- 25 –
______________________________________________________________________
in_hypalm_aw1_v3.doc
Appendix 1: Practical Advice for Measuring Aw
When measuring water activity, the main concerns should be (a) the water activity and
temperature stability of the product sample and (b) the stability of temperature at the probe.
Beware of product samples that have been refrigerated and beware of product sample that
have been freshly taken from the production line. In both cases, the sample temperature may
require a few hours to stabilize. At a water activity of 0.50 aw, a temperature difference of 1
q
C
(1.8
q
F) between the probe and the product typically results in an error of 0.03 aw. At higher
values of water activity, the same temperature difference may cause an error of 0.006 aw.
Factors that influence the accuracy of the AwQuick mode
On average, the difference between the AwQuick mode and full product equilibrium is less than
±0.005 aw (±0.5 %RH). However, the difference can be sometimes much higher.
A large difference between the AwQuick mode and full equilibrium does not necessarily mean
that the AwQuick mode is unreliable or inaccurate. A large difference means that factors have
intervened after the completion of the AwQuick mode that could not be anticipated while this
mode was running:
a) Lack of temperature stability during the measurement
Without temperature control of the probe and environment, temperature tends to change during
the relatively long time required by the AWE mode. In turn this affects any comparison between
the result of the AWQ mode and the result of the AWE mode. Samples with a high water activity
are particularly sensitive to the effect of a change in temperature both regarding the probe and
the sample itself.
An inversion in the trend of temperature causes particular problems. When the trend of
temperature changes during the measurement, a false equilibrium may appear for some time and
this can potentially fool the AwQuick mode.
A frequent problem has to do with the measurement of product samples that have been
refrigerated. Refrigerated product samples can take several hours to come to real equilibrium with
room temperature and the water activity of theses samples varies as the sample temperature
changes.
b) Inherently unstable product samples
Some product samples are inherently unstable and are problematic to measure. Typically the
water activity of the sample goes through successive stages of quasi equilibrium while the actual
end value of water activity keeps creeping at a very slow rate.
Products with a high oil content are an example of this: during the measurement, the oil tends to
separate from the rest of the sample and to float at the top of the sample, thus changing the water
activity of the sample (example: pesto sauce).
Another example are products with a high affinity for water. Even during a full equilibrium
measurement, water activity may keep creeping very slowly up from an initial apparent
equilibrium.
Other products that are hard to measure are non homogeneous products such as puffed rice,
cookies with fruit filling and sausage. Unless they are pre-homogenized, such products equilibrate
very slowly and the AwQuick results tend to be less accurate than with other products.
