WATER ACTIVITY: REACTION RATES,

PHYSICAL PROPERTIES, AND SHELF LIFE

Leonard N. Bell, Ph.D.

Nutrition and Food Science
Auburn University, AL 36849

1
 SHELF LIFE AND STABILITY
• Consumers expect quality, freshness, nutritive
value, and accurate label information.
• Shelf life indicates the extent of allowable product
change due to
– Time
– Environment
• Understanding stability and factors that affect
stability, such as moisture, can lead to improving
shelf life and shelf life predictions.
2
TYPES OF STABILITY

• Microbial
• Sensory
• Chemical
• Physical

3
FOOD PRODUCT TYPES

• Dry (powders, cereals)

• Intermediate Moisture (bars)
• Solutions (beverages)

4
 THE STABILITY QUESTION
• A food has a sensitive “attribute”
– Loss of sensitive ingredient
– Formation of an undesirable product
– Undesirable change in texture
• How does this “attribute” change due to:
– Processing
– Distribution
– Storage
• Commercial
• Consumer

• Answers come from modeling kinetic data

5
 REACTION KINETIC OVERVIEW

Active Compound Degradation Product(s)

6
 REACTION TYPES

• Hydrolysis
• Oxidation
• Rearrangement reactions
• Amine-Carbonyl (Maillard reaction)
• Enzymatic

7
 KINETIC DATA

• zero order model: [A] = [A]o - kt

• first order model: ln([A]/[A]o ) = -kt

• Terms
– k = rate constant
– t = time
– [A], [A]o = concentration
8
 ZERO ORDER PLOT
70

-slope = k = conc/day
Concentration

60

50

40

30
0 2 4 6 8 10 12 14 16 18 20
9
Time
 FIRST ORDER PLOT
100
% Vitamin C Remaining

90
80 -slope = k = 1/day
70
60

50

40

Lee & Labuza. 1975. J. Food Sci. 40: 370.

30
0 10 20 30 40 50 60 70 80 90 100 110

Time (h) 10
FACTORS INFLUENCING STABILITY

• Temperature
• Moisture
• pH
• Ingredients
• Oxygen

11
 TEMPERATURE

• Reaction rates increase with increasing

temperature
• Models
– Arrhenius Equation
ln(k) = ln(A) - EA /RT
log(k) = log(A) - EA /(2.303RT)
– Q10 Approach
Q10 = (θs,T) /(θs,T+10)
12
 ARRHENIUS PLOT FOR VITAMIN C
DEGRADATION AT aW 0.32
0.5
Lee & Labuza. 1975. J. Food Sci. 40:370.
Rate Constant (day -1)

0.3

0.1
0.07
0.05
0.03

-slope = EA /2.303R
0.01
3.1 3.2 3.3 3.4
1/T (Kelvin-1) 13
 SHELF LIFE PLOT FOR VITAMIN C
DEGRADATION AT aW 0.32
70
50
-slope = log (Q10)/10
30
Half-life (days)

10
7
5
3

Lee & Labuza. 1975. J. Food Sci. 40:370.

1
20 30 40 50

Temperature (°C) 14
 SHELF LIFE PLOT FOR THIAMIN
Time for 10% Loss (days) DEGRADATION AT aW 0.75
100

-slope = log (Q10)/10

10

Arabshahi & Lund. 1988. J. Food Sci. 53:199.

0.1
25 35 45 55
Temperature (°C) 15
 SHELF LIFE AND TEMPERATURE

• Define shelf life as the time for 10% thiamin loss.

• Shelf life at 25ºC is 38 days.
• Shelf life increases by 3.3 for each 10ºC decrease
in temperature.
• Thus, the predicted shelf life at 20ºC is ≈80 days.
• Small temperature changes can have large effects
on product shelf life.

16
 TEMPERATURE FLUCTUATIONS
40
Temperature

30
Tx

20

time interval x

10
0 10 20 30 40 50 60 70 80
Time 17
 TEMPERATURE FLUCTUATIONS
Fraction of Shelf Life Consumed (fcon)

Σ
time at temperature xi
fcon =
total shelf life at temperature xi

°C Shelf Life Time f

20 150 d 33 d 0.22
30 75 d 7d 0.093
40 38 d 2d 0.053
fcon = 0.366 18
 TIME TEMPERATURE INDICATORS

• Gives a visual indication of net exposure

to time and temperature

INDEX/ INDICE
3M
Monitor
Mark 0 1 2 3 4 5

MonitorMark is a registered trademark of 3M. 19

 TIME TEMPERATURE INDICATORS

Quality Quality
Indicator Indicator
Use or Use or
freeze freeze
before before
center is darker center is darker
than outer ring. than outer ring.

20
 WATER
&
CHEMICAL STABILTIY

21
MOISTURE MOISTURE
DIFFUSION DIFFUSION

CHEMICAL PHYSICAL
DETERIORATION DETERIORATION
22
 WATER

OPEN

CHEMICAL PHYSICAL
DETERIORATION DETERIORATION
23
 PERSPECTIVES ON WATER

• Moisture Content
• Water Activity (relative vapor pressure)
• Glass Transition (plasticizer effect)
• Water Mobility

24
 WATER ACTIVITY

• Determines direction of moisture transfer

• Most reaction rates increase with increasing
water activity
• Most rates correlate better with water activity
than moisture content
• Moisture sorption isotherms are useful

25
 GLASS-RUBBER TRANSITION
Amorphous Amorphous
Glass Rubber Crystal

• Transition due to:

– Temperature
– Moisture
• Depicted by state diagram
• Relates to physical properties of system
– Rigid glassy matrix converts into soft rubbery matrix
• Affects reactant mobility
26
 QUESTION OF THE 1990’s

Does moisture affect reactions by water

activity or by increasing reactant mobility with
its plasticizing ability?

27
 GENERAL EFFECT OF WATER
ACTIVITY ON REACTION RATES
1.4

1.2
Reaction Rate

1.0

0.8

0.6

0.4

0.2

0.0
0.0 0.2 0.4 0.6 0.8 1.0
Water Activity 28
 MOISTURE SORPTION ISOTHERM FOR
MALTODEXTRIN (DE 25) AT 25°C
30
Roos & Karel. 1991. Biotechnol. Prog. 7: 49.
25
Moisture (%db)

20

15

10

0
0.0 0.2 0.4 0.6 0.8
Water Activity 29
MODELING MOISTURE SORPTION
ISOTHERM DATA
• GAB equation

mokCaW
m=
(1 - kaW)(1 - kaW + kCaW)

• Terms
– mo = monolayer
– k = multilayer factor
– C = heat constant 30
 GENERAL EFFECT OF WATER
ACTIVITY ON REACTION RATES
1.4

1.2
Reaction Rate

1.0

0.8

0.6

0.4
mo
0.2

0.0
0.0 0.2 0.4 0.6 0.8 1.0
Water Activity 31
GLUCOSE MOVEMENT IN DRIED CARROTS
MONOLAYER = 7.2%(db)
Moisture Movement
% (db) Detected

5.9 No
8.1 Yes
10.2 Yes
13.9 Yes
18.1 Yes
21.7 Yes
26.3 Yes

Duckworth & Smith. 1963. Recent Advances in Food Science-3.

London: Butterworths. pp 230-238.
32
GLUCOSE MOVEMENT IN DRIED CARROTS
MONOLAYER = 7.2%(db)
Moisture Movement
% (db) Detected

5.9 No
mo = 7.2
8.1 Yes
10.2 Yes
13.9 Yes
18.1 Yes
21.7 Yes
26.3 Yes

Duckworth & Smith. 1963. Recent Advances in Food Science-3.

London: Butterworths. pp 230-238.
33
 ASPARTAME DEGRADATION

• Rearrangement reaction at pH>6

• Hydrolysis reaction at pH<4
• Both reaction types between pH 4 and 6
• Catalyzed by buffer salts
• Follows pseudo-first order kinetics

34
 ASPARTAME DEGRADATION AS A FUNCTION
OF WATER ACTIVITY AT pH 3 AND 30ºC
8
Bell & Labuza. 1994. J. Food Eng. 22:291.
100 x Rate Constant

6
(day –1)

0
0.2 0.4 0.6 0.8 1.0

Water Activity 35
 ASPARTAME DEGRADATION

100 x Rate Constant

• Water activity 0.3 to 0.8

(day –1)
– Solutes dissolving
– Mobility increasing
– Reactivity increasing
0
0.2 0.4 0.6 0.8 1.0
Water Activity

36
 ASPARTAME DEGRADATION

100 x Rate Constant

• Water activity 0.8 to 1.0

(day –1)
– Solutes becoming diluted
– Reactivity decreasing
0
0.2 0.4 0.6 0.8 1.0
Water Activity

37
 EFFECT OF Tg ON ASPARTAME DEGRADATION IN
POLYVINYLPYRROLIDONE (PVP) AT 25ºC AND pH 7
Bell & Hageman. 1994. J. Agric. Food Chem. 42:2398.
0.15
Glassy Tg Rubbery
Rate Constant (day –1)

0.1

0.05
aw 0.33
aw 0.54
aw 0.76
0
-60 -40 -20 0 20 40 60
T - Tg 38
 QA PLOT FOR ASPARTAME DEGRADATION
IN PVP AT 25°C AND pH 7
4
ln(QA) = -slope/0.1
QA = 1.4
ln (half-life)

Bell & Hageman. 1994. J. Agric. Food Chem. 42:2398.

1
0.2 0.3 0.4 0.5 0.6 0.7 0.8
Water Activity 39
 QA CONCEPT

half-life at aw
QA =
half-life at aw + 0.1

• QA = 1.4 means half-life decreases by 40% for

each 0.1 aw increase
• If aw decreases by 0.1, half-life increases by 1.4
– 33 days at aw 0.33 (experimental)
– 46 days at aw 0.23 (predicted)
– 65 days at aw 0.13 (predicted)
40
 RELATION BETWEEN ACTIVATION ENERGY OF
ASPARTAME DEGRADATION AND WATER ACTIVITY
30
pH 3
pH 5
Activation Energy
(kcal/mole)

25

20

Bell & Labuza. 1991. J. Food Sci. 56:17.

15
0.2 0.4 0.6 0.8 1
Water Activity 41
WATER ACTIVITY EFFECTS ON Q10 VALUES
OF ASPARTAME DEGRADATION AT pH 5

aW Q10

0.33 4.3
0.55 4.2
0.65 3.7
0.99 2.6

42
WATER ACTIVITY AND TEMPERATURE

• Activation energies often decrease as water

activity increases
• Sensitivity of a reaction to temperature
changes as aW changes
• A temperature increase has a larger effect
(higher Q10) on a reaction in low moisture
food than in a high moisture food
• However, the “dryness” adds stability,
compensating for the food being more
temperature sensitive
43
 MAILLARD REACTION
(NONENZYMATIC BROWNING)

• Bimolecular reaction involving:

– unprotonated amines (amino acid)
– carbonyl compounds (reducing sugars)
• Reactant loss follows second order kinetics
• Brown pigment formation modeled via
pseudo-zero order kinetics

44
 BROWN PIGMENT FORMATION IN
GLUCOSE-GLYCINE MODEL SYSTEM AT 37ºC
0.25
Eichner & Karel. 1972. J. Agric. Food Chem. 20:218.
Extent of Browning

0.2

0.15

0.1

0.05

0
0.2 0.4 0.6 0.8 1
Water Activity 45
 BROWN PIGMENT FORMATION IN
GLUCOSE-GLYCINE MODEL SYSTEM AT 37ºC
0.25
Eichner & Karel. 1972. J. Agric. Food Chem. 20:218.
Extent of Browning

0.2

0.15

0.1

0.05 Dissolution,
Higher Mobility

0
0.2 0.4 0.6 0.8 1
Water Activity 46
 BROWN PIGMENT FORMATION IN
GLUCOSE-GLYCINE MODEL SYSTEM AT 37ºC
0.25
Eichner & Karel. 1972. J. Agric. Food Chem. 20:218.
Extent of Browning

0.2
Dilution
0.15

0.1

0.05

0
0.2 0.4 0.6 0.8 1
Water Activity 47
EXPERIMENTAL DESIGN TO ELIMINATE
CONCENTRATION EFFECT
• Prepare each system with pre-determined
reactant concentration so that after
moisture sorption, all have the same
reactant concentration in the aqueous
phase (Bell et al. 1998. J. Food Sci. 63:625).
• For example
– 10% moisture = 10 mg added reactants/g
– 20% moisture = 20 mg added reactants/g
48
 GLYCINE LOSS IN PVP-K15 AT 25ºC AS A
FUNCTION OF WATER ACTIVITY
1.4
100 x Rate Constant

1.2

1
(M –1 d –1)

0.8

0.6

0.4

0.2
Bell et al. 1998. J. Food Sci. 63:625.
0
0 0.2 0.4 0.6 0.8
Water Activity 49
 GLYCINE LOSS IN PVP-K15 AT 25ºC AS A
FUNCTION OF GLASS TRANSITION
Bell et al. 1998. J. Food Sci. 63:625.
1.4
Glassy Tg Rubbery
100 x Rate Constant

1.2

1
(M –1 d –1)

0.8

0.6

0.4

0.2

0
-60 -40 -20 0 20 40 60
T - Tg 50
GLYCINE LOSS VIA MAILLARD REACTION

• Concentration effects eliminated

• Water activity 0.1 to 0.4
– Mobility increasing
– Reactivity increasing
• Water activity 0.4 to 0.7
– Pass through glass transition
– Structural collapse
– Reactivity decreasing
51
GLUCOSE LOSS IN PVP AT aW 0.33 AND 25ºC
AS AFFECTED BY POROSITY AND COLLAPSE
White & Bell. 1999. J. Food Sci. 64:1010.
4
Porous (Φ>0.7)
1/[Glucose] (M -1)

Collapsed (Φ<0.2)
3

1
0 10 20 30 40 50 60 70 80 90 100
Time (days) 52
EFFECT OF GLASS TRANSITION ON BROWNING IN
PVP AT pH 7, aW 0.54, and 25°C
3.0
rubbery
(O.D./g dry solids)

glassy
Browning

2.0

1.0

Bell. 1995. Food Res. Intl. 28:591.

0.0
0 20 40 60 80
Time (days) 53
 BROWNING RATES IN PVP AT 25ºC AS
AFFECTED BY THE GLASS TRANSITION
Bell et al. 1998. J. Food Sci. 63:625.
15
Glassy Tg Rubbery
100 x Rate Constant
(OD g –1 mL–1 d –1)

12

0
-40 -20 0 20 40 60
T - Tg 54
WATER AND THE MAILLARD REACTION
• Moisture directly impacts this reaction primarily
by dissolving and/or diluting the reactants
• Moisture indirectly impacts this reaction by its
effect on the glass transition.
• Various researchers have shown the glass
transition affects the Maillard reaction
– Karmas et al. 1992. J. Agric. Food Chem. 40:873.
– Buera & Karel. 1995. Food Chem. 52:167.
– Lievonen et al. 1998. J. Agric. Food Chem. 46:2778.

55
VITAMIN STABILITY

• Ascorbic acid (vitamin C)

• Riboflavin
• Thiamin

56
 VITAMIN C DEGRADATION AT 35ºC AS
INFLUENCED BY MOISTURE
100
% Vitamin C Remaining

90
80
70
60

50 aW 0.32, 5%M (db)

aW 0.67, 18%M (db)
40

Lee & Labuza. 1975. J. Food Sci. 40:370.

30
0 10 20 30 40 50 60 70 80 90 100 110
Hours 57
QA PLOT FOR VITAMIN C DEGRADATION AT 23°C
100
QA = 1.7
Half-life (days)

10

Lee & Labuza. 1975. J. Food Sci. 40: 370.

1
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Water Activity 58
 VITAMIN C DEGRADATION IN DRIED
TOMATO JUICE AT 20ºC
1000 x Rate Constant (d –1)
125
Riemer & Karel. 1977. J. Food Proc. Preserv. 1:293.

100

75

50

25

0
0 0.2 0.4 0.6 0.8
Water Activity 59
QA PLOT FOR VITAMIN C DEGRADATION
AT 20ºC IN DRIED TOMATO JUICE
8
Below mo; not included in QA determination
7
6
ln (Half-life)

5
4 QA = 1.6
3
2
1
Riemer & Karel. 1977. J. Food Proc. Preserv. 1:293.
0
0 0.2 0.4 0.6 0.8
Water Activity 60
RIBOFLAVIN DEGRADATION AT 37ºC IN MODEL
SYSTEMS AS AFFECTED BY WATER ACTIVITY
7
1000 x Rate Constant

Dennison et al. 1977. J. Food Proc. Preserv. 1:43.

6

5
(day –1)

0
0 0.2 0.4 0.6 0.8
Water Activity 61
 QA PLOT FOR RIBOFLAVIN
DEGRADATION AT 37ºC
8

Below mo; not included in QA determination

7
ln (Half-life)

6
QA = 1.3

Dennison et al. 1977. J. Food Proc. Preserv. 1:43.

4
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Water Activity 62
 THIAMIN DEGRADATION AT 45ºC IN MODEL
SYSTEMS AS AFFECTED BY WATER ACTIVITY
Dennison et al. 1977. J. Food Proc. Preserv. 1:43.
12
1000 x Rate Constant

8
(day –1)

0
0 0.2 0.4 0.6 0.8
Water Activity 63
QA PLOT FOR THIAMIN DEGRADATION
IN PASTA AT 35ºC
8
Kamman et al. 1981. J. Food Sci. 46:1457.
ln (Half-life)

QA = 1.6
5
0.3 0.4 0.5 0.6 0.7
64
Water Activity
EFFECT OF MOISTURE ON THE ACTIVATION
ENERGY OF THIAMIN DEGRADATION

aW EA (kcal/mole)

0.44 30.8
0.54 29.8
0.65 26.6

Kamman et al. 1981. J. Food Sci. 46:1457.

65
THIAMIN DEGRADATION IN PVP AT 20ºC
AS AFFECTED BY WATER ACTIVITY
20
1000 x Rate Constant

15
(day –1)

10

Bell & White. 2000. J. Food Sci. 65:498

0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Water Activity 66
 THIAMIN DEGRADATION IN PVP AT
20ºC AS AFFECTED BY Tg
Bell & White. 2000. J. Food Sci. 65:498
0.03
Rate Constant (day –1)

Glassy Tg Rubbery

PVP-LMW
0.02 PVP-K30

0.01

0
-100 -75 -50 -25 0 25 50
T - Tg 67
 ENZYMES AND WATER
• Enzyme activity
– Water can:
• Dissolve substrate
• Increase substrate mobility
• Be a reactant
• Enzyme stability
– Moisture can influence “denaturation”
• Hydrolysis
• Deamidation
• Oxidation 68
 EXTENT OF LECITHIN HYDROLYSIS
BY PHOSPHOLIPASE AFTER 12 DAYS
100
Acker. 1963. Recent Advances in Food Science-3.
London: Butterworths. pp 239-247.
Percent Hydrolysis

80

60

40

20

0
0.2 0.4 0.6 0.8 1
Water Activity 69
SUCROSE HYDROLYSIS BY INVERTASE
AS A FUNCTION OF WATER ACTIVITY
0.2
Rate Constant (day –1)

Lactose/Sucrose at 24ºC (Kouassi & Roos.

0.16 2000. J. Agric. Food Chem. 48:2461.)
PVP-LMW at 30ºC (Chen et al. 1999.
J. Agric. Food Chem. 47:504.)
0.12 PVP-K30 at 30ºC (Chen et al. 1999.
J. Agric. Food Chem. 47:504.)
0.08 x T=Tg

0.04

0 xx x

0.2 0.3 0.4 0.5 0.6 0.7 0.8

Water Activity 70
QA PLOT FOR INVERTASE STABILITY
IN PVP AT 30ºC
5
ln (Invertase Half-life)
Chen et al. 1999. J. Agric. Food Chem. 47:504

QA=1.3
2
0.2 0.3 0.4 0.5 0.6 0.7 0.8
Water Activity 71
INVERTASE STABILITY IN PVP AT 30ºC AS
AFFECTED BY THE GLASS TRANSITION
Chen et al. 1999. J. Agric. Food Chem. 47:504
0.07
Rate Constant (day –1)

Glassy Tg Rubbery
0.06

0.05

0.04

0.03

0.02

0.01
-40 -20 0 20 40 60
T - Tg 72
RESIDUAL TYROSINASE ACTIVITY IN PVP
AFTER 3 DAYS EQUILIBRATION AT 20ºC
1500
Chen et al. 1999. Food Res. Intl. 32:467.
Activity (U/g solid)

1200

900

600

300 PVP-LMW
PVP-K30
0
0.2 0.4 0.6 0.8
Water Activity 73
RESIDUAL TYROSINASE ACTIVITY IN PVP
AFTER 3 DAYS EQUILIBRATION AT 20ºC
Chen et al. 1999. Food Res. Intl. 32:467.
1500
Rubbery Glassy
Activity (U/g solid)

1200

900

600

300 PVP-LMW
PVP-K30
0
-40 -20 0 20 40 60 80
Glass Transition Temperature (ºC) 74
 RATE CONSTANTS FOR TYROSINASE
ACTIVITY LOSS AT 20ºC AS AFFECTED BY aW
0.14
Rate Constant (day –1)

PVP-LMW
0.12 PVP-K30
0.1

0.08

0.06

0.04
Chen et al. 1999. Food Res. Intl. 32:467.
0.02
0.2 0.3 0.4 0.5 0.6 0.7 0.8
Water Activity 75
 RATE CONSTANTS FOR TYROSINASE
ACTIVITY LOSS AT 20ºC AS AFFECTED BY Tg
Chen et al. 1999. Food Res. Intl. 32:467.
0.14
Glassy Tg Rubbery
Rate Constant (day –1)

0.12

0.1

0.08

0.06

0.04 PVP-LMW
PVP-K30
0.02
-50 -40 -30 -20 -10 0 10 20 30 40 50
76
T - Tg
 LIPID OXIDATION

• Initiated by metal ions

• Measured by
– Oxygen consumption
– Peroxide values
– Hexanal formation
• Complex kinetic modeling

77
 OXYGEN UPTAKE IN CELLULOSE/LINOLEATE
MODEL SYSTEMS AT 37ºC
(moles O2 /mole linoleate)½
0.4

aW <0.01
Oxygen Uptake

0.3 aW 0.52

0.2

0.1

Labuza et al. 1971. JAOCS. 48:86.

0
0 40 80 120
Time (h) 78
 RATE CONSTANT FOR OXYGEN UPTAKE BY
LINOLEATE IN CELLULOSE MODEL SYSTEMS AT
37ºC AS AFFECTED BY WATER ACTIVITY
(moles O2 /mole linoleate)½ (h) –1

1.0
Labuza et al. 1971. JAOCS. 48:86.
100 x Rate Constant

0.75

0.5

0.25

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
79
Water Activity
LIPID OXIDATION IN POTATO CHIPS AT 37ºC
Quast & Karel. 1972. J. Food Sci. 37:584.
Rate of Oxygen Uptake 1.6
(µL O2 g –1 h –1)
1.2

0.8

0.4

0
0 0.2 0.4 0.6 0.8
Water Activity 80
 WATER AND LIPID OXIDATION

• Increase aW from 0 to 0.3

Oxidation Rate
– Less available metal ions
due to hydration spheres
– Reduced oxygen diffusion
– Free radical quenching
• Rate decreases

0 0.2 0.4 0.6 0.8

Water Activity
81
 WATER AND LIPID OXIDATION

• Increase aW from 0.3 to 0.8

Oxidation Rate
– Increased dissolution of
catalysts
– Increased mobility of oxygen
and metal ions
• Rate increases

0 0.2 0.4 0.6 0.8

Water Activity
82
LIPID OXIDATION AND GLASS TRANSITION

• Glassy encapsulants decrease oxidation by

reducing oxygen diffusion
• Conversion to a rubbery matrix can increase
oxygen diffusion and thus oxidation
• Structural collapse can:
– Reduce oxidation of entrapped lipid by eliminating
pores
– Enhance oxidation if lipid becomes exuded
• For additional information about water and
lipid oxidation:
– Nelson & Labuza. 1992. Lipid Oxidation in Food.
Washington, D.C.: ACS, pp. 93-103. 83
 pH EFFECTS

• Reactions affected by pH
– Aspartame degradation
– Maillard reaction
• non-enzymatic browning
• amino acid destruction
– Vitamin degradation
• The importance of pH is not limited to solutions
– Reducing water activity/moisture content can result
in a pH change
– pH changes can affect food chemical stability
84
pH ESTIMATION OF REDUCED-MOISTURE
SOLIDS USING A CHEMICAL MARKER
• Bell & Labuza. 1991. Cryo-Letters. 12:235.
– At aW 0.99, 0.1 M phosphate buffer in agar/MCC has a
pH of 6.5
– After lyophilization and equilibration to aW 0.34, system
behaves as pH 5
– Generally attributed to increased concentration of
buffer salts and their selective precipitation
• Results supported by other methods of estimating
reduced-moisture pH
– Bell & Labuza. 1992. J. Food Proc. Preserv. 16:289.
85
ASPARTAME DEGRADATION IN 0.1 m PHOSPHATE
BUFFER AT 25°C AS AFFECTED BY pH
Bell & Chuy. 2001. IFT Poster.
0.08
Rate Constant (day –1)

0.07 Control
Added 2 m sucrose
0.06
0.05
0.04
0.03
0.02
0.01
0
4.5 4.7 4.9 5.1 5.3 5.5 5.7 5.9
Initial pH 86
ASPARTAME DEGRADATION IN 0.1 m PHOSPHATE
BUFFER AT 25°C AS AFFECTED BY pH
Bell & Chuy. 2001. IFT Poster.
0.08
Rate Constant (day -1)

0.07 Control
0.06 Added 2 m sucrose
0.05
0.04
0.03
0.02
0.01
0
4.5 4.7 4.9 5.1 5.3 5.5 5.7 5.9
Actual pH 87
 STABILITY TESTING
• System as close to commercial product as possible
– formulation
– processing
– packaging
• Method of analysis
• Number of data points/extent of reaction
– 8 points minimum
– 50% change in reactant concentration
• Three temperatures
• Three water activities
• Extreme conditions may be used to accelerate
testing 88
 SHELF LIFE TESTING EXAMPLE
Given the following aspartame degradation data,
estimate the shelf life at 20ºC and aW 0.15
Conditions Rate Constant (d-1) t10% loss (days)

aW 0.34, 30ºC 0.00548 19.2

aW 0.35, 37ºC 0.01538 6.9
aW 0.42, 45ºC 0.04828 2.2

aW 0.34, 30ºC 0.00548 19.2

aW 0.57, 30ºC 0.01574 6.7
aW 0.66, 30ºC 0.02224 4.7
Bell. 1989. M.S. Thesis. University of Minnesota. 89
 SHELF LIFE PLOT FOR ASPARTAME
DEGRADATION AT aW 0.34
4
ln(Shelf Life)

1 Q10 = 4.25

0
20 25 30 35 40 45 50
Temperature (ºC) 90
 QA PLOT FOR ASPARTAME
DEGRADATION AT 30ºC
3
ln(Shelf Life)

QA = 1.55

1
0.2 0.4 0.6 0.8
Water Activity 91
 SHELF LIFE CALCULATION

• From kinetic data:

− At aW 0.34 and 30ºC, t10% = 19.2 days
− Q10 = 4.25
− QA = 1.55
• Want to know shelf life at aW 0.15 and 20ºC
− (t10%, aw 0.34, 30ºC ) x ( Q10 )∆T/10 x ( QA )∆a/0.1
− ∆T = change in temperature from 30ºC
− ∆a = change in water activity from 0.34
− (19.2) x (4.25)10/10 x (1.55)0.19/0.1 = 188 days

92
PROBLEMS OF ACCELERATED SHELF LIFE
TESTING AT EXTREME TEMPERATURES

• Water activity changes with temperature

• pH changes with temperature
• Solubility of reactants change
• Phase changes
– Glass transition
– Crystallization
• Competing chemical reactions
93
 SIMULTANEOUS DEGRADATIVE
REACTIONS WITHIN A FOOD
-2
ln (Rate Constant)

Reaction 1

-3
Reaction 2

Crossover at 25ºC

-4
3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7
1000/Temperature (K-1) 94
 WATER
&
PHYSICAL STABILITY

95
 PHYSICAL STABILITY
• Solutions
– separation
– crystallization
• Low and intermediate moisture systems
– hardening or softening*
– caking and sticking of powders*
– crystallization*

*Influenced by conversion of a glassy

matrix into a rubbery matrix
96
SENSORY CRISPNESS INTENSITY OF SALTINE
CRACKERS AS A FUNCTION OF aW
Katz & Labuza. 1981. J. Food Sci. 46:403
1.4 Very crisp
Crispness Intensity

1.2

1
Moderately Crisp
0.8
Minimum acceptability
0.6
Slightly crisp
0.4

0.2 aW,c

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Water Activity 97
 TEXTURAL ISSUES
• Staling (softening) of cereal products
– Crispness, crunchiness lost if aW > 0.4
– Aim to keep aW < 0.4

• Hardening of fruit pieces

– Softness, chewiness lost if aW < 0.5-0.6
– Aim to keep aW > 0.5-0.6

• Contradictory goals
for a fruit-containing
cereal product
98
 CAKING AND STICKING OF POWDERS
• Powders in the glassy state become rubbery from:
– Moisture sorption
– Temperature abuse
• Rubbery powders flow together under the force of
gravity
• Over long periods of time, the net result is formation
of a caked mass
• Supporting literature
– Aguilera et al. 1993. Biotech. Prog. 9:651.
– Chuy & Labuza. 1994. J. Food Sci. 59:43.
– Lloyd et al. 1996. J. Food Eng. 31:305.
– Netto et al. 1998. Int. J. Food Prop. 1:141. 99
MOISTURE SORPTION ISOTHERM OF
FREEZE-DRIED LACTOSE AT 34ºC
8
Berlin et al. 1970. J. Dairy Sci. 53:146.
Moisture Content (%db)

7
6
5
4
3
2
1
0
0 0.2 0.4 0.6 0.8 1
Water Activity 100
EFFECT OF EXPOSURE TIME AT aW 0.3 AND 24ºC
ON MOISTURE CONTENT AND CRYSTALLINITY OF
AMORPHOUS SUCROSE
8 120
Moisture Content (%)

Palmer et al. 1956. J. Agric. Food Chem. 4: 77.

7
100

Crystallinity (%)
6
80
5
4 60
3
40
2
20
1
0 0
0 5 10 15 20 25 30
Time (days) 101
 EFFECTS OF CRYSTALLIZATION
• Amorphous ingredients may crystallize from:
– Moisture gain
– Temperature abuse
– Storage (time)
• Crystalline materials hold less moisture so upon
crystallization water is released.
• This water is redistributed in the food, causing:
– Water activity to increase
– Glass transition temperature of remaining amorphous
regions to decrease
– Further chemical and physical deterioration!
• Literature 102
– Roos & Karel. 1990. Biotech. Prog. 6:159.
 EFFECTS OF CRYSTALLIZATION
• Amorphous ingredients may crystallize from:
– Moisture gain
– Temperature abuse
– Storage
• Crystalline materials hold less moisture so upon
crystallization water is released.
• This water is redistributed in the food, causing:
– Water activity to increase
– Glass transition temperature of remaining amorphous
regions to decrease
– Further chemical and physical deterioration!
• Literature 103
– Roos & Karel. 1990. Biotech. Prog. 6:159.
USING MOISTURE SORPTION ISOTHERMS
AND PHASE DIAGRAMS TO ESTIMATE
CRITICAL WATER ACTIVITY VALUES

104
 APPLICATION OF PHASE TRANSITION
TO A MALTODEXTRIN POWDER
• Caking of powders occurs due to the conversion of
glassy particles into rubbery particles which stick
together.

105
 APPLICATION OF PHASE TRANSITION
TO A MALTODEXTRIN POWDER
• Caking of powders occurs due to the conversion of
glassy particles into rubbery particles which stick
together.
• From state diagram, moisture content for Tg to be at
room temperature (20-25°C) is 8% (wb).

106
 GLASS TRANSITION STATE DIAGRAM FOR
MALTODEXTRIN (DE 25)
150
Roos & Karel, 1991, Biotech. Prog., 7: 49.
100

50
Tg (°C)

-50

-100

-150
0 10 20 30 40 50 60 70 80 90 100

Moisture (%wb) 107

 APPLICATION OF PHASE TRANSITION
TO A MALTODEXTRIN POWDER
• Caking of powders occurs due to the conversion of
glassy particles into rubbery particles which stick
together.
• From state diagram, moisture content for Tg to be at
room temperature (20-25°C) is 8% (wb).
• From moisture sorption isotherm, maltodextrin holds
8% water at aW 0.5.

108
 MOISTURE SORPTION ISOTHERM FOR
MALTODEXTRIN (DE 25) AT 25°C
30
Roos & Karel, 1991, Biotechnol. Prog., 7: 49.
25
Moisture (%db)

20

15

10

0
0.0 0.2 0.4 0.6 0.8
Water Activity 109
 APPLICATION OF PHASE TRANSITION
TO A MALTODEXTRIN POWDER
• Caking of powders occurs due to the conversion of
glassy particles into rubbery particles which stick
together.
• From state diagram, moisture content for Tg to be at
room temperature (20-25°C) is 8% (wb).
• From moisture sorption isotherm, maltodextrin holds
8% water at aW 0.5.
• Moisture gain from an environment when aW > 0.5
will promote caking of the maltodextrin.
110
 SUGGESTIONS FOR CONTROLLING
MOISTURE TRANSFER (GAIN OR LOSS)
• Understand moisture sorption isotherms
• Formulation approaches
– Humectants
– Anticaking agents (e.g. calcium silicate)
• Packaging approaches
– Select to minimize water permeation
– Resealable packaging
• Handling instructions 111
 OVERVIEW: WATER AND STABILITY
• Chemical Stability
– No universal physicochemical parameter
describes the effects of water
– Water activity and plasticization by water affects
chemical stability differently depending upon:
• Reaction type
• Matrix
• Physical Stability
– Water affects physical stability primarily by
plasticizing glassy systems into rubbery systems
112