Scaling theory correlation length

At first, one would tend to reconsider conventional crossover due to mean-field criticality associated with long-range interactions in terms of the refined theories. Conventional crossover conforms to the first case mentioned—that is, small u with the correlation length of the critical fluctuations to be larger than 0. However, in the latter case one expects smooth crossover with slowly and monotonously varying critical exponents, as observed in nonionic fluids. Thus, the sharp and nonmonotonous behavior cannot be reconciled with one length scale only. 

The correlation length is also found to scale in a power-law fashion, and it becomes very large at the transition temperature. One of the most significant results of renormalization group theory is to show that the behavior of the correlation length in the critical region is the basis of the power-law singularities observed in the other thermodynamic properties. 

The finite-size scaling theory combined with transfer matrix calculations had been, since the development of the phenomenological renormalization in 1976 by Nightingale [70], one of the most powerful tools to study critical phenomena in two-dimensional lattice models. For these models the partition function and all the physical quantities of the system (free energy, correlation length, response functions, etc) can be written as a function of the eigenvalues of the transfer matrix [71]. In particular, the free energy takes the form... 

Below the gel point, the system is self-similar on length scales smaller than the correlation length with a power law distribution of molar masses with Fisher exponent r = 5/2 [Eq. (6.78)]. Each branched molecule is a self-similar fractal with fractal dimension 27 = 4 for ideal branched mole-cules in the mean-field theory. The lower limit of this critical behaviour is the average distance between branch points (= ). There are very few... 

Determination of the critical exponents by Monte Carlo simulation is usually based on the aforementioned finite size scaling theory [123 - 126,131 - 134]. A detailed presentation of this theory is well beyond the scope of this chapter. Moreover, several excellent reviews concerning this theory are available [133,134]. Therefore, I confine the discussion to some basic facts showing that it is possible to obtain reliable information about the properties of macroscopic systems fi om the results obtained from computer simulations performed for finite systems. In a macroscopic system near the critical point the correlation length diverges to infinity and the following relationship is satisfied ... 

The relation A7 oc Rq obtained from the Ginzburg-Pitaevskii-Sobyanin theory for a finite system is related to the theory of second-order phase transitions with the experimental critical parameter, v = 0.67, for the superfluid fraction and for the correlation length scaling near the critical point of infinite systems [155, 193-197, 199]. This theory implies that the intensive properties of a system of size L(= Rq) depend on the ratio L/ T) Lf, where (T) = th bulk... 

From scaling theory it follows a concentration dependence of the correlation length It decreases with increasing concentration, with (47) therefore to... 

Arguments based on quite different scaling theory assumptions yield much the same order-of-magnitude estimates. For r - oo and fixed x= icr,2, where k is an inverse correlation length, Kcr/, we can use (4.82) with... 

---