Zeta Potential: What the Number Means and When It Lies

The plus-or-minus-30 rule is a rule of thumb

Almost every introduction to colloid characterisation states that a zeta potential more positive than +30 mV or more negative than -30 mV indicates a stable dispersion. It is a reasonable first filter and it is repeated as though it were a physical law. It is not. It is a heuristic for one stabilisation mechanism — electrostatic repulsion — measured in a particular kind of medium, and there are several common, entirely ordinary situations in which it gives the wrong answer in both directions.

This guide is about what the measurement actually is, so that you can tell which situation you are in.

What zeta potential is

The electrical double layer

A particle surface in an electrolyte acquires charge — by ionisation of surface groups, by adsorption of ions, or by lattice substitution. That charge attracts counter-ions from solution. The result is a structured region: a compact layer of ions held close to the surface, and beyond it a diffuse layer in which the counter-ion excess decays with distance into the bulk.

The slipping plane, and why zeta is not the surface potential

When the particle moves relative to the fluid, some of that ion atmosphere moves with it. The notional boundary between the fluid that travels with the particle and the fluid that does not is the slipping or shear plane, and zeta potential is the electrostatic potential at that plane — not at the surface.

The distinction matters more than it sounds. The surface potential is generally higher in magnitude than the zeta potential, and anything that adsorbs onto the particle and extends outward — a polymer, a protein, a non-ionic surfactant — pushes the slipping plane further out into the diffuse layer, where the potential has already decayed. That is why coating a highly charged particle with a neutral polymer reduces its measured zeta potential while making it more stable, which is the most common way the plus-or-minus-30 rule fails.

How it is measured

Electrophoretic light scattering

An electric field is applied across the sample, charged particles migrate, and the Doppler shift of light scattered from the moving particles gives their velocity. Velocity divided by field strength is the electrophoretic mobility. That is the measured quantity. Zeta potential is derived from it by a model, and the model is where the assumptions enter.

Mobility to potential: two limits and a function between them

The conversion depends on the ratio of particle radius to the thickness of the diffuse layer — conventionally written as the product of the inverse Debye length and the particle radius.

• Thin double layer, meaning the layer is thin compared with the particle: the Smoluchowski limit applies. This is the default in essentially all commercial software.
• Thick double layer, meaning the layer is large compared with the particle: the Hückel limit applies, and differs from Smoluchowski by a factor of one and a half in the derived potential.
• Between them, Henry's function interpolates, and neither limit is correct.

Small nanoparticles in low-ionic-strength media sit squarely in the awkward middle, and applying the Smoluchowski default there overestimates the magnitude of the zeta potential.

Where the double layer thickness comes from

The Debye length depends on ionic strength. In water at room temperature it is approximately 0.304 nanometres divided by the square root of the ionic strength in moles per litre for a symmetrical monovalent electrolyte. That gives, in round numbers, about 9.6 nm at 1 mM, about 3.0 nm at 10 mM, about 0.96 nm at 100 mM, and about 0.8 nm at physiological ionic strength.

Now apply it. A 50 nm particle has a radius of 25 nm. In 10 mM buffer, the radius-to-Debye-length ratio is a little over eight — marginal for Smoluchowski, and the error is modest. In 1 mM it is under three, where Smoluchowski is clearly the wrong limit and the software will apply it anyway without comment. For a 10 nm particle in 1 mM the ratio is around a half, and the default conversion is simply wrong. The instrument will still print a number to one decimal place.

The five situations where the number lies

1. No pH is reported

Surface charge on most oxides, and on any material bearing carboxyl or amine groups, is set by protonation equilibria. Zeta potential is therefore a function of pH, not a constant. A single value with no pH attached is not a property of the material; it is a property of whatever the medium happened to be.

The useful measurement is the whole curve, and specifically the isoelectric point — the pH at which the zeta potential passes through zero. That is a genuine material property, it identifies the surface chemistry, and it tells you which pH ranges to avoid, because a dispersion held near its isoelectric point will aggregate. For a functionalised particle, a shift in the isoelectric point relative to the bare material is direct evidence the functionalisation worked.

2. No ionic strength is reported

Added salt compresses the double layer, which reduces the magnitude of the measured zeta potential and simultaneously reduces the range over which electrostatic repulsion acts. A particle at -45 mV in deionised water may sit at -15 mV in buffer and aggregate on contact with cell culture medium. Quoting the deionised-water figure for a material intended for biological use is technically true and practically useless.

This is also why zeta measurements should be made in the medium of the intended application wherever possible, and why "diluted in deionised water for measurement" invalidates the result for exactly the samples where stability matters most.

3. The particle is sterically stabilised

Polymer-coated particles — polyethylene glycol being the standard example — are stabilised by the entropic cost of compressing the polymer layers, not by charge. Their zeta potential is often close to zero, and they are frequently more stable in high salt than any charged particle, because steric stabilisation is insensitive to ionic strength. Applying the plus-or-minus-30 rule to them predicts instability that does not occur.

The converse case is worth naming too: a highly charged particle in a medium containing multivalent counter-ions can aggregate rapidly despite a large zeta potential, because a small concentration of a multivalent ion collapses the double layer far more efficiently than a monovalent one.

4. The sample is not what the instrument thinks it is

• Too concentrated, and multiple scattering and particle-particle interaction distort the measured mobility.
• Too conductive, and the applied field drives Joule heating and electrode polarisation; the instrument compensates by reducing the field, which reduces the signal, and eventually by refusing. High-salt samples are genuinely hard to measure and the difficulty is physical, not a fault.
• Bubbles and electrode degradation produce plausible-looking nonsense. Zeta cells are consumables; an old one with a degraded electrode gives reproducible wrong answers.
• Sedimenting samples — large or dense particles — drift out of the measurement volume during acquisition.

5. It is being asked to predict stability on its own

Zeta potential quantifies one term in the balance. Aggregation behaviour follows from the sum of electrostatic repulsion, van der Waals attraction, steric effects and any specific chemical interaction, and the attractive term depends on the material's optical properties and on particle size — neither of which appears anywhere in a zeta measurement. Two dispersions with identical zeta potentials and different Hamaker constants behave differently.

The direct test of stability is stability: measure size over time, at the relevant temperature, in the relevant medium. Zeta potential is a fast, useful, indirect indicator that tells you why something is or is not stable, which is exactly what a stability time-course does not tell you. Use both. The sizing comparison covers the time-course half.

What a usable zeta report contains

• The electrophoretic mobility as well as the derived potential. Mobility is the measurement; potential is an interpretation, and a reader who disagrees with the model can redo it from the mobility.
• The model applied — Smoluchowski, Hückel, or Henry with the value used.
• pH, and how it was set and measured.
• Medium composition and ionic strength or conductivity.
• Temperature, and confirmation the sample was equilibrated.
• Concentration, and the number of runs with a spread rather than a single value.
• Ideally, a pH titration curve with the isoelectric point marked.

A certificate line reading "zeta potential: -35 mV" meets none of these. What to do about that is how to read a certificate of analysis.

Standards

The ISO 13099 series covers the determination of zeta potential: the first part sets out the theory and definitions, and later parts address the optical and acoustic measurement methods. It is the document to point at when a supplier's figure arrives without its conditions, because it makes explicit that the reporting conditions are part of the result.

Related guides

• Where colloid characterisation sits: the characterisation workflow.
• The long-form treatment with the scattering theory: DLS and zeta potential.
• Keeping a dispersion stable on the shelf rather than in a cuvette: storing and handling nanomaterials safely.