competing order in the fermionic hubbard model on the
play

Competing order in the fermionic Hubbard model on the hexagonal - PowerPoint PPT Presentation

Competing order in the fermionic Hubbard model on the hexagonal graphene lattice Southampton, 27 July 2016 Pavel Buividovich, Maksim Ulybyshev (Regensburg) Dominik Smith, Lorenz von Smekal (Giessen) Fachbereich 7 | Institut fr Theoretische


  1. Competing order in the fermionic Hubbard model on the hexagonal graphene lattice Southampton, 27 July 2016 Pavel Buividovich, Maksim Ulybyshev (Regensburg) Dominik Smith, Lorenz von Smekal (Giessen) Fachbereich 7 | Institut für Theoretische Physik | Lorenz von Smekal | 01

  2. Introduction [courtesy L. Holicki] Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 2

  3. Honeycomb Lattice - hexagonal Brillouin zone • triangular lattice (2 atoms per unit cell) graphene • single-particle energy bands E ± ( k ) = ± | Φ ( k ) | M structure factor: K' X e i k · δ i Φ (k) = t K i • massless dispersion around Dirac points K ± [Wallace, 1947] v f = 3 ta/ 2 ' 1 ⇥ 10 6 m / s ' c/ 300 E ( p ) = ± ~ v f | p | , Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 3

  4. Honeycomb Lattice • mass terms (gaps) Graphene Gets a Good Gap on SiC Nevis et al ., PRL 115 (2015) 136802 1 X a † k , σ a k , σ − b † � � H m = m σ k , σ b k , σ N 2 k , σ (pseudo-spin) staggered on-site potential Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 4

  5. Honeycomb Lattice • mass terms (gaps) Graphene Gets a Good Gap on SiC Nevis et al ., PRL 115 (2015) 136802 1 X a † k , σ a k , σ − b † � � H m = m σ k , σ b k , σ N 2 k , σ (pseudo-spin) staggered on-site potential � • spin (flavor) dependence m cdw = 1 2 ( m u + m d ) m sdw = 1 2 ( m u � m d ) Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 4

  6. Honeycomb Lattice • mass terms (gaps) Graphene Gets a Good Gap on SiC Nevis et al ., PRL 115 (2015) 136802 1 X a † k , σ a k , σ − b † � � H m = m σ k , σ b k , σ N 2 k , σ (pseudo-spin) staggered on-site potential � • spin (flavor) dependence m cdw = 1 2 ( m u + m d ) m sdw = 1 2 ( m u � m d ) • Coulomb interaction e 2 α g = 4 πε ~ v f effective coupling Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 4

  7. Honeycomb Lattice • mass terms (gaps) Graphene Gets a Good Gap on SiC Nevis et al ., PRL 115 (2015) 136802 1 X a † k , σ a k , σ − b † � � H m = m σ k , σ b k , σ N 2 k , σ (pseudo-spin) staggered on-site potential � with strong interactions: • spin (flavor) dependence Mott-insulator transition m → 0 m cdw = 1 2 ( m u + m d ) − → charge-density wave (CDW) m sdw = 1 2 ( m u � m d ) − → AF spin-density wave (SDW) • Coulomb interaction e 2 α g = 4 πε ~ v f effective coupling Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 4

  8. Honeycomb Lattice • mass terms (gaps) Graphene Gets a Good Gap on SiC Nevis et al ., PRL 115 (2015) 136802 1 X a † k , σ a k , σ − b † � � H m = m σ k , σ b k , σ N 2 k , σ (pseudo-spin) staggered on-site potential � with strong interactions: • spin (flavor) dependence Mott-insulator transition m → 0 m cdw = 1 2 ( m u + m d ) − → charge-density wave (CDW) m sdw = 1 2 ( m u � m d ) − → AF spin-density wave (SDW) • Coulomb interaction 4 QSH QSH e 2 α g = V 2 2 CDW SDW 4 πε ~ v f SM 0 effective coupling SDW 0 CDW 0 1 2 4 2 U V 1 Raghu et al ., PRL 100 (2008) 156401 Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 4

  9. Honeycomb Lattice • mass terms (gaps) Graphene Gets a Good Gap on SiC Nevis et al ., PRL 115 (2015) 136802 1 X a † k , σ a k , σ − b † � � H m = m σ k , σ b k , σ N 2 k , σ (pseudo-spin) staggered on-site potential � with strong interactions: • spin (flavor) dependence Mott-insulator transition m → 0 m cdw = 1 2 ( m u + m d ) − → charge-density wave (CDW) m sdw = 1 2 ( m u � m d ) − → AF spin-density wave (SDW) • Coulomb interaction 4 QSH QSH e 2 α g = V 2 2 CDW SDW 4 πε ~ v f SM 0 effective coupling SDW 0 CDW 0 1 2 4 2 U V 1 • sign-problem in HMC with m cdw > 0 Raghu et al ., PRL 100 (2008) 156401 Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 4

  10. Potentially Strong Interactions • suspended graphene remains conducting, semimetal Elias et al ., Nature Phys. 2049 (2011) e 2 ≈ 300 ε → 1 α g = 137 ≈ 2 . 19 4 π ~ v f • puzzle α crit ∼ 1 predictions at the time Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 5

  11. Potentially Strong Interactions • suspended graphene remains conducting, semimetal Elias et al ., Nature Phys. 2049 (2011) e 2 ≈ 300 ε → 1 α g = 137 ≈ 2 . 19 4 π ~ v f • puzzle α crit ∼ 1 predictions at the time Wehling et al ., PRL 106 (2011) 236805 • screening at short distances from σ - band electrons and localised higher energy states part. screened CRPA : • interpolate at intermediate distances ITEP screened std. Coulomb with dielectric thin-film model 10 U V(r) [eV] � 1 + 1 + ( � 1 − 1) e − kd k ) = 1 � − 1 ( ⃗ V 1 � 1 + 1 − ( � 1 − 1) e − kd � 1 V 2 ( � 1 = 2.4 and d = 2.8 ) 1 0 0.2 0.4 0.6 0.8 1 r [nm] Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 5

  12. HMC on Hexagonal Lattice • chiral extrapolation m sdw → 0 0.4 N x = 36 1 N x = 18 0.3 0.8 0.6 < Δ N > < Δ n> 0.2 0.4 α = 2.73 0.1 0.2 α = 3.36 α = 4.37 0 0 α = 4.86 0 0.1 0.2 0.3 0.4 0.5 0.6 2 2.5 3 3.5 4 4.5 5 m [eV] α eff • semimetal-insulator transition in unphysical regime Ulybyshev, Buividovich, Katsnelson, Polikarpov, α crit ≈ 3 > 2 . 19 PRL 111 (2013) 056801 Smith, LvS, PRB 89 (2014) 195429 Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 6

  13. Dyson-Schwinger Equations • hexagonal lattice, screened Coulomb Manon Bischoff, MSc, TU Da (2015) Katja Kleeberg, MSc, JLU Gi (2015) Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 7

  14. Dyson-Schwinger Equations • hexagonal lattice, screened Coulomb Manon Bischoff, MSc, TU Da (2015) Katja Kleeberg, MSc, JLU Gi (2015) Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 8

  15. Dyson-Schwinger Equations • hexagonal lattice, screened Coulomb Manon Bischoff, MSc, TU Da (2015) Katja Kleeberg, MSc, JLU Gi (2015) Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 9

  16. Dyson-Schwinger Equations • hexagonal lattice, screened Coulomb Manon Bischoff, MSc, TU Da (2015) Katja Kleeberg, MSc, JLU Gi (2015) Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 10

  17. Dyson-Schwinger Equations • hexagonal lattice, screened Coulomb graphene’s single-particle band structure α crit ≈ 1 . 5 • no Lindhard screening Π ( ! , ~ q ) = + + from π - band electrons Manon Bischoff, MSc, TU Da (2015) Katja Kleeberg, MSc, JLU Gi (2015) • what about CDW and the other insulating phases? Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 11

  18. Dyson-Schwinger Equations • hexagonal Hubbard model, Hartree-Fock − 1 − − 1 i Σ ( ~ p ) = = − fermion self-energy Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 12

  19. Dyson-Schwinger Equations • hexagonal Hubbard model, Hartree-Fock − 1 − − 1 i Σ ( ~ p ) = with on-site U and nearest-neighbor V = − first order fermion self-energy Katja Kleeberg et al. , in preparation Araki and Semenoff, PRB 86 (2012) 121402(R) Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 12

  20. HMC on Hexagonal Lattice m sdw → 0 • chiral extrapolation, SDW 12 only on-site U first N x = 6 N x = 12 11 10 max m = 0.0259259 κ 9 χ con 8 N x = 6 N t = 80 8 N x = 12 6 N x = 18 7 < χ dis > 4 6 6 N x = 6 0 0.01 0.02 0.03 0.04 N x = 12 2 m / κ 5 U peak / κ 0 4 4.4 4.8 5.2 5.6 4 U 00 / κ 3 2 0 0.01 0.02 0.03 0.04 m / κ Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 13

  21. HMC on Hexagonal Lattice m sdw → 0 • chiral extrapolation, SDW 12 only on-site U first N x = 6 N x = 12 11 10 max m = 0.0259259 κ 9 χ con 8 N x = 6 N t = 80 8 N x = 12 6 N x = 18 7 < χ dis > 4 6 6 N x = 6 0 0.01 0.02 0.03 0.04 N x = 12 2 m / κ 5 U peak / κ 0 4 4.4 4.8 5.2 5.6 4 U 00 / κ 3 Sorella, Tosatti, EPL 19 (1992) 699: U c ≈ 4 . 5 κ 2 0 0.01 0.02 0.03 0.04 Assaad, Herbut, PRX 3 (2013) 031010: U c ≈ 3 . 8 κ m / κ Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 13

  22. HMC with Geometric Mass • hexagonal Brillouin zone 8 × 8 lattice 12 × 12 lattice • removes Dirac points • preserves symmetries • improves invertibility Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 14

  23. Suitable Order Parameters for zero(geometric)-mass simulations, use s⌦� X 1 � 2 ↵ � 2 ↵ ⌦� X O = O i O i + L 2 i ∈ A i ∈ B with Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 15

  24. Suitable Order Parameters for zero(geometric)-mass simulations, use s⌦� X 1 � 2 ↵ � 2 ↵ ⌦� X O = O i O i + L 2 i ∈ A i ∈ B with • spin-density wave: ⇢ a i , ~ � σσ 0 i ∈ A O i → ~ X c † S i = c i, σ 0 c i = i, σ 2 b i , i ∈ B σ , σ 0 Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 15

  25. Suitable Order Parameters for zero(geometric)-mass simulations, use s⌦� X 1 � 2 ↵ � 2 ↵ ⌦� X O = O i O i + L 2 i ∈ A i ∈ B with • spin-density wave: ⇢ a i , ~ � σσ 0 i ∈ A O i → ~ X c † S i = c i, σ 0 c i = i, σ 2 b i , i ∈ B σ , σ 0 • charge-density wave: X c † � � O i → Q i = i, σ c i, σ − 1 σ Lattice 2016 27 July 2016 | Lorenz von Smekal | p. 15

Recommend


More recommend