Temp Dep

Heat of Sorption Measured using a Chilled Mirror Dew Point
Water Activity Meter

G. S. Campbell, E. M. Huffaker, and A. J. Fontana
The Third International Workshop on Water in Food was held at Lausanne, Switzerland, Mar
29 – 30, 2004. Dr. Gaylon Campbell, Vice President of Decagon Devices, Inc. presented a
paper at that conference on the measurement of heat of sorption. Following is a summary of
that presentation.
Heat of sorption is of interest to food scientists because it determines the temperature

dependence of water activity. If one knows the heat of sorption and the water activity at one
temperature, water activity at any other temperature can be computed. Typically, increasing
the temperature of a sample increases its water activity, but some samples (saturated salts, for
example) show decreased water activity with increasing temperature, and some show little if
any change (some unsaturated salt solutions). Knowing the temperature dependence of water
activity is critical if one measures water activity of a product at one temperature, but stores or
handles the product at a different temperature. For example, water activity of a product
sample might be measured to be 0.70 at 20C laboratory temperature. The product might then
be loaded in a truck where summer temperatures reach 40C. At this higher temperature the
water activity could be 0.80. The product might be stable and safe at a water activity of 0.70,
but might spoil at the higher water activity.
The heat of sorption is obtained by measuring the temperature dependence of the water
activity. The Clausius-Clapeyron equation from thermodynamics is used to compute
heat of sorption:
Qs 1 1
ln(a w 2 / a w1 ) =  − 
R  T2 T1 
Where aw1 is the water activity at Kelvin temperatre T1, aw2 is the water activity at temperature
T2, R is the gas constant, and Qs is the heat of sorption. Thus, if lnaw is plotted as a function
of RT, the slope of the resulting line will be the heat of sorption.
Measurements of water activity were made on samples of wheat flour, corn meal, ground
mustard, and milk powder at temperatures of 15, 30, 45 and 60 C. An AquaLab Series 3B TE
chilled mirror water activity meter was used for the measurements. Samples were adjusted to
water activities ranging from 0.15 to 0.75 prior to measurement, and allowed to equilibrate
overnight. These adjustments were made by vapor phase sorption or desorption.
Measurements typically started at the coldest temperature and proceeded to the hottest, but
some samples were measured with both increasing and decreasing temperature to assure that
systematic errors were not introduced. The results for wheat flour are shown in Fig. 1.
1 Decagon Devices, Inc.

1
0.8
Water Activity
0.6
0.4
0.2
0
15 30 45 60
Temperature (C)
Figure 1. Changes in water activity of wheat flour with temperature at four water
activities. The double line shows water activity on the same sample with increasing
and decreasing temperature. The upper line is from the decreased temperature, and
is higher than the lower values because of changes in the matrix resulting from
heating.
Note that the water activity increases with increasing temperature, and that the change is not
insignificant. For example, the sample with a water activity of 0.7 at 15 C has a water activity
of 0.82 at 60 C.
Figure 2 shows these same data transformed as suggested by the Claussius-Clapeyron

equation. Straight lines that best fit the data are also shown. The slopes of these lines
-0.5
ln(a w)
-1
-1.5
-2
-2.5
0.003 0.0031 0.0032 0.0033 0.0034 0.0035
Reciprocal Temperature (1/K)

Figure 2. Transformed data from Fig. 1. The lines are from linear regression, and
the slopes of the lines are used to compute heat of sorption.
are used to compute the heats of sorption. Lines similar to those in Fig. 2 were fit to all of the
data, so heats of sorption were obtained at four water activities for wheat flour, corn meal,
ground mustard, and milk powder. In general, a single line fit the data well over the whole
range of temperatures, indicating that the heat of sorption is approximately constant with

temperature. Heats of sorption ranged from -14 kJ/kg to near zero, with heat of sorption
decreasing with decreasing water activity.
Water activity is a relative measure of the energy status of water in food. An absolute
measure can be obtained using the Kelvin equation:
ψ = RT ln(aw )
Here, ψ is the water potential (J/mol), and is an absolute measure of the energy of the food
water. The energy status is defined with respect to a reference, the reference being pure, free
water. Thus, at a water activity of 1.00, the water potential is zero, meaning that no work
would be required to remove water from the sample and transport it (in a thermodynamic
sense) to a pool of pure, free water. Only pure water is completely free, or unbound in the
sense that no work is required to remove it from the sample. Any water activity below 1.00
indicates that the water is bound, to some extent, and the lower the water activity, the more
work is required to remove the water from the sample. For, example, the energy of water in a
sample at room temperature and a water activity of 0.50 is -1.7 kJ/mol. This is the minimum
energy required to remove water from the sample and make it free water.
The water potential is less widely used in food science than water activity, but it is directly
related to water activity, and is a good measure for correlating the results of our experiments.
Figure 3 shows the heat of sorption for all samples as a function of water potential of those
samples.
0
Heat of Sorption (J/mol)
-4
-8
-12 y = 2.4852x - 0.0874

R2 = 0.8182
-16
-6 -5 -4 -3 -2 -1 0
Water Potential (J/mol)
Figure 3. Heat of sorption as a function of water potential for wheat flour, corn meal,
ground mustard and milk powder.

Obviously the two measures are highly correlated. As expected, heat of sorption becomes
more negative as the sample becomes drier, and is zero for pure, free water.
Using the Clausius-Clapeyron equation, the definition of water potential, and the regression
from Fig. 3, we can compute the change in water activity with temperature for samples like
those used in this study, for any water activity. The equation is
 T − T21 
a w 2 = a w1 exp 2.5 ln(a w ) 1 
 T 
The graph of this equation is shown in Fig. 4 for a one degree change in sample temperature.
0.004
Water Activity Change per Degree
0.003
0.002
0.001
0
0 0.2 0.4 0.6 0.8 1
Water Activity
Figure 4. Change in water activity of samples like flour, milk powder and ground corn
for a one degree temperature change.
The change with water activity with temperature is greatest for samples at mid-range water
activity. At high water activity, and at very low water activity, the change is small. As water
activity approaches 1.00 the change with temperature goes to zero.
One practical application of this information is in determining whether, for a particular

application, one needs a water activity meter with temperature control, or whether one that
operates at laboratory temperature is sufficient. If, for example, an application were for
measurements in the range 0.80 to 0.90, and machine temperatures were always in the range
of 22 to 25 C (typical of an air conditioned laboratory), then water activity measurement
variation due to sample temperature variation would be 3 C x 0.001 C-1 = 0.003. This range
of variation is likely negligible for most applications, so an instrument without temperature
control would be adequate. On the other hand, assume an application required measurement
of samples around water activity of 0.4 in an at-line environment where temperatures varied
from 20 to 35C. The expected water activity variation would then be 15 C x 0.003 C-1 =
0.045, which would be significant for most applications. One would therefore want to use a
temperature controlled meter.

Temp Dep

Uploaded by

Copyright:

Available Formats

Temp Dep

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Temp Dep

Uploaded by

Copyright:

Available Formats

Heat of Sorption Measured using a Chilled Mirror Dew Point

Water Activity Meter

Heat of sorption is of interest to food scientists because it determines the temperature

1 Decagon Devices, Inc.

Figure 2 shows these same data transformed as suggested by the Claussius-Clapeyron

Reciprocal Temperature (1/K)

2 Decagon Devices, Inc.

-12 y = 2.4852x - 0.0874

3 Decagon Devices, Inc.

One practical application of this information is in determining whether, for a particular

4 Decagon Devices, Inc.

You might also like