Normally hyperbolic invariant manifolds orbits

The point q = p = 0 (or P = Q = 0) is a fixed point of the dynamics in the reactive mode. In the full-dimensional dynamics, it corresponds to all trajectories in which only the motion in the bath modes is excited. These trajectories are characterized by the property that they remain confined to the neighborhood of the saddle point for all time. They correspond to a bound state in the continuum, and thus to the transition state in the sense of Ref. 20. Because it is described by the two independent conditions q = 0 and p = 0, the set of all initial conditions that give rise to trajectories in the transition state forms a manifold of dimension 2/V — 2 in the full 2/V-dimensional phase space. It is called the central manifold of the saddle point. The central manifold is subdivided into level sets of the Hamiltonian in Eq. (5), each of which has dimension 2N — 1. These energy shells are normally hyperbolic invariant manifolds (NHIM) of the dynamical system [88]. Following Ref. 34, we use the term NHIM to refer to these objects. In the special case of the two-dimensional system, every NHIM has dimension one. It reduces to a periodic orbit and reproduces the well-known PODS [20-22]. 

An unstable periodic orbit is one-dimensional, being of dimension two less than the energy surface in systems with two DOFs. In an n-DOF system the energy surface is of dimension 2n — 1. In such systems, Wiggins showed that the analog of unstable periodic orbits is the so-called normally hyperbolic invariant manifold (NHIM) of dimension 2n — 3 [20,21]. Trajectories slightly displaced from an NHIM can be analyzed using a many-dimensional stability analysis. The... 

Recently, Wiggins et al. [15] provided a firm mathematical foundation of the robust persistence of the invariant of motion associated with the phase-space reaction coordinate in a sea of chaos. The central component in RIT that is, unstable periodic orbits, are naturally generalized in many DOFs systems in terms of so-called normally hyperbolic invariant manifold (NHIM). The fundamental theorem on NHIMs, denoted here by M, ensures [21,53] that NHIMs, if they exist, survive under arbitrary perturbation with the property that the stretching and contraction rates under the linearized dynamics transverse to jM dominate those tangent to M. Note that NHIM only requires that instability in either a forward or backward direction in time transverse to M is much stronger than those tangential directions of M, and hence the concept of NHIM can be applied to any class of continuous dynamical systems. In the case of the vicinity of saddles for Hamiltonian problems with many DOFs, the NHIM is expressed by a set of all (p, q) satisfying both q = p = Q and o(Jb) + En=i (Jb, b) = E, that is. 

We shall make more use of the notion of normally hyperbolic invariant manifold (NHIM). This invariant surface is the n-DOF generalization of the periodic orbit dividing surface, even if originally defined in a much more general framework (a bibliography may be found in Ref. 24). Its correct definition is put forward in Section IV.A and is used in all examples coming thereafter. 

Figure 3.25(a) shows a case where a strong correlation exists between processes of crossing two neighboring barriers. This correlation results from direct intersection between the stable and unstable manifolds of the normally hyperbolic invariant manifolds located on the tops of neighboring barriers. Then, some of the orbits starting from one of the normally hyperbolic invariant manifolds directly reach the other one without falling into the potential well. Some other orbits may fall into the well and would take some time to reach the other manifold. The ratio of these two types of orbits depends on how steep the intersection is. 

As for the term invariant, it means that orbits starting on a NHIM stay on it at least locally in time for both forward and backward directions. However, due attention must be paid to the following. In general, a NHIM will have boundaries where orbits starting on it flow off the manifold. This is because the flow on it could reach those locations where normal hyperbolicity breaks down. Later in this chapter we will mention an example of this behavior. 

The surface t v-1 is a particular example of what Wiggins has referred to as a hyperbolic manifold and what De Leon and Ling have termed a normally invariant hyperbolic manifoldd Hyperbolic manifolds are unstable and constitute the formal multidimensional generalization of unstable periodic orbits. Hyperbolic manifolds, like PODS, can be either repulsive or attractive. - If motion near a hyperbolic manifold falls away without recrossing it in configuration space, the hyperbolic manifold is said to be repulsive. On the other hand, it is often the case that motion near a hyperbolic manifold will cross it several times in configuration space as it falls away, and in this case it is said to be attractive. 

---