Skew-Symmetric Chern–Simons System

• The system is a two-component singular elliptic PDE where each equation features exponential coupling driven by the other component.
• It arises in both relativistic and non-relativistic Chern–Simons gauge theories, incorporating vortex singularities and topological solution concepts.
• Analytical techniques include variational methods, bubbling analysis, and discrete maximum principles to overcome non-coercivity and indefiniteness.

Searching arXiv for the cited skew-symmetric Chern–Simons papers to ground the article. A skew-symmetric Chern–Simons system is a two-component singular elliptic system in which each equation is driven by the exponential nonlinearity of the other component rather than by a self-interaction term. In the relativistic planar formulation, a representative model is

$\begin{cases} \Delta u_1 + e^{u_2}(1-e^{u_1}) = 4\pi \displaystyle\sum_{j=1}^{N_1}\delta_{p_{1j}},\[1ex] \Delta u_2 + e^{u_1}(1-e^{u_2}) = 4\pi \displaystyle\sum_{j=1}^{N_2}\delta_{p_{2j}}, \end{cases} \qquad \text{in }\mathbb R^2,$

while on the lattice graph $\mathbb Z^n$ the corresponding topological system takes the form

$$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$

with topological boundary condition $u(x)\to0$, $v(x)\to0$ as $d(x)\to\infty$ (Huang et al., 2014, Liu, 18 Sep 2025). Across Euclidean, toroidal, compact-surface, and lattice settings, the subject is organized around vortex singularities, topological versus non-topological asymptotics, and the analytical consequences of a coupling structure that is non-coercive or indefinite (Huang et al., 2014, Jevnikar et al., 11 Jan 2026).

1. Defining structure and model classes

In the literature covered here, “skew-symmetric” refers to a coupling pattern in which the first equation contains $e^{u_2}(1-e^{u_1})$ and the second contains $e^{u_1}(1-e^{u_2})$. One source describes this as asymmetric coupling, and another makes the structure explicit through the purely off-diagonal matrix

$K=\begin{pmatrix}0&1\1&0\end{pmatrix},$

so that each component interacts only with the other, never with itself (Huang et al., 2014, Jevnikar et al., 11 Jan 2026).

The principal settings appearing in the current literature are not identical. On $\mathbb R^2$ and $\mathbb Z^n$0, the system arises as a singular elliptic PDE with vortex Dirac masses and a coupling parameter $\mathbb Z^n$1 entering through $\mathbb Z^n$2 (Huang et al., 2014). On a compact Riemannian surface $\mathbb Z^n$3, the problem is rewritten in mean-field form with parameters $\mathbb Z^n$4, smooth positive weights $\mathbb Z^n$5, and singular factors

$\mathbb Z^n$6

(Jevnikar et al., 11 Jan 2026). On the lattice graph $\mathbb Z^n$7, $\mathbb Z^n$8, the discrete Laplacian is

$\mathbb Z^n$9

and the sources are finite sums of weighted Dirac masses,

$$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$0

(Liu, 18 Sep 2025).

These formulations share a common vortex interpretation: $$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$1 or $$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$2 are prescribed singular points, $$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$3 denotes Dirac mass at $$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$4, and the unknowns are logarithmic field variables such as $$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$5 in the gauge-theoretic derivation (Huang et al., 2014). A plausible implication is that the term “skew-symmetric Chern–Simons system” now denotes a family of closely related two-component problems rather than a single canonical PDE.

2. Gauge-theoretic and Liouville origins

One relativistic origin is a self-dual $$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$6 Chern–Simons model with two Higgs fields $$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$7 and two Abelian gauge fields $$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$8, governed by the Lagrangian

$$\left\{ \begin{aligned} \Delta u &=\lambda e^{v}(e^{u}-1)+4\pi\sum_{j=1}^{k_1}m_j\delta_{p_j},\ \Delta v &=\lambda e^{u}(e^{v}-1)+4\pi\sum_{j=1}^{k_2}n_j\delta_{q_j}, \end{aligned} \right.$$9

with Higgs potential

$u(x)\to0$0

After BPS reduction,

$u(x)\to0$1

and setting $u(x)\to0$2 yields the skew-symmetric elliptic system with vortex singularities (Huang et al., 2014).

A non-relativistic origin is a $u(x)\to0$3 Chern–Simons model with purely mutual interaction. Its Lagrangian density is

$u(x)\to0$4

and at the critical coupling $u(x)\to0$5 the self-dual equations are

$u(x)\to0$6

with Gauss laws

$u(x)\to0$7

Setting $u(x)\to0$8, $u(x)\to0$9 leads to a singular Liouville system with mutual coupling (Jevnikar et al., 11 Jan 2026).

The compact-surface reduction yields

$v(x)\to0$0

together with the Euler–Lagrange functional

$v(x)\to0$1

(Jevnikar et al., 11 Jan 2026). This surface formulation places the subject at the intersection of Chern–Simons gauge theory, singular Liouville equations, and mean-field problems.

3. Topological and non-topological solution concepts

The basic dichotomy is between vacuum asymptotics and collapse to $v(x)\to0$2. On $v(x)\to0$3, a topological solution satisfies

$v(x)\to0$4

whereas a non-topological solution satisfies

$v(x)\to0$5

(Huang et al., 2014). In the 2014 non-topological existence theory, the sought asymptotics are

$v(x)\to0$6

so that $v(x)\to0$7 and each component decays to $v(x)\to0$8 logarithmically (Huang et al., 2014).

On the lattice graph $v(x)\to0$9, a solution $d(x)\to\infty$0 is topological if

$d(x)\to\infty$1

and the relevant topological solution is nonpositive: $d(x)\to\infty$2 (Liu, 18 Sep 2025). On $d(x)\to\infty$3, the corresponding small-$d(x)\to\infty$4 dichotomy is that topological solutions satisfy $d(x)\to\infty$5 a.e., while non-topological solutions satisfy $d(x)\to\infty$6 a.e. (Huang et al., 2014).

A common misconception is that “topological” merely refers to the presence of vortices. In these works, the term is defined by the asymptotic state at infinity or, on compact domains, by the limiting vacuum behavior as the singular perturbation parameter tends to zero (Huang et al., 2014, Liu, 18 Sep 2025). By contrast, non-topological solutions are distinguished by their decay to $d(x)\to\infty$7, not by the absence of vortices (Huang et al., 2014).

4. Existence theory across domains

For the lattice graph system on $d(x)\to\infty$8, the main existence theorem states that for any fixed vortex data $d(x)\to\infty$9, the system has a topological solution

$e^{u_2}(1-e^{u_1})$0

with $e^{u_2}(1-e^{u_1})$1, $e^{u_2}(1-e^{u_1})$2, and maximality in the sense that any other solution $e^{u_2}(1-e^{u_1})$3 satisfies $e^{u_2}(1-e^{u_1})$4, $e^{u_2}(1-e^{u_1})$5 (Liu, 18 Sep 2025). The proof uses exhaustion of $e^{u_2}(1-e^{u_1})$6 by finite connected subgraphs $e^{u_2}(1-e^{u_1})$7, a monotone iteration scheme on each finite $e^{u_2}(1-e^{u_1})$8, the discrete maximum principle, a uniform lower bound obtained via a combinatorial/isoperimetric argument, and passage to the limit as $e^{u_2}(1-e^{u_1})$9 (Liu, 18 Sep 2025).

For non-topological planar solutions, one existence theorem states that if

$e^{u_1}(1-e^{u_2})$0

then, provided $e^{u_1}(1-e^{u_2})$1 avoids the finite exceptional set of curves

$e^{u_1}(1-e^{u_2})$2

there exists a solution $e^{u_1}(1-e^{u_2})$3 of the planar system satisfying the logarithmic asymptotics above (Huang et al., 2014). A radial collapsed-vortex model,

$e^{u_1}(1-e^{u_2})$4

is treated first and then used as the base case in a homotopy argument for general vortex configurations (Huang et al., 2014).

On compact surfaces, existence and multiplicity are obtained for the singular mean-field system under the parameter regime

$e^{u_1}(1-e^{u_2})$5

where

$e^{u_1}(1-e^{u_2})$6

(Jevnikar et al., 11 Jan 2026). If $e^{u_1}(1-e^{u_2})$7 and $e^{u_1}(1-e^{u_2})$8 for all $e^{u_1}(1-e^{u_2})$9, there is at least one solution; if $K=\begin{pmatrix}0&1\1&0\end{pmatrix},$0 has genus $K=\begin{pmatrix}0&1\1&0\end{pmatrix},$1 and $K=\begin{pmatrix}0&1\1&0\end{pmatrix},$2, then for generic $K=\begin{pmatrix}0&1\1&0\end{pmatrix},$3,

$K=\begin{pmatrix}0&1\1&0\end{pmatrix},$4

(Jevnikar et al., 11 Jan 2026).

These three existence theories illustrate distinct mechanisms. The lattice problem is constructive and order-theoretic, the non-topological planar problem is based on bubbling analysis and Leray–Schauder degree, and the compact-surface problem is variational but requires a constrained reduction because the original functional is indefinite (Liu, 18 Sep 2025, Huang et al., 2014, Jevnikar et al., 11 Jan 2026).

5. Asymptotics, maximality, and uniqueness

Several sharp asymptotic regimes are known for the lattice topological problem. Every topological solution satisfies the exponential decay estimate

$K=\begin{pmatrix}0&1\1&0\end{pmatrix},$5

for any $K=\begin{pmatrix}0&1\1&0\end{pmatrix},$6, with

$K=\begin{pmatrix}0&1\1&0\end{pmatrix},$7

(Liu, 18 Sep 2025). If

$K=\begin{pmatrix}0&1\1&0\end{pmatrix},$8

then

$K=\begin{pmatrix}0&1\1&0\end{pmatrix},$9

and consequently $\mathbb R^2$0, $\mathbb R^2$1 as $\mathbb R^2$2 (Liu, 18 Sep 2025). For $\mathbb R^2$3, the behavior depends sharply on dimension. In $\mathbb R^2$4, there are four cases: $\mathbb R^2$5 for $\mathbb R^2$6,

$\mathbb R^2$7

pointwise, where $\mathbb R^2$8 solves $\mathbb R^2$9 with $\mathbb Z^n$00 as $\mathbb Z^n$01 (Liu, 18 Sep 2025).

Uniqueness also exhibits more than one regime. On $\mathbb Z^n$02, there exist constants $\mathbb Z^n$03 and $\mathbb Z^n$04 such that the topological solution is unique if either $\mathbb Z^n$05 or $\mathbb Z^n$06 (Liu, 18 Sep 2025). The argument is qualitative in the high-dimensional case and uses the key lemma

$\mathbb Z^n$07

(Liu, 18 Sep 2025).

On $\mathbb Z^n$08 and $\mathbb Z^n$09, the small-$\mathbb Z^n$10 topological problem has a different uniqueness mechanism. There exists $\mathbb Z^n$11, depending on the vortex locations, such that for every $\mathbb Z^n$12 the torus system has exactly one topological solution, and that solution is also the unique maximal solution; similarly, there exists $\mathbb Z^n$13 such that for each $\mathbb Z^n$14, there is a unique topological entire solution on $\mathbb Z^n$15 (Huang et al., 2014).

The small-$\mathbb Z^n$16 proof relies on bubbling analysis near vortex points and on non-degeneracy of the linearized equation at the unique radial entire topological profile. After rescaling around a blow-up point, the normalized difference of two candidate solutions converges to a bounded solution of

$\mathbb Z^n$17

and Theorem B from Chern–Chen–Lin states that every bounded solution is trivial (Huang et al., 2014). This produces the contradiction needed for uniqueness.

6. Analytical mechanisms and recurrent proof techniques

A recurring feature of skew-symmetric systems is loss of coercivity. In the planar non-topological problem, after subtracting singular background functions the system is rewritten variationally, but the quadratic form

$\mathbb Z^n$18

is not coercive, so standard minimization does not apply directly (Huang et al., 2014). On compact surfaces, the change of variables

$\mathbb Z^n$19

transforms the energy into

$\mathbb Z^n$20

which has a positive minus negative kinetic structure and is therefore indefinite (Jevnikar et al., 11 Jan 2026).

The compact-surface remedy is a constrained reduction. Writing

$\mathbb Z^n$21

with

$\mathbb Z^n$22

and using that $\mathbb Z^n$23 is coercive and convex in $\mathbb Z^n$24, one defines the unique minimizer $\mathbb Z^n$25 and the reduced functional

$\mathbb Z^n$26

(Jevnikar et al., 11 Jan 2026). The resulting Morse-theoretic analysis uses a deformation lemma, high-sublevel contractibility, improved Moser–Trudinger inequalities, and low-sublevel topology described by formal barycenters

$\mathbb Z^n$27

(Jevnikar et al., 11 Jan 2026).

On the lattice graph, the main tools are discrete rather than variational. The paper establishes a maximum principle of the form

$\mathbb Z^n$28

uses discrete Green identities,

$\mathbb Z^n$29

and exploits the isoperimetric inequality

$\mathbb Z^n$30

for finite $\mathbb Z^n$31 (Liu, 18 Sep 2025). The monotone iteration on finite graphs starts from $\mathbb Z^n$32 and produces decreasing sequences satisfying

$\mathbb Z^n$33

(Liu, 18 Sep 2025).

For the planar non-topological problem, bubbling analysis, generalized Brezis–Merle alternatives, Pohozaev identities, and Leray–Schauder degree theory are central (Huang et al., 2014). The finite exceptional family of curves arises precisely where the possible quantized concentration patterns of mass are compatible with the Pohozaev balance equations in a degenerate way (Huang et al., 2014). This suggests that parameter exclusion is not an artifact of proof technique alone, but is tied to the internal mass-quantization structure of the system.

Taken together, these results show that the mathematical identity of skew-symmetric Chern–Simons systems is determined as much by analytical structure as by formal PDE appearance. The systems may be relativistic or non-relativistic, continuous or discrete, and topological or non-topological, but across these variants the decisive features are mutual coupling, vortex singularities, and the need to overcome either non-coercivity or indefiniteness (Huang et al., 2014, Liu, 18 Sep 2025).