Feb. 04, University of Virginia
1 2008 of IRON Superconductors Outline
Introduction Point Contact (PC) spectroscopy Spin Transfer Torque ( STT ) Effects Point Contact Andreev Reflection ( PCAR ) spectroscopy Spin polarization Superconducting gap Pseudogaps ? Summary
4 Discovery of Superconductivity
Liquefied Helium (1908), b.p. = 4.17 K
Discovered superconductivity (1911) in Hg T C = 4.21 K
Nobel prize (1913)
Heike Kamerlingh Onnes ρ
T 0 C T The saga of superconductivity began. 5 Perfect Conductivity and Perfect Diamagnetism
Zero resistance Perfect conductivity I Persistent current I over 10 5 years
Meissner effect Perfect diamagnetism
No field inside SC
6 Type I and Type II Superconductors
Type I (Pb, Al, etc) Type II (NbTi, Nb 3Sn, etc)
B ≠ 0
B = 0 B = 0
Meissner (H < H C1 ) Meissner (H < H C) Mixed state (H C1 < H < H C2 ) HC2 (0) very large HC(0) ≈ 0.1 Tesla
7 Vortices in Mixed State
Abrikosov Vortex Lattices in Superconductors (1957)
Nobel Prize (2003)
2 ΦΦΦοοο = hc/2e ≈ 20 G-µm
Hess (1989) (STM, NbSe ) Essmann (1967) (Bitter, Pb) 2
1962 Theoretical prediction of Josephson effect 1973 Nobel prize
SC1 SC2
= φ Brian D. Josephson I(t) IC sin( (t)) φ: phase difference
h − Φ = = .2 067833636 ×10 15 Wb 0 2e
Superconducting Quantum Interference Device (SQUID ) Most sensitive detector 9 Bardeen-Cooper-Schrieffer (BCS) Theory (1957)
e ξ e
Cooper pairs Bardeen-Cooper-Schrieffer Nobel Prize, 1972
Superconducting Gap ∆∆∆(0)(0)(0) ∆∆∆ 2 /kBTC = 3.53 Eg = 2 10 Very large H c2 for Type II Superconductors
High H C2 is necessary Superconducting magnets Magnetic levitation
High TC is always beneficial
11 TC vs. Discovery Year
12 Discovery of Cuprates (1986)
Discovered (La-Ba)2CuO 4 in 1986, T C = 30 K
Nobel prize, 1987 Alex Müller and Georg Bednorz
CuO 2 Plane
La 2CuO 4
(La-Sr)2CuO 4
(La-Ba)2CuO 4 ∼ TC 30 – 40 K
13 Cuprates
YBa Cu O T = 90 K 2 3 7 C Cuprates TC
Cu chains (La-Sr)2CuO 4 35 K
YBa 2Cu 3O7 90 K
Bi 2Sr 2Ca 2Cu 3O10 105 K Cu planes TlBa 2Ca 2Cu 3O9 125 K
Cu planes HgBa 2Ca 2Cu 3O8 140 K
Cu chains Cu O Cu
CuO 2 plane Responsible for high T and a host of unusual properties C 14 TC vs. Discovery Year
15 Mind the Gap
16 s-wave and p-wave Superconductor
∆ s-wave ( isotropic ) Fermi sea Conventional SC (Nb, Pb, Al,…) TC < 25 K
+ p-wave 3 Fermi sea Superfluid He, Sr 2RuO 4(?) ∼ TC 1 mK - 1 K -
17 d-wave Superconductor
Fisher et al., RMP 79 , 353 (2007)
+
- Fermi sea -
+
d 2 2 x − y d-wave symmetry Parent compound: Mott insulator
Cuprates: CuO 2 plane TC < 160 K
18 New Oxypnictides Fe Superconductors (2008)
“Single layer” SmFeAsO Pnictogen (P, As, Sb)
(SmO) +
(FeAs) -
(SmO) +
No CuO 2 plane Contains Fe Contains As
19 Discovery of Fe-Superconductors (2008) (SDW)
TSDW
TC
New Fe-SC: ∼ SmFeAsO 1-xFx (1111) TC 55K Parent compound: SmFeAsO metallic Spin density wave ( SDW ) ∼ 160K
20 Fe -Superconductors New Fe-SC: ∼ SmFeAsO 1-xFx (1111), TC 55K Parent compound: SmFeAsO metallic Spin density wave ( SDW ) at 160K ∆ = ? What value ? What structure ? What temperature dependence ? Are there pseudogaps ?
21 Point Contact
I+ ρ = V+ RMaxwell a >> l, diffusive I- 2a 4ρl 3π a V- R = 1( + ) Wexler 3πa2 8 l a 4ρl R = Sharvin 3πa2 a << l, ballistic
Maxwell ρρρ: resistivity a : contact size l : mean free path
Different regime with different physics Sharvin ρ = 1 µΩ-cm
22 Point Contact Spectroscopy
Phonon spectra Phonon I+ Spin transfer torque effect V+ Contact size: a ∼ nm I- Current: I = 1 mA 10 2 V- Current density: j = 10 A/cm Point Contact Andreev Reflection a Spin polarization Superconducting gap Thermal transport ballistic, diffusive …
23 Spin Transfer Torque Effect
Contact size: a ∼∼∼ nm Current: I = 1 mA Current density: j = 10 10 A/cm 2
T. Y. Chen, Y. Ji and C. L. Chien, Appl. Phys. Lett. 84 , 380 (2004). T. Y. Chen, Y. Ji, C. L. Chien and M. D. Stiles, Phys. Rev. Lett. 93 , 026601 (2004). T. Y. Chen, S. X. Huang, C. L. Chien and M. D. Stiles, Phys. Rev. Lett. 96 , 207203 (2006).
24 Spin Polarization ( P)
N (E ) − N (E ) P = ↑ F ↓ F 0 + N(E F): density of states at Fermi energy N↑ (EF ) N↓ (EF )
P: crucial for spintronics
25 Point Contact Andreev Reflection ( PCAR )
I+ Tip: V+ Conducting, semiconducting, I- superconducting V-
a Sample: Conducting, semiconducting, superconducting
Normal metal/superconductor interface : Andreev Reflection
Measures spin polarization (P) and superconducting gap (∆∆∆)
26 Determination of Spin Polarization by PCAR Normal metal /SuperconductorHalf metal /Superconductor Any metal/ Superconductor
-∆∆∆ +∆∆∆
-∆∆∆ +∆∆∆ -∆∆∆ +∆∆∆ 0 < P < 1
G(0)/ Gn = 2(1-P) G(0)/ G = 2 P = 0G(0)/ G = 0 P = 1 P of metal,n and ∆∆∆of SC cann be determined.
Ideal contact, clean limit 27 Interfacial Scattering ( Z) and inelastic scattering ( ΓΓΓ)
Interfacial: Z suppresses in-gap conductance
Inelastic: ΓΓΓ affects conductance at gap
Thermal: T smears out conductance
Analyzing entire conductance
Blonder-Tinkham-Klapwijk (BTK) theory 28 Point Contact Andreev Reflection Spectroscopy
Using known SC (Nb, Pb) to determine P of metal • S. X. Huang, T. Y. Chen, and C. L. Chien, Appl. Phys. Lett. 92 , 242509 (2008). • T. Y. Chen, C. L. Chien and C. Petrovic, Appl. Phys. Lett. 91 , 142505 (2007). • C. Leighton, M. Manno, A. Cady, J. W. Freeland, L. Wang, K. Umemoto, R. M. Wentzcovitch, T. Y. Chen, C. L. Chien, P. L. Kuhns, M. J. R. Hoch, A. P. Reyes, W. G. Moulton, E. D. Dahlberg, J. Checkelsky and J. Eckert, J. Phys.: Condens. Matter 19 , 315219(2007) (invited review). • L. Wang, T. Y. Chen, C. L. Chien, and C. Leighton, Appl. Phys. Lett. 88 , 232509 (2006). • L. Wang, T. Y. Chen, C. L. Chien, J. G. Checkelsky, J. C. Eckert, E. D. Dahlberg, K. Umemoto, R. M. Wentzcovitch , and C. Leighton, Phys. Rev. B 73 , 144402 (2006). • L. Wang, K. Umemoto, R.M. Wentzcovitch, T. Y. Chen, C.L. Chien, J.G. Checkelsky, J.C. Eckert, E.D. Dahlberg and C. Leighton, Phys. Rev. Lett. 94 , 056602 (2005). • L. Wang, T. Y. Chen and C. Leighton, Phys. Rev. B 69 , 094412 (2004). • Lance Ritchie and Gang Xiao, Y. Ji, T. Y. Chen, C. L. Chien, Ming Zhang, Jinglan Chen, Zhuhong Liu, Guangheng Wu, and X. X. Zhang, Phys. Rev. B 68 , 104430 (2003).
Using known metal (Au, Co) to determine Gap of SC
Nb, MgB 2, SmFeAsO 0.85 F0.15 T. Y. Chen, Z. Tesanovic, R. H. Liu, X. H. Chen, and C. L. Chien, Nature 453 , 1224 (2008)
29 AR Spectroscopy Measurements of Gaps
One gap of Nb ( TC = 9.27 K) ∆ Nb = 1.42 meV
Two gaps of MgB 2 (TC = 36.5 K) ∆ L = 6.75 meV ∆ S = 2.75 meV
30 Superconducting SmFeAsO 1-xFx (x=0.15, 0.18, 0.30)
Gold tip I+ V+ I- V-
Fe-SCs
31 SmFeAsO 0.85 F0.15 (T C = 42 K) Small T=4.5K ∆ contact Same 2∆∆ for all G(V) Structure: ∼ isotropic Value: 2∆ = 13.34 ± 0.71 meV ∆ 2 /kBTC = 3.68 close to 3.53 (BCS s -wave)
Complications: o Extra features to ± 100 mV due to SC o Slanted background nothing to do with SC mismatch of Fermi energy o Emergence of ZBA due to SC Large contact o Small effect in 9 T field at 4.5K 32 Emergence of Zero Bias Anomaly ( ZBA )
• Emerges systematically • ZBA obscures gap • ZBA due to SC • H field has negligible effect
ZBA is a signature of d-wave pairing (???). ZBA depends strongly on B field, whereas gap does not (???). (A scheme to obtain gap structure by subtraction) 33 ZBAs in s-wave Superconductors
ZBA is a signature of d-wave pairing (???).
d-wave may give ZBA.
But, ZBA does not imply d-wave since s-wave also gives ZBA.
P. Xiong, G. Xiao, R. B. Laibowitz, Phys. Rev. Lett. 71 , 1907 (1993) 34 Neither Gap nor ZBA depends strongly on H
∆∆∆H = 0.5 T
Subtraction does not reveal the superconducting gap 35 Zero Bias Anomaly ( ZBA )
ZBA is a signature of d-wave pairing (???).
d-wave may give ZBA.
But, ZBA does not imply d-wave since s-wave also gives ZBA.
ZBA depends strongly on B field, whereas gap does not (???). (A scheme to obtain gap structure by subtraction)
Neither ZBA nor gap depends strongly on B field.
Subtraction method does not work.
36 A real gap is independent of contacts
SmFeAsO 0.70 F0.30 Ω every 0.3 Small R
Large R
37 SmFeAsO 0.82 F0.18 (T C = 46.5 K), SmFeAsO 0.70 F0.30 (T C = 53.75 K)
SmFeAsO 0.82 F0.18: TC = 46.5 K ∆ = 7.26 meV ∆ 2 /k BTC = 3.62
SmFeAsO 0.70 F0.30: TC = 53.75 K ∆ = 8.16 meV ∆ 2 /k BTC = 3.52
38 Superconducting Gap
Theory: s-wave This work: 68.3 SmFeAsO 0.85 F0.15 53.3 ∆ 2∆ )0( 2 = = 06.4 p-wave 62.3 SmFeAsO 0.82 F0.18 k T kBTc B C SmFeAsO F 28.4 d-wave 52.3 0.70 0.30
SmFeAsO 1-xFx
MgB 2
SmFeAsO 1-xFx
Rohlf, Modern Physics Fisher et al., Rev. Mod. Phys. 79, 000353 (2007) 39 Temperature-Dependence of Gap
every 0.8 K Double-peak (gap): decreasing splitting decreasing intensity
vanishes at T C = 42 K
Extra features (due to superconductivity)
vanishes at T C = 42 K
Asymmetrical background (due to normal state)
remains at T > T C
SmFeAsO 0.85 F0.15
T. Y. Chen, Z. Tesanovic, R. H. Liu, X. H. Chen, and C. L. Chien, Nature 453 , 1224 (2008) 40 BCS-Like Temperature Dependence
BCS s-wave
At each T: Gap value from analyzing data using modified BTK theory
41 Gaps from Other Measurements ARPES
NdFeAsO 0.9 F0.1 (0807.0815.Kondo) Fully gapped, nearly isotropic s-wave symmetry
Nearly isotropic, fully gapped
Penetration Depth
PrFeAsO 1-y (0806.3149. Hashimoto) NdFeAsO 0.9 F0.1 (0807.0876.Martin) A single, fully gapped nearly isotropic gap A single, fully gapped nearly isotropic gap 42 Pseudogaps ?
Pseudogap in both non-SC and SC cuprates
No pseudogaps found in SC samples
Non-SC samples ?
43 Characteristics of Pseudogaps
Pseudogap may induce conductance peaks .
Pseudogap should be non -hysteretic .
Pseudogap must be contact-independent .
44 Conductance Peaks in parent SmFeAsO
Conductance peaks “suggesting” pseudogaps
hysteretic
45 Temperature Dependence of Conductance Peaks
∆∆∆T ≈ 5 K
Vd
Structure disappears above 160 K
SmFeAsO with SDW at 160 K
Pseudogap for SDW ???
46 Systematic Contact-Dependent Conductance
0.14 Various contact resistance: ΩΩΩ ΩΩΩ R = 7.1 ΩΩΩ from 7 to 536
0.12 Various conductance peaks: 0.10 from 3 mV to 300 mV
0.08 dI/dV dI/dV 0.06 Strongly depend on 0.04 contact resistance
0.02
R = 536.4 ΩΩΩ NOT pseudogaps! 0.00 -1000 -500 0 500 1000 V (mV) 47 Pseudogaps?
Pseudogap may induce conductance peaks But, conductance peaks do not imply pseudogaps since resistance anomaly also gives conductance peaks . Pseudogap should be non-hysteretic. Spin density wave ( SDW ) results in hysteretic behavior in magnetic structure, crystalline structure, and resistance. Pseudogap must be contact-independent Contact-dependent peaks are not pseudogaps
Not pseudogaps, what are they ?
48 Conductance Peaks due to Ballistic Heating !
SDW
SmFeAsO with SDW at 160 K
Point contact on SmFeAsO Temperature dependence of “Pseudogap”
SmFeAsO with SDW transition at 160 K
2 ∼∼∼ V (TSDW - T)
Ballistic heating effect
NOT pseudogaps, but SDW transition!
50 Summary ∆∆∆ Superconducting gap ( ) of SmFeAsO 1-xFx (x = 0.15, 0.18, 0.30) : ♦Gap value ∆ 2 /k BTC = 3.68, 3.62, 3.52 close to 3.53 (BCS s-wave) ♦ Gap structure ≈ Isotropic single gap ♦ Gap temperature dependence BCS-like ♦ No pseudogaps
Perhaps due to interband S± mechanism
T. Y. Chen, Z. Tesanovic, R. H. Liu, X. H. Chen, and C. L. Chien, Nature 453 , 1224 (2008) 51 Acknowledgements
Prof. Chia-Ling Chien , Advisor, Johns Hopkins University Sunxiang Huang, JHU Dr. Mark Stiles , NIST Prof. Zlatko Tesanovic , Johns Hopkins University Prof. Xianhui Chen , University of Science and Technology of China Prof. Yi Ji , University of Delaware
52 Fe Superconductors
Large varieties of materials – 1111 : LaFeAsO, CeFeAsO, SmFeAsO, …
– 122 : BaFe 2As 2, CaFe 2As 2, … – 111 : LiFeAs, … – 11 : FeSe, FeTe, … – More to come…
Highest Tc~55K achieved in SmFeAsO 1-xFx (“single layer”) Metallic parent compound with common SDW feature Fully gapped , nearly s-wave symmetry FeAs plane can be doped with Co
Very different from Cuprates!
53 Saga of Superconductivity
Bronze age
Iron age
?
The saga of superconductivity continues… Stone age
54 Fermi Surface of Fe -SC
5 sheets: 2 electron sheets at MA 2 hole sheets at Γ 1 hole pocket at Z
55 Spin-Transfer Torque: Single Electron Scenario
θ θ ψ = cos |↑> +sin |↓> 2 2 FM torque ∼∼∼ sin θθθ θθθ e-
I
Slonczewski, JMMM 159 , L1 (1996) 56 Waintal et al ., PRB 62 , 12317 (2000) Switching via Spin-Transfer Torque
M < M M1 < M 2 1 2 fixed fixed
e - e -
favor parallel favor anti-parallel
low R high R I I
detect switching through GMR effect
57 Slonczewski, JMMM 159, L1 (1996); Waintal et al., PRB 62, 12317 (2000) Switching via Spin Transfer Torque (STT) Effect
∼∼∼ 7 2 jC 10 A/cm Nano-pillar, point contact
switching
GMR
T. Y. Chen, Y. Ji and C. L. Chien, Appl. Phys. Lett. 84 , 380 (2004). 58 STT in Magnetic Nanostructures
New form of STT effect Inverse of the domain wall MR
T. Y. Chen, Y. Ji, C. L. Chien and M. D. Stiles, Phys. Rev. Lett. 93 , 026601 (2004). 59 New Spin Structure Realized by STT
New spin structure, largest MR of over 400% T. Y. Chen, S. X. Huang, C. L. Chien and M. D. Stiles, Phys. Rev. Lett. 96 , 207203 (2006). 60 Cryogenic Dewar
Temperature control: 1.5 K to 400 K Magnetic field: 0 - 9 T Rotating magnetic field without mechanical motion Nitrogen jacket
Sensitive transport system Two lock -in from Stanford research 10-channel switching box Current source, nano-voltmeter
Budget: 150K 61 Purposes
Point contact spectroscopy Transport study of nanostructure Spin injection and detection studies
62 Sputtering System
Specific features: Tilt gun, co-sputtering Evaporator
63 Vibrating Sample Magnetometer
Thin film Study oxides for energy
64 Dependence of Contact Resistance
65 66 ZBA shape is different for different contacts
67 68 69 70 71 72 Various Contacts
Various contact, various peak values from 5 mV to 500 mV
A real pseudogap must be intrinsic independent of contact
73