Dukhin Number¶
A dimensionless electrokinetic ratio comparing conduction along a charged interface with conduction through its surrounding electrolyte.
Core Idea¶
The Dukhin number is a dimensionless comparison of electrical conduction associated with an interface and conduction through the adjacent bulk electrolyte. For a charged particle with a characteristic radius a, one common definition is Du = K_s/(K_b a), where K_s is specific surface conductance (S) and K_b is bulk electrolyte conductivity (S/m). K_b a also has units S, so their quotient expresses the relative importance of interfacial and bulk paths without dependence on the chosen unit system. Squires and Bazant use this form for their particle geometry in an electrokinetic analysis.[1]
The ratio is the abstraction; a model's response to it is conditional. A non-negligible value signals that neglecting surface conduction can distort some mobility or streaming-potential calculations. A small value suppresses that correction under the specified geometry and physical assumptions, but does not certify every classical formula. In Bolève and colleagues' granular-medium convention, effective macroscopic surface conductivity σ_S already incorporates grain geometry and has units S/m, like pore-fluid conductivity σ_f; their Du = σ_S/σ_f requires no further length divisor. They analyze it with Reynolds number and medium texture for streaming potentials.[1][2]
Structural Signature¶
Sig role-phrases: charged interface → surface-conduction numerator → compatible bulk-conduction denominator → explicit-or-absorbed geometry convention → dimensionless comparison.
- Charged interface and surface path: a particle boundary or pore/grain surface carries mobile charge and an associated tangential excess conductance. Without this comparator, the number loses its electrokinetic identity.[1][2]
- Bulk electrolyte path: a nonzero conductivity for the adjacent fluid supplies the reference volume-conduction route. The measurement and model must refer to compatible conditions.[1][2]
- Unit-compatible geometry convention: particle radius a converts K_b (S/m) to a conductance scale (S) for specific surface conductance K_s (S). In Bolève's porous-medium formulation, geometry is already absorbed in effective macroscopic σ_S (S/m), which is divided directly by σ_f (S/m). Adding another length there would count geometry twice.[1][2]
- Normalized result: the quotient is dimensionless and asks whether surface conduction is negligible, comparable or dominant relative to bulk conduction in the stated setup. Specific response equations require additional hypotheses.[1][2]
Zeta potential, mobility, permittivity and streaming potential may be affected outputs or inferred quantities, not constitutive terms in every definition of Du.
What It Is Not¶
It is not surface charge density alone: charge can influence surface conduction, but the Dukhin comparison is a conduction ratio. It is not a measured zeta potential or electrophoretic mobility. Nor is it any arbitrary quotient of conductivities: the numerator must represent interfacial excess conduction in a convention compatible with the bulk comparator. Specific surface conductance K_s (S) needs a length with K_b (S/m); Bolève's macroscopic σ_S and σ_f are both S/m already.[1][2]
It is not a universal switch saying that all classical electrokinetics works below Du = 1 and fails above it. Squires and Bazant discuss surface conduction alongside double-layer thickness, applied field and other approximations. In Bolève and colleagues' granular-media model, the familiar streaming-potential limit requires both a small Dukhin number and a small Reynolds number. A different violated assumption can matter even when Du is small.[1][2]
Scope of Application¶
For colloids, a charged particle and surrounding electrolyte give a concrete surface and bulk conduction comparison. Squires and Bazant define Du = K_s/(K_b a) for their geometry and show how finite surface conduction can alter electrophoretic mobility through concentration polarization and associated processes. This is a particle-specific interpretation, not a statement that every particle has the same surface conductance or that mobility is a function of Du alone.[1]
For granular porous media, pore-wall or grain-surface contributions compete with current through pore water. Bolève and colleagues define effective macroscopic surface conductivity σ_S using grain geometry and specific surface conductance, then compare it directly with pore-water conductivity σ_f as Du = σ_S/σ_f. Both are S/m. Their streaming-potential analysis also uses medium texture and a Reynolds comparison. The abstract ratio transfers, but its convention and observable consequence must be derived for that material model.[2]
Clarity¶
Du separates which conduction path matters from which quantity is inferred downstream. A mobility reading is not an independent Du measurement, and a zeta-potential calculation can be wrong for reasons other than surface conduction. Naming the surface and bulk terms, their units and whether geometry is explicit or absorbed makes the proposed correction inspectable.[1][2]
It also makes dimensions a diagnostic. Specific surface conductance K_s (S) and bulk conductivity K_b (S/m) cannot simply be divided to make a number; K_s/(K_b a) is dimensionless in the particle convention. In Bolève's porous medium, effective macroscopic surface conductivity σ_S and pore-water conductivity σ_f are both S/m, so σ_S/σ_f is dimensionless without another length. One should not compare reported Du values across studies before checking those conventions.[1][2]
Manages Complexity¶
The number compresses interface transport, bulk transport and geometry into one model parameter. Instead of tracking each term independently in every regime statement, an analyst can first ask whether the surface current can plausibly be ignored relative to bulk current. That is a powerful reduction for choosing which electrokinetic terms deserve attention.[1]
Compression has a limit. Bolève and colleagues need Reynolds number as well as Dukhin number to recover their simple streaming-potential law; texture/formation parameters remain part of the granular-medium calculation. A single dimensionless ratio can identify one neglected mechanism without carrying the entire system model.[2]
Abstract Reasoning¶
For a particle of radius a, define Du = K_s/(K_b a). Holding K_s and K_b fixed, decreasing a increases Du: interface conductance becomes relatively more important at smaller size. Holding geometry fixed, increasing bulk conductivity decreases Du, while increasing surface conductance raises it. These deductions concern the normalized comparison, not a guaranteed direction of every measured mobility or potential because the full electrokinetic response may contain additional terms.[1]
For a porous material, the same reasoning begins only after defining an effective macroscopic surface conductivity and the geometry absorbed in it. In Bolève's convention, compare σ_S with σ_f directly; do not divide by an additional pore length. A researcher can then ask whether a bulk-only streaming-potential approximation is justified by Du, and separately test Reynolds and other assumptions. If the model fails despite small Du, the failure should not automatically be blamed on surface conduction.[2]
Knowledge Transfer¶
The colloid and granular-medium cases literally transfer surface path / bulk path / unit-compatible normalization / dimensionless ratio. They do not transfer an identical radius, pore geometry, electrophoresis formula or streaming-potential law. The particle convention places a length explicitly in the denominator; Bolève's granular convention absorbs geometry into macroscopic σ_S. In the first case the observable is particle mobility; in the second it is a flow-induced electrical potential. Both use the ratio to decide whether an interfacial current can be neglected in a specified model.[1][2]
At the broader level, live Dimensionless Quantity is the proposed immediate genus and prime Ratio supplies the ordered normalization logic. Neither broader node alone identifies an electrolyte double layer or the Dukhin scaling convention. A resemblance to a surface-to-volume ratio in another field is analogy until its numerator, denominator and physical transport roles are shown to be the same.[1]
Examples¶
Charged colloidal particle. Mapped back: charged interface = particle double layer; surface path = specific surface conductance K_s (S); bulk path = surrounding electrolyte conductivity K_b (S/m); geometry = particle radius a (m); result = Du = K_s/(K_b a). In Squires and Bazant's model, a non-negligible value warns that electrophoretic mobility may be altered by surface-conduction-related concentration polarization. That effect is a model-dependent consequence, not part of the quotient's definition.[1]
Granular porous sample. Mapped back: charged interface = grain or pore walls; surface path = effective macroscopic surface conductivity σ_S (S/m), derived from specific surface conductance and grain geometry; bulk path = pore-water σ_f (S/m); geometry = already incorporated in σ_S; result = Du = σ_S/σ_f, used with Reynolds number to interpret streaming potential. Bolève and colleagues recover a simple limit only when both ratios satisfy their smallness assumptions.[2]
Structural Tensions¶
Simpler bulk-only model versus retained interface current. Dropping surface conduction simplifies measurement inversion, but can misattribute a significant interface contribution to another parameter. Retaining it demands credible particle-specific K_s or macroscopic σ_S estimates and can increase parameter uncertainty. Diagnostic: Is Du demonstrably small under the same electrolyte, geometry convention and interface conditions as the formula being used?[1][2]
One portable ratio versus several regime conditions. Du isolates surface against bulk conduction, making unlike systems comparable after normalization. The same economy hides Reynolds, double-layer and textural assumptions. Treating Du as a universal validity switch risks false confidence; checking the additional groups costs work but localizes the actual failure mode. Diagnostic: Which other model conditions must hold before the intended mobility or streaming-potential inference follows?[1][2]
Explicit particle length versus absorbed porous geometry. A spherical-particle denominator K_b a is easy to calculate, while a porous body needs effective σ_S derived from its grain geometry and specific surface conductance. Importing the particle formula, or dividing Bolève's σ_S/σ_f by another length, creates a precise-looking but dimensionally wrong number; the effective-medium definition requires more material characterization. Diagnostic: Does the surface numerator have units S or S/m, and has geometry already entered its definition?[1][2]
Structural–Framed Character¶
Evaluative weight. Du is a descriptive ratio, not a value judgment; whether its magnitude is acceptable depends on the accuracy sought from a particular electrokinetic model. Human-practice dependence. Investigators choose whether geometry is explicit in a particle-length denominator or absorbed into effective macroscopic surface conductivity, but the physical competition between charge-transport paths is not created by that choice.[1][2]
Institutional origin. Electrokinetics gives the name and reporting conventions; no specific laboratory or manufacturer defines all admissible cases. Vocabulary travel. “Surface,” “bulk” and “ratio” travel broadly, while electrical double-layer conduction and K_s/(K_b a) are specialist. Import versus recognition. A new electrokinetic interface can be tested by its conduction paths and geometry; borrowing “Dukhin-like” for a non-electrical budget tradeoff is analogy rather than literal reuse.[1][2]
Its character: structurally precise within electrokinetics, with framed choices of geometry and model-validity threshold. The dimensionless comparison is stable; the inferred observable and correction regime are not universal.
Structural Core vs. Domain Accent¶
Portable skeleton. Live prime Ratio supplies ordered normalization against a nonzero reference, while live Dimensionless Quantity gives a more immediate physical genus: Du is an interpretable dimension-one derived quantity invariant under coherent unit changes. The staged strict subsumption edge is to Dimensionless Quantity, not a forced second direct edge to Ratio.[1]
Domain-bound mechanism. Its numerator is interface-associated electrical conduction; the compatible bulk comparator is K_b a for specific surface conductance K_s, or σ_f directly for effective macroscopic σ_S. Charge transport, double-layer behavior and the chosen particle or porous-medium convention make this a specialist identity. Those commitments do not transfer merely because some other field also compares surface and volume effects.[1][2]
Why not prime. A prime Ratio can compare any aligned quantities, and a Dimensionless Quantity can normalize any coherent physical relation. “Dukhin number” specifically names the electrokinetic surface-to-bulk conduction comparison. Its use in colloids and porous media is breadth within that domain, not evidence that the named number travels unchanged across unrelated substrates.
Instantiates / Related Primes¶
This entry is a kind of Dimensionless Quantity.
The staged typed relation is strict subsumption under Dimensionless Quantity. Prime Ratio is a broader ancestor or related comparison operation; live Physical Quantity is inherited through Dimensionless Quantity. Neither relationship is being applied to the canonical DAG in this campaign.
Relationships to Other Abstractions¶
Current abstraction Dukhin Number Domain-specific
Parents (1) — more general patterns this builds on
-
Dukhin Number is a kind of Dimensionless Quantity Domain-specific
The Dukhin number is a dimension-one physical comparison of interface with bulk electrical conduction.The live Dimensionless Quantity genus requires an interpretable dimension-one derived relation. The Dukhin number compares unit-compatible surface and bulk conduction: a particle-specific conductance needs a bulk-conductivity-times-length denominator, whereas effective macroscopic surface conductivity already has bulk-conductivity units.
Hierarchy path (1) — routes to 1 parentless root
- Dukhin Number → Dimensionless Quantity → Physical quantity → Measurement
Neighborhood in Abstraction Space¶
Dukhin Number sits in a sparse region of the domain-specific corpus (85th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Electromagnetic Fields & Responses (11 abstractions)
Nearest neighbors
- Sintering — 0.83
- COSMO solvation model — 0.82
- Oncotic Pressure — 0.81
- Adsorption — 0.81
- Diffuson — 0.81
Computed from structural-signature embeddings · 2026-10-08
Not to Be Confused With¶
Reynolds number compares inertial and viscous effects and may enter the same porous-media model without measuring electrical surface conduction.[2] Zeta potential is an interfacial potential often inferred through an electrokinetic model, not Du itself. Surface conductance or conductivity is a numerator-type input, not the normalized ratio. Any surface/volume ratio is too broad: the conduction terms, electrolyte and compatible geometry convention must all be identified.[1][2]
References¶
[1] Todd M. Squires and Martin Z. Bazant, “Induced-charge electro-osmosis”, Journal of Fluid Mechanics 509 (2004), 217–252, especially §2.2, eqs. (2.7)–(2.10). registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x
[2] Arnaud Bolève et al., “Streaming potentials of granular media: Influence of the Dukhin and Reynolds numbers”, Journal of Geophysical Research: Solid Earth (2007), abstract and model-parameter sections. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v