Bound-State Poles, Fusion, and Coleman–Thun Alternatives

A simple pole in the physical rapidity strip can represent a stable bound state, but only after its sheet, channel, residue, quantum numbers, and competing on-shell singularities are checked. When the interpretation is valid, the pole fixes the bound-state mass and three-particle coupling and generates fusion equations for further scattering. Coleman–Thun anomalous thresholds show why pole position alone is not enough.

Required background. Exact S-matrix bootstrap, CDD freedom, and completeness supplies the scalar functional equations, physical strip, and spectrum assumptions. Helpful background. Landau equations and physical singularities supplies the general on-shell origin of anomalous thresholds.

Physical-strip and residue conventions

For stable particles $a$ and $b$ of masses $m_a$ and $m_b$, set

$$p_a^\mu=m_a(\cosh\theta_a,\sinh\theta_a), \qquad p_b^\mu=m_b(\cosh\theta_b,\sinh\theta_b),$$

and $\theta=\theta_a-\theta_b$. Then

$$s(\theta) =(p_a+p_b)^2 =m_a^2+m_b^2+2m_am_b\cosh\theta.$$

We use the physical sheet reached from real positive-energy scattering with the Feynman prescription. Its direct-channel rapidity strip is

$$0<\operatorname{Im}\theta<\pi.$$

Suppose a scalar eigenchannel of $S_{ab}(\theta)$ has a simple pole at $\theta=iu_{ab}^c$ with $0<u_{ab}^c<\pi$. Our residue convention is

$$S_{ab}(\theta) \underset{\theta\to iu_{ab}^c}{\sim} \frac{i\bigl(g_{ab}^c\bigr)^2} {\theta-iu_{ab}^c}, \qquad -i\,\underset{\theta=iu_{ab}^c}{\operatorname{Res}}S_{ab} =\bigl(g_{ab}^c\bigr)^2>0.$$

The positivity statement applies to this normalized scalar eigenchannel. With internal indices, the residue is a projector or product of on-shell couplings and may carry convention-dependent charge-conjugation and phase factors.

At the pole, $\cosh(iu)=\cos u$, so the proposed bound-state mass is

$$\boxed{ m_c^2 =m_a^2+m_b^2+2m_am_b\cos u_{ab}^c }.$$

Because $0<u<\pi$, this lies below $(m_a+m_b)^2$. The pole must also occur in a channel with the quantum numbers of $c$, and $c$ must be stable against all allowed decays. A crossed-channel image can occupy the same strip at a related location; the amplitude and channel convention decide which interpretation applies.

Deriving the fusion shifts

The pole admits a useful on-shell geometric representation. Let the bound state have rapidity $\theta_c$, and write its constituents at complex rapidities

$$\theta_a=\theta_c+i\alpha, \qquad \theta_b=\theta_c-i\beta, \qquad \alpha+\beta=u_{ab}^c.$$

Momentum conservation $p_a(\theta_a)+p_b(\theta_b)=p_c(\theta_c)$ gives, in the rest frame of $c$,

$$m_a\sin\alpha=m_b\sin\beta, \qquad m_c=m_a\cos\alpha+m_b\cos\beta.$$

Squaring these equations reproduces the pole mass formula. For equal constituent masses, $\alpha=\beta=u_{ab}^c/2$ and $m_c=2m_a\cos(u_{ab}^c/2)$.

Now scatter a stable particle $d$ from the fused state. In a diagonal theory, factorization gives

$$S_{cd}(\theta_c-\theta_d) = S_{ad}(\theta_c-\theta_d+i\alpha)\, S_{bd}(\theta_c-\theta_d-i\beta).$$

This is the scalar fusion equation. In a matrix theory, the constituent product is projected with the on-shell couplings $g_{ab}^c$ and $g_c^{ab}$, and its ordering follows the declared tensor convention. Fusion is closed only when every pole generated by the resulting amplitudes is classified and every accepted new particle is included consistently.

The pole and fusion logic is part of the factorized bootstrap of Zamolodchikov and Zamolodchikov 1979, §§ 3–4, pp. 264–282.

First sine-Gordon breather

Use the sine-Gordon coupling parameter

$$\xi=\frac{\beta_{\rm SG}^2}{8\pi-\beta_{\rm SG}^2}, \qquad 0<\xi<1$$

in the attractive regime, and let $M$ be the soliton mass. The soliton–antisoliton channel contains a pole for the $n$th breather at

$$u_n=\pi(1-n\xi), \qquad n\xi<1.$$

For equal masses $m_a=m_b=M$, the pole formula gives

$$\begin{aligned} m_n^2 &=2M^2\bigl(1+\cos u_n\bigr) \\ &=4M^2\sin^2\!\left(\frac{n\pi\xi}{2}\right), \end{aligned}$$

and hence

$$m_n=2M\sin\!\left(\frac{n\pi\xi}{2}\right).$$

For $n=1$, this is the first breather. The condition $\xi<1$ both places the pole inside the strip and makes $m_1<2M$. As $\xi\to1^-$, the pole approaches the two-particle threshold and the first breather leaves the stable spectrum. This is a direct consistency check between strip geometry and the mass formula.

The exact sine-Gordon spectrum and its fusion structure are developed in Zamolodchikov and Zamolodchikov 1979, §§ 4–5, pp. 269–287.

Coleman–Thun alternatives

An S-matrix singularity can also arise when a network of already-known particles goes on shell at a special external rapidity. In $1+1$ dimensions, the restricted kinematics can make such anomalous thresholds poles rather than ordinary higher-dimensional branch points. The Coleman–Thun mechanism explains double poles of the sine-Gordon S matrix through on-shell diagrams without adding a new asymptotic particle Coleman and Thun 1978, §§ 2–3, pp. 32–39.

A pole-classification test therefore asks:

1. Is the pole on the declared physical sheet and in the correct channel?
2. Does a simple-pole residue factorize with the sign and projector expected for a positive-norm stable particle?
3. Does the inferred mass lie below the relevant threshold and carry allowed conserved quantum numbers?
4. Do fusion equations close after the particle is added?
5. Can an on-shell diagram built entirely from the existing spectrum and couplings reproduce the pole position and order?
6. Do zeros of internal amplitudes reduce or cancel the naïve singularity?

The fifth and sixth questions are decisive for Coleman–Thun processes. Counting propagators alone is not enough; one must solve the on-shell kinematics and include numerator, Jacobian, and internal S-matrix zeros.

Bound state, resonance, and anomalous threshold

These three analytic phenomena should not be conflated.

• A stable bound state is a normalizable asymptotic particle below threshold. In the declared eigenchannel it produces a physical-sheet pole with the appropriate factorized residue.
• A resonance is unstable and is represented by a pole on an unphysical sheet at complex energy. It is not an external state in the asymptotic charge argument.
• A Coleman–Thun singularity is generated by a kinematically allowed on-shell network of existing particles. It need not enlarge the stable spectrum.

The integrability exact-data chain marks this pole-classification decision before finite-volume or observable calculations. The exact and rigorous status comparison keeps a closed on-shell bootstrap distinct from a separate local construction.

Common pitfalls

Reading every physical-strip pole as a particle. Check the residue, quantum numbers, fusion closure, and on-shell alternatives. Pole location is necessary but not sufficient.

Using the mass formula on the wrong sheet. The substitution $\theta=iu$ assumes the declared physical-strip pole. A resonance pole on another sheet does not define a stable mass through the same inference.

Forgetting crossed images. Crossing relates direct- and crossed-channel singularities. The amplitude convention must state which pole is being interpreted.

Exercises

1. For equal constituent masses $M$, derive $m_c=2M\cos(u/2)$ from the physical-strip pole and show that the sine-Gordon choice $u_n=\pi(1-n\xi)$ gives the breather mass displayed above.

Solution

The pole formula gives

$$m_c^2=2M^2(1+\cos u)=4M^2\cos^2(u/2).$$

Since $0<u<\pi$, the positive root is $m_c=2M\cos(u/2)$. For $u_n=\pi(1-n\xi)$, $\cos[u_n/2]=\sin(n\pi\xi/2)$, giving $m_n=2M\sin(n\pi\xi/2)$.

2. For unequal masses, use $m_a\sin\alpha=m_b\sin\beta$, $m_c=m_a\cos\alpha+m_b\cos\beta$, and $\alpha+\beta=u$ to recover the pole mass formula.

Solution

Square the energy equation:

$$m_c^2 =m_a^2\cos^2\alpha+m_b^2\cos^2\beta +2m_am_b\cos\alpha\cos\beta.$$

The momentum equation implies $m_a^2\sin^2\alpha=m_b^2\sin^2\beta$. Adding the corresponding sine terms allows the first two terms to become $m_a^2+m_b^2$, while $\cos\alpha\cos\beta-\sin\alpha\sin\beta=\cos(\alpha+\beta)=\cos u$. The result is $m_c^2=m_a^2+m_b^2+2m_am_b\cos u$.

References

• Coleman, Sidney, and H. J. Thun. “On the Prosaic Origin of the Double Poles in the Sine-Gordon S-Matrix.” Communications in Mathematical Physics 61 (1978): 31–39. DOI.
• Zamolodchikov, Alexander B., and Alexey B. Zamolodchikov. “Factorized S-Matrices in Two Dimensions as the Exact Solutions of Certain Relativistic Quantum Field Models.” Annals of Physics 120 (1979): 253–291. DOI.