Tyndall Effect

A beam you can see from the side

Shine a laser pointer through a glass of salt water in a dark room and look at it from the side. You see nothing — just a bright dot where the beam hits the far wall. Now shine the same beam through a glass of water with a single drop of milk stirred in. The beam springs to life: a glowing shaft hangs in the liquid, visible from every angle. Nothing was added to make light; the beam simply revealed itself. That side-visible beam is the Tyndall effect, and it is the single quickest way to tell a colloid from a true solution.

The reason is particle size. In salt water the dissolved species are individual Na⁺ and Cl⁻ ions, each well under one nanometre. They are so much smaller than the 400–700 nm wavelength of visible light that they barely perturb the passing wave — the beam slides through untouched. In the milky water the dispersed objects are fat globules and casein protein clusters of roughly 100–1000 nm. These are comparable to the wavelength of light, and a particle that size acts like a tiny antenna: it intercepts the oscillating electric field, its electrons slosh in sympathy, and it re-radiates light in all directions. The light you see from the side is that re-radiated, scattered light.

The Irish physicist John Tyndall studied this in the 1860s while investigating why the sky is blue, passing light through tubes of dust, smoke, and condensing vapours. He found that perfectly dust-free air gave no beam at all — only when fine particles were present did the path light up. The phenomenon now carries his name, though the underlying small-particle scattering law had been worked out in parallel by Lord Rayleigh.

The three size classes of a mixture

Chemistry sorts liquid mixtures into three families by the size of the dispersed particles, and the Tyndall effect cleanly separates the middle one from the others.

	True solution	Colloid	Coarse suspension
Particle size	< 1 nm (ions, molecules)	1 – 1000 nm	> 1000 nm (1 µm+)
Tyndall effect	No — beam invisible	Yes — beam glows	Yes (but cloudy/opaque)
Settles on standing	Never	No (kept up by Brownian motion)	Yes — sediments
Passes filter paper	Yes	Yes	No
Passes a dialysis membrane	Yes	No	No
Appearance	Clear, transparent	Often translucent / milky	Turbid, opaque
Examples	Salt water, sugar water, CuSO₄(aq)	Milk, fog, jelly, ink, blood plasma	Muddy water, chalk in water

The Tyndall test and the settling test together pin down the class. A true solution never scatters and never settles. A colloid scatters but does not settle — its particles are small enough that random thermal kicks (Brownian motion) keep them suspended indefinitely against gravity. A coarse suspension scatters strongly but also settles out and is caught by filter paper. The colloid is the only one of the three that is simultaneously stable, unfilterable, and Tyndall-active.

The physics: Rayleigh's 1/λ⁴ law

For particles much smaller than the wavelength of light, the scattered intensity is given by the Rayleigh scattering law. For a single small sphere of radius r, the fraction of light scattered per particle (the scattering cross-section) is:

        2 π⁵       d⁶     ( n² − 1 )²
σ  =  ─────── · ───── · ( ─────── )
        3        λ⁴      ( n² + 2 )

where d = 2r is the particle diameter, λ is the wavelength of the light, and n is the ratio of the refractive index of the particle to that of the medium. Two dependencies do almost all the work here:

• Sixth power of size (σ ∝ d⁶). Double the diameter and the scattering jumps by 2⁶ = 64×. Shrink a colloidal particle tenfold toward molecular size and you lose a factor of 10⁶ in scattering. This is precisely why a 0.3-nm ion scatters roughly 10¹⁸ times less — a billion billion — than a 300-nm colloid particle (a 1000× size gap raised to the sixth power), and why solutions look perfectly clear.
• Inverse fourth power of wavelength (σ ∝ 1/λ⁴). Blue light (≈450 nm) scatters (700/450)⁴ ≈ 5.8× more strongly than red light (≈700 nm). Side-scattered light from a fine colloid therefore looks bluish, while the transmitted beam emerging straight ahead is reddened.

That same 1/λ⁴ factor, acting on the air's own molecular density fluctuations, is what makes the daytime sky blue and the setting Sun red. The Tyndall effect in a beaker and the colour of the sky are the same equation evaluated for different particles.

Putting numbers on it

How big is the size gap the Tyndall effect exploits? Compare a sucrose molecule in true solution with a fat globule in milk:

Sucrose molecule:   d ≈ 0.9 nm
Milk fat globule:   d ≈ 300 nm
Size ratio:         300 / 0.9 ≈ 330×
Scattering ratio:   330⁶ ≈ 1.3 × 10¹⁵

A single milk-fat globule scatters roughly a quadrillion times more light than a single sucrose molecule. Even though a sugar solution contains vastly more dissolved molecules than a dilute milk contains globules, the per-particle sixth-power penalty overwhelms the head-count, and the sugar solution stays optically silent.

The wavelength dependence is just as concrete. Take violet light at 400 nm and deep red at 700 nm scattering off the same small particle:

I(violet) / I(red) = (700 / 400)⁴
                   = 1.75⁴
                   ≈ 9.4×

Violet scatters nearly ten times more than red. The crossover that ends this colour selectivity occurs when the particle diameter grows to roughly the wavelength of light — around 400–700 nm. Above that, scattering becomes nearly wavelength-independent (the Mie regime), and the colloid scatters white. That is the boundary between a faint-blue cigarette smoke (particles ~100 nm) and a flat-white fog bank (droplets ~10,000 nm).

Where the Tyndall effect shows up

A colloid is any two-phase dispersion with the dispersed phase in the 1–1000 nm window, and the effect appears across every combination of phases:

Colloid type	Dispersed phase in medium	Everyday example
Aerosol (liquid)	liquid droplets in gas	Fog, mist, clouds, hairspray
Aerosol (solid)	solid in gas	Smoke, dust haze
Emulsion	liquid in liquid	Milk, mayonnaise, butter
Sol	solid in liquid	Ink, paint, blood plasma, gold sol
Gel	liquid in solid	Jelly, gelatin, cheese, agar
Foam	gas in liquid/solid	Whipped cream, shaving foam

Some vivid real-world appearances of the effect:

• Sunbeams through trees and church windows. The visible shafts of light ("crepuscular rays" or "God rays") are the Tyndall effect off dust, pollen, and water-droplet aerosols suspended in the air. In genuinely clean air the beam would be invisible — exactly Tyndall's original observation.
• Headlamps in fog. The cone of glowing light ahead of a car is the beam scattering off micron-scale fog droplets. Because the droplets are in the Mie regime, the glow is white, not blue — and it is why fog lamps point low to avoid bouncing scattered glare back into the driver's eyes.
• Faraday–Tyndall gold sols. Michael Faraday's 1857 ruby-red colloidal gold preparations — gold nanoparticles ~10–50 nm dispersed in water — still sit in vials at the Royal Institution and still show a clean Tyndall cone. Their colour and scattering come from the same small-particle physics; the field of nanoparticle plasmonics grew directly out of these samples.
• Opalescence and the blue of diluted milk. Skim milk viewed from the side against black looks faintly blue (small casein micelles, ~100 nm, favouring blue scatter), while light transmitted through it looks creamy-yellow (the blue having been scattered away). Whole milk, with larger fat globules, scatters more whitely.

Turning scattering into a measurement

Once you accept that scattered intensity encodes particle size and concentration, the Tyndall effect becomes a quantitative instrument.

• Nephelometry and turbidimetry. A nephelometer measures light scattered at 90° to the incident beam to gauge how many particles are present. Drinking-water turbidity is reported in NTU (Nephelometric Turbidity Units); regulatory limits are typically below 1 NTU for treated water, and a reading climbing past that flags a failing filter. Turbidimetry instead measures the transmitted (attenuated) beam, the complementary signal described by the Beer–Lambert law.
• Dynamic light scattering (DLS). Point a laser at a colloid and the scattered intensity does not sit still — it flickers as particles wander in and out of phase by Brownian motion. Small particles diffuse fast and make the signal flicker quickly; large particles diffuse slowly. Fitting the autocorrelation of that flicker gives a diffusion coefficient D, and the Stokes–Einstein relation d = k_B·T / (3πηD) converts it to a hydrodynamic diameter. DLS routinely sizes proteins, micelles, and nanoparticles from ~1 nm to a few µm.
• Tyndallometry of reactions. Because a precipitate first appears as a colloid before the particles grow and settle, the onset of a Tyndall beam can time the very start of a precipitation reaction — useful for studying nucleation kinetics before any cloudiness is visible to the eye.

Common misconceptions and pitfalls

• "The Tyndall effect proves a colloid; a true solution can't scatter at all." A true solution scatters extraordinarily weakly, not literally zero — Rayleigh scattering off molecular density fluctuations is what makes a kilometre of clean atmosphere blue. In a centimetre-wide beaker that residual scattering is utterly undetectable, which is what makes the test practical, but the physics is continuous, not a hard on/off switch.
• "Tyndall scattering and Rayleigh scattering are different phenomena." They are the same mechanism. Rayleigh scattering is the small-particle limiting law; the Tyndall effect is its visible manifestation on colloidal particles. Calling them rivals is a textbook artefact.
• "Bigger particles always scatter more visibly, so a coarse suspension shows the cleanest beam." A coarse suspension scatters a lot, but it is turbid, opaque, and settles within minutes, so you cannot see a clean beam through it — and it fails the stability and filtration tests for a colloid. The cleanest, most persistent Tyndall cone comes from a stable colloid, not the largest particles.
• "The scattered light is a different colour because the particles glow." Nothing fluoresces or emits new light. The particles re-radiate the same incident light; the apparent bluish tint of side-scatter is purely the 1/λ⁴ weighting favouring shorter wavelengths, not any emission.
• "Concentration is what matters — dilute enough and any colloid stops showing it." Dilution lowers the beam's brightness but not the qualitative result. The defining variable is particle size, not number density. A single drop of milk in a litre of water still gives a faint but real Tyndall cone, whereas a saturated sugar solution gives none.
• "It only works with lasers." Any collimated, intense beam works — a slide-projector slit, a torch through a card aperture, even a sunbeam. A laser is just convenient because it is bright, narrow, and monochromatic, making the cone sharp.