The Mott critical point between a metal and a correlated insulator has usually been studied via density or spin density bosonic mode fluctuations according to the standard Ginzburg-Landau-Wilson phase transition paradigm. A moment's reflection leads to increasing doubts that such an approach should work as the transition is nonmagnetic, voiding the relevance of spin density modes. Charge density modes are irrelevelant since the long range Coulomb interaction leads to a large plasmon gap and their incompressibility. In solidarity with these doubts, recent measurements of the Mott critical point in low dimensional organic materials yield critical exponents in violent diasagreement with the bosonic mode criticality lore. We propose that fermionic fluctuations control the behavior of the Mott transition. The transition thus has an intrinsic quantum aspect despite being a finite temperature phase transition. We develop this hitherto unexplored physics, obtain experimental predictions and find agreement with one of the novel unexplained experimental exponents. We conclude that this Mott transition corresponds to a new universality class of finite temperature critical points that contains quantum effects and cannot be accounted for by conventional Ginzburg-Landau-Wilson wisdom.
A fundamental issue in the renormalization-group (RG) theory of critical phenomena concerns the allowed values of critical exponents that are consistent with the continuous nature of a phase transition. Here we conjecture a lower bound for the length-scale exponent
ν, which should hold for the large class of continuous transitions associated with
d-dimensional Landau-Ginzburg-Wilson (LGW)
Φ4 theories with a multicomponent scalar field
φ and a unique
φ⋅φ quadratic term (including some extensions with fermionic and gauge fields), describing many universality classes of critical phenomena. If
Δφ=(d−2+η)/2 is the dimension of the order-parameter field
φ, and
Δε=d−1/ν is the RG dimension of the energy operator
ε, which can be identified with
[φ⋅φ] (the squared field with a proper subtraction of the mixing with the identity), we conjecture the inequality
Δε≥2Δφ, which implies
ν≥(2−η)−1 and
γ=(2−η)ν≥1. These inequalities are supported by general arguments for ferromagnetic lattice models, by
ε-expansion results for generic LGW
Φ4 theories close to four dimensions, exact relations for two-dimensional minimal conformal field theories, and are consistent with all further known (numerical, perturbative, and exact) results for LGW
Φ4 theories. In particular, since unitarity requires
η≥0, the above inequality implies
ν≥1/2 for unitary theories. This lower bound is more restrictive than
ν>1/d, derived by noting that
ν=1/d characterizes the singular finite-size behavior at first-order transitions.