Open core of the Atiyah–Hitchin low-energy monopole scattering problems: the general geodesic flow on the two-monopole moduli space and explicit metrics on higher-charge monopole moduli spaces

Atiyah and Hitchin (1985) determined the natural $L^2$ metric on the four-dimensional centered moduli space $M_2^0$ of two SU(2) BPS monopoles (the Atiyah–Hitchin manifold) and analyzed monopole scattering through geodesics on its two totally geodesic surfaces, obtaining right-angle scattering in head-on collisions and the conversion of orbital angular momentum into electric charge. Their concluding discussion listed five unresolved issues: the general geodesic flow on the full $M_2^0$; the existence of closed geodesics (classical bound states); whether every geodesic arriving from infinity returns to infinity (completeness of geodesic scattering); the explicit metrics on the higher-charge moduli spaces $M_k^0$ for $k>2$; and a rigorous justification of the geodesic approximation. Two of the five issues are settled and excluded from this problem — closed geodesics do exist (Bates–Montgomery 1988, extended by Bielawski 1996 to all charges), and the geodesic approximation has been rigorously proven for slow monopole motion (Stuart 1994) — while the general geodesic flow on the full manifold is known only from numerical studies (differential cross sections and escape plots suggesting chaotic structure), and for $k\ge3$ only asymptotic metrics and metrics on symmetry submanifolds are known. The surviving open core is therefore: (1) the global geodesic dynamics of the exact Atiyah–Hitchin metric — in particular the completeness of geodesic scattering, i.e. the existence or absence of trapped orbits arriving from infinity; and (2) the explicit $L^2$ metric on $M_k^0$ for $k\ge3$.

Background

In SU(2) Yang–Mills–Higgs theory on $\mathbb{R}^3$ (a gauge field coupled to a scalar Higgs field) in the BPS or Prasad–Sommerfield limit — the limit in which the Higgs self-interaction is switched off and the potential energy vanishes — static finite-energy fields satisfy the first-order Bogomolny equation $B_i = D_i\Phi$, equating the non-Abelian magnetic field with the covariant derivative of the Higgs field. Its finite-energy solutions are magnetic monopoles, sorted by an integer magnetic charge $k$; the solutions of charge $k$ form a smooth moduli space $M_k$ of dimension $4k$ (roughly the positions and internal phases of $k$ monopoles), carrying a natural Riemannian metric given by the $L^2$ inner product of tangent zero-modes (the field variations that change the solution to first order); this metric is precisely the kinetic energy of time-dependent fields. Manton (1982) proposed that the dynamics of slowly moving monopoles is approximated by geodesic motion on $M_k$ with this metric. Factoring out the center of mass — translations ($\mathbb{R}^3$) and an overall phase ($S^1$) — leaves the centered moduli space $M_k^0$ of dimension $4k-4$; for $k=2$ this is four-dimensional.

Atiyah and Hitchin (1985) computed the metric on $M_2^0$, now called the Atiyah–Hitchin manifold. It is hyperkähler (Ricci-flat and Kähler for a whole sphere of complex structures; equivalently a self-dual Einstein metric), with an $SO(3)$ isometry acting with three-dimensional orbits. In left-invariant one-forms $\sigma_1,\sigma_2,\sigma_3$ on $SO(3)$ it takes the form

$ds^2 = (abc)^2\,d\eta^2 + a^2\sigma_1^2 + b^2\sigma_2^2 + c^2\sigma_3^2,$

where $a,b,c$ are functions of $\eta$ satisfying the first-order system $\frac{2}{a}\frac{da}{d\eta} = (b-c)^2 - a^2$ and its cyclic permutations, with the monopole boundary conditions determined by a linear equation solved by the complete elliptic integral $K$. At large separation the metric approaches the negative-mass Taub–NUT metric (a complete self-dual metric with a circle fiber over $\mathbb{R}^3$, the prototype of an ALF — asymptotically locally flat — gravitational instanton), with corrections that are exponentially small in the separation. The manifold contains two distinguished totally geodesic surfaces (submanifolds whose geodesics are geodesics of the ambient space), of types I and II. Geodesics on them encode two-monopole scattering: exact right-angle ($90^\circ$) deflection in a head-on collision on the type I surface, and on the type II surface, for small angular momentum, the emergence of a pair of dyons (monopoles carrying electric charge) — the conversion of orbital angular momentum into electromagnetic angular momentum.

The paper's concluding discussion listed what was then unresolved: the geodesic flow on the full four-dimensional manifold had not been investigated beyond the two surfaces; whether closed geodesics (which would model classical bound states of two monopoles) exist, the authors expecting not; whether every geodesic arriving from infinity returns to infinity (completeness of geodesic scattering), a conjecture left unproven; the fact that for $k>2$ the spaces $M_k^0$ are known parametrically — through Donaldson's classification of monopoles via Nahm's equations (an ODE system on an interval equivalent, by the Nahm transform, to monopoles) and their spectral curves in the mini-twistor space of $\mathbb{R}^3$ (the complex surface parametrizing oriented lines in $\mathbb{R}^3$, on which each monopole is encoded as an algebraic curve) — while their explicit metrics had not been found; and the absence of a rigorous justification of the geodesic approximation for the infinite-dimensional configuration space of the Yang–Mills–Higgs equations. Later work settled the closed-geodesic and approximation issues: Bates and Montgomery proved the Atiyah–Hitchin metric does admit closed geodesics (two monopoles orbiting perpetually), and Stuart rigorously proved the geodesic approximation for slow monopole motion. What remains open — the subject of this problem — is the global geodesic dynamics on the full Atiyah–Hitchin manifold (in particular the completeness of scattering) and the explicit metric on the higher-charge centered moduli spaces.

Problem Statement

The source question asked five things about the low-energy dynamics of SU(2) BPS monopoles through their moduli-space geometry; two have since been settled — the existence of closed geodesics on $M_2^0$ (they do exist; Bates–Montgomery 1988, Bielawski 1996) and the rigorous justification of the geodesic approximation (Stuart 1994) — and are excluded here. The problem is the surviving open core. Its two components serve one objective, that of the 1985 paper: making the geodesic picture of low-energy multi-monopole dynamics globally and quantitatively complete.

(1) Geodesic flow on the Atiyah–Hitchin manifold. Determine the global qualitative behavior of the geodesic flow of the exact Atiyah–Hitchin metric on the full four-dimensional centered two-monopole moduli space $M_2^0$, beyond the two totally geodesic surfaces already analyzed in the source. The central precise target is the source's conjecture: prove or refute that every geodesic that escapes to the unique negative-mass Taub–NUT-type end of $M_2^0$ as $t\to-\infty$ escapes again to that end as $t\to+\infty$ (completeness of geodesic scattering) — equivalently, decide whether trapped geodesics arriving from infinity exist, for example geodesics asymptotic in forward time to the closed geodesic of Bates–Montgomery. A full qualitative characterization of the general geodesic flow is a distinct, broader question.

(2) Explicit higher-charge metrics. Compute the explicit $L^2$ metric on the centered charge-$k$ monopole moduli spaces $M_k^0$ for $k>2$, whose points are parametrized by Donaldson's Nahm-equation/spectral-curve data, following the model of the Atiyah–Hitchin construction for $k=2$ (a closed form via elliptic-integral/Nahm data) — going beyond the known asymptotic (Gibbons–Manton) metric for well-separated monopoles and beyond metrics on totally geodesic symmetry submanifolds. Here 'explicit' means a closed-form expression in Nahm/spectral-curve data analogous to the Atiyah–Hitchin ODE-plus-elliptic-integral presentation for $k=2$, in contrast to the asymptotic (Gibbons–Manton) and symmetry-submanifold metrics excluded above.

The two components are separate open items of the source's concluding discussion; each may be addressed independently of the other.

References

1. M. F. Atiyah and N. J. Hitchin, 'Low-energy scattering of non-Abelian magnetic monopoles', Philosophical Transactions of the Royal Society of London A 315 (1985) 459-469. DOI: 10.1098/rsta.1985.0052, https://doi.org/10.1098/rsta.1985.0052
2. M. F. Atiyah and N. J. Hitchin, 'The Geometry and Dynamics of Magnetic Monopoles' (M. B. Porter Lectures), Princeton University Press, 1988. INSPIRE literature record 269754, https://inspirehep.net/literature/269754
3. N. S. Manton, 'A remark on the scattering of BPS monopoles', Physics Letters B 110 (1982) 54-56. DOI: 10.1016/0370-2693(82)90950-9, https://doi.org/10.1016/0370-2693(82)90950-9
4. S. K. Donaldson, 'Nahm's equations and the classification of monopoles', Communications in Mathematical Physics 96 (1984) 387-407. DOI: 10.1007/BF01214583, https://doi.org/10.1007/BF01214583
5. G. W. Gibbons and N. S. Manton, 'Classical and quantum dynamics of BPS monopoles', Nuclear Physics B 274 (1986) 183-224. DOI: 10.1016/0550-3213(86)90624-3, https://doi.org/10.1016/0550-3213(86)90624-3
6. L. Bates and R. Montgomery, 'Closed geodesics on the space of stable two-monopoles', Communications in Mathematical Physics 118 (1988) 635-640. DOI: 10.1007/BF01221112, https://doi.org/10.1007/BF01221112
7. D. Stuart, 'The geodesic approximation for the Yang-Mills-Higgs equations', Communications in Mathematical Physics 166 (1994) 149-190. DOI: 10.1007/BF02099305, https://doi.org/10.1007/BF02099305
8. M. Temple-Raston and D. Alexander, 'Differential cross sections and escape plots for low-energy solitonic SU(2) BPS magnetic monopole dynamics', Nuclear Physics B 397 (1993) 195-213. arXiv:hep-th/9209063, DOI: 10.1016/0550-3213(93)90341-L, https://arxiv.org/abs/hep-th/9209063
9. R. Bielawski, 'Existence of closed geodesics on the moduli space of k-monopoles', Nonlinearity 9 (1996) 1463-1467. DOI: 10.1088/0951-7715/9/6/004, https://doi.org/10.1088/0951-7715/9/6/004
10. N. J. Hitchin, N. S. Manton and M. K. Murray, 'Symmetric monopoles', Nonlinearity 8 (1995) 661-692. arXiv:dg-ga/9503016, DOI: 10.1088/0951-7715/8/5/002, https://arxiv.org/abs/dg-ga/9503016
11. P. M. Sutcliffe, 'The moduli space metric for tetrahedrally symmetric 4-monopoles', Physics Letters B 357 (1995) 335-341. arXiv:hep-th/9507173, DOI: 10.1016/0370-2693(95)00944-G, https://arxiv.org/abs/hep-th/9507173
12. G. W. Gibbons and N. S. Manton, 'The moduli space metric for well-separated BPS monopoles', Physics Letters B 356 (1995) 32-38. arXiv:hep-th/9506052, DOI: 10.1016/0370-2693(95)00813-Z, https://arxiv.org/abs/hep-th/9506052
13. C. J. Houghton and P. M. Sutcliffe, 'Inversion symmetric 3-monopoles and the Atiyah-Hitchin manifold', Nonlinearity 9 (1996) 1609-1622. arXiv:hep-th/9609136, DOI: 10.1088/0951-7715/9/6/013, https://arxiv.org/abs/hep-th/9609136
14. G. W. Gibbons, P. Rychenkova and R. Goto, 'HyperKähler Quotient Construction of BPS Monopole Moduli Spaces', Communications in Mathematical Physics 186 (1997) 581-599. DOI: 10.1007/s002200050121, https://doi.org/10.1007/s002200050121; preprint arXiv:hep-th/9608085, https://arxiv.org/abs/hep-th/9608085
15. R. Bielawski, 'Asymptotic Metrics for SU(N)-Monopoles with Maximal Symmetry Breaking', Communications in Mathematical Physics 199 (1998) 297-325. DOI: 10.1007/s002200050503, https://doi.org/10.1007/s002200050503
16. A. Hanany and B. Pioline, '(Anti-)Instantons and the Atiyah-Hitchin Manifold', Journal of High Energy Physics 07 (2000) 001. arXiv:hep-th/0005160, DOI: 10.1088/1126-6708/2000/07/001, https://arxiv.org/abs/hep-th/0005160
17. R. Fariello, H. Forkel and G. Krein, 'Regular and chaotic interactions of two BPS dyons at low energy', Physical Review D 72 (2005) 105015. arXiv:hep-th/0508233, DOI: 10.1103/PhysRevD.72.105015, https://arxiv.org/abs/hep-th/0508233
18. R. Maldonado and R. S. Ward, 'Dynamics of monopole walls', Physics Letters B 734 (2014) 328-332. arXiv:1405.4646, DOI: 10.1016/j.physletb.2014.05.072, https://arxiv.org/abs/1405.4646