Bubble structure

formation in bread and

cakes

Martin WhitworthBritish Society of Baking Spring Meeting,

10th April 2019

Bread and cake structure

• Bubbles are a key aspect of the

product structure.

• Important for:

– Product volume (~75-80% air)

– Softness

– Whiteness (diffuse scattering)

• Recipe, ingredient and process

variations can be used to achieve a

wide range of structures for different

product styles.

Bread structure

• Expanded foam structure

• Bubble nuclei created in mixer

• Expanded by gas production in

proof and baking

• Converted to set sponge structure

during baking

Factors affecting structure formation

• Bubble nuclei

– Mixing

• Gas production

– Proof

• Gas retention

– Proof and baking

Gas volume measurement

• Calculated from

dough density

• Buoyancy method

• Dynamic

measurement for

doughs proved in

warm oil.

Gas volume production

• Differences in

initial gas

entrainment in

mixer

0

20

40

60

80

100

120

140

160

180

200

0 5 10 15 20 25 30 35 40 45

Gas

vo

lum

e /

solid

do

ugh

vo

lum

e (%

)

Time (minutes)

0

20

40

60

80

100

120

140

160

180

200

0 5 10 15 20 25 30 35 40 45

Gas

vo

lum

e /

solid

do

ugh

vo

lum

e (%

)

Time (minutes)

0

20

40

60

80

100

120

140

160

180

200

0 5 10 15 20 25 30 35 40 45

Gas

vo

lum

e /

solid

do

ugh

vo

lum

e (%

)

Time (minutes)

0

20

40

60

80

100

120

140

160

180

200

0 5 10 15 20 25 30 35 40 45

Gas

vo

lum

e /

solid

do

ugh

vo

lum

e (%

)

Time (minutes)

2% yeast

5% yeast

Tweedy mixer

Spiral mixer

1.5 bar

Final mixer pressure

1.0 bar

0.5 bar

Effect of mixer headspace

pressure

• Lower final

pressure gives

less gas entrained

in dough.

• This results in

finer bread

structure.

Low

pressure

High

pressure

Measurement of bubbles

in dough

• Samples frozen in liquid nitrogen

• Microscopy

– Imaging of cross-sections

• X-ray micro CT

– Non-destructive 3D imaging

Effect of mixer pressure

(a) (b)

0.5 bar 1.5 bar

Bubbles in freshly mixed dough

Bubble size distribution - effect of

pressure• Similar range

of sizes

• More bubbles

for higher final

mixer

pressure.

Mixed dough

0

50

100

150

200

250

300

350

400

450

Bubble diameter (mm)0.01 0.1 1 10

1.0, 0.5 bar

1.0 bar

1.0, 1.5 bar

Num

ber

of

bubble

s / m

m3/log

10(b

in w

idth

)

Measurement of bubbles in bread

• …

X-ray micro CT

Cell

are

a (

% o

f sli

ce a

rea)

Cell diameter (mm)

C-Cell bread imaging system

– Image analysis of bread slices

Bubble size distribution - effect of

pressure• Many bubbles

lost during

processing

• More bubble

nuclei

coarser

bread

Mixed dough

0

50

100

150

200

250

300

350

400

450

Bubble diameter (mm)0.01 0.1 1 10

1.0, 0.5 bar

1.0 bar

1.0, 1.5 bar

Num

ber

of

bubble

s / m

m3/log

10(b

in w

idth

)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Bread

(N.B. different y

scale)

Mechanisms of bubble loss

• Damage

– e.g. during moulding

• Ostwald ripening

– Large bubbles grow

preferentially

• Coalescence

– Bubbles merge due to rupture

of walls between them

Freshly mixed7 minutes14 minutes

50µm

Why do more bubble nuclei give

coarser structure?More bubble nuclei Fewer bubble nuclei

Thinner walls Thicker walls

Same gas

volume

Greater likelihood of

coalescence

Stable expansionCoarser eventual

structure

Uniform structure

Hypothesis

Bubble structure during proof and baking

• X-ray CT scanning

• Oven placed inside scanner

Structure development during bread

production

• Gas retention

important

• Determined by

flour quality and

dough

development

Breadmaking flour Non-breadmaking flour

0

20

40

60

80

100

120

140

160

180

200

0 5 10 15 20 25 30 35 40 45

Gas

vo

lum

e /

solid

do

ugh

vo

lum

e (%

)

Time (minutes)

Control DATEM Lipase

Effect of lipid ingredients

aaa

Funded by DSM and CSM

Tweedy

Spiral mix

Control DATEM Lipase

Effect of lipid ingredients

• Effects of Datem and

lipase apparent

during baking.

• They help stabilise

bubbles at this stage.

Tweedy

Spiral mix

Funded by DSM and CSM

Dough moulding

Dough moulding: sheeting

Dough moulding: curling

Air trapped

between layers of

dough sheet

Elongated

bubblesRows of bubbles

within dough

sheet

Single piece moulding effects

Start of proof End of proof Bread

Single piece moulding effects

Start of proof End of proof Bread

Four-piece moulding

• sss

Time

Cell elongation in a 4-piece loaf

• C-Cell measurements

• Varies with slice position

Horizontal

elongation

within pieces

Join between

pieces

Radical bread processCampden BRI patented process

• Combine ingredients to an

underdeveloped dough

• Dough lamination

• Cutting and orientation of dough

pieces in pan

• Proof, baking and cooling

Benefits

• Finer structure

• Increased softness

• Increased volume

CBP control Radical process

C-Cell contrast:

0.7895 ± 0.005 0.832 ± 0.008

153 ± 24 g 116 ± 18 g

4.21 ± 0.07 ml/g 4.36 ± 0.08 ml/g

Cake structure during baking

High ratio yellow cake• Oven temperature = 180°C

• Baking time = 47 minutes

0.1

0.2

0.3

0.4

0.5

0.7

0.6

Relative

attenuation

0.8

TemperatureStart

• Cold, dense batter

0-20 minutes

• Convection

25-35 minutes

• Temperature gradient

• Low density zone

moves upwards

40-45 minutes

• Contraction

45 minutes

• End of baking

Temperature /°C

20

30

40

50

60

70

80

90

100

Measured with thermocouples 20mm from the imaged plane

40 39 42 24

27 25 26 25

0 minutes

29 29 27

39 39 43

5 minutes

43 40 38

48 45 47

10 minutes

62 51 52

59 49 54

76 65 76

15 minutes

78 68 75

72 59 69

78 65 76

20 minutes

91 89 91

87 83 86

86 76 82

85 77 85

25 minutes

99 98 99

96 95 95

92 90 90

88 83 86

30 minutes

102 101 102

100 100 100

97 98 96

93 91 91

35 minutes

102 102 103

102 101 101

100 100 99

97 95 95

40 minutes

102 102 103

102 102 102

101 101 99

98 96 97

45 minutes

Pressure in a high ratio cake

• Foam to sponge

conversion occurs

as the temperature

reaches ~95°C,

which is the starch

gelatinisation

temperature in this

system.

20

30

40

50

60

70

80

90

100

110

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

-10 0 10 20 30 40 50 60

Tem

per

atu

re (º

C)

Pre

ssu

re (

kPa)

Time (minutes)

Pressure

Temperature

Batter added to pan

End of baking

Sponge

• aaa

0.1

0.2

0.3

0.4

Relative

attenuation

• 190°C

• 20 minutes

Muffins

• aaa

0.1

0.2

0.3

0.4

0.5

0.7

0.6

Relative

attenuation

0.8

0.9

1.0

1.1

• 180°C

• 24 minutes

Model for formation of tunnel holesMuffin

• Bubbles are large enough to span a

steep viscosity gradient.

• High viscosity at one end of the

bubble prevents migration.

• Low viscosity at the other end

facilitates growth,

• This results in tunnel holes.

• Hypothesis: Tunnel holes form if the

rate of bubble growth is similar to

the speed of a setting front.

Growth

Immobile

Tunnel holes

Fruit cakesIncreased water

Increased water,

no tartaric acid

Fruit sinking

Control

Conclusions

• Methods such as X-ray tomography enable us to study the

mechanisms of structure formation in bread and cakes.

• Bread:

– The number of bubble nuclei formed in mixing is critical;

– Gas retention during proof and baking depends on flour quality

and dough development;

– Ingredients such as Datem and Lipase have their effect during

baking.

• Cakes:

– The structure depends on the balance between the rates of

bubble growth and setting. Tunnel holes occur when the rates are

similar.