Possibility of magnetic field induced Composite Fermi sea in neutral Graphene

G. Baskaran
DOI: https://doi.org/10.48550/arXiv.cond-mat/0702420
2007-02-18
Abstract:Neutral graphene in strong magnetic fields is believed to be an (exchange stabilized) integer Hall state of completely filled up spin (say) and empty down spin bands of n = 0, two fold valley degenerate Landau levels. We suggest that correlation energy gain from a nearly SU(4) singlet (2 spin $\times$ 2 valley) composite fermi sea formation, at $\nu \approx$ \hlf filling for each component, destabilizes ferromagnetic integer quantum Hall state. This radically different scenario is consistent with a dissipative gapless state seen in experiments in neutral graphene. Interesting paired Hall states are possible, from residual interactions, in this SU(4) fermi sea with a small Zeeman spin polarization.
Mesoscale and Nanoscale Physics,Strongly Correlated Electrons
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to calculate the discrete eigenvalues in the essential spectral gaps of the perturbed periodic Schrödinger operators by rapidly decaying potentials, while avoiding spectral pollution. Specifically: 1. **Problem Background**: - For the periodic Schrödinger operator \(H = -\Delta+V\), where \(V = V_p+V_d\), \(V_p\) is the periodic potential and \(V_d\) is the rapidly decaying perturbation potential. - The essential spectrum is determined by the periodic part \(V_p\), forming a series of continuous spectral bands, and there are spectral gaps between the bands. - When \(V_d\neq0\), discrete eigenvalues may appear in these spectral gaps. 2. **Challenges**: - When using standard projection methods (such as the Galerkin method) to approximate these discrete eigenvalues located in the spectral gaps, the phenomenon of spectral pollution is likely to occur. That is, spurious eigenvalues will appear, and these spurious eigenvalues do not belong to the real spectrum. - This phenomenon makes it very difficult to numerically locate these discrete eigenvalues. 3. **Solutions**: - The paper introduces a new method - the quadratic projection method - to overcome the spectral pollution problem. - This method constructs a quadratic matrix polynomial \(P_L(z)\), which has the form: \[ P_L(z):=Q_L - 2zA_L+z^2B_L \] where \(A_L\), \(B_L\), \(Q_L\) are the stiffness matrix, the mass matrix and the bending matrix respectively. - The core advantage of the quadratic projection method is that it does not produce spectral pollution and can provide two - way estimates of eigenvalue errors. 4. **Theoretical Basis and Applications**: - The paper describes in detail the theoretical basis of the quadratic projection method and demonstrates its effectiveness through specific numerical examples. - This method is not only applicable to one - dimensional models, but also successfully applied to two - dimensional models, verifying its applicability and accuracy in different situations. In summary, this paper aims to solve the spectral pollution problem in the eigenvalue calculation in the spectral gaps of the periodic Schrödinger operator through the quadratic projection method, thereby providing a more reliable and accurate numerical method.