Recently, the symmetry of many single-particle non-Hermitian Hamiltonians has been conjectured to be sufficient to keep their spectrum real. We demonstrate this and how the similar concept of a `weakened Hermiticity' can be extended to some exactly solvable two- and three-particle models.
The coupling of non-Hermitian -symmetric Hamiltonians to standard Hermitian Hamiltonians, each of which individually has a real energy spectrum, is explored by means of a number of soluble models. It is found that in all cases the energy remains real for small values of the coupling constant, but becomes complex if the coupling becomes stronger than some critical value. For a quadratic non-Hermitian -symmetric Hamiltonian coupled to an arbitrary real Hermitian -symmetric Hamiltonian, the reality of the ground-state energy for small enough coupling constant is established up to second order in perturbation theory.
Ordinarily, one demands a Hamiltonian to satisfy H † = H, where † represents the mathematical operation of complex conjugation and matrix transposition. This conventional hermiticity condition ensures that H has a real spectrum. Recently, it was discovered that replacing this mathematical condition by the weaker and more physical requirement H ‡ = H, where ‡ represents combined parity reflection and time reversal , one obtains new classes of complex Hamiltonians whose spectra are often real and positive. However, the condition of symmetry alone is not strong enough to guarantee that the spectrum of H is real. In this paper evidence is presented to support the conjecture that the spectrum of the complex Hamiltonian H = p 2 + V(x)(ix) ∊ (∊ > 0) will be real, positive, and discrete if (1) V(x) is real when x is real; (2) the spectrum of H = p 2 + V(x) is positive and discrete; (3) the function V(x) is even [V(x) = V(-x)]; (4) the function V(x) is an entire function of complex x. We believe that if these conditions are satisfied then the complex deformation H = p 2 + V(x)(ix) ∊ of H = p 2 + V(x) will have a real, positive, and discrete spectrum.
We introduce the notion of pseudo-Hermiticity and show that every Hamiltonian with a real spectrum is pseudo-Hermitian. We point out that all the PT-symmetric non-Hermitian Hamiltonians studied in the literature belong to the class of pseudo-Hermitian Hamiltonians, and argue that the basic structure responsible for the particular spectral properties of these Hamiltonians is their pseudo-Hermiticity. We explore the basic properties of general pseudo-Hermitian Hamiltonians, develop pseudosupersymmetric quantum mechanics, and study some concrete examples, namely the Hamiltonian of the two-component Wheeler–DeWitt equation for the FRW-models coupled to a real massive scalar field and a class of pseudo-Hermitian Hamiltonians with a real spectrum.
Although parity-time (PT)-symmetric Hamiltonians are often associated with real energy spectra, PT symmetry is neither a sufficient nor a necessary condition for a real spectrum. More generally, real spectra are associated with the broader class of pseudo-Hermitian Hamiltonians, of which PT-symmetric Hamiltonians constitute a simple subclass. Here, we investigate the nonequilibrium dynamics of the imbalanced-pairing Kitaev model, a prototypical pseudo-Hermitian system, under a linearly time-dependent chemical potential. The dynamics are analyzed within the biorthogonal framework using the concept of dynamical quantum phase transitions (DQPTs). We show that, under a linear ramp protocol, DQPTs occur only when the post-ramp Hamiltonian possesses a real energy spectrum. For positive values of the non-Hermiticity parameter (γ>0), where the energy spectrum remains entirely real, a ramp crossing a single quantum critical point gives rise to a single family of critical times, analogous to the Hermitian case. Furthermore, for ramps crossing two critical or exceptional points, the critical sweep velocity above which DQPTs disappear decreases as the non-Hermiticity parameter is reduced and vanishes in the staggered-pairing limit, γ=−1.