Fundamentals Updated July 7, 2026 · 15 min read

Fick’s Laws of Diffusion in Metals: First Law, Second Law and Applications

Diffusion is the thermally activated migration of atoms through a crystal lattice, and it is the rate-controlling mechanism behind carburizing, nitriding, homogenisation, sintering, and countless other metallurgical processes. This article develops Fick’s first and second laws from first principles, explains the Arrhenius temperature dependence of the diffusion coefficient, and applies the error-function solution to practical case-hardening problems.

Key Takeaways

  • Fick’s first law, J = -D(dC/dx), applies to steady-state diffusion where the concentration profile is unchanging in time.
  • Fick’s second law, ∂C/∂t = D(∂²C/∂x²), governs transient diffusion and is the basis for essentially all practical case-hardening calculations.
  • The diffusion coefficient D follows an Arrhenius relation, D = D0·exp(-Q/RT), so diffusion rate rises exponentially, not linearly, with temperature.
  • For a semi-infinite solid with constant surface concentration, the error-function solution shows penetration depth scales with the square root of Dt, not with time directly.
  • Interstitial diffusion (e.g. C or N in iron) is orders of magnitude faster than substitutional diffusion because no adjacent vacancy is required for an atomic jump.
  • Grain boundary and dislocation core diffusion paths are faster than lattice diffusion and can dominate mass transport at lower homologous temperatures.

Diffusion Coefficient and Case Depth Calculator

Compute the diffusion coefficient via the Arrhenius equation and estimate case penetration depth from Fick’s second law.

D (cm²/s)
Characteristic depth 2√(Dt) (mm)
Depth at C/C0=0.5 (mm)
Transient Concentration Profiles (Error Function Solution) Depth below surface, x Concentration, C Cs (surface, held constant) C0 (bulk, initial) t1 (short time) t2 (medium time) t3 (long time), t3 > t2 > t1 x1 x2 x3 x ∝ √(D·t)
Fig. 1 — As treatment time increases from t1 to t3, the penetration depth of a given concentration level advances in proportion to the square root of time, per the error-function solution of Fick’s second law. © metallurgyzone.com

Fick’s First Law: Steady-State Diffusion

Fick’s first law states that the diffusive flux of atoms is proportional to the local concentration gradient and acts down that gradient, from regions of high concentration toward regions of low concentration.

J = -D · (dC/dx)

where:
  J = diffusive flux (atoms or mass per unit area per unit time)
  D = diffusion coefficient (m²/s or cm²/s)
  dC/dx = concentration gradient (concentration per unit distance)
  (negative sign: flux runs opposite to the gradient, i.e. down-hill in concentration)

When Fick’s First Law Applies

Fick’s first law is strictly valid only under steady-state conditions, meaning the concentration profile C(x) does not change with time anywhere in the system. This condition is closely approximated in thin-membrane permeation problems, such as hydrogen diffusing through a steel wall of constant thickness under a constant upstream partial pressure once an initial transient has decayed, or in diffusion couples analysed far from their transient early stage. Most heat-treating processes of interest to metallurgists, however, are inherently transient and require Fick’s second law instead.

Diffusion Mechanisms and the Arrhenius Temperature Dependence

Vacancy (Substitutional) Diffusion

Substitutional atoms diffuse by exchanging position with an adjacent vacant lattice site. Because equilibrium vacancy concentration itself rises exponentially with temperature, and because the atom must additionally possess enough thermal energy to squeeze past its neighbours during the jump, substitutional diffusion coefficients carry a comparatively high activation energy and are markedly slower than interstitial diffusion at a given temperature.

Interstitial Diffusion

Small solute atoms such as carbon, nitrogen, boron, and hydrogen occupy interstitial sites, principally octahedral interstices in iron, and can jump directly to an adjacent interstitial site without requiring a vacancy. This removes one major kinetic barrier, so interstitial diffusion coefficients are typically several orders of magnitude larger than substitutional coefficients, which is precisely why carburizing (interstitial carbon diffusion) can build a useful case in hours, while homogenising a substitutionally alloyed casting can require many hours at high temperature to achieve comparable penetration.

D = D0 · exp(-Q / (R·T))

where D0 is a pre-exponential (frequency) factor characteristic of the diffusing species and lattice, Q is the activation energy for diffusion (J/mol), R is the universal gas constant (8.314 J/mol·K), and T is absolute temperature (K). Because Q sits in the exponent, diffusion rate is exquisitely sensitive to processing temperature: raising a carburizing temperature by only 50°C can more than double the diffusion coefficient. This exponential relationship also governs the composition-dependent transformation kinetics discussed in the iron-carbon phase diagram and the diffusional transformations covered under the eutectoid reaction.

Diffusion coupleD0 (cm²/s)Q (kJ/mol)Mechanism
Carbon in FCC austenite (γ-Fe)0.23148Interstitial
Carbon in BCC ferrite (α-Fe)0.007983.6Interstitial
Nitrogen in FCC austenite (γ-Fe)0.003477.9Interstitial
Self-diffusion, Fe in γ-Fe49.0284Vacancy (substitutional)
Cu in Al (interdiffusion)0.647136Vacancy (substitutional)

Values are representative literature averages for illustrative and educational use; production process design should reference certified data for the specific alloy grade and temperature range in service.

Fick’s Second Law: Non-Steady-State Diffusion

Combining Fick’s first law with a mass balance on a thin differential slab, requiring that any net flux imbalance across the slab must accumulate as a local concentration change with time, yields Fick’s second law:

∂C/∂t = D · (∂²C/∂x²)

This partial differential equation, not the simpler first law, is the correct starting point for essentially all case-hardening, homogenisation, and decarburization calculations, because the concentration profile in these processes evolves continuously with time.

Error Function Solution for a Semi-Infinite Solid

For the practically important case of a semi-infinite solid initially at uniform concentration C0, with its surface held at a constant concentration Cs from time zero onward (the boundary condition of carburizing and nitriding furnaces), Fick’s second law has a closed-form solution in terms of the Gaussian error function:

(Cx - Cs) / (C0 - Cs) = erf( x / (2√(D·t)) )

where Cx is the concentration at depth x and time t. Because the argument of the error function is x divided by 2√(Dt), any fixed target concentration Cx corresponds to a fixed value of x/2√(Dt); therefore, the depth at which that concentration is reached scales with √(Dt), i.e. with the square root of time at fixed temperature. This is the origin of the well-known heat-treater’s rule of thumb that doubling case depth requires roughly quadrupling treatment time.

Worked Example: Carburizing Depth Estimate

A steel part is gas carburized at 920°C (1193 K). Using D0 = 0.23 cm²/s and Q = 148,000 J/mol for carbon in austenite, D = 0.23·exp(-148000 / (8.314·1193)) ≈ 1.4 × 10-7 cm²/s. Over a 6-hour (21,600 s) cycle, a characteristic diffusion depth of 2√(Dt) ≈ 2√(1.4×10-7 × 21600) ≈ 0.35 cm (3.5 mm) gives an order-of-magnitude estimate of the case depth achievable; use the calculator above to reproduce and adjust this result for other times and temperatures.

Grain Boundaries, Dislocations and Short-Circuit Diffusion Paths

The Arrhenius relation above describes lattice (volume) diffusion through the bulk crystal, but atoms also diffuse faster along grain boundaries, free surfaces, and dislocation cores, collectively called short-circuit diffusion paths, because the locally disordered atomic packing lowers the effective activation energy. Grain boundary diffusion coefficients can exceed lattice diffusion coefficients by three to six orders of magnitude at a given temperature, though the effect on overall mass transport depends on the fraction of the diffusion path actually occupied by boundaries or dislocations, so grain boundary diffusion becomes progressively more important, relative to lattice diffusion, as temperature decreases and as grain size decreases. This is one reason fine-grained microstructures (see the grain boundaries guide) can show accelerated diffusion-controlled degradation phenomena at temperatures where lattice diffusion alone would be negligible.

Industrial Applications and Significance

Carburizing and nitriding are the most direct industrial applications of the error-function solution developed above: carburizing builds a hard, wear-resistant, compressively stressed case by diffusing carbon into austenite followed by quenching and tempering, while nitriding diffuses nitrogen directly into ferrite at lower temperature to form hard alloy nitrides with minimal distortion. Homogenisation heat treatments rely on the same second-law framework, but in reverse, using extended high-temperature soaks to flatten dendritic microsegregation left over from casting solidification through long-range substitutional diffusion. In welding metallurgy, diffusion-controlled processes govern hydrogen redistribution and the risk of hydrogen-induced cracking in the heat-affected zone (see also HAZ microstructure), where post-weld heat treatment or controlled cooling is specified explicitly to allow diffusible hydrogen time to escape before it can concentrate at susceptible microstructural sites. Diffusion bonding, sintering of powder metallurgy compacts, and the decarburization that must be controlled during annealing and normalising of high-carbon and spring steels are further processes governed directly by the same first- and second-law framework.

Frequently Asked Questions

What is the difference between Fick’s first law and Fick’s second law?
Fick’s first law describes steady-state diffusion, where the concentration profile does not change with time and the diffusive flux is simply proportional to the local concentration gradient. Fick’s second law describes non-steady-state (transient) diffusion, where the concentration at a given position changes with time, and is derived by combining Fick’s first law with a mass balance on a differential volume element. Most practical heat-treating processes such as carburizing are transient and require the second law, while first law analysis is reserved for cases with a constant, unchanging concentration gradient, such as thin-membrane permeation.
Why does the diffusion coefficient follow an Arrhenius temperature dependence?
Diffusion in crystalline solids proceeds by discrete atomic jumps between lattice sites, and each jump requires the diffusing atom to overcome an energy barrier, the activation energy for diffusion, associated with squeezing past neighbouring atoms or, for substitutional diffusion, forming and moving into an adjacent vacancy. The probability that an atom possesses enough thermal energy to clear this barrier follows Boltzmann statistics, which produces the exponential Arrhenius form D equals D0 times exp of minus Q over RT, where small increases in absolute temperature produce large increases in diffusion rate.
Why is interstitial diffusion generally much faster than substitutional diffusion?
Interstitial diffusion, such as carbon or nitrogen moving through the octahedral interstices of the iron lattice, requires the diffusing atom only to squeeze between neighbouring host atoms without needing a vacancy to be present nearby. Substitutional diffusion, such as one metal atom exchanging with another in a solid solution, requires an adjacent vacancy to be available before a jump can occur, and vacancies are themselves present only in low equilibrium concentrations. As a result, interstitial diffusion coefficients are typically several orders of magnitude larger than substitutional diffusion coefficients at a given temperature, and activation energies for interstitial diffusion are correspondingly lower.
What does the error function solution to Fick’s second law represent physically?
The error function solution describes the concentration profile that develops with time when a semi-infinite solid at an initial uniform concentration C0 is exposed at its surface to a constant surface concentration Cs, which is exactly the boundary condition of a gas carburizing or nitriding furnace. The solution shows that the concentration at any depth x and time t depends only on the combined variable x divided by two times the square root of Dt, meaning that a given concentration level penetrates deeper in proportion to the square root of time, not linearly with time.
Why does case depth in carburizing scale with the square root of time rather than time itself?
Fick’s second law and its error-function solution show that penetration depth for a fixed target concentration is proportional to the square root of the product of diffusion coefficient and time. Physically this arises because the concentration gradient driving further diffusion becomes progressively shallower as the diffusion zone widens, so the flux into deeper material continuously slows. Consequently, doubling the case depth in a carburizing cycle requires approximately four times the treatment time at a fixed temperature, which is why case depth specifications are often paired with tight time and temperature control rather than time alone.
How does grain boundary diffusion compare with lattice (volume) diffusion?
Grain boundary diffusion coefficients are typically several orders of magnitude higher than lattice diffusion coefficients because the disordered, more open atomic packing at a grain boundary offers a lower energy pathway for atomic jumps. At low homologous temperatures, grain boundary diffusion can dominate the overall mass transport even though grain boundaries occupy only a small volume fraction of the material, which is why fine-grained materials can show accelerated diffusion-controlled degradation, such as certain forms of intergranular corrosion or sensitisation, compared with coarse-grained material of the same composition.
What is the activation energy for diffusion and how is it determined experimentally?
The activation energy for diffusion, Q, is the energy barrier an atom must overcome to complete a single diffusive jump, and it appears in the exponent of the Arrhenius relation for the diffusion coefficient. It is determined experimentally by measuring the diffusion coefficient at several temperatures, plotting the natural logarithm of D against the reciprocal of absolute temperature (an Arrhenius plot), and taking the slope of the resulting straight line, which equals minus Q divided by the universal gas constant R.
What is the difference between interdiffusion and self-diffusion?
Self-diffusion refers to the random thermal migration of atoms within a single-component (pure) material, measurable using radioactive or stable isotope tracers even though there is no net concentration gradient or compositional change. Interdiffusion refers to the diffusion of atoms across a concentration gradient between two different chemical species, such as carbon diffusing into iron, and is the process actually responsible for net mass transport and composition change in most practical metallurgical processes.
How is Fick’s first law applied to hydrogen permeation through steel?
Hydrogen permeation through a steel membrane of constant thickness under a constant upstream hydrogen partial pressure eventually reaches steady state, where the hydrogen concentration profile across the wall thickness becomes time-invariant and Fick’s first law applies directly: the steady-state flux is proportional to the diffusion coefficient and to the concentration difference across the membrane divided by its thickness. This analysis underlies hydrogen permeation barrier design and underpins standard test methods, such as those used to evaluate susceptibility to hydrogen-induced cracking in pipeline and pressure vessel steels.
Why do carburizing and nitriding use different temperature ranges given that both are diffusion processes?
Gas carburizing is typically performed in the austenite range, around 850 to 950 degrees C, because carbon solubility in face-centred cubic austenite is far higher than in body-centred cubic ferrite, allowing a useful carbon case to be built up without exceeding the solubility limit. Gas nitriding is performed at lower temperatures, around 500 to 550 degrees C, within the ferrite range, because nitrogen forms hard alloy nitrides directly in ferrite without requiring austenitisation, which allows nitriding to be applied to finished, hardened, and tempered components with minimal additional distortion.

Recommended Reference Texts

Diffusion in Solids (Shewmon)

The classic graduate reference on diffusion mechanisms, Fick’s laws, and solutions for practical boundary conditions.

View on Amazon

Phase Transformations in Metals and Alloys (Porter & Easterling)

Covers diffusion-controlled transformations, kinetics, and the mathematics of Fick’s second law in alloy systems.

View on Amazon

ASM Handbook Vol. 4: Heat Treating

Industry-standard reference for carburizing, nitriding, and diffusion-based case-hardening process data.

View on Amazon

Callister’s Materials Science and Engineering

A widely used bridge text with a clear introduction to Fick’s laws and diffusion coefficient calculations.

View on Amazon

Disclosure: MetallurgyZone participates in the Amazon Associates programme. If you purchase through these links, we may earn a small commission at no extra cost to you. This helps support free technical content on this site.

Further Reading

garg5917@gmail.com

← Previous
Slip Systems in FCC, BCC and HCP Metals: Burgers Vector and Dislocations
Next →
Sand Casting vs Die Casting vs Investment Casting: Process and Application Comparison