Quantum Entanglement, Strange metals, and black holes. Subir Sachdev, Harvard University

Transcription

1 Quantum Entanglement, Strange metals, and black holes Subir Sachdev, Harvard University

2 Quantum entanglement

3 Quantum Entanglement: quantum superposition with more than one particle Hydrogen atom: Hydrogen molecule: = _ = 1 2 ( )

4 Quantum Entanglement: quantum superposition with more than one particle _

5 Quantum Entanglement: quantum superposition with more than one particle _

6 Quantum Entanglement: quantum superposition with more than one particle _

7 Quantum Entanglement: quantum superposition with more than one particle _ Einstein-Podolsky-Rosen paradox (1935): Measurement of one particle instantaneously determines the state of the other particle arbitrarily far away

8 Quantum entanglement

9 Quantum entanglement Black holes

10 Black Holes Objects so dense that light is gravitationally bound to them. In Einstein s theory, the region inside the black hole horizon is disconnected from the rest of the universe. Horizon radius R = 2GM c 2

11 On September 14, 2015, LIGO detected the merger of two black holes, each weighing about 30 solar masses, with radii of about 100 km, 1.3 billion light years away 0.1 seconds later!

12 LIGO September 14, 2015

13 Black Holes + Quantum theory Around 1974, Bekenstein and Hawking showed that the application of the quantum theory across a black hole horizon led to many astonishing conclusions

14 Quantum Entanglement across a black hole horizon _

15 Quantum Entanglement across a black hole horizon _ Black hole horizon

16 Quantum Entanglement across a black hole horizon _ Black hole horizon

17 Quantum Entanglement across a black hole horizon There is long-range quantum entanglement between the inside and outside of a black hole Black hole horizon

18 Quantum Entanglement across a black hole horizon Hawking used this to show that black hole horizons have an entropy and a temperature (because to an outside observer, the state of the electron inside the black hole is an unknown) Black hole horizon

19 Quantum Entanglement across a black hole horizon The Hawking temperature k B T H = ~2 8 M`2P and the Bekenstein-Hawking (BH) black hole entropy S BH k B = A 4`2P where `P = p ~G/c 3 in the Planck length, and A is the surface area of the black hole. Note the entropy is proportional to the surface area rather than the volume. Black hole horizon

20 LIGO September 14, 2015 The Hawking temperature, T H influences the radiation from the black hole at the very last stages of the ring-down (not observed so far). The ring-down (approach to thermal equilibrium) happens very rapidly in a time ~ = 8 GM k B T H c 3 8 milliseconds.

21 Quantum entanglement Black holes Black holes have an entropy and a temperature, T H. The entropy is proportional to their surface area. They relax to thermal equilibrium in a time ~/(k B T H ).

22 Quantum entanglement Strange metals

23 High temperature superconductors YBa 2 Cu 3 O 6+x

24 Slide by J. C. Seamus Davis Power Efficiency/Capacity/Stability Power Bottlenecks Efficient Rotating Machines Information Technology HE Accelerators Ultra-High Magnetic Fields Accommodate Renewable Power Next Generation HEP Science / Medicine Medical Transport

25 High temperature superconductors YBa 2 Cu 3 O 6+x

26 Antiferromagnet Strange metal SM Superconductor FL p (hole/cu) Figure: K. Fujita and J. C. Seamus Davis

27 Antiferromagnet Strange metal SM FL Spins of electrons on Cu sites p (hole/cu) Figure: K. Fujita and J. C. Seamus Davis

28 Square lattice of Cu sites

29 Square lattice of Cu sites Remove density p electrons

30 Square lattice of Cu sites Electrons entangle in ( Cooper ) pairs into chemical bonds =

31 Square lattice of Cu sites Superconductivity = Cooper pairs form quantum superpositions at different locations: Bose-Einstein condensation in which all pairs are everywhere at the same time

32 Square lattice of Cu sites Superconductivity = Cooper pairs form quantum superpositions at different locations: Bose-Einstein condensation in which all pairs are everywhere at the same time

33 Square lattice of Cu sites Superconductivity = Cooper pairs form quantum superpositions at different locations: Bose-Einstein condensation in which all pairs are everywhere at the same time

34 Square lattice of Cu sites Superconductivity = Cooper pairs form quantum superpositions at different locations: Bose-Einstein condensation in which all pairs are everywhere at the same time

35 Square lattice of Cu sites High temperature superconductivity! Electrons entangle by exchanging partners, and there is longrange quantum entanglement in the strange metal. =

36 Square lattice of Cu sites High temperature superconductivity! Electrons entangle by exchanging partners, and there is longrange quantum entanglement in the strange metal. =

37 Square lattice of Cu sites High temperature superconductivity! Electrons entangle by exchanging partners, and there is longrange quantum entanglement in the strange metal. =

38 Square lattice of Cu sites High temperature superconductivity! Electrons entangle by exchanging partners, and there is longrange quantum entanglement in the strange metal. =

39 Square lattice of Cu sites High temperature superconductivity! Electrons entangle by exchanging partners, and there is longrange quantum entanglement in the strange metal. =

40 Square lattice of Cu sites High temperature superconductivity! Electrons entangle by exchanging partners, and there is longrange quantum entanglement in the strange metal. =

41 Square lattice of Cu sites High temperature superconductivity! Electrons entangle by exchanging partners, and there is longrange quantum entanglement in the strange metal. =

42 Strange metal Entangled electrons lead to strange SM temperature dependence of resistivity and FL other properties p (hole/cu) Figure: K. Fujita and J. C. Seamus Davis

43 Almost all many-electron systems are described by the quasiparticle concept: a quasiparticle is an excited lump in the many-electron state which responds just like an ordinary particle. R.D. Mattuck

44 Almost all many-electron systems are described by the quasiparticle concept: a quasiparticle is an excited lump in the many-electron state which responds just like an ordinary particle. Quasiparticles are additive excitations: The low-lying excitations of the many-body system can be identified as a set {n } of quasiparticles with energy " E = P n " + P, F n n +...

45 Almost all many-electron systems are described by the quasiparticle concept: a quasiparticle is an excited lump in the many-electron state which responds just like an ordinary particle. Quasiparticles eventually collide with each other. Such collisions eventually leads to thermal equilibration in a chaotic quantum state, but the equilibration takes a long time. In a Fermi liquid, this time is of order ~E F /(k B T ) 2 as T! 0, where E F is the Fermi energy.

46 Quantum matter without quasiparticles The complex quantum entanglement in the strange metal does not allow for any quasiparticle excitations.

47 Quantum matter without quasiparticles The complex quantum entanglement in the strange metal does not allow for any quasiparticle excitations. Systems without quasiparticles, like the strange metal, reach quantum chaos much more quickly than those with quasiparticles. There is an lower bound on the phase coherence time ( ' ), and the time to many-body quantum chaos ( L ) in all many-body quantum systems as T! 0: ' C ~ k B T ~ L 2 k B T (SS, 1999) (Maldacena, Shenker, Stanford, 2015) So e.g. we cannot have ' ~/ p Jk B T where J is a microscopic coupling. In the strange metal the inequalities become equalities as T! 0, and the time ~/(k B T ) influences numerous observables.

48 Quantum matter without quasiparticles The complex quantum entanglement in the strange metal does not allow for any quasiparticle excitations. Systems without quasiparticles, like the strange metal, reach quantum chaos much more quickly than those with quasiparticles. There is an lower bound on the phase coherence time ( ' ), and the time to many-body quantum chaos ( L ) in all many-body quantum systems as T! 0: ' C ~ k B T ~ L 2 k B T (SS, 1999) (Maldacena, Shenker, Stanford, 2015) So e.g. we cannot have ' ~/ p Jk B T where J is a microscopic coupling. In the strange metal the inequalities become equalities as T! 0, and the time ~/(k B T ) influences numerous observables.

49 Quantum matter without quasiparticles The complex quantum entanglement in the strange metal does not allow for any quasiparticle excitations. Systems without quasiparticles, like the strange metal, reach quantum chaos much more quickly than those with quasiparticles. There is an lower bound on the phase coherence time ( ' ), and the time to many-body quantum chaos ( L ) in all many-body quantum systems as T! 0: ' C ~ k B T ~ L 2 k B T (SS, 1999) (Maldacena, Shenker, Stanford, 2015) So e.g. we cannot have ' ~/ p Jk B T where J is a microscopic coupling. In the strange metal the inequalities become equalities as T! 0, and the time ~/(k B T ) influences numerous observables.

50 Quantum entanglement Strange metals have no quasiparticle description. Their entropy is proportional Strange metals to their volume. They relax to local thermal equilibrium in a time ~/(k B T ).

51 Quantum entanglement Black holes Black holes have an entropy and a temperature, T H. The entropy is proportional to their surface area. They relax to thermal equilibrium in a time ~/(k B T H ).

52 Quantum entanglement Black holes Strange metals A toy model which is both a strange metal and a black hole!

53 A simple model of a metal with quasiparticles Pick a set of random positions

54 A simple model of a metal with quasiparticles Place electrons randomly on some sites

55 A simple model of a metal with quasiparticles Electrons move one-by-one randomly

56 A simple model of a metal with quasiparticles Electrons move one-by-one randomly

57 A simple model of a metal with quasiparticles Electrons move one-by-one randomly

58 A simple model of a metal with quasiparticles Electrons move one-by-one randomly

59 A simple model of a metal with quasiparticles 1 NX H = (N) 1/2 i,j=1 t ij c i c j +... c i c j + c j c i =0, c i c j + c j c i = ij 1 N X i c i c i = Q t ij are independent random variables with t ij = 0 and t ij 2 = t 2 Fermions occupying the eigenstates of a N x N random matrix

60 A simple model of a metal with quasiparticles p Let " be the eigenvalues of the matrix t ij / N. The fermions will occupy the lowest NQ eigenvalues, upto the Fermi energy E F.Thedensity of states is (!) =(1/N ) P (! " ). (!) E F!

61 A simple model of a metal with quasiparticles There are 2 N many body levels with energy Many-body level spacing 2 N E = NX =1 n ", Quasiparticle excitations with spacing 1/N where n =0, 1. Shown are all values of E for a single cluster of size N = 12. The " have a level spacing 1/N.

62 The Sachdev-Ye-Kitaev (SYK) model Pick a set of random positions

63 The Sachdev-Ye-Kitaev (SYK) model Place electrons randomly on some sites

64 The Sachdev-Ye-Kitaev (SYK) model Entangle electrons pairwise randomly

65 The Sachdev-Ye-Kitaev (SYK) model Entangle electrons pairwise randomly

66 The Sachdev-Ye-Kitaev (SYK) model Entangle electrons pairwise randomly

67 The Sachdev-Ye-Kitaev (SYK) model Entangle electrons pairwise randomly

68 The Sachdev-Ye-Kitaev (SYK) model Entangle electrons pairwise randomly

69 The Sachdev-Ye-Kitaev (SYK) model Entangle electrons pairwise randomly

70 The Sachdev-Ye-Kitaev (SYK) model The SYK model has nothing but entanglement

71 The Sachdev-Ye-Kitaev (SYK) model This describes both a strange metal and a black hole!

72 The Sachdev-Ye-Kitaev (SYK) model H = 1 (2N) 3/2 NX i,j,k,`=1 J ij;k` c i c j c k c` µ X i c i c i c i c j + c j c i =0, c i c j + c j c i = ij Q = 1 X c N i c i i J ij;k` are independent random variables with J ij;k` = 0 and J ij;k` 2 = J 2 N!1yields critical strange metal. S. Sachdev and J. Ye, PRL 70, 3339 (1993) A. Kitaev, unpublished; S. Sachdev, PRX 5, (2015)

73 The Sachdev-Ye-Kitaev (SYK) model Many-body level spacing 2 N = e N ln 2 Non-quasiparticle excitations with spacing e S GP S There are 2 N many body levels with energy E, which do not admit a quasiparticle decomposition. Shown are all values of E for a single cluster of size N = 12. The T! 0 state has an entropy S GP S with S GP S N = G + ln(2) 4 < ln 2 = where G is Catalan s constant, for the half-filled case Q =1/2. GPS: A. Georges, O. Parcollet, and S. Sachdev, PRB 63, (2001) W. Fu and S. Sachdev, PRB 94, (2016)

74 A simple model of a metal with quasiparticles There are 2 N many body levels with energy Many-body level spacing 2 N E = NX =1 n ", Quasiparticle excitations with spacing 1/N where n =0, 1. Shown are all values of E for a single cluster of size N = 12. The " have a level spacing 1/N.

75 The Sachdev-Ye-Kitaev (SYK) model Many-body level spacing 2 N = e N ln 2 Non-quasiparticle excitations with spacing e S GP S There are 2 N many body levels with energy E, which do not admit a quasiparticle decomposition. Shown are all values of E for a single cluster of size N = 12. The T! 0 state has an entropy S GP S with S GP S N = G + ln(2) 4 < ln 2 = where G is Catalan s constant, for the half-filled case Q =1/2. GPS: A. Georges, O. Parcollet, and S. Sachdev, PRB 63, (2001) W. Fu and S. Sachdev, PRB 94, (2016)

76 SYK and black holes The SYK model has a nonzero entropy, S GP S / N as T! 0. A. Georges, O. Parcollet, and S. Sachdev, PRB 63, (2001) The SYK model has a phasecoherence time ' ~/(k B T ) O. Parcollet and A. Georges, PRB 59, 5341 (1999) These properties indicate that SYK model holographically realizes a black hole, and the black hole entropy S BH = S GP S. S. Sachdev, PRL 105, (2010)

77 SYK and black holes T 2 ~x T 2 ) two-dimensional torus

78 SYK and black holes Black hole horizon SS, PRL 105, (2010) T 2 ~x The SYK model has dual description in which an extra spatial dimension,, emerges. The curvature of this emergent spacetime is described by Einstein s theory of general relativity

79 SYK and black holes GPS BH entropy entropy AdS 2 T 2 ds 2 =(d 2 dt 2 )/ 2 + d~x 2 Gauge field: A =(E/ )dt charge density Q T 2 = 1 Z S = d 4 x p ĝ ~x ˆR +6/L ˆF µ µ ˆF SS, PRL 105, (2010) The BH entropy is proportional to the size of T 2, and hence the surface area of the black hole. Mapping to SYK applies when temperature 1/(size of T 2 ).

80 SYK and black holes Same long-time e ective action for energy and number fluctuations, involving Schwarzian derivatives of time reparameterizations f( ). AdS 2 T 2 ds 2 =(d 2 dt 2 )/ 2 + d~x 2 Gauge field: A =(E/ )dt charge density Q T 2 = 1 ~x Einstein-Maxwell theory + cosmological constant A. Kitaev, unpublished, J. Maldacena, D. Stanford, and Zhenbin Yang, arxiv: ; K. Jensen, arxiv: ; J. Engelsoy, T.G. Mertens, and H. Verlinde, arxiv:

81 SYK and black holes GPS BH entropy entropy AdS 2 T 2 ds 2 =(d 2 dt 2 )/ 2 + d~x 2 Gauge field: A =(E/ )dt charge density Q T 2 = 1 ~x An extra spatial dimension emerges from SS, PRL 105, (2010) quantum entanglement!

82 Tensor network of hierarchical entanglement ~x D-dimensional space depth of entanglement B. Swingle

83 String theory near a D-brane ~x D-dimensional space Emergent depth of spatial direction entanglement of SYK model or string theory

84 String theory near a D-brane ~x D-dimensional space Quantum entanglement can lead to an emergent spatial dimension Emergent depth of spatial direction entanglement of SYK model or string theory

85 Quantum entanglement Black holes Strange metals A toy model which is both a strange metal and a black hole!

86 Quantum Entanglement, Strange metals, and black holes Subir Sachdev, Harvard University