Moisture Content and Water Activity in Botanical Powders: Two Numbers, Not One

Almost every botanical specification carries a moisture figure, and almost none of them carries a water activity figure. That single omission explains a large share of the complaints that arrive after delivery: a lot that conformed on paper, that caked in the sachet, lost assay faster than the stability data predicted, or grew mould in a customer’s warehouse. The two numbers are not alternative ways of expressing the same property, and one does not convert into the other. This guide sets out what each measurement actually tells you, why the same powder can read 5% moisture and still behave badly, and how to write both into a specification so that incoming material is judged the same way every time.

Two measurements that answer different questions

Moisture content is a quantity. It expresses how much water the material contains, as a percentage of its mass, and the common methods are loss on drying in an oven, Karl Fischer titration, near-infrared measurement against a calibration, and gas chromatography for specific cases.

Water activity is a condition. It describes how available that water is, expressed on a scale from zero to one as the ratio between the vapour pressure above the sample and the vapour pressure above pure water. It is measured with a chilled mirror or a capacitive sensor, and because it is a thermodynamic property it is only defined at a stated temperature.

The practical difference is that moisture tells you how much water is there and water activity tells you what that water can do. A powder built on gums, fibres or starches can hold a substantial amount of water in a bound form and still sit at a modest water activity, because the water is not free to move. A powder that has taken up water only at its surface, or that contains a crystalline hydrate, can show a low moisture content and a water activity high enough to matter. Neither number predicts the other, and quoting one as a proxy for the other is the root of most moisture-related misunderstandings in the ingredient trade.

Why a certificate can show 5% moisture and tell you nothing useful

The moisture figure on a certificate of analysis is often reported without the method, and that is where it loses its value. Loss on drying is not a water-specific measurement: it drives off whatever is volatile at the conditions used, which includes residual solvent, essential oil and volatile flavour components. A botanical extract carrying a percent of volatile oil will read high on loss on drying without being wet in any sense that affects its stability. Karl Fischer measures water specifically, and the two methods routinely return different figures for the same lot.

The conditions matter as much as the method. An oven determination at 105 °C for two hours, a vacuum dry, and a Karl Fischer titration are three different measurements with three different answers, and a specification that says only “moisture: not more than 5%” allows any of them. Add the question of basis — as-is or calculated on a dried basis — and two suppliers can both be reporting truthfully while describing materials that behave quite differently in the same sachet.

This is why the method is part of the number rather than a footnote to it. When moisture is the parameter that decides whether your blend can be weighed accurately, whether your tablet presses cleanly or whether your stick pack stays free-flowing, a figure without a method is not a specification.

What water activity actually predicts

Water activity maps onto the behaviours that moisture content only hints at. The growth thresholds of spoilage organisms — moulds, yeasts and bacteria — are conventionally expressed against water activity rather than moisture content, because the organisms respond to available water and not to total water. Most moulds stop growing below a water activity in the region of 0.6, but the exact thresholds vary by organism and product, so the convention is a starting point for setting limits rather than a figure to copy.

It also predicts the physical behaviour of the powder. Caking and loss of flow happen when surface moisture allows particles to bridge and the bridges then set; a hygroscopic material can pass a moisture specification at packing and still cake after a few weeks if the water it takes up is available rather than bound. Chemical degradation follows the same logic: hydrolysis, enzymatic activity in materials with residual enzyme, browning reactions and the loss of labile actives all track water activity more closely than they track the percentage of moisture. Where a stability programme has been designed on moisture content alone, it has been designed around the wrong variable.

Moisture that arrives from somewhere else

A great deal of the water that causes trouble was not in the material when it was released. It arrived in transit, in storage, or inside the customer’s own plant.

Temperature cycling is the classic route. A container crossing several climates takes the powder and its packaging through repeated condensation cycles, and the water that condenses on the inside of the liner is then absorbed at the surface of the powder and at the seam where the liner is folded. The result is a lot that conformed at the supplier’s gate, arrived with a soft crust at the top of the drum, and is measured at a different moisture content depending on where the sample was taken from — which is one of the reasons sampling discipline matters as much as the analytical method.

Blending creates a second route that is easy to miss. Two powders with different water activities, packed together, will equilibrate: the component with the lower activity gains water and the component with the higher one loses it. Where a hygroscopic carrier such as maltodextrin or a gum sits alongside a salt or a sugar in one blend, the equilibrium that establishes inside the pack can raise the risk of caking above what either component suggested on its own. A blend that was free-flowing at packing can be a solid block at arrival in a warm warehouse without any packaging failure at all.

The third route is the one in your own facility. A part-used drum that is resealed carelessly, or left open between weighing operations, is the most common source of caking complaints that get attributed to the supplier. The head space of a drum in a humid room is an efficient way to raise the surface moisture of everything underneath it, and the effect is not uniform, which is why the moisture measured from the top layer of an opened drum can be badly misleading.

Methods, sampling and why two laboratories disagree

Disagreement between laboratories on moisture usually has three explanations, and only one of them is the analytical method. The first is sampling: a sample taken from the surface of a drum is not a sample of the drum, and a composite built from increments taken at inconsistent depths will not be reproducible. The second is sample handling after sampling. A powder in a jar that has been opened twice, weighed slowly, or left on a bench in a humid room will gain or lose water before the determination, and the result reflects the journey rather than the lot. The third is the method itself, including the drying temperature, the drying time, and whether the technique measured water specifically or everything volatile.

Water activity has its own version of this. Because the measurement is temperature-dependent, a result quoted without the temperature at which it was taken is only partly a specification, and results read at 25 °C and at 40 °C are not the same figure. The instrument’s calibration and the equilibration time allowed before reading both affect the outcome, and a sample that has not reached equilibrium with the sensor head will read low.

The response is procedural rather than technical: name the method on your own specification, require the same method on every certificate, fix the sample handling in writing, and ask for the temperature alongside any water activity figure. Where a supplier cannot state the method, the number is not comparable with anything.

Specifying both, and setting limits that mean something

The two figures belong on the specification together, each with the information needed to interpret it. Moisture content should be stated with its method, its conditions and its basis. Water activity should be stated with the temperature at which it is determined. Both should be supported by a sampling instruction, because a limit cannot be enforced on a sample that was not representative.

The limits themselves should be set against the failure you are trying to prevent rather than copied from a generic range. Where the risk is microbial, water activity is the controlling figure. Where the risk is caking, flow or accurate weighing in a low-dose blend, both figures matter and the packaging barrier is part of the answer. Where the risk is loss of assay through hydrolysis, the combination of water activity and storage temperature is what drives the reaction rate. The finished product’s own requirement — a tablet, a stick pack, a clear beverage that must not cloud — sets the acceptable window, and it is usually tighter than any default a supplier would propose.

Where a supplier can report only one of the two, treat that as a gap to close at qualification rather than a detail to accept. A water activity figure on the qualification lot plus a moisture figure on every certificate is a workable compromise; neither figure alone is a specification of the material’s condition.

What to write into the purchase specification

State the moisture content with its method, its conditions and whether the basis is as-is or dried, and state the water activity with the temperature of determination. State the sampling instruction, including the number of increments, the depths and the sample size, so that a result can be reproduced by a second party. State the packaging and the liner that the limit assumes, because the limit is only achievable in a defined barrier. State the storage conditions under which the limit applies, and whether the certificate is issued per lot or on a periodic basis for the water activity figure. State the retention requirement and the container the retain sample is held in. And state the opened-container rule for your own operation, because the largest single cause of a moisture complaint is a part-used drum rather than a failed lot.

Frequently Asked Questions

Is moisture content the same as water activity?

No. Moisture content is the quantity of water present, expressed as a percentage of mass. Water activity is a measure of how available that water is, on a scale from zero to one, and it is defined at a stated temperature. A material can have a high moisture content with a modest water activity if the water is bound, and a low moisture content with a water activity that matters if the water is free.

My powder meets the moisture specification but cakes in storage. Why?

Because the moisture specification does not describe where the water is or how available it is. Caking is driven by surface moisture and by the migration of water between components, not by the total percentage. Check the water activity, the packaging barrier, the temperature history in transit, and how the container is resealed after part use, which is a frequent cause in the buyer’s own warehouse.

Why do two laboratories report different moisture figures for the same lot?

Usually because they used different methods or different conditions, because the sample taken was not representative, or because the sample gained or lost water between opening and analysis. Loss on drying, Karl Fischer and near-infrared are not interchangeable, and a drying temperature or time that differs will give a different figure. Name the method in the specification so that results stay comparable.

What water activity limit should I specify?

Set it against the failure you are preventing rather than from a generic table. The growth thresholds of spoilage organisms, the onset of caking and the rate of hydrolysis all sit at different water activities for different materials. Confirm the limit with your own microbiology and stability data for the finished product, and state it with the temperature at which it is measured.

Does loss on drying measure water?

Not specifically. Loss on drying measures everything volatile under the conditions used, which can include residual solvent and volatile oils as well as water. That is why a botanical material containing essential oil can show an elevated loss on drying without being wet, and why the method has to be stated with the figure.