Fundamentals

The science of drying

Every practical drying decision — temperature, airflow, slice thickness, when to stop — is a consequence of a handful of physical principles. This chapter explains water activity, moisture content, the shape of the drying curve, and where the energy actually goes.

What drying really moves

Drying removes water, but the useful mental model is a coupled heat-and-mass-transfer problem. Three resistances sit in series between the water inside a food and the vapour in the air:

Internal resistance — water must diffuse through cell walls, starch, sugar, protein and air pockets to reach the surface.
The surface — water must change phase from liquid to vapour, absorbing latent heat.
External resistance — vapour must diffuse through the still boundary layer of air clinging to the food, then be swept away by the bulk airflow.

Early in drying, resistance 3 dominates: the surface is wet, water is plentiful, and the rate depends mostly on airflow and temperature. Later, resistance 1 dominates, and drying becomes a slow, diffusion-limited process that no amount of extra heat can accelerate much.

Free water, bound water and solids

Water in food is not a uniform puddle. It exists in several states, and only some of it can be removed by drying:

Free water: Occupies pores, vacuoles and intercellular space. Behaves like ordinary liquid water and is removed first and fastest. It is what supports microbial growth.
Capillary water: Held in fine pores by surface tension. Removed comparatively easily, but with more resistance than free water.
Physically bound water: Adsorbed in the monolayer onto hydrophilic groups of starch, protein and sugars. Removal requires appreciably more energy and cannot be driven off by ambient air alone.
Chemically bound water: Water of crystallisation in hydrates and sugar molecules. Effectively part of the molecular structure; not removable by drying and not relevant to spoilage.

The practical consequence: there is a floor. No dryer can drive moisture to zero, and a product left in humid air will always return towards equilibrium. Drying is about reaching a stable point, not about total dehydration.

Moisture content: wet basis vs dry basis

Two conventions exist, and confusing them is the most common arithmetic mistake in drying.

Wet basis (wb) — water mass ÷ total mass. This is what recipes, labels and most extension guides use. Fresh broccoli is “about 90 % moisture”.
Dry basis (db) — water mass ÷ dry-solids mass. This is what engineering models, sorption isotherms and most journal papers use. It can exceed 100 %.

wet basis Mw = Ww / (Ww + Ws) × 100 % dry basis Md = Ww / Ws × 100 % convert: Md = 100 · Mw / (100 − Mw) Mw = 100 · Md / (100 + Md) example — fresh apple at 85 % wb: Md = 100 × 85 / 15 = 567 % dry basis (i.e. 5.67 kg water per kg of dry solids) dry to 18 % wb → Md = 100 × 18 / 82 = 22 % dry basis → 96 % of the original water has been removed

Wet basis is human-friendly; dry basis is physics-friendly. Convert before comparing numbers from different sources.

🧮

Do the arithmetic for me

The yield calculator converts between the two bases and tells you the final weight and water loss for any batch, and the converter handles single values.

Water activity (aw) — the master variable

Water activity is the ratio of the vapour pressure of water in the food to the vapour pressure of pure water at the same temperature:

aw = p / p₀ = ERH / 100 p vapour pressure of water in the food p₀ vapour pressure of pure water at the same temperature ERH equilibrium relative humidity (%) Pure water → aw = 1.00 Bone-dry product → aw ≈ 0.10–0.30 Shelf-stable dried food → aw ≤ 0.60

Water activity is bounded between 0 and 1, and it is a direct measure of how available the water is to microbes, enzymes and chemical reactions.

Aw is not the same as moisture content. A food can have 30 % moisture and still be stable if most of that water is bound to sugar or salt; a food with 15 % moisture can spoil if the water is free. Aw captures what moisture content only approximates, and it is why aw — not water content — is the regulatory and microbiological benchmark.

Aw is also the reason a single number can govern enzymatic activity, lipid oxidation, Maillard browning, texture crispness and microbial growth simultaneously. Most quality reactions have their own optimum aw, and they are not the same optimum. This is why “driest” is not always “best”: at very low aw, oxidation of fats actually accelerates.

The same ladder the table below spells out. Water activity is not moisture content: two foods at 12 % water can sit at opposite ends of this scale, because what matters is how tightly the water is held.

Water activity thresholds

Shelf stability is defined at aw ≤ 0.60. In practice, products with fat or high sugar are also limited by oxidation, caking and texture, so they are often specified lower or packaged differently.
aw range	What can grow	Typical products	Stability
0.99–1.00	All bacteria, yeasts, moulds	Fresh produce, milk	Perishable — days
0.90–0.99	Most bacteria (Salmonella, E. coli, Listeria)	Fresh meat, some fruit	Perishable — days
0.86–0.90	Staphylococcus aureus, most moulds and yeasts	Some sausage, syrups	Needs refrigeration or another hurdle
0.80–0.86	Yeasts, most moulds	Intermediate-moisture fruit, some jerky at the wet end	Weeks–months refrigerated
0.70–0.80	Xerophilic moulds, osmophilic yeasts	Dried fruit, fruit leather	Months, if packaged dry
0.65–0.70	A few xerophilic moulds only	Raisins, dates, dried figs	Months–a year
0.60–0.65	No growth (some spores survive but do not germinate)	Low-moisture dried fruit	Stable
< 0.60	No microbial growth	Vegetables, herbs, spices, jerky (with cure)	Shelf-stable — 1–2 years+

The drying curve

Plot moisture content against time and any drying process shows the same three phases. Learn the shape and you can diagnose almost any problem.

The drying curve. The critical moisture content marks the transition from surface-limited to diffusion-limited drying — the point where increasing temperature stops helping much.

Warm-up phase (A)

The food heats towards the drying air temperature. Short, but it matters for microbial safety: if the food starts cold and warm humid air surrounds it, this is when spoilage organisms are most comfortable.

Constant-rate period (B)

Free water migrates to the surface as fast as it can evaporate. The surface behaves as if it were a wet bulb, and its temperature sits close to the wet-bulb temperature of the drying air — often 15–25 °F below the dry-bulb temperature. Drying rate depends on airflow, air temperature and humidity, and hardly at all on the food’s internal structure. This phase is linear in the curve above.

Falling-rate period (C)

The critical moisture content is reached, the surface dry layer forms, and water must diffuse outward through the food. Rate now depends on thickness, temperature, and the effective moisture diffusivity of the material. Drying becomes slow and asymptotic; this phase is where most of the time in a batch is spent, and where case hardening, browning and nutrient loss mostly happen.

Equilibrium plateau (D)

Further drying is limited by the vapour pressure of the air. The product approaches the equilibrium moisture content for the current air conditions, and no amount of extra time will make it drier while the air stays the same. If your food stalls here, change the air, not the clock.

💡

Diagnostic value of the curve

Weigh the batch every hour and plot it. A batch that goes flat early but is still soft inside is at the critical moisture content with a dry surface — case hardening. A batch that goes flat while still obviously wet is at equilibrium for the current air — fix humidity or airflow, not temperature.

Heat and mass transfer

How heat arrives

Convection — hot air flowing over the food. Dominant in most dehydrators, ovens, solar and hot-air tunnel dryers.
Conduction — contact with a hot surface. Important in drum dryers, tray dryers and freeze-dryer shelves.
Radiation — infrared or solar radiant energy absorbed at the surface. Dominant in sun drying, infrared dryers and refractance window drying.
Dielectric / volumetric heating — microwaves and radio frequency heat from the inside out. Expensive and hard to control, but very fast.

Where the energy goes

Evaporating water is expensive: about 2.26 MJ/kg at 100 °C, rising to roughly 2.45 MJ/kg around 40–60 °C. That is roughly 0.63–0.68 kWh per kilogram of water removed, before any losses. A domestic dehydrator typically performs at 20–50 % efficiency — the rest heats the room, the cabinet, and the exhaust air.

Energy balance of a dryer air heater ───────► hot dry air │ ▼ ┌──────────────────────────────┐ │ food: water + dry solids │ │ absorbs latent heat →→→ │──── water vapour (leaves with air) └──────────────────────────────┘ │ ▼ warm humid air ──► exhaust (and losses) useful energy = water removed × latent heat specific energy = total energy input / water removed (kWh per kg)

The gap between latent heat and measured specific energy is your dryer's inefficiency — almost always dominated by exhaust losses.

The boundary layer and why airflow beats heat

A thin film of slow-moving air sits against the food. For vapour to escape it must diffuse across this film, and the concentration gradient across it is proportional to the difference in humidity between the food surface and the bulk air. Increasing airflow thins the film and replenishes dry air, which raises the rate; increasing temperature raises the food's vapour pressure, which also raises the rate but also raises quality damage. In practice, the cheapest fix for slow drying is usually better ventilation — not more heat.

Psychrometrics: the properties of drying air

You cannot understand slow drying without understanding air. Five quantities matter:

Dry-bulb temperature: Ordinary air temperature, what a normal thermometer reads.
Wet-bulb temperature: The temperature of a wetted thermometer in the same airflow. The gap between dry and wet bulb measures how much evaporative cooling is available — a big gap means air can accept lots of water.
Relative humidity (RH): Vapour present as a percentage of saturation at that temperature. Air at 100 % RH cannot accept any more water, so drying stops no matter how hot it is.
Dew point: The temperature at which air becomes saturated and water condenses. If the dew point approaches the food temperature, no net evaporation occurs.
Absolute humidity: Mass of water vapour per unit mass of dry air. This is the quantity that actually changes as air passes through a dryer: air enters dry and leaves humid.

Drying-air state	RH	Effect on drying
Hot, dry desert air	10–25 %	Excellent. Even unheated air dries quickly.
Temperate dry day	30–50 %	Good. Solar or low-heat drying works well.
Humid summer day	60–80 %	Poor. Raising temperature 15–20 °F is usually necessary; expect long runs.
Tropical rainy season	85–100 %	Hopeless without a dehumidifier, heat pump or refrigeration-assisted dryer. Ambient drying will spoil food.
Sealed jar with damp food	→ ~100 %	Mould. This is exactly what conditioning is designed to detect.

Heating air has a double benefit: it lowers relative humidity (so the same absolute humidity now represents a drier atmosphere) and raises the vapour pressure of the water in the food. That is why the standard advice in humid climates is “heat the air 15–20 °C above ambient, and make sure the humid exhaust actually leaves the box”.

A drying process on the psychrometric chart. Heating moves the air's state point vertically (same water, lower RH); drying moves it diagonally up and to the right as the air picks up the water your food loses.

Sorption isotherms and the GAB model

A sorption isotherm plots equilibrium moisture content against water activity at constant temperature. It is the map that connects “how wet is it?” to “is it safe and will it stay crisp?”

The classic sigmoid (Type II) isotherm. Zone A water is bound and immobile, zone B is available but not free, zone C is bulk water that causes caking, sogginess and growth.

The widely used GAB (Guggenheim–Anderson–de Boer) model fits this shape with three parameters — monolayer moisture Xm, and constants C and K:

Xm · C · K · aw X = ────────────────────────────────── (1 − K·aw) · (1 + (C − 1)·K·aw) X moisture content, dry basis Xm monolayer moisture (strongly bound, non-solvent water) C energy of binding in the monolayer vs the bulk K energy difference of multilayer water vs pure liquid (0.7 ≲ K ≲ 1 for most foods)

Given representative Xm, C and K for a food class, GAB converts between moisture content and water activity in both directions. This knowledge base implements exactly that in the water-activity estimator.

🧮

Estimate aw from your moisture reading — or the reverse

Open the water-activity estimator. It uses GAB with representative parameters per food class and tells you both the aw and the classification, so you can see whether “dry enough” really is.

Case hardening

Case hardening is the formation of a dense, dry, sealed shell around a still-wet interior. It happens when the surface loses water much faster than the interior can supply it: surface cells collapse and shrink, soluble solids (sugars, salts, proteins) migrate outward, and the outer layer sets into a barrier.

Consequences: drying time stretches dramatically, the interior may stay wet enough to spoil, texture becomes unpleasantly tough, and rehydration is poor because the shell resists water entry.

Causes

Temperature too high at the start of the run
Very strong direct airflow onto one face
Thick, dense pieces or large whole fruits
Sugar-rich or high-solids foods
Hot dry air entering before the interior warms

Prevention

Start lower than your target, then raise temperature
Slice thinner and uniformly (3–6 mm for most foods)
Blanch to open the tissue structure
Use moderate airflow, spread across the load
Equalise the load: pack thick pieces on one tray, thin on another

Case hardening in one picture. The surface dries and glassifies first; once it does, the interior water has to diffuse through a solid film instead of evaporating freely. Cooler air and real airflow keep the surface from getting ahead of the core.

Glass transition and stickiness

Dried foods with lots of sugar are amorphous glasses, not crystals. Like all glasses they have a glass transition temperature, Tg. Below Tg the material is a brittle, stable glass; above Tg it becomes rubbery, sticky and prone to collapse and caking.

Two practical implications:

Stickiness in a warm dryer. If the product temperature exceeds its sticky-point temperature, pieces fuse together and stick to trays. This is common with fruit leathers, fruit juice powders and honey-rich products. Lower the temperature, or dry to a lower final moisture so the residual water does not plasticise the matrix.
Collapse in storage. A product stored warmer than its Tg slowly caves in and cakes. Freeze-dried fruit stored in a hot car or a warehouse loft will do this even at very low moisture.

Water is a powerful plasticiser: adding a little water lowers Tg steeply. That is precisely why “a little moisture” is so damaging to crispness, and why an oxygen absorber plus a desiccant in a sealed pouch preserves texture so well.

Shrinkage, porosity and rehydration

As water leaves, the structure usually shrinks. Typical results:

Air drying → strong shrinkage, dense product, moderate rehydration.
Freeze drying → almost no shrinkage, high porosity, the best and fastest rehydration.
Vacuum / puff drying → expansion and puffing, crisp and porous.
Osmotic pretreatment before drying → less shrinkage but solutes impregnated, altering taste and texture.

Rehydration is a two-stage process: water wets and fills the pores and capillaries quickly, then diffuses into the cell interiors slowly. Products that suffered case hardening or collapse rehydrate badly because the pores have closed. This has a direct quality consequence: what you do during drying determines what happens in the soup pot weeks later.

Drying kinetics: the models

Engineers summarise a drying curve with a moisture ratio and a rate model. The simplest and most useful is first-order:

moisture ratio: MR = (M − Me) / (M₀ − Me) first-order (Newton): MR = exp(−k·t) Page: MR = exp(−k·tⁿ) Lewis / logarithmic: MR = a·exp(−k·t) + c Arrhenius temperature dependence: k = k₀ · exp(−Ea / (R·T)) M moisture at time t (wet basis) Me equilibrium moisture content at those air conditions M₀ initial moisture content k drying rate constant (per hour), temperature dependent Ea activation energy (J/mol); R = 8.314 J/(mol·K); T in kelvin n empirical exponent > 1 for most foods (Page model)

The Page model usually fits foods better than pure first-order, but the first-order model is accurate enough for planning and is what the calculator uses.

Indicative ranges from the drying literature. Treat them as order-of-magnitude and calibrate against your own equipment.
Material	Approx. Ea (kJ/mol)	Ea/R (K)	Note
Fruit (apple, banana, mango)	20–40	2,400–4,800	Higher for sugar-rich tissue
Vegetables (carrot, potato)	15–35	1,800–4,200	Lower for cellular, porous tissue
Herbs and leafy material	10–25	1,200–3,000	Very thin, surface-limited
Meat and fish	25–45	3,000–5,400	Dense protein matrix; strong internal resistance
Pastes and purees	15–30	1,800–3,600	Depends strongly on layer thickness

Three identical loads at three temperatures. Temperature buys time — roughly a factor of two from 50 °C to 70 °C here — and spends colour, aroma and vitamins to do it. The shape is the point: fast at the start, slow at the end, so the tail is where judgement decides quality.

What actually controls drying speed

In rough order of leverage, based on typical home and small-scale equipment:

Airflow and exhaust Get humid air out and dry air in. Doubling airflow across a wet surface can double the rate in the constant-rate period. Most home dryers are poor here.
Piece size and uniformity Time scales roughly with thickness — doubling thickness roughly quadruples drying time. Uniformity prevents the frustrating “one piece is still wet” problem.
Temperature Roughly doubles the rate for every 10 °C increase in the diffusion-limited stage, but costs quality and risks case hardening. Use it, don't lean on it.
Ambient humidity Sets the equilibrium moisture content you can ever reach. Above about 65 % RH, mechanical drying or dehumidification becomes necessary.
Load density Overloading starves the inner trays of airflow and creates a humid microclimate inside the stack.
Pretreatment Blanching and osmotic dips alter structure and permeability, and can shorten or lengthen drying depending on the food.

Energy: where it goes and what it costs

Drying is an energy-intensive unit operation. Knowing the two numbers below tells you whether your process is sane:

Theoretical energy: Water removed × latent heat ≈ 0.7 kWh/kg water at 50 °C.
Specific energy consumption: Measured energy input ÷ water removed. For a well-run small dryer this is 1.5–3 kWh/kg water; for an inefficient cabinet it can be 5+.

Dryer type	Typical efficiency	Energy per kg water	Comment
Solar cabinet / greenhouse	20–40 %	0 (fuel) plus fan	Almost free energy; slower and weather dependent
Home electric dehydrator	20–35 %	2–3.5 kWh	Cheap cabinet, poor insulation, much heat exhausted
Domestic oven (door ajar)	10–20 %	4–8 kWh	Worst option energetically, but universally available
Heat-pump dryer	50–70 %	1.0–1.6 kWh	Condenses and reuses latent heat; excellent for small producers
Industrial tunnel / belt dryer	40–60 %	1.3–2.0 kWh	Heat recovery, staged temperatures, large scale
Freeze dryer	n/a (vacuum + refrigeration)	5–15 kWh	Energy-rich but quality-rich; justified for high-value products

🧮

Cost your own batch

The energy calculator turns watts, hours and your tariff into a cost per kilogram of dried product. The solar sizing calculator does the equivalent for a sun-driven dryer.

Microbiology: what drying does to microbes

Drying is a stress, not a sterilisation step. Understanding the difference is the foundation of dry-food safety.

Vegetative cells (Salmonella, E. coli, Listeria, yeasts) die off progressively during drying, but the rate depends on temperature, and a meaningful reduction requires a genuine time–temperature combination. This is why jerky gets a lethality step rather than relying on dryness.
Spores (Clostridium botulinum, Bacillus cereus) survive drying essentially intact. They cause no harm while the food stays dry — but if moisture returns, especially anaerobically (vacuum-sealed, oil-packed), they can germinate.
Moulds and yeasts are inhibited by low aw, and xerophilic moulds remain the limiting organism for many dried fruits. Their spores are everywhere in kitchen air and on trays.
Mould toxins (mycotoxins such as aflatoxin and ochratoxin) are heat-stable and persist in the food. The correct response to visible mould is always to discard the lot.

Drying is therefore a hurdle technology: reduced aw does the heavy lifting, pretreatment contributes acid or salt, heat contributes a partial kill, and packaging keeps the whole system stable. Remove one hurdle — for example re-wetting a cured but unheated product — and the safety margin shrinks.