Lumps, Textures, and Skyrmions

Sigma-model topology labels maps from compactified space into a target manifold, but topology does not select a finite size. In two spatial dimensions the two-derivative $O(3)$ model has degree-$Q$ lumps whose energy is scale invariant; their size is a static-solution modulus. In three dimensions a two-derivative texture lowers its energy by shrinking. A Skyrmion acquires a finite size only after a four-derivative term supplies the opposing Derrick scaling, subject to the validity of that higher-derivative model.

Required background. Finite-energy boundary data distinguishes whole-space maps from defect boundary maps, and Derrick scaling supplies the virial test. Helpful background. EFT power counting is needed before interpreting higher-derivative stabilization as a controlled approximation.

Two-dimensional O(3) lumps

Let $\mathbf n(x^1,x^2)$ be a unit vector, $\mathbf n\mathbin{\cdot}\mathbf n=1$, with static energy

$$E_2 =\frac{1}{2g^2}\int_{\mathbb R^2}\mathrm d^2x\, \partial_i\mathbf n\mathbin{\cdot}\partial_i\mathbf n .$$

Finite energy requires $\mathbf n(\mathbf x)\to\mathbf n_\infty$ independent of angle. Adding the point at infinity compactifies the plane to $S^2_{\mathrm{space}}$, so

$$\mathbf n:S^2_{\mathrm{space}}\longrightarrow S^2_{\mathrm{target}} .$$

Choose the spatial orientation $\epsilon_{12}=+1$. The degree is

$$Q =\frac{1}{4\pi} \int\mathrm d^2x\, \mathbf n\mathbin{\cdot} \left(\partial_1\mathbf n\times\partial_2\mathbf n\right) \in\mathbb Z .$$

For positive $Q$, completing a square gives

$$E_2 =\frac{1}{4g^2}\int\mathrm d^2x\, \left| \partial_i\mathbf n +\epsilon_{ij}\mathbf n\times\partial_j\mathbf n \right|^2 +\frac{4\pi}{g^2}Q.$$

For negative $Q$, reverse the sign inside the square. The two branches give $E_2\geq4\pi|Q|/g^2$.

Define the stereographic coordinate and its inverse by

$$w=\frac{n^1+in^2}{1+n^3}, \qquad \mathbf n =\frac{\left(2\operatorname{Re}w,\,2\operatorname{Im}w,\,1-|w|^2\right)}{1+|w|^2}.$$

With these orientation choices, holomorphic meromorphic $w(z)$ saturates the positive-$Q$ branch. For degree one, a representative is

$$w(z)=\frac{\rho}{z-Z}, \qquad z=x^1+ix^2,$$

where $w(\infty)=0$ and $Q=+1$. The complex parameter $Z$ is the position, $|\rho|$ is the size, and $\arg\rho$ is the residual $U(1)$ target rotation that preserves the fixed value $\mathbf n_\infty$. Its energy is $4\pi/g^2$, independent of $|\rho|$. This is the Belavin–Polyakov lump Belavin and Polyakov 1975, pp. 245–247.

Scale independence is not a stable size. The family contains arbitrarily small profiles with the same classical two-derivative energy. Perturbations, higher derivatives, a potential, curvature, a lattice, or quantum running can lift $\rho$ and may favor collapse or expansion. The topological charge remains meaningful for smooth maps with the boundary condition, but the limiting zero-size configuration is singular and lies outside that configuration space.

Textures and the dimension test

“Texture” is used for a nonsingular whole-space configuration whose charge comes from the compactified domain rather than a sphere linking a localized defect. In $D$ spatial dimensions, a two-derivative sigma-model energy scales as

$$E_2[\mathbf n_\lambda]=\lambda^{2-D}E_2[\mathbf n], \qquad \mathbf n_\lambda(\mathbf x)=\mathbf n(\lambda\mathbf x).$$

Thus $D=2$ is scale invariant, whereas in $D=3$ shrinking the configuration lowers $E_2$. A nontrivial element of $\pi_3(\mathcal M)$ does not prevent concentration into a singular limit. A static particle-like texture therefore needs an additional scale-setting term, background, or conserved constraint. This is a direct example of why homotopy classification is not an existence or stability theorem.

The three-dimensional Skyrme model

Let $U(\mathbf x)\in SU(2)$ with $U(\infty)=\mathbf 1$. Compactified space is $S^3$, and $SU(2)\simeq S^3$, so the field has an integer degree. With

$$L_i=U^\dagger\partial_iU,$$

choose $\epsilon_{123}=+1$ and the oriented charge

$$B =-\frac{1}{24\pi^2} \int\mathrm d^3x\, \epsilon_{ijk}\operatorname{tr}(L_iL_jL_k).$$

The standard massless static energy is

$$E =\frac{f_\pi^2}{16} \int\mathrm d^3x\, \big[-\operatorname{tr}(L_iL_i)\big] +\frac{1}{32e^2} \int\mathrm d^3x\, \big[-\operatorname{tr}([L_i,L_j][L_i,L_j])\big].$$

The traces are nonpositive because $L_i$ is anti-Hermitian, so both displayed contributions are positive. Under $U_\lambda(\mathbf x)=U(\lambda\mathbf x)$,

$$E(\lambda)=\lambda^{-1}E_2+\lambda E_4 .$$

Stationarity requires $E_2=E_4$, and the positive second scale derivative $2E_2$ shows stability against uniform rescaling. This is one direction in configuration space, not a proof that the full physical Hessian has no negative modes. The four-derivative term is the scale-setting mechanism.

For the hedgehog

$$U(\mathbf x) =\cos F(r) +i\,\widehat{\mathbf x}\mathbin{\cdot}\boldsymbol\sigma\,\sin F(r),$$

regular unit charge uses

$$F(0)=\pi,\qquad F(\infty)=0 .$$

With the stated spatial orientation, the left current and the minus sign in the charge definition, substitution gives

$$B =-\frac{2}{\pi} \int_0^\infty\mathrm dr\,F'(r)\sin^2F(r)=1 .$$

The profile equation is nonlinear and is normally solved numerically. Its characteristic radius is $R\sim(ef_\pi)^{-1}$ up to an order-one number. Reversing spatial orientation, changing the sign in the charge definition, or defining a current with an extra minus sign reverses the displayed charge. The original construction is Skyrme 1961, pp. 127–138; Manton and Sutcliffe 2004, ch. 9, pp. 349–415 develop the classical solutions and collective coordinates.

What the higher-derivative term does not prove

The Skyrme term demonstrates a mathematically consistent classical stabilization mechanism. If the model is interpreted as a truncation of a derivative expansion, one must also compare

$$\frac{1}{R}\sim ef_\pi$$

with the EFT cutoff and estimate omitted six- and higher-derivative operators on the soliton. When $R^{-1}$ is not parametrically below the cutoff, the classical model can remain useful phenomenologically but its truncation error is not small by power counting alone. Nuclear properties and fitted low-energy constants belong to the phenomenology treatment, not to the general stability argument here.

The shared stability taxonomy makes this distinction visible: a topological degree, a virial balance, a positive Hessian, and a controlled EFT expansion are four different claims. The soliton boundary and stability comparison records the scale modulus of a lump and the scale-setting Skyrme term without promoting either classical result to quantum protection.

Family comparison and collective data

Lumps and Skyrmions use compactified whole space rather than a sphere transverse to a defect. The shared boundary-family map shows why they should not be assigned a codimension charge by analogy alone, and why position, scale, and internal orientation are only candidate moduli until their zero-mode norms are checked.

For a charge-one lump on $\mathbb R^2$, translations have finite kinetic norm, whereas size and phase variations have logarithmically divergent norms on the infinite plane. They are exact parameters of static solutions but not automatically finite-inertia collective coordinates; finite volume or deformations change this conclusion. For a Skyrmion, translations and combined spatial/internal orientations generate collective coordinates, while the size is fixed by $E_2=E_4$ rather than remaining a modulus. Adding a pion-mass potential further changes the tail and virial identity.

Checks and limitations

Charge normalization. Evaluate the degree on an explicit representative and verify that the chosen orientation gives $+1$. Reversing spatial orientation changes the sign but not the energy.

Dimensions. In $2+1$ dimensions, $[g^2]=-1$, so $4\pi/g^2$ has the dimension of energy. In $3+1$ dimensions, $[f_\pi]=1$ and the Skyrme parameter $e$ is dimensionless; both $E_2$ and $E_4$ have mass dimension one, and $ef_\pi$ is an inverse length. Here $e$ is not an electric charge.

Scale residual. A numerical Skyrmion profile must satisfy $-E_2+E_4-3E_0=0$ if a nonnegative potential $E_0$ is included.

Regularity. A sequence of shrinking smooth maps can approach a singular limit. Charge conservation applies along smooth finite-energy paths, not after the configuration space has been enlarged to include singular maps.

Quantum scope. Classical collective-coordinate quantization may organize spin and isospin states, but operator ordering, loop corrections, radiation, and EFT counterterms are additional inputs.

Common pitfalls

Calling the lump size “stable” because the energy is constant. A flat scale direction is a modulus, not a restoring force. Perturbations that lift it determine whether the lump expands or collapses.

Attributing Skyrmion size to topology. Degree prevents smooth unwinding. The opposing scaling of $E_2$ and $E_4$ fixes the size.

Assuming a four-derivative term is automatically controlled. It stabilizes the classical equations, but predictivity requires the profile’s gradients to suppress every omitted operator.

Exercises

1. Verify that the $O(3)$ charge is invariant under the rescaling $\mathbf n_\lambda(\mathbf x)=\mathbf n(\lambda\mathbf x)$ and that $E_2$ is scale invariant in two dimensions.

Solution

Each derivative contributes $\lambda$, while $\mathrm d^2x$ contributes $\lambda^{-2}$. Both the charge density integral and $E_2$ therefore have net exponent zero. The rescaling changes the size of the map but neither its degree nor its two-derivative energy.

2. Derive the hedgehog charge from its radial formula.

Solution

Use $u=F(r)$:

$$B =-\frac{2}{\pi} \int_\pi^0\sin^2u\,\mathrm du =-\frac{2}{\pi}\left(-\frac{\pi}{2}\right)=1.$$

If the endpoint values or the spatial orientation are reversed, the sign changes.

3. Expand the positive-$Q$ lump square and recover the bound with the orientation used on this page.

Solution

Because $\mathbf n\mathbin{\cdot}\partial_i\mathbf n=0$ and $\mathbf n^2=1$,

$$\sum_i|\epsilon_{ij}\mathbf n\times\partial_j\mathbf n|^2 =\sum_i|\partial_i\mathbf n|^2.$$

The cross term is

$$2\epsilon_{ij}\partial_i\mathbf n\mathbin{\cdot} (\mathbf n\times\partial_j\mathbf n) =-4\mathbf n\mathbin{\cdot} (\partial_1\mathbf n\times\partial_2\mathbf n).$$

Hence the square equals $2\int(\partial_i\mathbf n)^2-16\pi Q$. Multiplying by $1/(4g^2)$ and moving the charge term to the right gives $E_2\geq4\pi Q/g^2$ for $Q>0$. The opposite square gives the $Q<0$ branch.

4. Show that the charge-one lump’s size parameter is a static modulus but has a logarithmically divergent kinetic norm on the infinite plane.

Solution

Take $Z=0$ and real $\rho>0$. The profile has polar angle $F(r)=2\arctan(\rho/r)$ on the target sphere, so

$$|\partial_\rho\mathbf n|^2 =(\partial_\rho F)^2 =\frac{4r^2}{(r^2+\rho^2)^2}.$$

With an infrared cutoff $R$, its norm contains

$$2\pi\int_0^R r\,\mathrm dr\, \frac{4r^2}{(r^2+\rho^2)^2} =4\pi\left[ \log\!\frac{R^2+\rho^2}{\rho^2} +\frac{\rho^2}{R^2+\rho^2}-1 \right].$$

This grows as $8\pi\log(R/\rho)$, so changing $\rho$ costs no static energy but has infinite inertia as $R\to\infty$. The phase of $\rho$ has the same logarithmic infrared problem.

Continue

Bogomolny Bounds and First-Order Equations develops the general logic of square completion, saturation, and residual positive terms used here. Moduli-Space Dynamics and Collective Quantization states when positions and orientations can be quantized.

References

• Belavin, Alexander A., and Alexander M. Polyakov. “Metastable States of Two-Dimensional Isotropic Ferromagnets.” JETP Letters 22 (1975): 245–247. CERN record.
• Manton, Nicholas, and Paul Sutcliffe. Topological Solitons. Cambridge Monographs on Mathematical Physics. Cambridge University Press, 2004, chs. 6 and 9. DOI.
• Skyrme, T. H. R. “A Non-Linear Field Theory.” Proceedings of the Royal Society of London A 260 (1961): 127–138. DOI.