Paint has to be thin enough to brush and thick enough not to run. No ordinary liquid can be both

Consider what a wall paint is asked to do. Under a brush it must spread easily and thinly. One second later, clinging to a vertical surface, it must stop moving entirely. And in the minutes after that, it must flow just enough to let the brush marks level out into a smooth film — but not so much that it slides down the wall.

Spread easily, then don't move, then move a little, then stop. For an ordinary liquid those are contradictory instructions, because an ordinary liquid has one thickness and obeys it regardless of circumstance.

Paint is not an ordinary liquid. Neither is toothpaste, ketchup, shampoo, blood, drilling mud, screen-printing ink or most of what is manufactured and sold in a container. Their resistance to flow depends on how hard they are being pushed, and on how long ago the pushing stopped — and engineering that dependence is a discipline in its own right.

Viscosity that is not a number

Water is Newtonian: its viscosity is a fixed property at a given temperature. Stir twice as hard and it flows twice as fast, and that is the end of the story.

Most formulated liquids are not like that. The important behaviour is shear thinning — viscosity falls as the shear rate rises. The liquid is thick when left alone and thin when worked.

The reason is structural. In a shear-thinning fluid, something inside is arranged in a way that resists flow at rest: polymer chains coiled and entangled, or fine particles held in a loose network by weak attractive forces. Start shearing and that arrangement is dismantled — chains stretch and align with the flow, the particle network breaks into fragments that slide past each other. Less structure means less resistance, so the harder you push, the easier it gets.

Stop pushing and the structure reassembles. Which introduces the variable that turns out to matter most.

The clock is the hard part

If the structure rebuilds instantly, the fluid is simply shear thinning. If it takes measurable time, the fluid is thixotropic, and that recovery time is the single most important number in a great many products.

Paint makes the tension concrete. After the brush passes, two things must happen in sequence and they want opposite properties.

Levelling — brush marks flowing sideways into a smooth surface — needs the paint to stay fluid for a while. Sag resistance — the film not sliding down a vertical wall — needs it to become solid-like quickly. Recover too fast and the brush marks freeze in place as visible ridges. Recover too slowly and the whole coat slides and forms curtains and runs.

The formulator's job is to place the recovery time in the narrow window between those failures. When a paint leaves brush marks or when it sags, it is almost never because it was "too thick" or "too thin" in any single-number sense. It is because the clock was set wrong.

The same tension appears wherever a material is applied and then expected to stay put. An adhesive or sealant in a vertical joint must not slump before it cures — which is a distinct requirement from its bond strength, and a common reason a technically strong product performs badly in use. A screen-printing ink must flow through the mesh and then stop dead so the printed edge stays sharp.

Yield stress: the fluids that are solid until provoked

A related and even more useful property is a yield stress — a threshold below which the material does not flow at all, behaving as a weak solid, and above which it flows readily.

This is why toothpaste sits on the brush in a neat ribbon instead of spreading out. Gravity applies a stress; the stress is below the yield point; nothing happens. Squeeze the tube and you exceed the threshold, and it flows.

It is also the correct explanation of the ketchup bottle. The contents are not merely thick — they are genuinely not flowing at all under their own weight. Striking the bottle applies a stress well above the yield point, the structure collapses, and the now-thin liquid arrives in a quantity nobody wanted. The traditional remedy of tapping the neck at an angle works because it applies enough stress to yield the contents near the opening without dumping momentum into the entire bottle.

There is a serious version of this in industry. Drilling fluid circulating in a borehole must be fluid enough to pump, and must carry rock cuttings up to the surface. When pumping stops — for a pipe connection, or a fault — an ordinary fluid would let every suspended cutting settle, and restarting into a bed of solids is how expensive things go wrong downhole. A yield-stress fluid simply holds the cuttings in suspension, motionless, until pumping resumes. The property is not a convenience; it is what makes the operation safe.

For these liquids, "viscosity" is not a property you can look up. It is a curve — viscosity against shear rate — and often a curve that also depends on how long ago the material was last disturbed. A single number quoted without a shear rate is not a specification, it is a rounding of something that had a shape.

And the opposite: liquids that fight back

A smaller class does the reverse. Shear thickening fluids become more viscous the harder they are sheared, and the demonstration everyone has seen is cornflour stirred into water: pour it slowly and it flows, punch it and it resists like a solid.

The mechanism is jamming. In a densely packed suspension, slow flow lets particles move around each other with lubricating fluid between them. Force them to rearrange quickly and they cannot get out of the way in time — they lock into transient clusters pressing directly against one another, and the resistance climbs sharply.

The obvious application is impact protection, and fabrics impregnated with shear-thickening fluid are a real and researched approach to flexible armour: limp while you move, stiff when struck. The honest position is that this works and is not a replacement for conventional armour; it is used as a supplement, and the engineering problems are weight, durability and what repeated impacts and laundering do to the suspension.

Shear thickening is more often a nuisance than a feature. A concentrated slurry that thickens in a pump can stall it, which is why high-solids formulations are checked across the shear rates their process will actually impose.

The nanomaterial that almost nobody sees

Now the part that connects this directly to the catalogue, because one of the highest-volume applications of a nanomaterial anywhere is one almost no consumer has heard of.

Fumed silica — silica made by flame processes, the route described in our article on how nanoparticles are made — is sold in enormous quantities as a rheology modifier. Dispersed into a liquid at a few per cent, its fine aggregated particles link into a three-dimensional network held together by hydrogen bonding between the silanol groups on their surfaces.

That network is exactly the structure described above. At rest it gives the liquid a yield stress and stops it flowing. Under shear it breaks, the liquid thins and spreads. At rest again the hydrogen bonds re-form and the structure rebuilds — and the rate at which it rebuilds can be tuned by the grade of silica, its surface treatment and the loading.

So when a paint does not sag, when a sealant stays in a vertical joint, when a resin does not drain off a mould, when a cosmetic holds its texture, there is a reasonable chance that a nanoscale oxide network is doing the work. Organoclays — layered silicates surface-modified to disperse in organic media — do the same job by a related mechanism, as do cellulose nanofibrils in water-based systems.

Two consequences follow, and both are familiar from elsewhere in this catalogue. The effect depends entirely on dispersion: the network requires particles distributed through the liquid, and a poorly dispersed batch delivers lumps instead of structure, so the mixing protocol is part of the formulation rather than an afterthought. And it is the surface chemistry that matters, not the bulk material — the same silica with a hydrophobic surface treatment builds a different network, or none at all, in the same liquid. As with gas sensors and catalysis, what the particle's outside is doing decides what the product does.

Why one viscosity number tells you almost nothing

The practical consequence for anyone specifying or comparing these materials is worth stating bluntly.

Real processes span an enormous range of shear rates. Levelling and sagging happen at very low shear, of order a hundredth to a tenth of a reciprocal second. Brushing and stirring are around a hundred to a thousand. Spraying and high-speed coating reach tens or hundreds of thousands. A fluid's behaviour at one end of that range says almost nothing about the other, and a product can be correctly formulated at the shear rate it is applied at and completely wrong at the shear rate at which it must then hold still.

So a useful specification is a flow curve — viscosity measured across the relevant shear rates — plus, for thixotropic materials, a recovery measurement showing how the structure rebuilds with time after shearing stops. Anyone comparing two products on a single quoted viscosity is comparing two points on two curves that may cross.

This is the same lesson this site has now reached from several directions: a strength without a test geometry, a particle size without a technique, a shielding figure without a thickness — and now a viscosity without a shear rate. The number is not wrong. It is under-specified, which is worse, because it looks usable.

Why it matters for students and researchers

Rheology governs an enormous share of manufacturing and is taught to a small fraction of the people who will need it. Coatings, adhesives, inks, ceramics processing, foods, cosmetics, pharmaceuticals, cement and oilfield operations are all rheology problems wearing different labels, and the same three concepts — shear thinning, yield stress, thixotropy — explain most of what goes wrong in all of them.

It is also a good example of a property that belongs to a formulation and a process together rather than to a substance. The same resin with and without two per cent of fumed silica is, functionally, two different products, and which one is correct depends on whether the part is coated horizontally or vertically. Students who learn to ask "at what shear rate, and how long after?" acquire a habit that makes them immediately useful in any formulating industry.

The open problems are substantial. Predicting the rheology of a concentrated nanoparticle suspension from the particle properties remains difficult, so formulation is still largely empirical. Measurement at very high shear rates, and of structure recovery in the first fractions of a second, is genuinely hard and matters for spraying and printing. Shear thickening in dense suspensions is an active research area where the microscopic mechanism is still being argued over. And for additive manufacturing, inks and pastes that flow through a nozzle and then hold a printed shape without slumping are a rheology specification more than a materials one.

Frequently asked questions

Why does ketchup refuse to pour and then come out all at once?

Because it has a yield stress. Under its own weight the applied stress is below the threshold, so it behaves as a soft solid and does not flow at all. When you hit the bottle you apply a stress well above the threshold, the internal structure collapses, and what is now a thin liquid arrives with momentum behind it. Tapping the neck at an angle works better than hitting the base because it yields only the material near the opening.

Why does my paint show brush marks, or sag?

Both are the recovery clock being wrong for the conditions. Brush marks mean the paint regained its structure before it had time to level; sagging means it stayed fluid too long on a vertical surface. Thinning a paint beyond what the manufacturer specifies usually causes sagging, because it weakens the very network that was supposed to stop the flow. Applying thinner coats, and respecting the recommended dilution, is more effective than any brushing technique.

Is cornflour body armour a real thing?

The effect is real and shear-thickening fluid impregnated fabrics are a genuine research and development area. They are treated as a supplement to conventional armour rather than a replacement, and the practical obstacles are added weight, durability of the suspension over time, and behaviour after repeated impacts and cleaning. The kitchen demonstration is honest physics; the product is an engineering problem that is not finished.

Why does a medicine bottle say "shake well before use"?

Because it is a suspension, and if the solid drug particles have settled, the first dose poured is weaker than the label and the last is stronger. Formulators fight this with suspending agents that create a weak yield stress, so the particles are held in place rather than settling — but the structure is deliberately weak enough that gentle shaking restores uniformity. Shaking is not a ritual; skipping it is a dosing error.

Why do some paints say stir rather than shake?

Because shaking beats air into the paint, and entrained bubbles produce pinholes and a poor finish, while the shear from stirring is enough to break down the thixotropic structure anyway. For a thixotropic paint there is a second reason: what you want is to restore uniform structure and disperse settled pigment, and slow thorough stirring does that without aeration or splitting an emulsion.