Food Science & Technology · Food Science Core · also Food Physics, Food Rheology, Engineering Properties of Foods, Food Texture
Measuring and modelling how foods flow, deform, break and transfer heat: rheometry of fluids and gels, viscoelasticity, texture profile analysis, and thermal, electrical and optical properties. Agricultural Engineering's engineering-properties-biomaterials covers size, shape, density and friction of grains and produce; this subject goes deeper into processed foods and their measurement.
In the exams: ICAR AIEEA-PG Code 18 Units I–II (physical properties of foods; engineering properties of food materials); ICAR AICE 37 Unit 5 (rheology of dairy foods)
Physical properties decide how a food behaves in a process and in the mouth. Density and porosity set how much product fits in a silo or a pack and how fast it dries; thermal conductivity and specific heat set heating and cooling times; viscosity and yield stress set how a sauce pumps, coats and pours; viscoelastic properties set how a dough rises, a cheese stretches or a gel breaks; texture, measured instrumentally, stands in for what a consumer feels when biting; colour is often the first quality judgement a buyer makes; and dielectric properties decide how a food heats in a microwave oven. These properties are measurable, and measuring them is a routine part of product development, quality control and process design.
This subject covers the physical properties of foods and their measurement: density, porosity and particle size; thermal properties; the fundamentals of rheology — stress, strain and viscosity; Newtonian and non-Newtonian flow models; time-dependent flow; viscometers and rheometers; viscoelasticity, creep, stress relaxation and oscillatory testing; texture profile analysis and empirical texture tests; the rheology of doughs and dairy gels; colour measurement; electrical and dielectric properties; and the glass transition and state diagrams. Agricultural Engineering's engineering-properties-biomaterials covers size, shape, density, friction and aerodynamic properties of grains and produce for machine design; this subject concentrates on processed foods and on the instruments of a food laboratory.
The subject connects to several others. The flow models defined here are used to calculate pressure drop and pumping in food-transport-phenomena, which also uses the thermal properties for heat transfer; the molecular origins of viscosity, gelation and glass transition — starch pasting, protein gels, fat crystal networks — are explained in food-chemistry-major-constituents; and texture links to sensory-science. It is examined in ICAR AIEEA-PG Code 18 (physical properties of foods, colloidal systems, gels and emulsions in Unit I; engineering properties in Unit II) and ICAR AICE-JRF/SRF (rheology of dairy foods), and supports the rheology parts of GATE XE-G Section 4 and the quality-control work of food technologists and analysts. Values quoted are typical; measured data for the actual product should be used in design.
Density is mass per unit volume, but for foods the volume can be defined in several ways, and the distinction matters. True (solid or substance) density is the mass divided by the volume of the solid material alone, excluding all pores; it is measured by gas pycnometry, in which helium or nitrogen penetrates open pores, or estimated from composition, since water, protein, fat, carbohydrate and ash each have characteristic densities. Apparent (particle) density includes the internal pores of each piece but not the spaces between pieces; it is measured by liquid displacement (using a liquid that does not penetrate, such as toluene for some products, or by coating the sample) or by geometric measurement of regular shapes. Bulk density is the mass of a collection of particles divided by the total volume they occupy, including spaces between particles; it depends on how the material is filled and whether it has been tapped or compacted, so loose and tapped bulk densities are quoted separately.
Bulk density decides the size of silos, bins, hoppers and packages, the filling weight of a carton and the cost of transport; tapped bulk density and the ratio between tapped and loose values (Hausner ratio or Carr index) indicate flowability of powders. Liquid densities are measured with pycnometers, hydrometers or oscillating-tube density meters, and densities of sugar solutions and milk are linked to their composition, as in the measurement of total solids from density.
Porosity is the volume fraction of air or voids. Using the densities defined above, the total porosity of a bulk material is ε = 1 − ρ_bulk/ρ_true; inter-particle (bulk) porosity is 1 − ρ_bulk/ρ_apparent, and the internal porosity of the pieces is 1 − ρ_apparent/ρ_true. For example, a breakfast cereal with true density 1,400 kg/m³, apparent density 350 kg/m³ and bulk density 140 kg/m³ has internal porosity 1 − 350/1,400 = 0.75, inter-particle porosity 1 − 140/350 = 0.60 and total porosity 1 − 140/1,400 = 0.90.
Porosity has large effects: it lowers thermal conductivity (air conducts poorly), affects drying and frying rates, controls the crispness and crunch of puffed and extruded products, and governs how much air is in whipped products (overrun). During drying, foods shrink and their porosity changes: air-dried fruits and vegetables shrink greatly and become dense, while freeze-dried and puffed products keep or gain porosity (evaporation-dehydration). Overrun of whipped cream or ice cream, the percentage increase in volume due to incorporated air, is calculated as (volume of product − volume of mix)/volume of mix × 100, or equivalently from the densities of the mix and the aerated product.
Powders, grains, crystals, droplets and fat globules are described by size distributions rather than a single size. Measurement methods include sieving (for particles larger than about 40–50 µm), laser diffraction, image analysis, sedimentation and electrical sensing; each measures a different property and gives a distribution on a different basis (number, surface or volume), so results from different methods are not directly interchangeable. Mean diameters such as the Sauter mean d32 and the volume mean d43, and percentile values such as d10, d50 and d90, summarise distributions; the span (d90 − d10)/d50 describes width. The calculation of these means and of screen effectiveness is covered in mechanical-unit-operations.
Shape is described by sphericity (the ratio of the surface area of a sphere of the same volume to the actual surface area), roundness, aspect ratio and, for irregular particles, by image analysis. Particle size and shape affect powder flow and caking, dissolution and dispersibility, mouthfeel (grittiness of chocolate and ice cream above roughly 20–30 µm), emulsion stability and the rheology of suspensions, whose viscosity rises sharply as the volume fraction of particles approaches the maximum packing fraction.
The specific heat (specific heat capacity) cp of a food is the heat needed to raise the temperature of 1 kg by 1 K. It depends mainly on composition, because water has a much higher specific heat (about 4.18 kJ/kg K) than protein, fat, carbohydrate and ash (roughly 1.5 to 2.0 kJ/kg K). Simple models weight the components: Siebel's equations use water content alone (cp ≈ 3.35w + 0.84 kJ/kg K above freezing), while the more detailed Choi–Okos model uses temperature-dependent specific heats for water, protein, fat, carbohydrate, fibre and ash, combined in proportion to their mass fractions: cp = Σ xᵢ cpᵢ. For example, a food with 75 per cent water, 20 per cent protein and 5 per cent fat, taking cp of 4.18, 1.6 and 2.0 kJ/kg K for the components, has cp ≈ 0.75 × 4.18 + 0.20 × 1.6 + 0.05 × 2.0 = 3.56 kJ/kg K.
Below the freezing point, latent heat release makes the apparent specific heat rise to a sharp peak and then fall (chilling-freezing-frozen-foods), so enthalpy–temperature data or tables are used instead. Specific heat and enthalpy changes are measured by differential scanning calorimetry (DSC), which records the heat flow into a small sample relative to a reference as both are heated or cooled at a controlled rate, and by method-of-mixtures calorimetry.
Thermal conductivity k, the property in Fourier's law of conduction, measures how readily heat is conducted through a material. Unfrozen foods have values mostly between about 0.2 and 0.6 W/m K, close to but below that of water (about 0.6), with lower values for fatty and porous foods; frozen foods have higher values, because ice conducts about four times better than liquid water; dry, porous materials such as flour, powders and puffed products have very low values because air (about 0.025 W/m K) fills the pores. Because foods are mixtures, their conductivity depends on structure as well as composition: heat flowing parallel to fibres in meat meets less resistance than heat crossing them. Models bound the possible values: the parallel model, k = Σ vᵢ kᵢ (volume fractions), gives an upper limit, and the series model, 1/k = Σ vᵢ/kᵢ, a lower limit; real foods lie between, and empirical or structural models such as Maxwell–Eucken are used for dispersed phases.
Conductivity is usually measured by transient methods, chiefly the line-heat-source (thermal conductivity probe) method: a thin needle containing a heater and a temperature sensor is inserted into the sample, a constant heating power is applied, and the conductivity is calculated from the slope of temperature rise against the logarithm of time. Steady-state guarded-hot-plate methods are slower and used for slabs and insulating materials.
Thermal diffusivity, α = k/(ρ cp), governs how fast temperature changes spread through a material during unsteady heating or cooling; it appears in the Fourier number used in heat-penetration and freezing calculations (food-transport-phenomena). Most moist foods have diffusivities of about 1.2 to 1.6 × 10⁻⁷ m²/s, close to that of water (about 1.4 × 10⁻⁷ m²/s at room temperature). For example, a food with k = 0.50 W/m K, ρ = 1,050 kg/m³ and cp = 3,600 J/kg K has α = 0.50/(1,050 × 3,600) ≈ 1.32 × 10⁻⁷ m²/s. Diffusivity can be measured directly from temperature–time data in a sample of simple shape, or calculated from measured k, ρ and cp. Because diffusivity is low, large foods heat and cool slowly, and heating time increases with the square of size — the reason thin-profile packages and small pieces process so much faster than large cans or joints. Diffusivity changes sharply on freezing, because ice has both a higher conductivity and a lower specific heat than water; frozen foods have diffusivities several times those of the same foods unfrozen, which is one reason freezing proceeds faster than thawing.
Rheology is the study of the deformation and flow of materials under applied forces. Stress is force per unit area (Pa). When the force acts perpendicular to a surface it is a normal stress (tension or compression); when it acts parallel to the surface it is a shear stress, τ. Strain is the relative deformation: normal strain is the change in length divided by original length; shear strain γ is the displacement of one surface relative to another divided by the distance between them. Strain is dimensionless. The rate of shear strain, the shear rate γ̇ (s⁻¹), is the velocity gradient across the flowing material: for a liquid between two plates a distance h apart, one moving at velocity V, γ̇ = V/h.
Typical shear rates in food processes and use vary enormously: about 10⁻⁶ to 10⁻³ s⁻¹ for sedimentation of particles and sagging under gravity, roughly 10 to 100 s⁻¹ for pouring and mixing, about 10 to 500 s⁻¹ in chewing and swallowing (commonly quoted as around 50 s⁻¹ for sensory 'thickness' of fluids, though values vary), 10 to 1,000 s⁻¹ or more in pipe flow and up to 10⁵ s⁻¹ or more in homogenisers and spraying. A property measured at one shear rate may therefore say little about behaviour at another — a key reason for measuring flow curves rather than single-point viscosities.
An ideal (Hookean) solid deforms instantly under stress and returns fully when the stress is removed; stress is proportional to strain, τ = G γ, where G is the shear modulus (or σ = E ε in tension, with E Young's modulus). An ideal (Newtonian) liquid flows as long as stress is applied, and stress is proportional to shear rate: τ = μ γ̇, where μ is the dynamic viscosity (Pa s). The energy stored in elastic deformation is recovered; the energy used in viscous flow is dissipated as heat. Real foods often combine both behaviours (viscoelasticity, chapter 7) and many liquids do not follow Newton's law (chapter 4).
For a non-Newtonian fluid, the apparent viscosity η = τ/γ̇ is the ratio of stress to shear rate at a given shear rate, and must be quoted with that shear rate. Kinematic viscosity ν = μ/ρ (m²/s or centistokes) is measured directly by gravity-driven capillary viscometers. Viscosity of liquids falls with temperature, often following an Arrhenius-type relation, μ = A exp(Ea/RT): honey and syrups thin dramatically with warming, which is why they are heated for pumping and filling. Viscosity of solutions and dispersions rises with concentration — steeply for polymers such as hydrocolloids above the overlap (critical) concentration at which molecules begin to entangle, and for suspensions as the particle volume fraction approaches maximum packing.
Rheological measurements serve four purposes. In process engineering they provide the flow properties needed to calculate pressure drops, pump power, mixing power, heat transfer coefficients and holding-tube residence times. In quality control they check consistency of products such as ketchup, yoghurt, sauces and chocolate against specifications. In product development they relate composition and structure — starch type, hydrocolloid level, fat content, protein gelation — to behaviour, and they help design products with target textures. In sensory research they are correlated with perceived thickness, creaminess, spreadability and firmness. The structural basis of these properties — entangled polysaccharides, protein gel networks, fat crystal networks, emulsion droplets and starch granules — is explained in food-chemistry-major-constituents. Rheological data are also needed for the design of fillers and dosing pumps, for predicting how quickly particles will settle in a beverage or a sauce, and for scaling up mixing and heating from pilot to factory. Without measured flow properties, these calculations rest on guesses that often fail when a recipe or supplier changes.
A flow curve plots shear stress against shear rate. For Newtonian fluids — water, milk, clear juices, sugar syrups, edible oils and dilute solutions — it is a straight line through the origin, and viscosity is independent of shear rate. Most other liquid foods are non-Newtonian. Shear-thinning (pseudoplastic) fluids have an apparent viscosity that decreases with increasing shear rate, because shearing aligns or disentangles polymer molecules, breaks up aggregates or deforms droplets. Fruit purées, concentrated juices with pulp, sauces, dressings, hydrocolloid solutions such as xanthan and guar, and many dairy products are shear-thinning. Shear-thinning is often desirable: a dressing is thick in the bottle (low shear) but pours and spreads easily (high shear), and a sauce coats food without being slimy. Shear-thickening (dilatant) fluids, whose apparent viscosity rises with shear rate, are rare in foods; concentrated suspensions of raw corn starch are the familiar example. Because apparent viscosity varies so much with shear rate, specifications for shear-thinning products must state the shear rate or the instrument conditions, otherwise two laboratories can report very different 'viscosities' for the same sample.
Several empirical models describe flow curves. The power-law (Ostwald–de Waele) model, τ = K γ̇ⁿ, has a consistency coefficient K (Pa sⁿ) and a flow behaviour index n: n < 1 for shear-thinning, n = 1 for Newtonian (K = μ) and n > 1 for shear-thickening. The apparent viscosity is η = K γ̇^(n−1). For example, a purée with K = 10 Pa sⁿ and n = 0.4 has η = 10 × 50^(−0.6) ≈ 0.96 Pa s at 50 s⁻¹ and 10 × 500^(−0.6) ≈ 0.24 Pa s at 500 s⁻¹. Taking logarithms, log τ = log K + n log γ̇, so a log–log plot of the flow curve gives n as the slope and K from the intercept.
Some foods have a yield stress, τ0: a minimum stress below which they do not flow, behaving like a soft solid. The Bingham plastic model, τ = τ0 + μp γ̇, has a constant plastic viscosity above the yield stress. The Herschel–Bulkley model, τ = τ0 + K γ̇ⁿ, combines a yield stress with power-law behaviour and fits many foods well — tomato paste, ketchup, mayonnaise, apple sauce, minced fish and meat batters. The Casson model, √τ = √τ0 + √(η∞ γ̇), was developed for pigment suspensions and has long been used for molten chocolate, where the Casson yield value and plastic viscosity are standard quality parameters; international methods for chocolate rheology have since moved towards reporting viscosity at specified shear rates and a yield stress from the flow curve, rather than relying on the Casson model alone.
Yield stress explains why ketchup stays on a chip, why mayonnaise keeps its shape when spooned, why particles stay suspended in a dressing and why a sauce must be shaken before it pours. It can be estimated by extrapolating flow-curve data to zero shear rate using a model, but the value depends on the model and on the shear-rate range used. Direct methods include stress ramps in a controlled-stress rheometer (the stress at which flow begins), creep tests at increasing stresses, and the vane method, in which a four- or six-bladed vane is inserted into the undisturbed sample and rotated slowly; the maximum torque gives the yield stress, avoiding wall slip and disturbance of the structure. Simple empirical tests, such as the Bostwick consistometer (how far a sample flows along a trough in 30 seconds, used for ketchup and baby foods) and slump tests, are used in quality control because they are fast, though they mix yield stress with viscosity and give instrument-specific numbers. Pipe-flow design for these materials is described in food-transport-phenomena.
Some fluids change their apparent viscosity not only with shear rate but also with the time of shearing. A thixotropic fluid shows a decrease in apparent viscosity with time at constant shear rate, followed by gradual (often partial) recovery when shearing stops. The cause is a structure — a network of aggregated particles, droplets, fat crystals or weakly bonded polymers — that is progressively broken down by shear and rebuilds at rest. Yoghurt, ketchup, mayonnaise, some salad dressings, honey of certain types, starch pastes, cream cheese and many gels show thixotropy. Stirred yoghurt is a typical example: its viscosity falls during pumping and filling and partly recovers in the cup, though structure broken during processing may never fully rebuild.
Thixotropy is distinct from shear-thinning, although many foods show both: shear-thinning is an immediate response to shear rate, while thixotropy is a time-dependent response at a given shear rate. Shear history therefore matters in sampling and testing: a sample stirred before measurement gives lower readings than an undisturbed one, so standard methods specify pre-shearing or resting times.
Thixotropy is detected and quantified in several ways. In a hysteresis loop test, shear rate is increased steadily from zero to a maximum and then decreased again; for a thixotropic material the down-curve lies below the up-curve, and the area between them indicates the degree of structural breakdown. In step tests, shear rate is held constant and stress (or apparent viscosity) is recorded over time until an equilibrium value is reached; the decay is fitted to structural kinetic models. Recovery is measured by resting the sample and then measuring again, or by oscillatory tests after shearing, which track the rebuilding of the elastic modulus. Results depend strongly on the test protocol, so comparisons are valid only between tests made the same way. Because thixotropic recovery can take minutes to hours, the time between sampling and measurement, and whether the sample was stirred, must be fixed in any quality-control method; otherwise the same batch can appear to pass or fail depending only on how long it waited on the bench.
Rheopexy (anti-thixotropy) is the opposite: apparent viscosity increases with time of shearing at constant shear rate, as gentle shear promotes structure formation. It is uncommon in foods; some whipped and concentrated protein systems and certain starch dispersions show it under specific conditions. In processing, thixotropy affects pump selection and line design (start-up pressures are higher after a product has rested), filling accuracy and dosing, mixing times, and the texture of the final product after pumping. Products such as yoghurt and fresh cheese are often handled with low-shear pumps and short lines to preserve structure, and their final viscosity specifications are set after a defined recovery time. In sensory terms, thixotropic breakdown during spreading or stirring makes products easier to spread or eat while they remain thick in the container. Recognising time-dependent behaviour also helps in troubleshooting: a product that tests correctly at the end of the line but is too thin when the consumer opens it may have lost structure in a pump or filler that does not recover.
Capillary viscometers measure the time for a fixed volume of liquid to flow through a narrow tube under gravity or applied pressure. In glass gravity viscometers such as the Ostwald and Ubbelohde types, the flow time is proportional to kinematic viscosity: ν = C t, where C is the instrument constant found by calibration with a standard liquid. Comparing a sample with water in the same viscometer gives μ_sample/μ_water = (ρ_sample t_sample)/(ρ_water t_water). For example, a sugar syrup of density 1,200 kg/m³ that takes 300 s where water (998 kg/m³, 1.00 mPa s) takes 60 s has a viscosity of about (1,200 × 300)/(998 × 60) × 1.00 ≈ 6.0 mPa s. Capillary viscometers are cheap and precise for Newtonian liquids such as oils, syrups and juices, but shear rate varies across the tube, so for non-Newtonian fluids they give only an apparent value unless the Rabinowitsch–Mooney correction is applied to obtain true wall shear rates.
Pressure-driven tube viscometers, measuring flow rate and pressure drop in a tube of known length and diameter, are used for viscous non-Newtonian foods and can be installed in-line in a plant. Falling-ball viscometers time a sphere falling through the liquid and calculate viscosity from Stokes' law, μ = d²(ρs − ρl)g/(18 v), suitable for transparent Newtonian liquids.
Rotational instruments measure the torque needed to rotate one surface relative to another with the sample between them, or the rotation produced by a set torque. Concentric-cylinder (Couette or cup-and-bob) geometries suit low- to medium-viscosity liquids and suspensions; when the gap is narrow, shear rate is nearly uniform across it. Cone-and-plate geometries, with a shallow cone (commonly 1° to 4°) above a flat plate, give a uniform shear rate throughout the sample, need only a small sample and are ideal for non-Newtonian liquids without large particles. Parallel-plate geometries have an adjustable gap, tolerate particles and are used for gels, pastes, doughs and cheese, though shear rate varies with radius. Vane geometries measure yield stress and pastes prone to slip.
Simple rotational viscometers with spindles rotating in a large container (often called spindle or 'Brookfield-type' viscometers) are widespread in quality control. They give a single-point apparent viscosity at a set speed, with undefined shear rate for non-Newtonian samples, so readings are comparable only between tests with the same spindle, speed, container and temperature. Modern rheometers, operating in controlled-rate or controlled-stress mode with precise temperature control, can measure flow curves, yield stresses, creep and oscillatory properties in one instrument.
Reliable rheological data require attention to several sources of error. Temperature must be controlled closely, because viscosity can change several per cent per degree. Wall slip — a thin, low-viscosity layer forming at smooth walls, common with emulsions, suspensions and gels — causes falsely low readings and is reduced by roughened or serrated geometries and vanes. Particles must be small relative to the gap (a common rule is that the gap should be several times larger than the largest particle). Samples must be loaded consistently, with a controlled rest time to allow recovery from loading damage, and evaporation must be prevented during long tests by solvent traps or oil seals. Inertia and secondary flows at high speeds, and instrument limits at low torque, define the valid range. Results should be reported with the geometry, temperature, shear-rate range, model fitted and any pre-shear used. Calibration with certified viscosity standards at the measurement temperature, and periodic checks of geometry dimensions and torque, belong to the laboratory's quality system like any other analytical method.
Many foods — doughs, cheese, gels, meat, bread crumb, chewing gum, many semi-solid desserts — are viscoelastic: they show both elastic (solid-like) and viscous (liquid-like) responses, depending on the time scale of deformation. A useful way to picture this is with mechanical models built from springs (elastic elements, stress proportional to strain) and dashpots (viscous elements, stress proportional to strain rate). The Maxwell model, a spring and dashpot in series, represents a viscoelastic liquid: under a fixed strain the stress decays exponentially, σ(t) = σ0 exp(−t/λ), with relaxation time λ = η/E (viscosity divided by modulus). The Kelvin–Voigt model, a spring and dashpot in parallel, represents a viscoelastic solid: under a constant stress the strain rises gradually to an equilibrium value (retarded elasticity). The Burgers model combines a Maxwell and a Kelvin–Voigt element in series and describes the creep of many foods: instantaneous elastic strain, retarded elastic strain and steady viscous flow. Real foods need several elements, giving spectra of relaxation or retardation times.
The Deborah number, the ratio of a material's relaxation time to the time scale of the observation, expresses why the same material can behave as a solid in a fast test and as a liquid in a slow one: a ball of dough bounces slightly when dropped but flows slowly into a pan when left.
In a creep test, a constant stress is applied suddenly and the strain is recorded over time; then the stress is removed and recovery is followed. The result is often expressed as creep compliance J(t) = γ(t)/σ0. An elastic solid shows an instantaneous strain that stays constant and recovers fully; a viscous liquid shows strain increasing linearly with no recovery; a viscoelastic food shows a combination, and the fractions of recoverable and non-recoverable strain characterise its structure. Creep tests at increasing stresses also locate the yield stress.
In a stress relaxation test, a constant strain is applied suddenly and the decay of stress is recorded. Viscoelastic liquids relax towards zero stress; viscoelastic solids relax to a non-zero equilibrium stress. For a single Maxwell element, the time to fall to 1/e (about 37 per cent) of the initial stress is the relaxation time; for example, a dough with modulus 10 kPa and viscosity 10⁵ Pa s would have λ = 10 s. Relaxation tests on cheese, bread and fruit tissue, often in compression, are used to compare firmness and elasticity of products and to model their behaviour.
Oscillatory tests apply a small sinusoidal strain (or stress) and measure the resulting stress (or strain). For a purely elastic material the stress is in phase with the strain; for a purely viscous material it is 90° out of phase; for a viscoelastic material the phase angle δ lies in between. The response is described by the storage modulus G′ (the elastic, in-phase component, representing energy stored and recovered), the loss modulus G″ (the viscous, out-of-phase component, representing energy dissipated), the loss tangent tan δ = G″/G′, the complex modulus G* = √(G′² + G″²) and the complex viscosity η* = G*/ω, where ω is the angular frequency. For example, a gel with G′ = 1,200 Pa and G″ = 150 Pa has tan δ = 0.125, indicating a predominantly elastic structure, and G* ≈ 1,209 Pa.
Tests are made within the linear viscoelastic region (LVR), where moduli are independent of strain amplitude, found first by a strain (or stress) sweep; within it the structure is not damaged, so oscillatory tests can follow structure formation non-destructively. Frequency sweeps characterise materials: in a true gel, G′ is greater than G″ and both are nearly independent of frequency; in a concentrated polymer solution, G″ exceeds G′ at low frequency and G′ crosses above G″ at higher frequency. Time sweeps at fixed frequency follow gelation — for example the set of a rennet or acid milk gel, a starch paste or a heat-set protein gel — and the gel point is often taken as the crossover where G′ = G″. Temperature sweeps follow melting of gelatin and fat or gelatinisation of starch. The empirical Cox–Merz rule, which states that η*(ω) often equals steady-shear viscosity η(γ̇) at equal ω and γ̇, holds for many polymer solutions but not for gels and structured dispersions.
Texture is the group of physical characteristics that arise from the structure of a food and are sensed by touch, related to deformation, disintegration and flow under force, and measured objectively as functions of force, time and distance. Instrumental texture measurement aims to give rapid, reproducible numbers that correlate with sensory attributes such as hardness, crispness, chewiness and tenderness. It uses universal testing machines and texture analysers — instruments that drive a probe or fixture into or across a sample at a controlled speed while recording force against time or distance. Tests fall into three groups: fundamental tests, which measure well-defined physical properties such as Young's modulus or fracture stress; empirical tests, which measure parameters known from experience to relate to quality but not defined in physical terms; and imitative tests, which imitate the conditions of eating or use, of which texture profile analysis is the best known. Sample preparation is critical in all of these tests: size, shape, temperature, orientation of fibres and time since cooking or removal from refrigeration all change the result, so methods specify them closely.
Texture profile analysis (TPA), developed from the General Foods texture profile in the 1960s and adapted for universal testing machines by Bourne, compresses a bite-sized sample twice in succession — a 'two-bite' test — commonly to a substantial fraction of its height, and derives several parameters from the force–time curve. Hardness is the peak force of the first compression. Fracturability (brittleness) is the force at the first significant break in the first compression, if any. Cohesiveness is the ratio of the area of work during the second compression to that of the first (A2/A1), indicating how well the sample withstands a second deformation. Springiness (elasticity) is the degree to which the sample recovers its height between the end of the first compression and the start of the second, usually as a ratio of distances or times. Adhesiveness is the negative force area when the probe is withdrawn after the first compression, representing the work to pull the probe away from a sticky sample. Derived parameters are gumminess = hardness × cohesiveness, used for semi-solid foods, and chewiness = gumminess × springiness (or hardness × cohesiveness × springiness), used for solid foods; a food is described by one or the other, not both.
For example, a cheese with first-peak force 40 N, area ratio A2/A1 = 0.6 and springiness 0.8 has gumminess 24 N and chewiness 19.2 (in N, as a dimensionless springiness ratio is used). TPA results depend strongly on sample size, compression percentage, test speed and the time between compressions, so these must be standardised and reported; comparisons between laboratories using different settings are not valid.
Many empirical tests are used in quality control and grading. Puncture tests drive a cylindrical probe into a product and record the peak force: the Magness–Taylor fruit firmness tester, with a plunger of specified diameter pressed to a set depth, is used to judge harvest maturity and ripeness of apples, pears and other fruits. Shear tests cut through samples: the Warner–Bratzler shear test cuts a cylindrical core of cooked meat with a triangular-notched blade and is the standard instrumental measure of meat tenderness; the Kramer shear-compression cell, with multiple blades forced through a box of sample, is used for peas, berries and other particulate foods. Three-point bending (snap) tests measure the breaking force and deformation of biscuits, crackers, chocolate bars and spaghetti. Penetrometers measure the depth a cone or needle penetrates under a set load, used for butter, margarine and fats (spreadability) and for gels. Gel strength of gelatin is measured as the Bloom value — the mass in grams needed for a standard plunger to depress the surface of a standard gel by 4 mm under specified conditions. Acoustic measurements recorded during compression of crisp foods quantify crispness and crunchiness. Results of empirical tests are instrument- and method-specific, and their meaning comes from correlation with sensory panels (sensory-science).
Wheat flour dough is a viscoelastic network of hydrated gluten proteins in which starch granules are embedded. Its rheology decides how much water the flour absorbs, how long dough must be mixed, how well it holds gas during proofing and baking, and so the volume and crumb of bread. Gliadins contribute viscous extensibility and glutenins elasticity (food-chemistry-major-constituents). The baking industry relies on a set of standardised empirical instruments. The farinograph records the torque (in Brabender units, BU) needed to mix flour and water: water absorption is the amount of water needed to bring the curve's peak to the 500 BU line; dough development time is the time to reach peak consistency; stability is the time the curve stays above the 500 BU line; and the mixing tolerance index is the drop in consistency after the peak. Strong bread flours have high absorption, longer development and longer stability. The mixograph is a similar recording mixer used widely in wheat breeding.
The extensograph stretches a rested, shaped dough piece until it breaks, recording resistance to extension and extensibility; the area under the curve represents the energy and overall strength, and the balance of resistance and extensibility indicates suitability for bread, biscuits or noodles. The alveograph inflates a sheet of dough into a bubble until it bursts, giving the tenacity P (related to resistance), the extensibility L, the P/L ratio and the deformation energy W; it is widely used in Europe for grading wheat for bread and biscuits. Fundamental tests — oscillatory rheometry, creep and biaxial extension — are increasingly used in research to relate gluten composition to baking performance.
Starch pasting properties, measured by the amylograph (viscoamylograph) or the Rapid Visco Analyser, record viscosity as a flour or starch slurry is heated, held and cooled, giving pasting temperature, peak viscosity, breakdown and setback, as explained in food-chemistry-major-constituents. In flour quality control, the falling number test measures α-amylase activity: a heated flour–water suspension is stirred, and the time in seconds for a plunger to fall through the gelatinised paste is recorded. High α-amylase activity, as in flour from sprouted wheat, liquefies the paste quickly and gives a low falling number and a sticky, poorly sliceable bread crumb; very high falling numbers indicate low enzyme activity, often corrected by adding malt flour or fungal amylase. Typical acceptance ranges are set by millers and bakers for each product. The same instruments are used to choose starches for sauces and fillings that must keep viscosity through retorting or freezing, and to check that modified starches meet their specifications for peak viscosity, breakdown and setback.
Milk forms gels in two main ways. In rennet coagulation, chymosin cleaves κ-casein, and the destabilised micelles aggregate into a network; in acid coagulation, as in yoghurt and fresh cheese, the pH falls towards the isoelectric point of casein (about 4.6) and the micelles aggregate. Oscillatory rheometry with small strains follows both processes non-destructively: the storage modulus G′ rises from near zero as the gel forms, and the gelation time, the rate of firming and the final firmness are taken from the curve; acid gels from heated milk are firmer because denatured whey proteins associate with casein. Industrial instruments such as gel-firmness sensors and the formagraph monitor curd firmness to decide when to cut curd in cheese vats.
Yoghurt texture is measured by apparent viscosity at specified conditions (stirred yoghurt is shear-thinning and thixotropic), by penetration or back-extrusion tests in the cup (set yoghurt), and by syneresis (whey separation). Cheese texture is measured by TPA, uniaxial compression to fracture and stress relaxation; its functional properties for cooking — meltability (the Schreiber or Arnott tests, measuring spread or height loss on heating), stretchability (the length of strands pulled from melted cheese, important for pizza cheese) and oiling-off — are measured by empirical tests and by temperature sweeps in an oscillatory rheometer, where loss of G′ and rise of tan δ mark softening and flow. Depth in dairy rheology is covered in the Dairy Technology track.
Colour is the perception produced when light reflected from or transmitted through an object reaches the eye; it depends on the light source, the object and the observer. Instrumental colour measurement standardises all three: a defined illuminant (most often D65, representing average daylight, or illuminant C or A for incandescent light), a defined standard observer (the CIE 2° or 10° observer, representing the colour-matching responses of the average eye for small or larger fields of view), and the measured spectral reflectance or transmittance of the sample. From these, the CIE tristimulus values X, Y and Z are calculated, Y corresponding to luminance.
Because XYZ values are not visually uniform — equal numerical differences do not look equally different — more uniform colour spaces are used. The CIE 1976 L*a*b* (CIELAB) space is the standard in food work. L* is lightness, from 0 (black) to 100 (white); a* is the red–green axis, positive for red and negative for green; b* is the yellow–blue axis, positive for yellow and negative for blue. The older Hunter L, a, b scale has the same axes with different formulae, and values from the two scales are not interchangeable. Colour can also be expressed in cylindrical coordinates: chroma C* = √(a*² + b*²), representing saturation or vividness, and hue angle h° = arctan(b*/a*), the 'colour' itself (0° red, 90° yellow, 180° green, 270° blue).
The total colour difference between a sample and a standard is calculated as ΔE*ab = √(ΔL*² + Δa*² + Δb*²). For example, a sample with L* = 52, a* = 18, b* = 30 compared with a standard of L* = 55, a* = 16, b* = 26 has ΔE = √(9 + 4 + 16) = √29 ≈ 5.4. As a rough guide, differences of about 1 are at the threshold of perception for trained observers, and larger differences become obvious, but acceptable tolerances depend on the product and on which direction the difference lies; newer formulae such as CIEDE2000 (ΔE00) correct for non-uniformities in CIELAB and are used where precise tolerances are needed. For the same tomato paste sample, a* = 30 and b* = 24 give chroma √(900 + 576) ≈ 38.4 and hue angle arctan(24/30) ≈ 38.7°.
Colour indices derived from L*, a* and b* are used for particular foods: the a*/b* ratio for tomato products, browning indices for heated and dried foods, and whiteness indices for flour, rice and milk powder. Colour measurement is used to grade and specify raw materials (tomatoes, citrus juices, meat), to follow processes such as blanching, browning in baking and frying, and drying, to monitor storage changes such as meat discoloration (oxymyoglobin to metmyoglobin) and chlorophyll loss in vegetables, and to check consistency between batches.
Spectrophotometers measure reflectance or transmittance at many wavelengths across the visible range (about 380 to 780 nm) and calculate colour values for any illuminant and observer; tristimulus colorimeters use filters to approximate the observer responses directly for one illuminant and are cheaper and simpler for routine control. Geometry matters: 0°/45° (or 45°/0°) instruments exclude gloss and approximate visual judgement, while diffuse/8° instruments with the specular component included measure colour independent of surface gloss. Samples must be presented consistently — in the same cell, with the same thickness, packing and background — because translucent and particulate foods are sensitive to these factors. Computer vision systems with calibrated cameras and lighting measure colour over whole surfaces, which suits non-uniform products such as fried snacks, meat cuts and fruit, and allow automated grading on processing lines. Instruments are calibrated against white and black standard tiles before use, and coloured reference tiles are used to check their accuracy; results from different instruments should be compared only after confirming that they use the same illuminant, observer and geometry.
Foods conduct electricity mainly through dissolved ions — salts, acids — in their water phase, so electrical conductivity σ (S/m) rises with salt content and moisture and falls with fat content. Conductivity also rises with temperature, roughly linearly over the processing range, because ion mobility increases as viscosity falls. In ohmic (resistive) heating, alternating current is passed through the food, which heats internally by its own electrical resistance; the volumetric heat generation rate is q = σE², where E is the electric field strength (V/m). Ohmic heating heats liquids and particles at similar rates if their conductivities are similar, which makes it attractive for particulate foods in continuous processing (novel-nonthermal-processing). Because conductivity rises with temperature, regions that warm first conduct more and heat faster, so conductivity matching between particles and carrier liquid, and control of field strength, are needed. Conductivity is also used as a rapid index of composition — for example in detecting mastitic milk, whose higher salt content raises conductivity, and in monitoring salt in brines and cleaning-solution strength in CIP.
In microwave and radio-frequency heating, foods are heated by an oscillating electromagnetic field that interacts with polar molecules (mainly water, rotating to follow the field) and with ions (moving back and forth). The relevant properties are the dielectric constant ε′ (the relative permittivity, describing the ability to store electrical energy) and the dielectric loss factor ε″ (describing the conversion of electrical energy to heat); their ratio is the loss tangent, tan δ = ε″/ε′. The power absorbed per unit volume is P = 2π f ε0 ε″ E², where f is frequency, ε0 the permittivity of free space (8.854 × 10⁻¹² F/m) and E the field strength inside the food. For example, with f = 2.45 × 10⁹ Hz, ε″ = 15 and E = 1,000 V/m, P = 2π × 2.45 × 10⁹ × 8.854 × 10⁻¹² × 15 × 10⁶ ≈ 2.0 × 10⁶ W/m³. Water at room temperature has ε′ of about 78 and, at 2.45 GHz, ε″ of about 10; dissolved salt raises ε″ through ionic conduction, especially at lower frequencies, and fat and dry solids have low values.
Penetration depth — the depth at which microwave power falls to about 1/e (37 per cent) of its surface value — decreases as ε″ increases; at 2.45 GHz it is typically a centimetre or so in moist foods, so large or salty items heat mostly near the surface, with the interior heated by conduction. Industrial and domestic microwave ovens use 2.45 GHz, with 915 MHz (896 MHz in some regions) also allocated for industrial use, giving deeper penetration; radio-frequency heating uses lower frequencies such as 13.56, 27.12 and 40.68 MHz, with still deeper penetration, which is why it is preferred for tempering and heating large blocks (chilling-freezing-frozen-foods).
Dielectric properties change with temperature and with state. Ice has a very low loss factor compared with liquid water, so in a partly frozen food, thawed regions absorb microwave energy much more strongly than frozen regions, heat faster, and become still more absorbing — the runaway heating that makes microwave thawing uneven. In salty foods, ε″ rises with temperature because ionic conduction increases, which can also cause runaway heating at edges and corners. Field concentration at the edges and corners of rectangular trays, and standing-wave patterns in the oven cavity, add to non-uniformity; rotating turntables, mode stirrers, round or ring-shaped containers, pulsed power and formulation (adjusting salt and moisture) are used to improve uniformity. Microwave-transparent packaging materials such as most plastics, glass and paper let energy through, while metals reflect it; susceptors — thin metallised films that absorb microwaves and become hot — are used in packs to brown and crisp the surface of foods such as microwave popcorn and pizza.
The glass transition — the change of an amorphous solid from a hard, brittle glass to a soft, rubbery state on heating or plasticisation by water — was introduced in food-chemistry-major-constituents with the Gordon–Taylor and WLF equations. In physical-property terms, the transition shows as large changes in measurable properties over a narrow temperature range: a step increase in specific heat (seen as a step in a DSC thermogram), a fall of several orders of magnitude in the storage modulus and in viscosity, a peak in tan δ in dynamic mechanical analysis, and changes in thermal expansion and dielectric properties. Because the transition occurs over a range, the onset, midpoint and endpoint temperatures, and the measurement method and heating rate, must be reported; values from DSC and from mechanical methods differ by several degrees. For practical work, the transition is often described by the temperature difference T − Tg between storage temperature and glass transition, because rates of physical change above Tg depend on this difference more than on temperature itself.
Differential scanning calorimetry is the standard method: the sample and an empty reference pan are heated at a constant rate, commonly 5 to 10 K/min, and the heat flow difference is recorded. The glass transition appears as a step change in heat flow (baseline shift) reflecting the change in heat capacity; melting and crystallisation appear as peaks, and their areas give enthalpies. Modulated DSC separates reversing (heat capacity) and non-reversing (relaxation, crystallisation) signals and resolves weak transitions in complex foods. Dynamic mechanical analysis (DMA) and dynamic mechanical thermal analysis (DMTA) apply an oscillating stress to a solid sample while heating and record E′, E″ and tan δ; the glass transition appears as a drop in E′ and a peak in tan δ, and mechanical methods are more sensitive than DSC for low-moisture cereal products and films. Rheometers with temperature control serve the same purpose for softer materials, and nuclear magnetic resonance relaxometry measures water and polymer mobility directly.
A state (supplemented phase) diagram plots temperature against solids concentration for a food system and shows, on one chart, the equilibrium freezing (melting) curve of ice, the solubility curve of the solute, the glass transition curve (Tg against concentration), and the point where the freezing curve meets the glass curve: the maximally freeze-concentrated glass at temperature Tg′ and concentration Cg′ (corresponding to unfrozen water Wg′). The diagram shows at a glance which state a product is in at any temperature and moisture content, and which way processing will move it.
State diagrams are used to design and control processes: in drying, to keep products below the stickiness curve (which lies a little above Tg) in spray dryers and to choose drying aids; in freeze drying, to keep the product below the collapse temperature near Tg′ during primary drying (evaporation-dehydration); in freezing and frozen storage, to choose storage temperatures that limit ice recrystallisation and reactions, ideally near or below Tg′ (chilling-freezing-frozen-foods); and in formulating low-moisture products such as biscuits, cereals and confectionery, to keep them glassy and crisp at storage conditions, or to make them deliberately rubbery and chewy. Combined with sorption isotherms, the state diagram gives the critical water activity and moisture content at which a product's Tg falls to room temperature — a key specification for crisp and powdered foods.
A student uses this subject to understand the property values that appear in every food-engineering calculation and to answer MCQs on flow behaviour, viscoelasticity, TPA, colour and glass transition. In the laboratory the same student measures bulk and true density and porosity of a cereal, determines a flow curve with a rotational viscometer and fits the power-law model, runs TPA on cheese or gels, measures L*a*b* of browned samples, and compares farinograph curves of strong and weak flours. They also learn to report measurements with the conditions that define them.
A quality or product-development technologist five years in uses these measurements daily: setting viscosity and Bostwick specifications for sauces and baby foods, checking yoghurt texture after filling, running farinograph and falling-number tests on incoming flour, TPA and Warner–Bratzler tests for product claims, colour checks against tolerances in ΔE units, and bulk-density and particle-size checks on powders. They also translate complaints — runny sauce, soggy biscuits, uneven microwave heating — into measurable properties and causes. They keep the instruments calibrated and the methods standardised so that results from different shifts and plants can be compared.
A senior R&D or quality manager uses rheology and physical properties to set specifications across product ranges and suppliers, to validate reformulations (fat, sugar or salt reduction, plant-protein replacement) by showing that texture and flow match the original, to transfer methods between plants and laboratories with consistent instruments and protocols, and to choose new process technologies such as ohmic or microwave heating on the basis of electrical and dielectric data. They decide which empirical tests remain useful and when fundamental measurements are needed.
A research leader works on structure–property relationships: large-amplitude oscillatory shear, extensional rheology, tribology and oral processing to link instrumental data to sensory perception, rheology of plant-protein gels and meat analogues, glass-transition-based stability models, and dielectric modelling of microwave processes. They develop methods, contribute to international standard methods, supervise doctoral research and design examinations that test understanding of the physics behind the instruments. They also judge which empirical industry tests can be replaced by fundamental measurements and how to relate new measurements to the long historical databases built on the old ones.
This manual does not cover extensional rheology, large-amplitude oscillatory shear, tribology or oral-processing measurements in depth, nor the full derivations of viscometer equations (such as the Rabinowitsch–Mooney and Margules equations). Tabulated property values for individual foods are not given, because they vary with composition, variety and method; questions supply the values they need. Standard-method details for farinograph, alveograph, falling number and chocolate rheology are summarised, not reproduced, and must be taken from the current official methods. Sensory correlation is in sensory-science; molecular explanations of gels and glass transition in food-chemistry-major-constituents.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions. Every numerical answer was computed by script. Items tagged to licensing, analyst or recruitment exams whose syllabi Laisha has not yet verified are quality-control applications written to the expected scope.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.