Stefan Number¶

Compare the sensible enthalpy available over a declared temperature interval with the latent enthalpy of a declared phase transition, yielding a dimensionless control parameter whose interpretation is valid only when its reciprocal convention, phase, temperatures, and properties are stated.

Version: v1 · 2026-08-30 · History
Domain-specific #: 2851
Origin domain: heat transfer
Subdomain: phase change modeling
Aliases: Stefan ratio

Core Idea¶

The Stefan number is a dimensionless energy-budget ratio for heat-transfer problems with phase change. In the convention most common in engineering heat-transfer and phase-change-material literature,

Ste = c_p ΔT / L,

where c_p is a phase-appropriate specific heat capacity, ΔT is the declared temperature departure from a transition reference, and L is the corresponding specific latent heat. Both numerator and denominator have units of energy per mass. The numerator estimates sensible enthalpy change across the selected temperature interval; the denominator is the enthalpy required or released per mass by the phase transition. Thus Ste compares sensible and latent thermal budgets, not two heat-transfer rates and not the amount of material already transformed.^[1]^[2]^[3]

The simple formula presumes a representative constant c_p. When heat capacity varies materially, the structurally faithful numerator is an enthalpy integral,

Ste = |∫_(T_ref)^(T_end) c_p(T) dT| / L,

with phase, path, and reference temperatures stated. For melting in a one-phase liquid model, one often writes Ste_l = c_l(T_b-T_m)/L, where T_b is a hot boundary temperature and T_m the melting temperature. For freezing or cooling newly formed solid, a solid-side form may be Ste_s = c_s(T_m-T_s)/L. A two-phase formulation can contain two distinct Stefan numbers because the solid and liquid sensible fields have different properties and temperature spans.^[2]

The name has a consequential reciprocal convention. Some Stefan-problem and cryophysical literature defines the named parameter as latent heat divided by sensible heat, S = L/(c_p ΔT), while other sources call that quantity the inverse Stefan number.^[4] The two conventions contain the same comparison but reverse all “large” and “small” statements. A bare symbol such as Ste = 0.1 is therefore incomplete evidence. A valid use declares the formula or says explicitly “sensible-to-latent convention” or “latent-to-sensible convention.” This node takes c_pΔT/L as its display convention and treats the reciprocal as a documented convention variant.

The Stefan number is autonomous because it does more than instantiate generic Ratio. It selects two specific, competing enthalpy scales; enters nondimensional moving-boundary or enthalpy equations; controls similarity solutions and approximation regimes; and supports comparison across materials and operating temperature spans. At the same time it remains domain-specific: without phase-change enthalpy, a phase-conditioned sensible interval, and a heat-transfer model, the name does not travel literally.

Structural Signature¶

A defensible Stefan-number statement contains these roles:

The phase-change system — a material and declared transition, commonly melting/freezing but potentially another thermally driven transition with a defined latent enthalpy.
The transition reference — a melting, freezing, equilibrium, solidus/liquidus, or model-defined reference temperature appropriate to the idealization.
The sensible-temperature interval — the two temperatures whose difference supplies the sensible enthalpy scale, with sign handled consistently.
The phase-specific heat capacity — c_s, c_l, an effective value, or an explicit integral ∫c_p(T)dT; its phase and averaging basis are named.
The latent-heat scale — specific latent heat L for the same material, transition, and mass or mole basis as the sensible term.
The orientation convention — sensible/latent or its reciprocal, stated rather than inferred from a symbol.
The dimensionless quotient — units cancel because both terms are enthalpy per the same amount of substance.
The model role — the quotient enters nondimensional equations, similarity relations, correlations, or regime arguments for phase-change heat transfer.

The condensed signature is: declared phase transition + phase-conditioned sensible enthalpy over a declared interval ÷ matched latent enthalpy + explicit reciprocal orientation → a dimensionless control parameter for the relative thermal budgets in a phase-change model.

The identity is invariant under changes of coherent units and under replacement of constant c_pΔT by the corresponding enthalpy integral. It is not invariant under silently changing the reference temperature, switching solid and liquid c_p, changing latent-heat basis, or reciprocating the quotient. Those changes may define legitimate Stefan numbers, but not the same numerical parameter.

What It Is Not¶

The Stefan number is not the Stefan condition. The condition is an interfacial energy balance: the jump in conductive heat flux supplies latent heat at a moving boundary. Nondimensionalizing that balance can introduce a Stefan number, but an equation governing interface velocity is not identical to the scalar ratio.

It is not a Stefan problem. A Stefan problem is a free- or moving-boundary heat equation coupled to an interface condition. Its solution also depends on geometry, initial and boundary conditions, thermal diffusivities, and sometimes convection, buoyancy, solute transport, kinetics, or density change. The Stefan number is one parameter within such a model.

It is not the Stefan–Boltzmann law or constant, which concerns thermal radiation proportional to absolute temperature to the fourth power. It is not latent heat alone, sensible heat alone, a phase fraction, a temperature difference, or a dimensional material property. It is not universally the Jakob number: heat-transfer nomenclature sometimes uses Stefan and Jakob numbers for the same c_pΔT/L ratio, especially in phase-change convection, but scopes and symbol conventions vary; an equivalence claim must be formula-qualified.^[1]

Scope of Application¶

The canonical scope is solid–liquid melting and freezing, including classical one- and two-phase Stefan problems, metal solidification, ice growth and ablation, cryopreservation, additive manufacturing, and phase-change-material thermal storage. Stefan’s historical sea-ice calculation gave the moving-boundary setting from which the eponym developed; modern accounts distinguish the dimensionless number from Stefan’s radiation law.^[3]^[5]

The abstraction also applies in numerical and experimental studies where the interface is diffuse rather than tracked sharply. In an enthalpy or apparent-heat-capacity formulation, the same quotient compares the sensible term with latent storage even if no explicit moving boundary is solved. A 2025 Journal of Fluid Mechanics framework for phase-change-material systems writes St = C_p ΔT_l/𝓛, identifies it as the ratio of sensible and latent terms, and uses the common St << 1 regime to justify neglecting sensible energy in part of a reduced model.^[6]

The simple form requires care for alloys, mixtures, broad mushy zones, temperature-dependent latent heat, hysteresis, or multiple transitions. There may be no single sharp T_m, and effective enthalpy can include composition and path dependence. A generalized ratio can still be useful, but the selected enthalpy interval and transition contribution must be defined. The node does not license copying a pure-material c_pΔT/L value into a multicomponent system without that closure.

Clarity¶

A Stefan-number claim can be audited with a short protocol. First, write the equation. Second, attach the phase label to c_p and give the temperature endpoints. Third, identify the latent heat and whether it is per kilogram, per mole, or per volume. Fourth, verify that numerator and denominator use the same extensive basis. Fifth, state where the number enters the nondimensional model or correlation.

Suppose a phase-change material has c_l = 2.0 kJ kg⁻¹ K⁻¹, a liquid superheat of 10 K, and latent heat L = 200 kJ kg⁻¹. Then the sensible-to-latent Stefan number is (2.0×10)/200 = 0.10; the reciprocal convention gives 10. Both describe a latent budget ten times the declared sensible budget. Without the defining formula, the number alone reverses interpretation. If c_p varies strongly, replacing it by a convenient room-temperature value is not clarity; the enthalpy integral or a justified effective value is needed.

Manages Complexity¶

Moving-boundary heat transfer couples conduction, heat storage, latent conversion, and an unknown interface location. Nondimensionalization groups material properties and forcing into a few parameters. The Stefan number compresses c_p, the relevant temperature span, and L into one comparison of thermal budgets. In the classical one-phase Neumann solution, the dimensionless interface coefficient is determined by a transcendental equation containing the Stefan number, so families of materials and boundary temperatures can be compared without solving every dimensional case from scratch.^[2]

The compression has limits. Stefan number does not encode the time available for diffusion (Fourier number), surface-to-internal thermal resistance (Biot number), or buoyancy and flow (Rayleigh and Prandtl numbers). Nor does it determine geometry or boundary waveform. It manages one axis of complexity: how much temperature adjustment accompanies a unit phase conversion relative to the latent budget.

Abstract Reasoning¶

Under the display convention, Ste << 1 means the specified sensible enthalpy is small relative to latent heat. Latent storage dominates that energy comparison; in suitable classical models the interface evolves slowly relative to rapid equilibration of the temperature field, motivating quasi-steady or small-Stefan-number approximations. This conclusion is conditional on the model assumptions, not a universal theorem about every convective or multicomponent system.^[2]^[6]

Ste ≈ 1 means comparable declared budgets. Ste >> 1 means the sensible scale exceeds the latent scale, so temperature-field evolution cannot ordinarily be discarded as a small correction. Under the reciprocal convention all inequalities reverse. Changing ΔT changes the operating condition even if the material is unchanged; changing L or c_p changes material response even at fixed forcing.

Two systems with equal Stefan number are similar only with respect to this one energy ratio. Their interface histories need not match unless the remaining dimensionless groups, geometry, initial state, and boundary conditions are also compatible. Conversely, if the nondimensional classical problem contains Stefan number as its sole parameter, equality of that number does organize the same dimensionless solution family. The model determines how much similarity the number licenses.

Knowledge Transfer¶

The exact quantity transfers across ice, metals, paraffins, salts, polymers, and other phase-change materials because the same roles recur: specific heat, temperature distance from transition, latent enthalpy, and a phase-front or enthalpy model. It permits comparison across laboratory experiments, numerical benchmarks, and device designs once convention and scope are aligned.

The broader reasoning pattern—compare two commensurate stores by a dimensionless ratio—belongs to prime:ratio. Literal Stefan-number transfer stops when there is no latent phase-transition enthalpy or no phase-conditioned sensible interval. A financial “latent versus sensible budget,” for example, is metaphor, not an instance. The domain-specific node remains useful precisely because it retains the heat-transfer semantics that make large/small limits predictive.

Examples¶

One-phase melting. A semi-infinite solid begins at its melting temperature and is heated at a boundary above T_m. With constant liquid properties, Ste_l = c_l(T_b-T_m)/L. In the nondimensional Neumann solution, this parameter selects the similarity constant and therefore the front position s(t) = 2λ√(α_l t) under the stated idealization.^[2]

Freezing with subcooled solid. Newly formed solid cools from T_m toward a boundary or initial solid temperature T_s. The relevant solid-side ratio is Ste_s = c_s(T_m-T_s)/L. Substituting liquid heat capacity would describe a different budget and can materially change the numerical value.

Phase-change thermal storage. For a PCM with high L and a narrow operating interval, Ste may be much less than one, signaling that latent storage dominates the declared sensible contribution. A system comparison must still include conductivity, geometry, heat-transfer coefficients, and convection; equal Ste is not equal charge time.^[6]^[7]

Sea-ice convention warning. A cryophysical treatment may define S = L/[c_i(T_m-T_B)] and call it the Stefan number.^[4] That value is the reciprocal of this node’s display convention. Translating formulas requires reciprocation before comparing “small” or “large” regimes.

Negative case. The ratio q_conv/q_rad between convective and radiative heat fluxes is dimensionless and thermally meaningful, but it contains no sensible-versus-latent phase-change comparison and is not a Stefan number.

Structural Tensions¶

Material property versus operating parameter. L and c_p are material properties, while ΔT depends on the chosen operation. Treating Stefan number as an immutable material constant hides this dependence.
Compact ratio versus scope specificity. A single number enables similarity and regime reasoning, but only after phase, reference temperature, and enthalpy basis are fixed. More compression creates more ambiguity.
Sharp transition versus distributed enthalpy. The textbook formula is clean for a sharp T_m; mixtures and hysteretic materials spread transition enthalpy across temperature. Generalization preserves the energy-ratio core but weakens a single reference-temperature interpretation.
Constant-property simplicity versus integrated accuracy. c_pΔT is transparent and convenient; ∫c_p(T)dT is more faithful when properties vary. The choice should follow the required accuracy, not habit.
One controlling number versus coupled similarity. Stefan number may be the only parameter in a classical reduction, yet real devices also depend on Fourier, Biot, Rayleigh, Prandtl, aspect-ratio, and kinetic groups. Its explanatory power is model-relative.
Community convention versus numerical comparability. Both reciprocal definitions are established. Preserving local notation aids source fidelity, while cross-study comparison requires explicit conversion.

Structural–Framed Character¶

Stefan number is strongly structural. Membership is determined by an energy-ratio equation, unit cancellation, and a role in phase-change nondimensionalization. Its value does not depend on an institution or evaluator. ISO standardizes names and symbols for transport characteristic numbers, but standardization records rather than creates the underlying ratio.^[1]

The structural–framed score is approximately 0.02: 0.05 for vocabulary travel because St, Ste, Stefan, and inverse-Stefan conventions vary, and zero for evaluative weight, institutional origin, human-practice dependence, and import-versus-recognize ambiguity. Convention variance is a documentation obligation, not evidence that the object is socially constituted.

Structural Core vs. Domain Accent¶

The structural core is dimensionless commensuration: divide one energy scale by another to expose relative dominance and similarity regimes. This is a literal instance of Ratio and supports parameter reduction.

The domain accent is constitutive: sensible enthalpy from phase-specific heat capacity over a temperature interval; latent enthalpy for a named transition; moving-interface or enthalpy heat-transfer equations; and asymptotic meanings tied to melting or freezing. Removing those terms leaves a generic quotient that the catalog already covers. Keeping them produces a stable specialist abstraction used across physical materials and model families. Stefan Number is therefore domain-specific rather than prime.

Stefan Number specializes Ratio, the sole proposed DAG parent. It is a division of commensurate nonzero-reference quantities whose numerator, denominator, units, and scope must be named. The child adds the compulsory sensible/latent enthalpy roles, phase-change reference, convention lock, and model consequences.

It relates to Thermodynamic Equilibrium because T_m or another transition reference is often an equilibrium temperature, but Stefan number can be used in nonequilibrium interface motion and is not a subtype of equilibrium. It relates to Second Law of Thermodynamics through phase-transition direction and entropy, but the ratio is fundamentally an energy-scale nondimensionalization, not an entropy-production law. It relates to Phase Diagram for mixtures whose transition temperatures and phase ranges must be selected, yet a phase diagram does not supply the ratio.

Relationships to Other Abstractions¶

Current abstraction Stefan Number Domain-specific

Parents (1) — more general patterns this builds on

Stefan Number is a kind of Ratio Prime

Stefan Number specializes Ratio, the sole proposed DAG parent.

Hierarchy path (1) — routes to 1 parentless root

Stefan Number → Ratio → Comparison → Self Checking

Neighborhood in Abstraction Space¶

Stefan Number sits in a sparse region of the domain-specific corpus (95^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (1565 abstractions)

Nearest neighbors

Isothermal Process — 0.78
Van der Waals Equation — 0.78
Adiabatic Process — 0.77
Tempering — 0.77
Continuous Cooling Transformation — 0.77

Computed from structural-signature embeddings · 2026-09-08

Not to Be Confused With¶

Stefan condition: moving-interface energy balance, not the dimensionless scalar.
Stefan problem: a free-boundary PDE problem that may contain Stefan number among several parameters.
Stefan’s equation/formula for ice thickness: an approximate growth relation, not the general dimensionless ratio.
Stefan–Boltzmann law or constant: thermal-radiation relation and physical constant.
Jakob number: sometimes formula-equivalent to c_pΔT/L, particularly for liquid–vapor change, but naming and scope must be verified in the source.
Fourier number: dimensionless diffusion time αt/ℓ², not an energy-budget ratio.
Biot number: internal versus surface thermal resistance, not sensible versus latent heat.
Phase fraction: extent transformed, whereas Stefan number is a parameter that can influence its evolution.
Inverse Stefan number: the reciprocal convention; numerically and asymptotically opposite unless explicitly converted.

References¶

[1] International Organization for Standardization, ISO 80000-11:2019, Quantities and units — Part 11: Characteristic numbers (2019; confirmed 2025), the standard for named characteristic numbers in transport and transfer phenomena. https://www.iso.org/standard/64982.html registry ↩a ↩b ↩c

[2] Vasilios Alexiades and Alan D. Solomon, Mathematical Modeling of Melting and Freezing Processes (Hemisphere/Taylor & Francis, 1993), §§2.1–2.2. ISBN 1-56032-125-3. Author-hosted contents and chapter material: https://web.math.utk.edu/~vasili/va/bk/toc.html registry ↩a ↩b ↩c ↩d ↩e

[3] John C. Crepeau, “Josef Stefan: His Life and Legacy in the Thermal Sciences,” Experimental Thermal and Fluid Science 31(7), 795–803 (2007). https://doi.org/10.1016/j.expthermflusci.2006.08.005 registry ↩a ↩b

[4] J. S. Wettlaufer, “The Stefan Problem: Polar Exploration and the Mathematics of Moving Boundaries,” in Die Zentralanstalt für Meteorologie und Geodynamik, 1851–2001: 150 Jahre Meteorologie und Geophysik in Österreich (Leykam, 2001), 420–435. Author manuscript: https://users.math.yale.edu/users/wettlaufer/articles/StefanFinal.pdf registry ↩a ↩b

[5] G. S. H. Lock, “On the Use of Asymptotic Solutions to Plane Ice—Water Problems,” Journal of Glaciology 8(53), 285–300 (1969). https://doi.org/10.3189/S0022143000031269 registry ↩

[6] Min Li and Lailai Zhu, “Theoretical Framework for Designing Phase Change Material Systems,” Journal of Fluid Mechanics 1015, A7 (2025). https://doi.org/10.1017/jfm.2025.10262 registry ↩a ↩b ↩c

[7] Jeanne Woods and coauthors, “Phase Change Material-Based Thermal Energy Storage,” Cell Reports Physical Science 2(8), 100540 (2021). https://doi.org/10.1016/j.xcrp.2021.100540 registry ↩