Water Activity Meter

Water activity is defined as the equilibrium relative humidity of air in contact with a sample, expressed as a fraction. A water activity meter is built directly around that definition: it seals a sample in a small chamber, waits for the air in that chamber to stop changing — the point of equilibrium — and measures the humidity of that stabilised air. The reading is not an inference about the sample's internal chemistry the way an oven-drying or near-infrared result is; it is a direct measurement of the one thing the definition actually specifies, the air.

Chilled-mirror dew-point sensor

Cools a mirror until dew forms on it and detects the temperature at which condensation begins, giving a direct physical measurement of dew point that is converted to water activity. This is the primary approach against which other sensor types are checked.

Capacitive polymer sensor

A polymer film absorbs water vapour from the headspace and changes electrical capacitance in response; faster and less expensive than a chilled mirror, but more prone to drift and to fouling by volatile compounds.

Resistive electrolytic sensor

Measures a change in electrical resistance caused by water vapour absorption; shares the capacitive sensor's susceptibility to fouling and drift.

Capacitive and resistive sensors are calibrated, like chilled-mirror instruments, against saturated salt standard solutions of known water activity, but they require more frequent checking because their sensing surfaces are chemically active and can be poisoned by volatile compounds such as propylene glycol, ethanol, or ammonia present in or near the sample. A poisoned sensor produces a persistent, wrong reading that looks like a normal, stable measurement until the sensor is cleaned, replaced, or re-verified against a known standard.

Because water activity is itself a temperature-dependent property, the sample and the measuring chamber must be at the same, correctly stated temperature for the reading to be meaningful. A sample introduced warmer or cooler than the chamber produces a reading that reflects the temperature gradient between them as much as the sample's underlying water activity, and a result obtained at one temperature cannot simply be applied at another temperature without a fresh measurement or a validated correction.

A water activity meter also does not report moisture content, and the two properties cannot be interconverted without the specific commodity's sorption isotherm — the curve relating that commodity's moisture content to its water activity at a given temperature, which differs from one commodity to the next. See Moisture Content and Water Activity for why these are genuinely separate properties, not two units for the same thing.

The instrument is also a single-sample measurement: a lot of any real size is heterogeneous, and one reading taken from one sample point does not represent the water activity of the whole lot, particularly where moisture or microbial activity is unevenly distributed.