Solid Diffusion and Rate Limits in Li-ion Electrodes

Introduction

Fast charging requires lithium to cross electrolyte-filled pores, pass an interphase, react at particle surfaces and redistribute inside active solids. A slow step anywhere can raise overpotential or leave material unused. Particle size is often highlighted because a shorter solid diffusion distance can reduce time, but shrinking particles also raises surface area and may intensify side reactions. A complete rate analysis joins solid transport to porous-electrode and interfacial kinetics.

Core explanation

For an idealised single-phase particle with a roughly constant diffusion coefficient D and characteristic length L, the redistribution time scales as τ L²/D . This is an order-of-magnitude relationship; the numerical prefactor depends on sphere, slab or cylinder geometry and boundary conditions. Halving L reduces the estimate by a factor of four. In a sphere, surface concentration can change before the centre responds, producing a gradient during high current. If surface lithium approaches a limit, further current may require a larger overpotential or become inaccessible.

The symbol D needs care. Tracer diffusion tracks motion of labelled ions without necessarily imposing a large composition gradient. Chemical diffusion describes relaxation of a composition gradient and includes how chemical potential changes with composition, sometimes expressed through a thermodynamic factor. Intercalation materials can have D that varies strongly with lithium fraction, crystal direction and phase state. A single value fitted from one method is not a universal material constant.

In a phase-separating host, ordinary Fickian diffusion in one homogeneous composition may be incomplete. A phase-field model starts with a free-energy functional that can include chemical free energy and a cost for composition gradients. It predicts movement or suppression of Li-rich and Li-poor domains while coupling to reaction at the surface. Coherent strain and particle shape can also affect domain evolution. The model helps explain why a flat equilibrium voltage does not imply uniform concentration inside every particle during operation.

A Newman-style porous-electrode model treats each electrode as a continuum of solid particles in electrolyte-filled pores. It solves for lithium concentration and potential in the electrolyte, lithium distribution within particles, electronic potential in solids and charge-transfer current at interfaces. Concentrated-solution transport and effective conductivity reflect porosity and tortuosity. At high rate, electrolyte depletion may dominate even if solid particles are small; in another design, solid diffusion or interface kinetics may dominate. Fitting one discharge curve with many free parameters may not uniquely identify the bottleneck.

The useful design scale depends on the target. Smaller particles shorten diffusion paths but increase total surface area per mass, potentially more SEI or cathode-electrolyte reaction. Thicker electrodes raise areal energy but lengthen pore-ion paths. A measured “fast material” in a thin, low-loading half cell may not retain that rate in a dense full cell.

Step-by-step reasoning

Map the lithium pathway from one electrode through electrolyte to the other electrode's particle interior. Estimate L²/D for the relevant particle dimension and check whether D is composition-dependent. Compare this time with interfacial reaction and electrolyte-transport times. For a plateau material, consider phase separation rather than forcing uniform Fickian concentration. Test a proposed bottleneck by changing one length scale, such as particle size or coating thickness, while holding other factors as constant as possible.

Visual explanation

Draw a porous electrode cross-section with separator, electrolyte channels, carbon network and spherical particles. Show concentration gradients through electrode thickness and from particle surface to centre. A second particle sketch has Li-rich and Li-poor domains separated by an interface, unlike a smooth single-phase gradient. Label each transport segment with a possible resistance or time scale.

Real-world analogy

Filling a stadium requires people to travel through city roads, gates, corridors and finally to seats. Enlarging the gates will not help if the city roads are the bottleneck; shortening corridors will not help if the gates are shut. Lithium movement likewise has coupled scales. The analogy does not capture electrochemical potential or phase separation, so quantitative models remain necessary.

Real-world example

A battery lab reduces cathode particle radius while keeping electrode thickness similar. High-rate capacity improves modestly, but low-temperature charging still shows a large voltage drop. This suggests that electrolyte transport, interfacial desolvation or electron-transfer kinetics may now dominate. A Newman-style model constrained by conductivity, diffusion and EIS measurements can test which change would help next.

Why?

Why is the L² scaling so influential? Diffusive redistribution over twice the distance requires roughly four times the time in the simple model. Why can diffusion appear composition-dependent? Ion mobility and the chemical-potential slope both change as sites fill or phases evolve. Why combine particle and electrode models? A particle never operates in isolation from pore electrolyte and electronic networks.

Common misconception

One fitted diffusion coefficient is not valid at all lithium contents and temperatures. A second misconception is that nanosizing always improves cell performance; added surface area may accelerate parasitic chemistry, and electrode packing density may fall. A third is to infer solid diffusion limitation from any voltage sag without excluding electrolyte depletion or resistance.

Worked example

Question: Estimate τ L²/D for a 1 µm diffusion length and D = 10⁻¹⁴ m² s⁻¹, then for a 0.25 µm length with the same D.

Reasoning: First L = 10⁻⁶ m, so τ 10⁻¹²/10⁻¹⁴ = 100 s. For L = 2.5×10⁻⁷ m, τ 6.25×10⁻¹⁴/10⁻¹⁴ = 6.25 s. The ratio is 16, consistent with the fourfold length reduction squared. These are characteristic estimates without geometry factors or composition changes.

Answer: About 100 s and 6.25 s, respectively, within the simple scaling model.

Quick check

1. If a diffusion length doubles while D stays fixed, how does the simple characteristic time change? Answer: It becomes four times longer because the time scale is proportional to L²/D.

Exam focus

Convert nanometres or micrometres to metres before calculating L²/D. State whether D is tracer, chemical or an effective fitted value. Link a cell's rate to particles, interfaces and pore electrolyte; do not assign a bottleneck from a single curve. Mention phase-field or two-phase behaviour when a homogeneous diffusion model is inappropriate.

Advanced insight

Chemical diffusion can become very large or small near phase boundaries depending on the thermodynamic factor and kinetic regime, so a simple concentration-gradient law can appear misleading. Phase-field models can couple strain, interfaces and electrochemical reaction, predicting moving fronts or mosaic transformation among particles. Newman models often use an empirical open-circuit voltage function and effective transport parameters; a physically consistent free energy becomes important when extending them to phase-separating materials.

Summary

Solid redistribution time scales approximately with particle length squared over a suitable diffusion coefficient, but that coefficient can vary with composition and phase. Phase-field models handle domain formation, while porous-electrode models couple particle behaviour to electrolyte and electronic transport. Faster cells require identifying the actual limiting scale rather than optimising one isolated material property.

Practice questions

1. If D is 4×10⁻¹⁴ m² s⁻¹ and L is 2 µm, estimate L²/D. Answer: (2×10⁻⁶)²/(4×10⁻¹⁴) = 100 s.

2. Why might a chemical diffusion coefficient differ from a tracer diffusion coefficient? Answer: Chemical diffusion includes the response of chemical potential to concentration as well as microscopic ion mobility.

3. Name two fields solved in a porous-electrode battery model. Answer: Examples include electrolyte lithium concentration, electrolyte potential, solid potential and lithium concentration within active particles.

4. Why may small cathode particles have an ageing disadvantage? Answer: More surface area per mass can increase contact with electrolyte and surface side reactions.

Sources: Newman and Tiedemann, porous-electrode theory; Bazant and colleagues, multiphase porous-electrode theory; Operando LiFePO₄ phase imaging.