Melt-Down: Effects of

Microstructure and Rheology

R W Hartel
Frozen Dessert Center
University of Wisconsin-Madison
 No-Melt Ice Cream?
• Japanese ”no-melt” ice cream

https://youtu.be/GFE91TTJfN8
 Ice Cream Sandwiches Don’t Melt?
• Walmart ice cream under scrutiny when
Cincinnati mom says it doesn’t melt.

What causes that?

Walmart says:
“Ice cream melts based on the ingredients including cream. Ice
cream with more cream will generally melt at a slower rate,
which is the case with our Great Value ice cream sandwiches.”
 Factors that Influence Meltdown
• Heat transfer
• Overrun, number and size of air bubbles
• Outside temperature, convective factors
• Ice content
• Thermal diffusivity – insulation effect
• Viscosity of serum phase
• Diluted by melted ice
• Gravity
• Ability of serum phase to flow 1
2

• Fat globule clusters

3

• Number and size 3

3
4
1

• Polyphenols
4
1

3
4
 Ice Cream Melting

Not all ice creams are created equal – or melt in the same way

Brand 2
Brand 1
 Melt-Down Test
• Ice cream sample on screen at room temperature
• Melts and drips through the screen
• Parameters measured
• Induction time
• Fastest melting rate
• Final drip-through weight
• Height change over time
• Analysis
• Remnant foam
• Dripped fluid

Koxholt et al. (2001)

 Melt-down Test
• Weight measurement • Height measurement
100 100
90 90
80 0%ST, 0%PS80, 50%OR 80

Height change (%)

Drip-through (%)

70 70
0.2%ST, 0.015%PS80, 75%OR
60 60
50 0.4%ST, 0.03%PS80, 100%OR 50
40 40
30 30
20 20
10 10
0 0
0 40 80 120 160 200 240 280 320 360 0 20 40 60 80 100 120
Time (min) Time (min)

Note: ST: stabilizer

PS80: polysorbate 80
OR: overrun
 Low Fat Destabilization,
Full Collapse and Drip-Through
1 - Ice crystals - Fat/destabilized fat
-Free water
2 - Air cells
-Serum phase

2
1
1
2

2 1

t = 0 minutes t = 60 minutes t = 70 minutes

 High Fat Destabilization,
Minimal Collapse
1 - Ice crystals - Fat/destabilized fat
-Free water
2 - Air cells
-Serum phase

2
1
1
2

t = 0 minutes t = 60 minutes t = 120 minutes

 Commercial Products 638
2 652

1.8 638 159

957
1.6 286
Drip-through Rate (g/min)

1.4 423
106 913
1.2 559

1 313
472
0.8
106

0.6 824
215
0.4
293 727
0.2 880
171
0
603
0 10 20 30 40 50 60
293
Fat Destabilization (%)
Warren & Hartel (2014)
 Objectives
• Evaluate effects of microstructure (air, ice, fat,
serum) on rheology of melting/melted ice cream and
melt-down attributes

Ingredients Sensory

Melt-down
Formulation Microstructure
Texture
Process
Conditions
Rheology
 Experimental Design
Mix formulation
(FPD of -2.72 ± 0.06°C)
• Serum phase viscosity è stabilizer level
Formulation (%)
• Mix viscosity (aged ~ 24h) measured at 50s-1
Fat 12.0
• Overrun è air flow in continuous freezer Milk Solids Nonfat 11.3
- Overrun was measured before collecting the ice Sucrose 16.9
cream in the container (473mL)
MDG 0.15
• Fat destabilization è polysorbate 80 level *Total Solids ~40.5
- 0 to 0.03% * Total solids varied as stabilizer
and PS80 levels varied.

Overrun(OR) Stabilizer(ST) Emulsifier(PS80)

50% 0.0% 0.000%

75% 0.2% 0.015%

100% 0.4% 0.030%

Mix viscosity Fat destabilization

 Results
• Fat destabilization
80

70
Induction time (min)

60

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80
Fat destabilization (%)

Ice cream with high extent of fat destabilization promoted a high yield stress,
which increased the resistance of water to flow against gravitational force.
Wu et al., 2019
 Results
• Fat destabilization
2.0
1.8
Drip-through rate (g/min)

1.6
1.4
1.2
1.0
0.8
0.6 0% stabilizer
0.4
0.2% stabilizer
0.2 0.4% stabilizer
0.0
0 10 20 30 40 50 60 70 80
Fat destabilization (%)
Without stabilizer, drip-through rate greatly depended on how the large fat clusters
collided with each other and resisted the drainage of serum phase.
When increasing mix viscosity to a certain degree, it became the dominant
factor in the drip-through rate instead of fat destabilization extent. Wu et al., 2019
Results
• Fat destabilization
100

90
Final drip-through weight (%)

80
0% stabilizer
70

60

50 0.2% stabilizer

40

30
0.4% stabilizer
20

10

0
0 10 20 30 40 50 60 70 80
Fat destabilization (%)
When the apparent viscosity of serum phase was low, fat clusters could readily collide with
each other during the drainage and maintained the foam structure on the mesh.
When the mix viscosity increased to 0.09 Pa • s or even 0.28 Pa • s, the mobility of fat clusters
was restricted by the viscous serum phase. Wu et al., 2019
 Results
• Fat destabilization
0.045

0.040
Height-change rate (cm/min)

0.035

0.030

0.025

0.020 0% stabilizer
0.015 0.2% stabilizer
0.010

0.005
0.4% stabilizer
0.000
0 10 20 30 40 50 60 70 80
Fat destabilization (%)
When ice cream contained a higher degree of fat destabilization at 0% stabilizer level, the foam
structure was held by the network of fat clusters and air cells, and resulted in less variation in height.
With the presence of stabilizers, the melted ice cream structure collapsed slowly due to the low mobility
of serum phase, which led to slow change in height. Wu et al., 2019
Results
• Fat destabilization

Effect of fat destabilization extent on shape retention has been suggested by Tharp and
others (1998), Bolliger and others (2000), Warren and Hartel (2018).
In this study, a positive correlation was found between fat destabilization and final height
across all ice creams (r=0.8571, p<0.0001). Wu et al., 2019
 Results
• Mix viscosity
2.0
Drip-through rate (g/min)

1.6

1.2

0.8

0.4

0.0
0.00 0.05 0.10 0.15 0.20 0.25 0.30
Mix viscosity (Pa • s)
Low viscosity of serum phase with a good mobility drained rapidly through the lamella,
resulting in a high drip-through rate.
In contrast, ice cream with high mix viscosity, above about 0.8 Pa • s (50-1 shear rate),
had a limited mobility for drainage and led to a low drip-through rate. Wu et al., 2019
Results
• Overrun

• There was no significant relationship between overrun and

height-change rate (r=-0.3187, p=0.1051), although a trend
of decreased height-change rate with increasing overrun was
seen in the ice creams without stabilizer (r=-0.8060,
p=0.0087).

• The correlation of overrun and final height was only seen

in ice creams without added stabilizer (r=0.9518, p<0.0001).
 Rheological Measurements
• Oscillatory Thermo-Rheometry (OTR) from
-15 to 25°C (Adapted from Wildmoser et al.,
2004 and Granger et al. 2005)
• Creep and Recovery Test at 0°C (adapted
from Steffe, 1996, and Dogan et al., 2013)
• Stress Growth Test at 0°C (Modified from
Elliott and Ganz, 1977 and Rao, 2007)
• Flow Ramp Test at 0°C
Oscillatory Thermo-Rheometry (OTR)

Angular frequency of 10 rad/s with strain 0.01% from

-15 to 0°C and 0.1% from 0 to 25°C
 Oscillatory Thermo-Rheometry (OTR)
Zone   Zone  II Zone  III
1.E+07
I
G’ at 0°C è Ln(G’0°C);
1.E+06 S1P1O1 G' R2 = 0.77
1.E+05 S1P1O1 G"
Term Estimate Prob>|t|
G' and G" (Pa)

S3P3O3 G'
1.E+04 Intercept 4.50 <.0001*
S3P3O3 G"
1.E+03 ViscPas 2.30 0.0367*

1.E+02
FD 0.0335 <.0001*
OR 0.004 0.4740
1.E+01

1.E+00
-15 -10 -5 0 5 10 15 20 25
Temperature (°C)

Behavior of storage modulus, G’, and loss modulus, G”, during

oscillatory thermo-rheometry of samples S1P1O1 and S3P3O3.
 Temperature at Tanδ Peak [TTan (δ) Peak]

OTR
TTan (δ) Peak è TTanδ Peak; R2 = 0.87

Term Estimate Prob>|t|

Intercept -4.183 <.0001*
Visc -8.170 <.0001*
FD -0.031 0.0001*
Zone  I Zone  II Zone  III
1.60 OR 0.0018 0.7690
1.40
1.20
1.00 Tan (δ) = G”/G’
Tan(Delta)

0.80
0.60 S1P1O1 Tan(delta)
S3P3O3 Tan(delta)
0.40
0.20
0.00
-15 -10 -5 0 5 10 15 20 25
Temperature (°C)

Behavior of Tan (δ) during oscillatory thermo-rheometry for samples S1P1O1 and S3P3O3.
 Structure and Rheology
10000.0 10000.0

1000.0 1000.0
G’0°C (Pa)

G’0°C (Pa)
100.0 100.0

G’at 0°C G’at 0°C

10.0 10.0

1.0 1.0
0.000 0.100 0.200 0.300 0 10 20 30 40 50 60 70 80
Mix Viscosity at 50 s-1 (Pa*s-1) Fat Destabilization Extent (%)
 Stress Growth Test at 0°C

Stress growth (increasing stress at constant shear rate of 0.01s-1 at 0°C) with steady
state sensing (5% tolerance within 30 seconds in three consecutive sampling)
 Stress Growth Test at 0°C
“Yield Stress” Ln (σYS0°C); R2 = 0.86
35
30 Term Estimate Prob>|t|

25 Intercept 0.288 0.4001

Visc 0.005 <.0001*
Stress (Pa)

20
15 FD 0.030 <.0001*

10 OR 0.0039 0.4257

5
Yield  Strain
0
0 100 200 300 400 500
Strain (%)

Stress growth for sample S3P3O2 with steady state sensing (5% tolerance within
30 seconds in three consecutive sampling) at constant shear rate of 0.01s-1 at 0°C.
 Correlations: Structure and Rheology
100 100
Yield Stress (Pa)

Yield Stress (Pa)

10 10

1 1
0.000 0.100 0.200 0.300 0 10 20 30 40 50 60 70 80
Mix Viscosity at 50 s-1 (Pa*s-1) Fat Destabilization Extent (%)

Yield Stress at 0°C vs. Fat

Yield Stress at 0°C vs. Mix Viscosity at 50 s-1
Destabilization (%)
 Flow Ramp Test

Vary shear rate from 0.001 to 100/s (Up and Down curves) at 0°C and
measure shear stress and instantaneous viscosity.
 Flow Ramp Test – Up Curves
10000 100
Maximum viscosity
Instantaneous Viscosity

Stress at maximum B
1000 viscosity

Stress (Pa)
100
(Pa.s)

10
10

1
Height  of  the  peak
0.1 1
0.001 0.1 10 0.001 0.01 0.1 1 10 100
Shear rate (1/s) Shear rate (1/s)

(A) Instantaneous viscosity and (B) shear stress curves against shear rate
(from 0.001 to 100/s) of an ice cream with higher fat destabilization extent.
 Flow Ramp (Up and Down Curves)
Thixotropic Loop Test
100
Thixotropy0°C; R2 = 0.89
Term Estimate Prob>|t|
10
Stress (Pa)

Intercept 4.677 0.9674

ViscPas 2484.1 <.0001*

1 Up FD 7.55 0.0005*
Down
OR 1.72 0.2970
0.1
0.001 0.01 0.1 1 10 100
Shear rate (1/s)

Up and down curves for sample S2P2O3 obtained from shear rate ramp
during 10 minutes at 0°C. Thixotropy measured as area between up and
down curves, related to extent of structure breakdown
 Structure and Rheology

1600 1600
1400 1400
Thixotropy (Pa*s-1)

Thixotropy (Pa*s-1)
1200 1200
1000 1000
800 800
600 600
400 400
200 200
0 0
0.000 0.100 0.200 0.300 0 10 20 30 40 50 60 70 80
Mix Viscosity at 50 s-1 (Pa*s) Fat Destabilization Extent (%)

Thixotropy at 0°C vs. Fat

Thixotropy at 0°C vs. Mix Viscosity at 50 s-1
Destabilization (%)
Correlations between Drip-Through
and Rheology
-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐

-­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐ -­‐
 Drip-Through and Rheology
70
2
1.8 60
1.6
50

Final Height (%)

1.4
DT (g/min)

1.2 DT (g/min) 40
1
30
0.8
0.6 20 Final height (%)
0.4 10
0.2
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000
G' (Pa) G' (Pa)

Drip-Through Rate vs. G’0°C Final Height (%) vs. G’0°C

 Drip-Through and Rheology
2 70
1.8
60
1.6

Final Height (%)

1.4 50
DT (g/min)
DT (g/min)

1.2 40
1
0.8 30
0.6 20
0.4 Final height (%)
10
0.2
0 0
0 20 40 60 0 20 40 60
Yield Stress (Pa) Yield Stress (Pa)

Drip-Through Rate vs. Final Height (%) vs. Yield Stress

Yield Stress (σYS0°C) (σYS0°C)
 Drip-Through and Rheology
2 70
1.8
60
1.6

Final Height (%)

1.4 50
DT (g/min)

1.2 DT (g/min)
40
1
0.8 30
0.6 20
0.4 Final height (%)
10
0.2
0 0
0 500 1000 1500 0 500 1000 1500
Thixotropy (Pa s-1) Thixotropy (Pa s-1)

Drip-Through Rate vs. Thixotropy0°C Final Height (%) vs. Thixotropy0°C

 Summary
• Melt-down and rheology of melted ice cream must
be governed by elements of microstructure
• Ice, air, fat, serum
• Correlations between attributes, however, remain
inconsistent, suggesting that numerous elements
work together to affect properties
• Search for the elusive “structure parameter”
continues
• Would allow prediction of ice cream properties from
structure measurements
 Ice Cream & the OJ Trial
• Prosecutors say the murders
happened about 10:15 p.m. But
police found a container of
melting Ben & Jerry ice cream at
the crime scene about 12:15 a.m.
Defense attorneys are suggesting
that, because the ice cream wasn't
totally melted by 12:15, the
murders had to have happened
after 11 p.m. - when O.J. was
already on his way to the airport.

Is it possible to tell time by ice cream “melting”?

As an expert witness, what would you tell prosecutors?
 Questions?

Thanks to CNPq “Brazilian National Council for Scientific and Technological

Development” (which provided scholarship for the Dieyckson Freire).