A note on the existence of soliton solutions in the Chern-Simons-CP(1) model

We study a gauged Chern-Simons-CP(1) system. We show that contrary to previous claims the model in the absences of a potential term cannot support finite size soliton solution in $R^2$.

💡 Research Summary

The paper conducts a thorough analytical investigation of a (2+1)-dimensional gauged Chern‑Simons–CP(1) model without any scalar potential term, addressing the long‑standing claim that such a theory admits finite‑size static soliton solutions on the infinite plane ℝ². The authors begin by writing the Lagrangian density as a sum of a pure Chern‑Simons term for a U(1) gauge field Aμ and a kinetic term for a CP(1) field z constrained by z†z = 1, with a Lagrange multiplier enforcing the constraint. No self‑interaction potential V(z) is present.

To explore possible static, radially symmetric configurations, they adopt the standard hedgehog‑type Ansatz: the CP(1) doublet is parametrized by a single radial profile f(r) and a winding number n, while the gauge field is taken purely azimuthal, Aθ(r)=a(r)/r, with Ar=0. This choice eliminates the electric field, leaving only the magnetic field B(r)=a′(r)/r. Substituting the Ansatz into the Euler‑Lagrange equations yields two coupled ordinary differential equations: one for the gauge profile a(r) derived from the Chern‑Simons equation, and one for the scalar profile f(r) from the CP(1) kinetic term.

A crucial observation is that, for static configurations, the temporal component of the covariant derivative vanishes, implying zero charge density J⁰. The Chern‑Simons field equation then reduces to κ B = J⁰, which forces B(r)=0 everywhere. Consequently a′(r)=0 and a(r) must be a constant. Finite‑energy boundary conditions require a(0)=0 (regularity at the origin) and a(∞)=n (to match the winding of the CP(1) field). These two conditions are mutually incompatible for a continuous a(r); the only way to satisfy them is to allow a discontinuous jump, which generates a δ‑function singularity in B and makes the magnetic energy ∫B² diverge.

The scalar equation simplifies dramatically when a(r)=n, because the term proportional to (n−a)² vanishes. The remaining equation is f″ + (1/r)f′ = 0, whose general solution is f(r)=C ln r + D. Finite energy demands that the gradient term (∂f)² be integrable, which forces C=0, leaving f constant. A constant f corresponds to a trivial CP(1) configuration that carries no topological charge or magnetic flux, and therefore cannot represent a non‑trivial soliton.

Putting these results together, the authors demonstrate that the model without a potential cannot simultaneously satisfy the regularity, finite‑energy, and topological boundary conditions required for a genuine soliton. Any attempt to construct a localized, finite‑energy solution inevitably leads to either a divergent magnetic energy or a trivial field configuration.

The paper contrasts this rigorous analytical conclusion with earlier numerical studies that reported soliton‑like solutions in the same model. The authors argue that those studies implicitly assumed boundary conditions that are incompatible with the exact Chern‑Simons constraint, or they employed regularization schemes that mask the underlying divergence.

In the concluding section, the authors suggest that adding a suitable potential V(z) (for example, a mass term or a Skyrme‑type quartic term) can break the strict relation κ B = 0 and allow non‑trivial magnetic profiles, thereby restoring the possibility of finite‑size solitons. They also mention alternative routes such as compactifying the spatial manifold to a sphere S², considering higher‑rank CP(N) models, or introducing higher‑derivative gauge dynamics. The work thus clarifies the precise conditions under which Chern‑Simons–CP(1) theories can support solitons and sets a clear direction for future investigations.