Engineered nanoparticles moving through saturated soil are removed by attaching to sand grains — the same colloid filtration theory (CFT) used to design groundwater remediation barriers and predict how far a spill of nanomaterial can travel before it is trapped.
Single-collector efficiency (Yao, 1971):
η₀ = η_D + η_I + η_G
η_D = 4.04·Pe⁻²ᐟ³ Pe = dc·v / D (diffusion)
η_I = 1.5·(dp/dc)² (interception)
η_G = g(ρp−ρf)dp² / (18μv) (gravity settling)
D = kT / (3πμ·dp) (Stokes–Einstein diffusion coeff.)
Filter coefficient: k_att = [3(1−θ)/(2dc)]·α·η₀
Breakthrough: C/C₀ = exp(−k_att·L)
- dp (particle size) — small particles diffuse fast (high η_D) and get caught; very large particles settle out (high η_G). Removal is weakest for particles around 100–500 nm, the classic CFT "minimum efficiency" window most engineered nanomaterials fall into.
- v (velocity) — faster flow gives particles less time to diffuse or settle onto a grain, so it lowers η₀ and pushes more mass through.
- α (attachment efficiency) — a stand-in for ionic strength / surface charge (DLVO). Low-salinity, highly-charged particles repel grains (α → 0, particles travel far); high ionic strength collapses that repulsion (α → 1, rapid capture near the inlet).
- dc (grain size) — fine sand has more surface area per volume and traps particles faster than coarse gravel.
- Each injected particle is assigned a capture depth drawn from the exponential decay predicted by k_att — the same statistics a real packed-bed column experiment produces — then animated falling through the grain bed until it attaches or exits at the bottom outlet.
Real-world relevance: this is the mechanism regulators model when assessing whether nanoparticles released from consumer products, industrial waste or remediation nanomaterials will stay trapped in topsoil or reach an aquifer.