintered glass disc

pressure gauge

milk (50 mL) flow meter Low form 250ml measuring cylinder

Kamath, 2007

Initial foam volume (mL)

Foamability of commercial milk samples 100

80 60 40 20

0 0

10

20

40

50

60

70

80

Temperature (Deg C) Full Cream milk

Kamath et al., 2008

30

Lite White

Skim milk

UHT full cream

UHT Skim

90

Foam stability of commercial milk samples

Half life (mins)

600 Cappuccino temperature

500 400 300 200 100 0 0

10

20

30

40

50

60

70

80

Temperature (o C) Full Cream milk

Lite White

Skim milk

UHT full cream

UHT Skim

90

Image analysis of foams Whole milk at frothing

Skim milk at frothing

Whole milk at frothing half -life

Skim milk at frothing half-life

Effect of milk fat and vegetable oils 100

pasteurised homogenised milk (3.4% fat)

Foamability (mL)

80

pasteurised homogenised milk (1.4% fat)

60

1.4% olive oil 3.4% olive oil

40 1.4% canola oil 1.4% sunflower oil

20

0 0

20

40

60

Temperature (°C)

Kamath and Deeth, 2011

80

100

Effect of adding oil to reconstituted SMP- foamability 140

Foamability (mL)

120

100 80 60 40 20 0

Kamath, 2007

Effect of adding oil to reconstituted SMP- foam stability 350 300

Half-life (mins)

250 200 150 100 50 0

Bubble ghost analyses  Bubble ghost material is the interfacial

material

 It remains after foam subsides  Is mostly micellar casein  Electron microscopy shows micelles

aggregated and spread over surface

Foam bubble surface – electron micrograph (em) INTERFACE

CASEIN MICELLES

Kamath et al. 2011

Bubble ghost em analyses Interfacial Membrane

Spreading and aggregation of casein micelles

Bubble ghost em analyses

Spread casein micelles

Interfacial material

Caseins are more effective than whey proteins in stabilising milk foams Casein micelles have ability to aggregate and spread over the bubble surface Whey proteins are very mobile and move quickly to the surface and create foam but are unable to spread and aggregate and stabilise the foam.

Caseins are responsible for foam stability - more evidence

Half life (mins)

800

casein

600 ultracentrifugal supernatant 400 defatted ultracentrifugal supernatant 200

milk 0 0

20

40

60

Temperature (oC)

80

100

Acknowlegements  Sapna Kamath  Bussarin Samarnphanchai  Ross Smith  Carolyn Fitz-Gerald  Trent Seeto  Erika Naranjo Martinez  Dairy Australia

Thank you for your attention