Type I and Type II Superconductors

⚡ Key Takeaways

One line: Superconductors are classified as Type I or Type II by how they respond to an applied magnetic field — Type I expels it completely until abruptly going normal at a single critical field $H_c$ , while Type II admits quantized flux vortices over a range $H_{c1} < H < H_{c2}$ , remaining superconducting up to much larger fields. Key result: The sign of the surface energy between normal and superconducting regions — set by the Ginzburg–Landau parameter $\kappa = \lambda/\xi$ — determines the type: $\kappa < 1/\sqrt{2}$ gives Type I, $\kappa > 1/\sqrt{2}$ gives Type II. Applies to: All known superconductors; pure elemental metals (Al, Pb, Hg) are mostly Type I, while alloys, transition metals (Nb), and all high- $T_c$ cuprates and conventional high-field magnet materials (Nb-Ti, Nb₃Sn) are Type II. Who/when: Meissner & Ochsenfeld (1933, expulsion); Ginzburg & Landau (1950); Abrikosov (1957, vortex lattice prediction); experimental vortex imaging (Essmann & Träuble, 1967).

💡 The Core Insight

The fate of a magnetic field at a normal/superconducting boundary is a balance between two energy costs. Excluding the field from a thickness $\sim\lambda$ saves magnetic energy; suppressing the condensate over a thickness $\sim\xi$ costs condensation energy. The surface energy per unit area is positive when $\xi > \lambda$ (the system minimizes interface, expels flux wholesale — Type I) and negative when $\lambda > \xi$ (the system maximizes interface, subdividing flux into a lattice of vortices — Type II). The crossover is exact at $\kappa = \lambda/\xi = 1/\sqrt{2}$ .

DefinitionCharacteristic length scales

The London penetration depth $\lambda$ sets the decay of magnetic field inside a superconductor:

1

\quad \lambda = \sqrt{\frac{m^*c^2}{4\pi n_s e^{*2}}}\quad(\text{Gaussian}),\qquad \lambda=\sqrt{\frac{m^*}{\mu_0 n_s e^{*2}}}\quad(\text{SI})

Derivation hint: Combine the London equation $\nabla\times \mathbf{J}_s=-(n_se^{*2}/m^*c)\mathbf{B}$ with Ampère's law $\nabla\times\mathbf{B}=(4\pi/c)\mathbf{J}_s$ to obtain $\nabla^2\mathbf{B}=\mathbf{B}/\lambda^2$ . Here $n_s$ is the superfluid density, $e^*=2e$ and $m^*=2m_e$ are the Cooper-pair charge and mass. The Ginzburg–Landau coherence length $\xi$ sets the scale over which the order parameter $\psi$ recovers:

2

\quad \xi(T)=\frac{\hbar}{\sqrt{2m^*|\alpha(T)|}},\qquad \alpha(T)=\alpha_0\left(\frac{T}{T_c}-1\right)

Derivation hint: Linearize the GL equation $\alpha\psi+\beta|\psi|^2\psi-(\hbar^2/2m^*)\nabla^2\psi=0$ about the uniform solution; the decay length of a small perturbation is $\xi$ . The Ginzburg–Landau parameter is their ratio:

3

\quad \kappa \equiv \frac{\lambda}{\xi}

Derivation hint: Both $\lambda$ and $\xi$ scale as $|\alpha|^{-1/2}\propto(1-T/T_c)^{-1/2}$ near $T_c$ , so $\kappa$ is approximately temperature-independent — a material constant.

DefinitionGinzburg–Landau free energy functional

The state is described by minimizing

4

\quad F=\int d^3r\left[\alpha|\psi|^2+\frac{\beta}{2}|\psi|^4+\frac{1}{2m^*}\left|\left(-i\hbar\nabla-\frac{e^*}{c}\mathbf{A}\right)\psi\right|^2+\frac{|\mathbf{B}|^2}{8\pi}\right]

Derivation hint: Expand the free energy in powers of the order parameter $\psi$ (allowed near a second-order transition), include gauge-invariant gradient coupling to $\mathbf{A}$ , and add the magnetic field energy $B^2/8\pi$ . with $\mathbf{B}=\nabla\times\mathbf{A}$ , $\alpha<0$ below $T_c$ , $\beta>0$ . The equilibrium uniform order parameter is $|\psi_\infty|^2 = -\alpha/\beta$ and the condensation energy density is $\alpha^2/2\beta = H_c^2/8\pi$ .

TheoremClassification criterion

Define the thermodynamic critical field $H_c$ via the condensation energy, $H_c^2/8\pi = \alpha^2/2\beta$ . The normal/superconducting interface energy $\sigma_{ns}$ satisfies:

5

\quad \sigma_{ns}>0\iff \kappa<\frac{1}{\sqrt{2}}\quad(\text{Type I}),\qquad \sigma_{ns}<0\iff\kappa>\frac{1}{\sqrt{2}}\quad(\text{Type II})

Derivation hint: Evaluate the GL free energy across a planar domain wall in equilibrium at $H=H_c$ ; the deficit/excess relative to the bulk yields $\sigma_{ns}\approx (H_c^2/8\pi)(\xi-\lambda)\cdot\text{(geometric factor)}$ , changing sign at $\xi=\lambda$ — refined analysis gives the threshold $\kappa=1/\sqrt{2}$ . The critical fields of a Type II superconductor are:

6

\quad H_{c1}\approx\frac{H_c}{\sqrt{2}\,\kappa}\ln\kappa,\qquad H_{c2}=\sqrt{2}\,\kappa H_c=\frac{\Phi_0}{2\pi\xi^2}

Derivation hint: $H_{c2}$ follows from linearizing the GL equation in field — it is the highest field at which a nonzero $\psi$ can nucleate, mathematically the lowest Landau level of the operator $(-i\hbar\nabla-e^*\mathbf{A}/c)^2$ . $H_{c1}$ is the field at which the Gibbs free energy of one vortex first becomes negative. with the flux quantum

7

\quad \Phi_0=\frac{hc}{e^*}=\frac{hc}{2e}=2.07\times10^{-7}\,\text{G}\,\cdot\,\text{cm}^2=2.07\times10^{-15}\,\text{Wb}

Derivation hint: Single-valuedness of $\psi=|\psi|e^{i\theta}$ around a vortex requires $\oint\nabla\theta\cdot d\mathbf{l}=2\pi n$ ; combining with the supercurrent equation and integrating the enclosed flux gives quantization in units of $hc/2e$ .

ProofSign of the surface energy

Step 1 — Set up the planar wall. Consider the half-space $x<0$ superconducting and $x>0$ normal, in equilibrium at the applied field $H=H_c$ (the only field at which the two phases coexist in a Type I material). Let $\psi(x)$ rise from $0$ to $\psi_\infty$ and $B(x)$ fall from $H_c$ to $0$ .

Step 2 — Two competing regions. The order parameter recovers over a length $\sim\xi$ ; in that region the condensate is suppressed, costing condensation energy density $\sim H_c^2/8\pi$ over thickness $\xi$ . The field penetrates over a length $\sim\lambda$ ; in that region magnetic energy $\sim H_c^2/8\pi$ is retained rather than expelled, lowering the energy over thickness $\lambda$ .

Step 3 — Net surface energy. Schematically,

8

\quad \sigma_{ns}\sim \frac{H_c^2}{8\pi}\left(\xi-\lambda\right)

Derivation hint: The condensation-energy deficit scales with the depleted thickness $\xi$ (positive contribution); the magnetic-energy gain scales with the penetration thickness $\lambda$ (negative contribution); their difference is the wall energy.

Step 4 — Exact threshold. Solving the coupled one-dimensional GL equations and integrating gives a delicate cancellation; the rigorous result (Ginzburg–Landau 1950) is that $\sigma_{ns}$ vanishes precisely at $\kappa=1/\sqrt{2}$ . For $\kappa<1/\sqrt{2}$ the energy is positive — interfaces are penalized, so the material expels flux entirely (Type I). For $\kappa>1/\sqrt{2}$ it is negative — the system spontaneously generates interfaces, subdividing the flux into the maximum number of singly-quantized vortices permitted (Type II), each carrying exactly $\Phi_0$ . $\blacksquare$

∎

ProofUpper critical field

H_{c2}

from Landau-level nucleation

Step 1 — Linearize. Just below $H_{c2}$ the order parameter is infinitesimal, so drop the $\beta|\psi|^4$ term:

9

\quad \frac{1}{2m^*}\left(-i\hbar\nabla-\frac{e^*}{c}\mathbf{A}\right)^2\psi = -\alpha\,\psi=|\alpha|\psi

Derivation hint: This is the Schrödinger equation for a charged particle of mass $m^*$ , charge $e^*$ in a uniform field $\mathbf{B}=H\hat z$ — eigenvalues are Landau levels $\hbar\omega_c(n+\tfrac12)$ with $\omega_c=e^*H/m^*c$ .

Step 2 — Lowest Landau level. Nucleation first occurs at the smallest eigenvalue $\tfrac12\hbar\omega_c$ . Setting $\tfrac12\hbar e^*H/m^*c=|\alpha|$ and solving for $H$ :

10

\quad H_{c2}=\frac{2m^*c|\alpha|}{\hbar e^*}=\frac{\Phi_0}{2\pi\xi^2}=\sqrt{2}\,\kappa\,H_c

Derivation hint: Substitute $\xi^2=\hbar^2/2m^*|\alpha|$ and $\Phi_0=hc/e^*$ ; the last equality uses $H_c=\Phi_0/(2\sqrt2\,\pi\lambda\xi)$ .

Step 3 — Interpretation. $H_{c2}$ corresponds to vortex cores (each of area $\sim\pi\xi^2$ ) packed so densely that they just touch — beyond this the superconductivity is destroyed in the bulk. Since $H_{c2}=\sqrt2\,\kappa H_c$ , Type II materials with large $\kappa$ remain superconducting to fields vastly exceeding $H_c$ . $\blacksquare$

∎

AA

Alexei Alexeyevich Abrikosov

1928–2017

A student in Landau's school, Abrikosov solved the Ginzburg–Landau equations in 1953–57 for the regime $\kappa>1/\sqrt2$ and found — against Landau's initial skepticism — that the energetically favored state is a periodic lattice of quantized flux lines, now called the Abrikosov vortex lattice. He initially obtained a square lattice; later refinement showed the triangular (hexagonal) lattice has marginally lower energy. The publication was delayed for years partly because Landau doubted the result. Abrikosov shared the 2003 Nobel Prize in Physics with Ginzburg and Leggett. His prediction was directly confirmed by Bitter-decoration imaging in 1967 and is the foundation of all high-field superconducting magnet technology.

🧠 Intuition

Think of the magnetic field as wanting to get into the superconductor, and the condensate as wanting to keep it out. Two zones form at any boundary between them:

A healing zone of width $\xi$ where the superconducting "fluid" rebuilds itself — building this costs energy.
A field-leakage zone of width $\lambda$ where the field seeps in before dying — allowing this saves energy (the field doesn't have to be fully excluded).

If the healing zone is the fatter one ( $\xi>\lambda$ , Type I), boundaries are expensive, so nature makes as few as possible: the sample is either fully superconducting or fully normal, switching abruptly at $H_c$ . If the leakage zone is fatter ( $\lambda>\xi$ , Type II), boundaries are cheap — in fact profitable — so nature makes as many as it can. But flux can only enter in indivisible packets of $\Phi_0$ (a topological constraint from the single-valued wavefunction), so it threads through as a dense array of microscopic tubes — vortices — each a tiny normal core wrapped in a whirlpool of supercurrent. The material stays superconducting around the cores, which is why it survives to enormous fields.

Figure

Magnetization

-4\pi M

vs applied field

H

for Type I (sharp cutoff at

H_c

) and Type II (gradual flux entry between

H_{c1}

and

H_{c2}

).

ExampleClassifying niobium

Given: Niobium has penetration depth $\lambda\approx 39\,\text{nm}$ and coherence length $\xi\approx 38\,\text{nm}$ near $T=0$ . Thermodynamic critical field $H_c\approx 1990\,\text{Oe}$ .

Step 1 — Compute $\kappa$ :

\kappa=\frac{\lambda}{\xi}=\frac{39}{38}\approx 1.03

Step 2 — Compare to threshold: $1/\sqrt2\approx0.707$ . Since $\kappa=1.03>0.707$ , niobium is Type II (marginally — it is the only elemental Type II superconductor at ambient pressure).

Step 3 — Estimate $H_{c2}$ :

H_{c2}=\sqrt2\,\kappa H_c=\sqrt2(1.03)(1990\,\text{Oe})\approx 2900\,\text{Oe}

\boxed{\kappa\approx1.03\Rightarrow\text{Type II},\quad H_{c2}\approx 0.29\,\text{T}}

(Experimental $H_{c2}\approx 0.27$ – $0.41$ T depending on purity — consistent.)

ExampleVortex density and spacing in Nb₃Sn at 10 T

Given: A Nb₃Sn magnet operates at applied field $B=10\,\text{T}$ , well within its mixed state ( $H_{c2}\approx 23\,\text{T}$ ). Find the areal density of vortices and the lattice spacing.

Step 1 — Each vortex carries one flux quantum: $\Phi_0=2.07\times10^{-15}\,\text{Wb}$ . The number density is

n_v=\frac{B}{\Phi_0}=\frac{10}{2.07\times10^{-15}}=4.83\times10^{15}\,\text{m}^{-2}

Step 2 — Triangular lattice spacing. For a hexagonal lattice, $B=\dfrac{2\Phi_0}{\sqrt3\,a^2}$ , so

a=\left(\frac{2\Phi_0}{\sqrt3\,B}\right)^{1/2}=\left(\frac{2(2.07\times10^{-15})}{1.732\times10}\right)^{1/2}

a=\sqrt{2.39\times10^{-16}}=1.55\times10^{-8}\,\text{m}

\boxed{n_v\approx4.8\times10^{15}\,\text{m}^{-2},\quad a\approx15.5\,\text{nm}}

This spacing (~16 nm) is comparable to $\xi$ , confirming vortices are densely packed near operating fields — hence the importance of flux pinning to prevent dissipative motion.

⚠️ Type II is not "weaker" or "less perfect" superconductivity

A common error is to regard the mixed state as a partially-failed superconductor. In the mixed state the bulk still carries supercurrent with zero resistance; only the microscopic vortex cores (radius $\sim\xi$ ) are normal. Type II materials are in fact more useful than Type I precisely because they remain superconducting to enormous fields. Another error: the Meissner state ( $H<H_{c1}$ ) of a Type II material is identical to a Type I material — perfect diamagnetism. The two types only differ once flux begins to penetrate.

⚠️

H_{c2}

is not a thermodynamic critical field

$H_c$ is defined by the condensation energy and is the true thermodynamic critical field of a Type I material. For Type II, $H_{c1}$ , $H_c$ , and $H_{c2}$ are all distinct, with $H_{c1}<H_c<H_{c2}$ . $H_{c2}$ is a nucleation field (where bulk $\psi$ vanishes), not where the free energies of normal and superconducting phases cross. There is even a surface superconductivity sheath that survives up to $H_{c3}\approx1.69\,H_{c2}$ .]

CounterexampleA "perfect conductor" is not a superconductor

Suppose one had a material with merely $R=0$ but no Meissner effect. Cooling it in a field would, by Lenz's law, trap the field (the induced currents oppose any change). A real superconductor instead actively expels a field that was present before cooling — this is the Meissner effect, a true thermodynamic equilibrium property, and it is what makes both Type I and Type II behavior a phase, not a kinetic accident. The distinction shows superconductivity is not "just" zero resistance.

connection

The Type I/II classification sits at the intersection of several frameworks. The phenomenology emerges from Ginzburg–Landau Theory as a macroscopic order-parameter theory; the microscopic justification (including expressions for $\lambda$ , $\xi$ , and pair charge $e^*=2e$ ) comes from BCS Theory, with Gor'kov's derivation showing GL is its $T\to T_c$ limit. The flux quantization and vortex topology connect to Topological Defects in Field Theory and the same mathematics describes vortices in Superfluid Helium. The dual scaling $\lambda,\xi\propto(1-T/T_c)^{-1/2}$ ties the subject to the broader theory of Critical Phenomena and Scaling. The Abrikosov lattice is a realization of Spontaneous Symmetry Breaking in a U(1) gauge theory, and the field-exclusion / photon-mass analogy underlies the Anderson–Higgs Mechanism.

🔬 Where this appears

⏳ History

- 1911 — Kamerlingh Onnes discovers superconductivity in mercury at 4.2 K (Leiden).

- 1933 — Meissner & Ochsenfeld discover active flux expulsion — the Meissner effect — proving superconductivity is a thermodynamic phase, not mere perfect conductivity.

- 1935 — F. & H. London formulate the London equations, introducing the penetration depth

\lambda

.

- 1937 — L. V. Shubnikov observes the "mixed state" magnetization in alloys (later understood as the Type II vortex state).

- 1950 — Ginzburg & Landau publish the phenomenological order-parameter theory, defining

\kappa

and identifying the

1/\sqrt2

threshold in the surface energy.

- 1957 — Abrikosov publishes the vortex-lattice solution for

\kappa>1/\sqrt2

, coining the modern Type I/Type II distinction; same year, BCS publish the microscopic theory.

- 1959 — Gor'kov derives the Ginzburg–Landau equations from BCS, confirming

e^*=2e

and microscopic

\lambda,\xi

.

- 1961 — Deaver & Fairbank and Doll & Näbauer measure flux quantization, confirming

\Phi_0=hc/2e

.

- 1967 — Essmann & Träuble image the triangular Abrikosov lattice directly by Bitter decoration; 2003 — Ginzburg, Abrikosov, Leggett share the Nobel Prize.

open-problem

Several refinements complicate the clean GL picture. (1) Type-1.5 superconductivity: multiband materials (e.g., MgB₂) possess two coherence lengths and can show vortex clustering — attraction at long range, repulsion at short range — defying simple Type I/II dichotomy. (2) The intermediate state of Type I: in non-ellipsoidal geometries, demagnetizing effects produce coexisting normal/superconducting domains even below $H_c$ — a genuinely complex pattern-forming system. (3) Vortex matter physics: thermal fluctuations in high- $T_c$ cuprates melt the vortex lattice into a vortex liquid over a wide region of the phase diagram, and disorder produces vortex glasses — the true equilibrium phase boundaries remain debated. (4) Pinning and critical current: the maximum dissipationless current of a real Type II magnet is set not by $H_{c2}$ but by how strongly defects pin vortices against the Lorentz force — engineering optimal pinning landscapes is still an active materials problem. (5) Microscopic $\kappa$ near $T_c$ vs $T=0$ : $\kappa$ acquires weak temperature dependence beyond GL, and strong-coupling/clean-limit corrections shift the precise $1/\sqrt2$ boundary.

Formal Statement

Why It Holds / Derivation

Building Intuition

Worked Examples

Common Misconceptions

Connections

Real-World Applications

Historical Development

Subtleties & Limits

See Also