Unstable stationary point

Exercise. Prove that inside a limit cycle there must be an unstable stationary point. 

More interesting aspects of stochastic problems are observed when passing to systems with unstable stationary points. Since we restrict ourselves to mono- and bimolecular reactions with a maximum of two intermediate products (freedom degrees), s — 2, only the Lotka-Volterra model by reasons discussed in Section 2.1.1 can serve as the analog of unstable systems. 

As was noted in Section 2.1.1, the concentration oscillations observed in the Lotka-Volterra model based on kinetic equations (2.1.28), (2.1.29) (or (2.2.59), (2.2.60)) are formally undamped. Perturbation of the model parameters, in particular constant k, leads to transitions between different orbits. However, the stability of solutions requires special analysis. Assume that in a given model relation between averages and fluctuations is very simple, e.g., (5NASNB) = f((NA), (A b)), where / is an arbitrary function. Therefore k in (2.2.67) is also a function of the mean values NA(t) and NB(t). Models of this kind are well developed in population dynamics in biophysics [70], Since non-linearity of kinetic equations is no longer quadratic, limitations of the Hanusse theorem [23] are lifted. Depending on the actual expression for / both stable and unstable stationary points could be obtained. Unstable stationary points are associated with such solutions as the limiting cycle in particular, solutions which are interpreted in biophysics as catastrophes (population death). Unlike phenomenological models treated in biophysics [70], in the Lotka-Volterra stochastic model the relation between fluctuations and mean values could be indeed calculated rather than postulated. 

The singularity of the system (0, 0), is an unstable stationary point for e > 0. Apparently, equations (3.71), (3.72) can be immediately solved, being the system of homogeneous first-order linear equations. The solutions have... 

The analysis of linearized sytem thus allows, when conditions (1)—(3) are met, us to find the shape of phase trajectories in the vicinity of stationary (singular) points. A further, more thorough examination must answer the question what happens to trajectories escaping from the neighbourhood of an unstable stationary point (unstable node, saddle, unstable focus). In a case of non-linear systems such trajectories do not have to escape to infinity. The behaviour of trajectories nearby an unstable stationary point will be examined in further subchapters using the catastrophe theory methods. 

Traditional reaction kinetics has dealt with the large class of chemical reactions that are characterised by having a unique and stable stationary point (i.e. all reactions tend to the equilibrium ). The complementary class of reactions is characterised either by the existence of more than one stationary point, or by an unstable stationary point (which could possibly bifurcate to periodic solutions). Other extraordinarities such as chaotic solutions are also contained in the second class. The term exotic kinetics refers to different types of qualitative behaviour (in terms of deterministic models) to sustained oscillation, multistationarity and chaotic effects. Other irregular effects, e.g. hyperchaos (Rossler, 1979) can be expected in higher dimensions. 

The necessary and suflScient condition of the stability of the stationary point is 5 < 1 + It can be shown for this system that an unstable stationary point implies a limit cycle. 

Based on the approximative calculations of Nitzan et al. (1974a) it was suggested that the stationary distribution can be multimodal, and the location of the maxima can be associated with the stable stationary points of the deterministic model, and the minimum corresponds to the unstable stationary point. This result does not hold for small systems and the question of multimodality will be discussed in Subsection 5.7.4. Matheson et al, (1975) obtained a qualitatively similar result, they also estimated the relaxation time of the process from the quasistationary state. Nicolis Turner (1977) calculated the variance at the critical point and in its vicinity. Under and beyond the critical point the variance is a linear function of the volume, but at the critical point it shows a stronger volume-dependence ... 

This solution is non-divergent only for K y (0)) 4= 0. Later it will be seen that the solution of (2.68) at an unstable stationary point where K(y = 0 and K (ym) = 7 > 0, leads to exponential fluctuation enhancement, see (2.93) later in this section. 

Case c) Fluctuation Initiated Motion. If motion is started from a distribution concentrated around an unstable stationary point Xq where K(xo) = 0 and K (xq) = y > 0, the expansion of (2.82, 84,85) fails since the initial fluctuations are enhanced exponentially, see (2.92) before the drift dominated motion sets in. The appropriate approximation here, developed by Haake [2.2, 3] and by Suzuki [2.4] essentially consists of two steps 1) Solve the Fokker-Planck equation for the first fluctuation dominated stage and 2) Find a smoothly fitting solution for the second drift dominated stage. 

An important outcome of these simulations is the location of HB points (largely ignored in previous work), which is important for the development of extinction theory. In particular, the turning point E lies on a locally unstable stationary solution branch and does not coincide with the actual extinction, as previously thought. The actual extinction point is the termination point of oscillations. Thus, local stability analysis is essential to properly analyze flame stability and develop extinction theory. 

With our example data, these give 0.9850 and 0.1020 respectively. These locations may be compared with two other values of /x (i) the crossing point for the stationary-state loci, which for small tcu is given by /z 1 — 2 Ku and which therefore lies slightly above /z and (ii) the maximum in the ass locus which occurs at /z = k 12 and therefore lies just below n. These conclusions about the relative positions of the various features just described in fact hold for all values of ku for which unstable stationary states can occur, not just for the limit ku - 0. The curves in the /z-ku parameter plane corresponding to each of these conditions are also shown in Fig. 3.6. 

As the dimensionless concentration of the reactant decreases so that pi just passes through the upper Hopf bifurcation point pi in Fig. 3.8, so a stable limit cycle appears in the phase plane to surround what is now an unstable stationary state. Exactly at the bifurcation point, the limit cycle has zero size. The corresponding oscillations have zero amplitude but are born with a finite period. The limit cycle and the amplitude grow smoothly as pi is decreased. Just below the bifurcation, the oscillations are essentially sinusoidal. The amplitude continues to increase, as does the period, as pi decreases further, but eventually attains a maximum somewhere within the range pi% < pi < pi. As pi approaches the lower bifurcation point /zf from above, the oscillations decrease in size and period. The amplitude falls to zero at this lower bifurcation point, but the period remains non-zero. 

The behaviour exhibited by this model is relatively simple. There is only ever one limit cycle. This is born at one bifurcation point, grows as the system traverses the range of unstable stationary states, and then disappears at the second bifurcation point. Thus there is a qualitative similarity between the present model and the isothermal autocatalysis of the previous chapter. The limit cycle is always stable and no oscillatory solutions are found outside the region of instability. 

When the Hopf bifurcation at p is supercritical (/ 2 < 0) the system has just a single stable limit cycle. This emerges at p and exists across the range p < p < p, within which it surrounds the unstable stationary-state solution. The limit cycle shrinks back to zero amplitude at the lower bifurcation point p%. This behaviour is qualitatively the same as that shown with the simplifying exponential approximation and is illustrated in Fig. 5.4(a). 

Even for the present simple case, for which the inflow does not contain the autocatalyst, we have seen a variety of combinations of stable and unstable stationary states with or without stable and unstable limit cycles. Stable limit cycles offer the possibility of sustained oscillatory behaviour (and because we are in an open system, these can be sustained indefinitely). A useful way of cataloguing the different possible combinations is to represent the different possible qualitative forms for the phase plane . The phase plane for this model is a two-dimensional surface of a plotted against j8. As these concentrations vary in time, they also vary with respect to each other. The projection of this motion onto the a-/ plane then draws out a trajectory . Stationary states are represented as points, to which or from which the trajectories tend. If the system has only one stationary state for a given combination of k2 and Tres, there is only one such stationary point. (For the present model the only unique state is the no conversion solution this would have the coordinates a,s = 1, Pss = 0.) If the values of k2 and tres are such that the system is lying at some point along an isola, there will be three stationary states on the phase... 

Fig. 8.8. Phase plane representations of the birth (or death) of limit cycles through homoclinic orbit formation. In the sequence (a)-fb)-(c) the system has two stable stationary states (solid circles) and a saddle point. As some parameter is varied, the separatrices of the saddle join together to form a closed loop or homoclinic orbit (b) this loop develops as the parameter is varied further to shed an unstable limit cycle surrounding one of the stationary states. The sequence (d)-(e)-(f) shows the corresponding formation of a stable limit cycle which surrounds an unstable stationary state. (In each sequence, the limit cycle may ultimately shrink on to the stationary state it surrounds—at a Hopf bifurcation point.)...

FlG. 8.14. The different phase plane portraits identified for cubic autocatalysis with decay (a) unique stable state (b) unique unstable stationary state with stable limit cycle (c) unique stable state with unstable and stable limit cycles (d) two stable stationary states and saddle point (e) stable and unstable states with saddle point (f) stable state, saddle point, and unstable state surrounded by stable limit cycle (g) two unstable states and a saddle point, all surrounded by stable limit cylcle (h) two stable states, one surrounded by an unstable limit cycle, and a saddle point (i) stable state surrounded by unstable limit cycle, unstable state, and saddle point, all surrounded by stable limit cycle (j) stable state, unstable state, and saddle point, all surrounded by stable limit cycle (k) stable state, saddle point, and unstable state, the latter surrounded by concentric stable and unstable limit cycles (1) two stable states, one surrounded by concentric unstable and stable limit cycles, and a saddle point. 

For the present scheme, when there is a unique stationary state, we find Apr < 0 and local stability. Under circumstances with multiple solutions, the highest and lowest states always have Apr < 0 and hence are stable the middle branch of solutions has Apr > 0 and hence is a branch of unstable saddle points. 

Fig. 13.1. (a) The phase line for a one-variable system showing three stationary points—two stable (s) and one unstable (u) (b) representation of the potential associated with points along the phase line showing the stationary points as extrema (c), (d) the disallowed motions on the phase line or potential curve which would correspond to oscillatory behaviour, but also to... 

We now have a total of six parameters four from the autonomous system (p, r0, and the desorption rate constants k, and k2) and two from the forcing (rf and co). The main point of interest here is the influence of the imposed forcing on the natural oscillations. Thus, we will take just one set of the autonomous parameters and then vary rf and co. Specifically, we take p = 0.019, r0 = 0.028, fq = 0.001, and k2 = 0.002. For these values the unforced model has a unique unstable stationary state surrounded by a stable limit cycle. The natural oscillation of the system has a period t0 = 911.98, corresponding to a natural frequency of co0 = 0.006 889 6. 

Fig. 2. Types of stationary points on the plane, (a), (c), (e) Stable nodes (b), (d), (f) unstable nodes (g) saddle point (h) stable focus (i) unstable focus, (k) whirl.

We now assume that the expansion of the potential is around a stationary point (stable or unstable, depending on the sign of the second-order derivatives), that is, all the first-order derivatives vanish. The energy is measured relative to the value at equilibrium, and we obtain... 

Figure 43 shows the stationary points for the reaction of EF2 with C2H4. Weakly bonded complexes for all reactions are predicted as minima on the PES which are unstable at... 

---