Skip to main content

Diffusion Coefficient Calculator

Ready to calculate
D = k_B · T / ξ.
7 Particle Shapes.
Brownian RMS Output.
100% Free.
No Data Stored.

How it Works

01Pick the Shape

7 options — sphere, three disk orientations, three ellipsoid modes — sets the friction-formula context

02Enter Temperature

K, °C, or °F — auto-converted to absolute K for the Einstein equation

03Enter Radius & Friction

a in 10 length units (Å→ft); ξ in kg/s (or s/kg mobility convention)

04Get D = k_B·T/ξ

Diffusion coefficient + 5-band classification + Brownian RMS over 1 μs to 1 min

What is a Diffusion Coefficient Calculator?

The diffusion coefficient (D) tells you how fast a particle spreads out by random thermal motion — and it's one of the most important transport parameters in physics, chemistry, and biology. Our Diffusion Coefficient Calculator applies the foundational Einstein-Smoluchowski relation D = k_B · T / ξ to compute D from absolute temperature and the particle's friction coefficient ξ. Pick from 7 particle shapes (sphere, three disk orientations, three ellipsoid modes), enter temperature in K, °C, or °F (auto-converted to absolute K), and supply the friction coefficient in kg/s or its inverse s/kg (mobility convention).

Albert Einstein derived this relation in 1905 in his Brownian-motion paper — the same paper that helped prove atoms exist. The equation links a microscopic property (friction in a viscous fluid) to a macroscopic observable (how far a particle drifts on average). For a sphere of radius a in a fluid of viscosity η, ξ = 6πηa (Stokes drag), giving the Stokes-Einstein equation D = k_B·T/(6πηa). The calculator handles the general case where you supply ξ directly, with the shape menu providing the appropriate friction formula for context.

The result panel shows D in SI (m²/s), CGS (cm²/s), and mm²/s; classifies the diffusion against five reference bands (small molecule → micron-scale particle); displays the calculation breakdown step-by-step; and computes the Brownian RMS displacement (√6Dt in 3D) at four timescales — 1 μs, 1 ms, 1 s, 1 minute — so you can see how far a typical particle drifts on each timescale.

Pro Tip: Pair this with our Molarity Calculator for solute concentration work, or the PPM to Molarity Calculator for environmental water-quality calculations.

How to Use the Diffusion Coefficient Calculator?

Pick the Shape: Sphere, three disk orientations (face-first, edge-first, rotating), or three ellipsoid modes (lengthways, sideways, tumbling). The shape sets the appropriate ξ formula for context — sphere uses Stokes drag (ξ = 6πηa), disks use Lamb-style closed forms, ellipsoids use Perrin friction factors.
Enter Absolute Temperature (T): In kelvins (K), Celsius (°C), or Fahrenheit (°F). The tool auto-converts to absolute K. Note that T must be above absolute zero — Kelvin numbers are always positive.
Enter Radius (a): The particle's characteristic radius in any of 10 length units (ångström, picometers, nanometers, micrometers, millimeters, centimeters, meters, mil/thou, inches, feet). The tool normalizes to meters internally.
Enter Friction Coefficient (ξ): In kg/s (standard physics convention) or s/kg (mobility μ = 1/ξ). The tool detects which convention you use and converts. If you don't know ξ, compute it from the shape formula using the surrounding fluid's viscosity (water ≈ 1 mPa·s at 20°C).
Press Calculate: The Einstein-Smoluchowski relation gives D in m²/s. The result panel shows D in three units, classifies against 5 reference bands, computes Brownian RMS displacement at multiple timescales, and provides the calculation worksheet.

How do I calculate the diffusion coefficient?

The diffusion coefficient comes directly from the Einstein-Smoluchowski relation — one of the most beautiful equations in statistical mechanics. Here's the complete derivation:

Think of it like a tug-of-war: the thermal energy k_B·T pushes the particle around randomly, while the friction ξ resists motion. The diffusion coefficient is the equilibrium result of this balance — k_B·T worth of "push" divided by ξ worth of "resistance" gives the rate at which the particle spreads.

The Einstein-Smoluchowski Relation

D = k_B · T / ξ

where k_B = 1.380649 × 10⁻²³ J/K is the Boltzmann constant, T is absolute temperature in kelvins, and ξ is the friction coefficient in kg/s. D has units of m²/s — area per unit time, which is exactly what you'd expect for a quantity that describes how a 2D area of probability density spreads over time.

Stokes-Einstein for Spheres

D = k_B · T / (6πηa)

For a sphere of radius a in a Newtonian fluid of viscosity η, Stokes' law gives the friction coefficient as ξ = 6πηa. Substituting into the Einstein-Smoluchowski relation gives the Stokes-Einstein equation, the most-used formula in soft-matter physics. For water at 25°C (η = 0.89 mPa·s) and a 1 nm radius particle, D ≈ 2.5 × 10⁻¹⁰ m²/s — typical for a small protein.

Friction Coefficients by Shape

  • Sphere: ξ = 6πηa (Stokes drag, ~1851)
  • Disk moving face-first: ξ = 16ηa
  • Disk moving edge-first: ξ = (32/3)ηa
  • Disk rotating about axis: ξ_rot = (32/3)ηa³ (rotational, units kg·m²/s)
  • Ellipsoid (lengthways translation): ξ = 6πηa·F_∥(p) — Perrin factor F_∥
  • Ellipsoid (sideways translation): ξ = 6πηa·F_⊥(p) — F_⊥ > F_∥, larger drag
  • Ellipsoid (tumbling rotation): ξ_rot = 6πηa·F_T(p) — Perrin's rotational factor

Brownian RMS Displacement

⟨r²⟩ = 6 D t (3D) → RMS = √(6Dt)

For a free Brownian particle in 3D, the mean-square displacement grows linearly with time at rate 6D. RMS displacement is the square root. The calculator shows this at 1 μs, 1 ms, 1 s, and 1 min so you can see the timescale-vs-distance trade-off characteristic of diffusion: doubling time only multiplies distance by √2.

Real-World Example

Diffusion Coefficient Calculator – Brownian Motion In Practice

Consider a 1 nm-radius spherical protein in water at 25°C. Standard Stokes-Einstein input:
  • Step 1: Identify the shape and inputs. Sphere, T = 298.15 K, a = 1 nm = 10⁻⁹ m. Water viscosity η ≈ 0.89 × 10⁻³ Pa·s at 25°C.
  • Step 2: Compute the Stokes friction. ξ = 6πηa = 6π × 0.89×10⁻³ × 10⁻⁹ ≈ 1.68 × 10⁻¹¹ kg/s.
  • Step 3: Compute k_B·T. k_B·T = 1.38×10⁻²³ × 298.15 ≈ 4.11 × 10⁻²¹ J (often called "kT thermal energy" — the natural unit for thermal phenomena).
  • Step 4: Apply D = k_B·T / ξ. D ≈ 4.11×10⁻²¹ / 1.68×10⁻¹¹ ≈ 2.4 × 10⁻¹⁰ m²/s.
  • Step 5: Read the band. D = 2.4×10⁻¹⁰ falls in the "Small Molecule / Protein-Class" region — consistent with a small folded protein like ubiquitin (8.5 kDa, hydrodynamic radius ~1.5 nm, measured D ≈ 1.5×10⁻¹⁰ m²/s).
  • Step 6: Compute the Brownian RMS. RMS = √(6 × 2.4×10⁻¹⁰ × 1) ≈ 38 μm in 1 second. After 1 ms it's ~1.2 μm — about a cell radius. Diffusion is fast on cellular distances but slow on macroscopic distances.

Now consider a 100 nm latex bead in water: ξ = 6π × 0.89×10⁻³ × 100×10⁻⁹ ≈ 1.68 × 10⁻⁹ kg/s, giving D ≈ 2.4 × 10⁻¹² m²/s — a hundred times smaller because friction scales with radius. RMS in 1 second is just 3.8 μm. This is why pollen grains in water look like they "jiggle" rather than drift far — visible Brownian motion but slow translation.

Who Should Use the Diffusion Coefficient Calculator?

1
Biophysicists & Cell Biologists: Predict protein diffusion in cells, validate FRAP and FCS measurements, design diffusion-limited reaction experiments.
2
Chemical Engineers: Estimate mass-transfer coefficients in reactors, predict mixing timescales, work with diffusion-limited regimes in catalysis and separation processes.
3
Soft Matter Physicists: Validate dynamic light scattering (DLS) measurements, model colloidal suspensions, study glassy dynamics where Stokes-Einstein breaks down.
4
Pharmaceutical Scientists: Drug delivery design — diffusion timescales determine when oral drugs reach the bloodstream and when topicals penetrate skin.
5
Environmental Scientists: Contaminant transport in groundwater and atmospheres, dispersion of pollutants, sediment-water interactions.
6
Physical Chemistry Students: Coursework in statistical mechanics and kinetic theory — Brownian motion is the textbook intro to fluctuation-dissipation and stochastic processes.

Technical Reference

Einstein-Smoluchowski (1905-1906): D = k_B·T/ξ. Derived independently by Albert Einstein (1905) and Marian Smoluchowski (1906). Linked the macroscopic diffusion coefficient (a phenomenological constant in Fick's laws) to the microscopic friction coefficient (a property of the particle and surrounding fluid).

Stokes-Einstein (sphere): Substituting Stokes' result ξ = 6πηa for a rigid sphere gives D = k_B·T/(6πηa). For water at 25°C: D × a ≈ 2.4 × 10⁻¹⁹ m³/s. Quick rule of thumb: D in m²/s ≈ 2.4 × 10⁻¹⁰ for a 1 nm particle, scaling as 1/a.

Perrin Friction Factors (1934): For ellipsoids of axis ratio p = c/a (c = semi-major, a = semi-minor):

  • Translational (parallel): F_∥ = (8/3)·(p²−1)^(3/2) / [(2p²−1)·ln(p+√(p²−1)) − p·√(p²−1)] for prolate
  • Translational (perpendicular): F_⊥ = (8/3)·(p²−1)^(3/2) / [(2p²−3)·ln(p+√(p²−1)) + p·√(p²−1)]
  • Average translation: F_avg = (1/3)·F_∥ + (2/3)·F_⊥
  • For p = 1 (sphere) all factors reduce to 1, recovering Stokes
  • For p >> 1 (rod-like), F_∥ ≈ p / [ln(2p) − 0.5] and F_⊥ ≈ 2p / [ln(2p) + 0.5]

Reference Diffusion Coefficients in Water at 25°C:

  • H⁺ ion: 9.31 × 10⁻⁹ m²/s (Grotthuss mechanism — anomalously fast)
  • OH⁻ ion: 5.27 × 10⁻⁹ m²/s
  • Na⁺: 1.33 × 10⁻⁹ m²/s
  • O₂ (dissolved): 2.0 × 10⁻⁹ m²/s
  • Glucose: 6.7 × 10⁻¹⁰ m²/s
  • Sucrose: 5.2 × 10⁻¹⁰ m²/s
  • Lysozyme (14 kDa): 1.1 × 10⁻¹⁰ m²/s
  • Hemoglobin (65 kDa): 7.0 × 10⁻¹¹ m²/s
  • BSA (66 kDa): 6.0 × 10⁻¹¹ m²/s
  • λ-DNA (48 kbp): ~5 × 10⁻¹³ m²/s

When Stokes-Einstein Fails. The relation breaks down for: (1) particles smaller than the solvent molecules, (2) supercooled liquids near the glass transition, (3) dense suspensions where hydrodynamic interactions dominate, (4) active particles (bacteria, ATP-driven motors) where Brownian-equilibrium assumptions fail, (5) anisotropic media or near interfaces. In these cases, D and ξ are still defined but the simple D = k_B·T/ξ relation is replaced by more complex theories (mode-coupling, Stokes-Einstein-Debye for rotation, etc.).

Key Takeaways

The diffusion coefficient is the natural language for describing how fast random thermal motion spreads out particles, molecules, and ions — and the Einstein-Smoluchowski relation D = k_B·T/ξ is one of the few exact, foundational results in non-equilibrium statistical mechanics. Use the ToolsACE Diffusion Coefficient Calculator to compute D for any combination of temperature, particle size, and friction coefficient, with shape-specific reference formulas (Stokes for spheres, Lamb for disks, Perrin factors for ellipsoids). The 5-band classification places your result in context — from small-molecule diffusion (~10⁻⁹ m²/s) down to micron-scale particles (<10⁻¹³). Bookmark it for biophysics, soft-matter physics, chemical engineering, and statistical-mechanics coursework.

Frequently Asked Questions

What is the Diffusion Coefficient Calculator?
The diffusion coefficient (D) measures how fast a particle spreads by random thermal (Brownian) motion. Our calculator applies the Einstein-Smoluchowski relation D = k_B · T / ξ — Boltzmann constant times absolute temperature divided by friction coefficient — to compute D from temperature and friction. The tool covers 7 particle shapes (sphere, three disk orientations, three ellipsoid modes), 10 length units for the radius, and 3 temperature units (K, °C, °F).

Output includes D in SI (m²/s), CGS (cm²/s), and mm²/s; a 5-band classification (small molecule → micron particle); a complete calculation breakdown; and the Brownian RMS displacement (√6Dt in 3D) at four timescales — 1 μs, 1 ms, 1 s, 1 minute — so you can see how far a typical particle drifts on each timescale.

Designed for biophysics, chemical engineering, and statistical-mechanics coursework, the tool runs entirely in your browser — no data is stored or transmitted.

Pro Tip: For more chemistry tools, try our Molarity Calculator.

What's the formula for the diffusion coefficient?
D = k_B · T / ξ — the Einstein-Smoluchowski relation (1905). For a sphere with Stokes drag (ξ = 6πηa), this becomes the Stokes-Einstein equation: D = k_B · T / (6πηa), where η is the surrounding fluid's viscosity and a is the particle radius. The formula is general — pick any shape, look up its friction coefficient, and the same Einstein-Smoluchowski equation gives D.
What units does ξ have — kg/s or s/kg?
Standard physics convention: ξ has units of kg/s (force per velocity, since friction = ξ × velocity). The inverse, 1/ξ, is the mobility μ with units s/kg. Some sources display the inverse and call it 'friction coefficient' — the calculator supports both via the unit dropdown. The math is the same; just match the convention to your input value.
Why does the calculator have 7 different shapes?
Because friction depends on shape and orientation. A sphere has the same friction in every direction (isotropic), but a disk has different drag depending on whether it moves face-first (high drag, ξ = 16ηa) or edge-first (lower, ξ = 32ηa/3). Ellipsoids are even richer — Perrin (1934) derived friction factors that depend on the axis ratio p. Picking the wrong shape gives the wrong friction and a wildly off D. The calculator keeps the math consistent with the geometry.
What's a 'typical' diffusion coefficient?
Order-of-magnitude reference for water at 25°C: small ions (H⁺, Na⁺) ~10⁻⁹ m²/s, small organic molecules (glucose) ~5×10⁻¹⁰, small proteins (lysozyme, ubiquitin) ~10⁻¹⁰, large proteins (hemoglobin, BSA) ~7×10⁻¹¹, ribosomes and DNA ~10⁻¹², sub-micron colloids ~10⁻¹², micron particles ~10⁻¹³. D scales as 1/radius for spheres, so a 10× larger particle has 10× smaller D.
What's Brownian RMS displacement and why does it scale with √t?
The mean-square displacement of a free Brownian particle grows linearly with time: ⟨r²⟩ = 6Dt (in 3D, factor 4 in 2D, 2 in 1D). The square root — RMS displacement — grows as √(6Dt). The √t scaling is fundamental: a million random steps gives only √1,000,000 = 1000 step-lengths displacement, not 1,000,000. Diffusion is a slow form of transport over long distances and a fast one over short distances.
Does temperature really matter that much?
Yes, but not as much as you might think. D scales linearly with absolute T, so going from 0°C (273 K) to 100°C (373 K) only increases D by ~37%. The bigger temperature effect comes through the viscosity η, which drops sharply with T (water viscosity halves between 25°C and 50°C). So in real solutions, D rises with T faster than linearly because both k_B·T grows and ξ ∝ η drops.
Why is the friction coefficient (ξ) so important?
Because it encapsulates everything about the particle's hydrodynamic interaction with the surrounding fluid — shape, size, surface texture, and the fluid's viscosity. The Einstein-Smoluchowski relation is exact, but it only works if you have the right ξ. For a rigid sphere in a Newtonian fluid at low Reynolds number, ξ = 6πηa (Stokes). For other shapes, you need Perrin factors. For deformable or porous particles (proteins with flexible loops, polymer coils), ξ depends on the conformation in subtle ways.
When does the Einstein-Smoluchowski relation fail?
Several regimes: (1) Particles smaller than solvent molecules — the continuum hydrodynamic picture breaks down. (2) Supercooled liquids near the glass transition — Stokes-Einstein fails because particles diffuse without rearranging the local cage. (3) Dense suspensions — hydrodynamic interactions between particles slow diffusion below the dilute-limit prediction. (4) Active particles — bacteria, motile colloids, and ATP-driven motors violate equilibrium assumptions. (5) Anisotropic media or near interfaces — confinement changes ξ in complex ways.
How is the diffusion coefficient measured experimentally?
Several techniques, depending on size scale: Dynamic Light Scattering (DLS) for nanoparticles and proteins (1 nm – 1 μm) — measures D via fluctuating scattered intensity. Fluorescence Correlation Spectroscopy (FCS) for fluorescent molecules in solution. Fluorescence Recovery After Photobleaching (FRAP) for membrane proteins and intracellular diffusion. NMR diffusion methods (DOSY, PFG-NMR) for small molecules and oligomers. Single-particle tracking for micron-scale particles under microscopy.
What's the relationship between diffusion coefficient and viscosity?
Inverse — D ∝ 1/η for a sphere. Increasing the solvent viscosity tenfold reduces D by tenfold. This is why diffusion is much faster in water than in glycerol or oil, and why intracellular diffusion (cytoplasm η ≈ 4–7× water) is several times slower than dilute aqueous diffusion. The ξ = 6πηa Stokes formula is the bridge.

Author Spotlight

The ToolsACE Team - ToolsACE.io Team

The ToolsACE Team

Our chemistry tools team implements the Einstein-Smoluchowski relation D = k_B · T / ξ — the foundational equation linking microscopic friction to macroscopic diffusion. The calculator handles 7 particle shapes (sphere · 3 disk orientations · 3 ellipsoid modes), 10 length units (Å to ft), 3 temperature units (K, °C, °F), and the friction coefficient in either kg/s (standard) or s/kg (mobility convention).

Statistical MechanicsBrownian Motion (Einstein 1905)Software Engineering Team

Disclaimer

The Einstein-Smoluchowski relation assumes a particle in dilute solution at thermal equilibrium with the bath, with friction independent of velocity (low Reynolds number). For dense suspensions, supercooled liquids, anisotropic media, or active particles, deviations from D = k_B·T/ξ can be significant. Pick the appropriate Stokes/Perrin friction formula from the shape menu.