Workshop on Principles and Design of Strongly Correlated Electronic Systems August 2010

Size: px

Start display at page:

Download "Workshop on Principles and Design of Strongly Correlated Electronic Systems August 2010"

Rosalyn Blake
5 years ago
Views:

1 Workshop on Principles and Design of Strongly Correlated Electronic Systems 2-13 August 2010 Low symmetry structures and hybridization: Key ingredients for finding new superconductors Pascoal G. PAGLIUSO GPOMS, Instituto de Fisica "Gleb Wataghin" UNICAMP. Campinas, SP Brazil

2 Workshop on Principles and Design of Strongly Correlated Electronic Systems Trieste - 08/10 Low symmetry structures and hybridization: key ingredients for finding new superconductors NFL T 0 PG AFM P c 0 P SC P* FL Pascoal G. Pagliuso Grupo de Propriedades Ópticas e Magnéticas dos Sólidos S (GPOMS) Instituto de FísicaF ``Gleb Wataghin'', UNICAMP Campinas -SP UC - Irvine CBPF

3 Area 8.5 millions km 2 Sea shore 8,500 km Population 176 millions inhab. São Paulo 15 millions inhab. CAMPINAS Rio de Janeiro ~ 6 millions inhab. Campinas ~ 100 km NW from SP km from Rio

4 Campinas City 1.5 million inhabitants

5 Universidade Estadual de Campinas UNICAMP ~4km 2-35,000 Students (40 years old)

6 Magnetic and Optics Facilities GPOMS PPMS-QD 9T, 0.3 K 14 T 50 mk Cp(T,H) M(T,H) ρ(t,h) e R H κ(t,h) 0 < P < 25 kbar. MPMS SQUID-QD 7T 1.8 < T < 800 K 0 < P < 12 kbar. RAMAN Lab 1.6<T<800K 0<H<7T.

7 10 K C X - R A Y 300 K V S M F U R N A C E S

8 ESR Spectrometers L (1.4 GHz); S (4.0 GHz); X (9.4 GHz) Q; (34.4 GHz) (400 G) (1.200 G) (3.200 G) ( G) Bruker-ELEXSYS E-500 VARIAN

9 CAMPINAS SYNCHROTRON NATIONAL LABORATORY EXAFS-XANES-XDR (structural and magnetic) ~ E = 8keV (ë ~1.5 ); D ~30 m

10 Outline I. Review of structurally related physical properties of HFS families Ce m M n In 3m+2 the role of CEF tetragonal symmetry layered structural II. CeRhIn 5 doped with La and Sn III. Cd-doped Ce 2 (Rh,Ir)In 8 IV. Possible relationship to FeAs-based, Yb-based HFS V. Implications in new materials search

12 Ce m M n In 3m+2n CeRh 1-x (Ir,Co) x In 5 [1,2] Ce 2 Rh 1-x Ir x In 8 [3,4,5] T N (c p ) T G (μsr) 2 AFM T * (K) AFM T C (ρ) 1 SC SC Co X SC 0.5 Rh 0.5 Ir? 0.5 Co 0.5 SC SG X -Ce 2 CoIn 8 T C 1.0 K γ 600 mj/k 2 mole Ce [1] Pagliuso et al,physicab,312, 129 (2002). [2] Pagliuso et al, Phys.Rev B, 64, (R) (2001). [3] E.N. Hering et al Physica B, 378, 423 (2006). [4] E.N. Hering, Physica B, 403, 780 (2008); [5] E. N. Hering, C. Adriano, et al, submited to PRB (2010).

Ce 2 (Rh,Ir) 1-x Co x In 8 Phase Diagrams 4 CeRh 1-x Ir x In 5 3 AFM T(K) 2 1 0 CeRhIn 5 0.5 X SC?

13 Ce 2 (Rh,Ir) 1-x Co x In 8 Phase Diagrams 4 CeRh 1-x Ir x In 5 3 AFM T(K) CeRhIn X SC? CeIrIn 5 4 CeRh 1-x Ir x In kbar E. N Hering, et al M. Nicklas et al. T(K) 2 1 SC T c,χ T c,ρ SC CeRhIn 5 X CeIrIn 5

7 μ B (in plane) 107 spiral layer-to-layer Curro et al.

14 Magnetic Structure CeRhIn 5 Ce 2 RhIn 8 AFMwithT N =2.8K ε = (1/2, 1/2, 0) and η =52º μ eff =0.55μ B 52º T N =3.8K μ ord =0.7 μ B (in plane) 107 spiral layer-to-layer Curro et al., PRB 62, R6100 (2000) Bao et al., PRB 62, R14621 (2000) Ce Rh In W. Bao et al Phys. Rev. B, R (2001)

15 CeMIn 5 - χ(t) summary 20x10-3 c-axis is the magnetic easy axis CeRhIn 5 H a; H c M/H (emu/mol) CeIrIn 5 H a; CeCoIn 5 H a; H c H c T(K) C. Petrovic et al., J. Phys.: Cond Mat. 13 (2001) L337

16 c/a as control parameter for T c 2.5 CeRhIn 5 (~ 16 kbar) ? CeCo 0.5 Ir 0.5 In 5 CeCo 0.5 Rh 0.5 In 5 CeCoIn 5 Is there an intrisic parameter in this plot, which depends on c/a and determines T C? T c 1.0 CeRh 0.5 Ir 0.5 In 5 Tuning: 0.5 CeCo 0.5 Ir 0.5 In 5 CeRh 0.25 Ir 0.75 In 5 -J RKKY 0.0 CeIrIn T K -CEF c/a P.G. Pagliuso et al. PRB, (R) (2002); Physica B , 129 (2002).

22 Ground state competition Competition between Kondo screening and magnetic order Tune J with doping or pressure Often find superconductivity where T N 0: quantum critical point. S. Doniach, in Valence Instabilities and Related Narrow Band Phenomena, ed. R.D. Parks (Plenum, New York, 1977) p. 169.

23 CEF effects on T k CEF effects in Ce-M-In 5 : Effective T K in the presence of CEF: T eff J = 5 / 2 K = TK / Δ 2 CEF T K CeRhIn 5 CeIrIn 5 CeCoIn 5 CeIn mev 18.3 mev 16,4 mev ~20meV 6.9 mev 6.7 mev 8,6 mev H.-U. Desgranges, J. W. Rasul, PRB 36, 328 (1987). A.D. Christianson et al PRB, (2004).

24 CeMIn 5 (M = Rh, Ir, Co) Family H tetrag CEF = B2O2 + B4O4 + B6 O6 + B4O4 + B6 O6

25 c/a as control parameter for T c 2.5 CeRhIn 5 (~ 16 kbar) ? CeCo 0.5 Ir 0.5 In 5 CeCo 0.5 Rh 0.5 In 5 CeCoIn 5 Is there an intrisic parameter in this plot, which depends on c/a and determines T C? T c 1.0 CeRh 0.5 Ir 0.5 In 5 Tuning: 0.5 CeCo 0.5 Ir 0.5 In 5 CeRh 0.25 Ir 0.75 In 5 -J RKKY 0.0 CeIrIn T K -CEF c/a P.G. Pagliuso et al. PRB, (R) (2002); Physica B , 129 (2002).

26 Tuning of J RKKY? Gd 4f 7 S=7/2, L=0 Pagliuso et al. PRB. 62, (2000); PRB (2001).

28 Tuning the CEF? Cp(T) and S(T) for Nd-based compounds 4000 S(Rln2) Amara et.al NdIn 3 Nd 2 RhIn 8 NdRhIn 5 Nd 2 IrIn 8 NdIrIn 5 S - decreasing C/T (mj/mol Nd K 2 ) T N for NdIn 3 NdIrIn 5 Nd 2 IrIn 8 NdRhIn 5 Nd 2 RhIn 8 T N increasing Temperature (K) Temperature (K) The larger Γ 8 CEF Splitting larger T N P.G. Pagliuso et al. Physical Review B. 62, (2000).

29 T N evolution for R m A n In 3m+2n (m = 1,2 ; n =0,1) (R = Nd, Tb, Gd, Sm) R-A-In T N /T N (1-0-3) 2 1 θ p /θ p (1-0-3) NdIr NdRh GdIr GdRh CeIr CeRh TbRh Pagliuso et al. PRB. 62, (2000); PRB (2001).

31 The model: Magnetic Interactions and Crystal Field Effects- KJ J Simplest Magnetic Interaction: (mimics RKKY interaction) i j Crystal Field (CFE) Interaction: B 20 O 20 B 40 O 40 B 44 O 44 Can this simple model account for change in direction of the order moment along the series and for the T N behavior? D. Garcia, E. Miranda, et al. JAP (2006) K

32 Cristal Field Effects: B 20 O 20 B 40 O 40 B 44 O 44 Results for (B44=5B40=0.25meV) θ

33 Interaction Effects on the order temperature T N Without CF: isotropic K CEF enhanced order CEF frustrated order each point is obtained for specific B20,B40 and B44 parameters Diminished low-t c-susceptibility Enhanced low-t c-susceptibility Isotropic susceptibility

34 R n MIn 3n+2 Magnetic Structures CeRhIn 5 NdRhIn 5

35 T N evolution for R m A n In 3m+2n (m = 1,2 ; n =0,1) R-A-In RE moments ordered along c-axis. T N /T N (1-0-3) 2 1 θ p /θ p (1-0-3) NdIr NdRh GdIr GdRh CeIr CeRh TbRh RE moments in the ab-plane. P.G. Pagliuso et al. Physical Review B. 62, (2000).

36 Extrapolation of CEF trend to Ce-M-In T K J eff (111) Γ 8 CeIn 3 Γ 7 ' T K Γ 6 Γ 7 isotropic Γ 7 ' g // >> gper T N decreases CeRhIn 5 CeIrIn 5 CeCoIn 5 CeIn mev 18.3 mev 16,4 mev ~20meV 6.9 mev 6.7 mev 8,6 mev A.D. Christianson et al PRB (2004).

37 INS and XAS CEF study in Ce-M-In T. Willers et al. A. Severing I14 SCES 2010

39 Extrapolation of CEF trend to Ce-based materials T RKKY T K (increasing of g-anisotropy) T N Strong magnetic-fluctuations?? Frustrated local moments + hybridization

c/a as control parameter for T c 2.5 CeRhIn 5 (~ 16 kbar) 2.0 1.5? CeCo 0.5 Ir 0.5 In 5 CeCo 0.5 Rh 0.

40 c/a as control parameter for T c 2.5 CeRhIn 5 (~ 16 kbar) ? CeCo 0.5 Ir 0.5 In 5 CeCo 0.5 Rh 0.5 In 5 CeCoIn 5 Is there an intrisic parameter in this plot, which depends on c/a and determines T C? T c 1.0 CeRh 0.5 Ir 0.5 In 5 Tuning: 0.5 CeCo 0.5 Ir 0.5 In 5 CeRh 0.25 Ir 0.75 In 5 -J RKKY 0.0 CeIrIn T K -CEF c/a P.G. Pagliuso et al. PRB, (R) (2002); Physica B , 129 (2002).

42 La- and Sn-doped CeRhIn 5 We have carefully chosen two specific compositions, Ce 0.90 La 0.10 RhIn 5 and CeRhIn 4.84 Sn (a) (b) H ab C/T (J/mol-K 2 ) χ (10-3 emu/mol-ce) CeRhIn 5 Ce 0.90 La 0.10 RhIn CeRhIn 4.84 Sn T (K) T (K) The ordering temperature T N = 3.8 K of undoped CeRhIn 5 was suppressed to the same T N 2.8 K for both La and Sn doping in the studied concentrations. L. Mendonça-Ferreira et al. PRL 101, (2008)

43 La- and Sn-doped CeRhIn 5 AC electrical resistivity under pressure 50 CeRhIn 4.84 Sn Ce 0.90 La 0.10 RhIn 5 T N ρ (μω.cm) P (kbar) T c T (K) T N ρ (μω.cm) 10 5 P (kbar) T c T (K) ρ (μω.cm) H 12 kbar T (K) H (10 koe) kbar kbar T (K) H // i 15.2 kbar ρ (μω.cm) kbar H // i 8 H H // i T (K) H (10 koe) kbar kbar T (K)

44 La- and Sn-doped CeRhIn 5 T* (K) P T phase diagram (a) (b) AFM Tmax /Tmin SC P (kbar) x=0 x = 0.16 (Sn) x=0.10(la) For the two doped samples, T max, T N and T c follow the same qualitative evolution as a function of pressure found for pure CeRhIn 5. However, by comparing the P T phase diagram of the three samples, one sees that the La-doping shifts the P T phase diagram of CeRhIn 5 to higher pressure while the Sndoping does exactly the opposite. T * Norm (K) 2 AFM 1 SC P (kbar)

45 x =0:P* = P x =0.16(Sn):P* = P +5kbar x =0.10(La):P* = P -2kbar 1.2 As the two doped samples have the same T N at ambient pressure and similar strength of the Ce-Ce inter-site magnetic coupling, it is the strength of the intra-site Kondo-coupling which mainly determines the pressure evolution of the CeRhIn 5 ground state. T cnorm (K) P* (kbar) The suppression of the magnetism has to be associated with an increasing of T K and a consequent crossover between localized and itinerant behavior of the Ce 4f. If a given control parameter tunes T N to zero by a local mechanism, such as dilution or magnetic frustration which are not necessarily associated with an increase of the Kondo-coupling, this tuning might in fact inhibit SC: SC would not be expected at ambient pressure in Ce 1-x La x RhIn 5, only at much higher P. However, SC was not found at ambient pressure for CeRhIn 5 x Sn x. Disorder effects: T cmax = 1.2 K for CeRhIn4.84Sn0.16 and T cmax = 1.6 K for Ce 0.90 La 0.10 RhIn 5 ) is roughly a factor of two smaller than that for CeRhIn 5 (T cmax =2.3K). L. Mendonça-Ferreira et al. PRL 101, (2008)

46 Cd doping in CeMIn 5 in block doping Cd has one p-electron less. Hole doping Fermi surface tuning L. D. Pham et al. PRL 97, (2006).

47 Ce 2 MIn 8-x Cd x phase diagram Specif-heat divided by temperature and phase diagram for Cd-doped samples C mag /T(mJ/mol K 2 ) C mag /T(mJ/mol K 2 ) x of Cd: (a) (b) Ce 2 RhIn 8-x Cd x x of Cd: Ce 2 IrIn 8-x Cd x T (K) C. Adriano et al, PRB 81, (2010) T N (K) 4 2 Ce 2 MIn 8-x Cd x M=Rh M=Rh 0.5 Ir 0.5 M=Ir Cd increases xt N for the 3 compounds; Cd induces long range order for M=Ir;

48 Resistivity results of Ce 2 MIn 8-x Cd x ρ(μω cm) ρ(μω cm) T MAX (K) Ce 2 RhIn 8-x Cd x T MAX T(K) Ce 2 IrIn 8-x Cd x x = T MAX x = 0 T MAX shows an increase not T(K) Ce 2 MIn 8-x Cd x M=Rh M=Ir x C. Adriano et al, PRB 81, (2010) T N obtained from ρ(t) is in good agreement with specific heat measurements; consistent with the decrease of the Kondo coupling.

51 Ce 2 RhIn 8-x Cd x under pressure Mag. Moment direction σ(q) (mb) (a) (b) (c) Ce 2 RhIn 7.79 Cd 0.21 P = 0 kbar S =0.90μ B η =47 o P=3kbar S =0.85μ B η =20 o P = 6 kbar S =0.80μ B η =0 o (½½l )(r.l.u) Moment direction, η (degree) η ab-plane P (kbar) No SC up to ~ 23 kbar!!

52 CeMIn 5 (M = Rh, Ir, Co) Family H tetrag CEF = B2O2 + B4O4 + B6 O6 + B4O4 + B6 O6

Extrapolation to â-ybalb 4 IV larget SF 20x10-3 CeRhIn 5 H a; H c M/H (emu/mol) 15 10 CeIrIn 5 H a; CeCoIn 5 H a; H c H c 5 0 0 50

54 Extrapolation to â-ybalb 4 IV larget SF 20x10-3 CeRhIn 5 H a; H c M/H (emu/mol) CeIrIn 5 H a; CeCoIn 5 H a; H c H c T(K) Nakatsuji S. et al. Nature 4, (2008). Andriy H. Nevidomskyy* and P. Coleman, PRL 102, (2009)

55 ESR g-anisotropy This remarkable and unprecedented ESR signal found in the HFS â-ybalb4 behaves as a CESR at high temperatures and acquires characteristics of Yb3+ LM ESR at low temperature. This dual behavior in same ESR spectra strikes as an in situ unique observation of the Kondo localization of the f-electron at the quantum critical point (QCP) of â- YbAlB4. cond-mat

56 PuCoGa 5 FeAs-intermetallics? M. S. Torikachvilli et al. PRB Marianne Rotter al. New J. of Phys. (2009).

Implications for designing materials ToincreaseT C for magnetically mediated SC: Tsf, characteristic T for spin fluctuations high Tunable carrier density Quasi 2-d helps a lot;

57 Implications for designing materials ToincreaseT C for magnetically mediated SC: Tsf, characteristic T for spin fluctuations high Tunable carrier density Quasi 2-d helps a lot; maybe not much 2-d needed increase c/a tune CEF GS Need bigger bandwidths to increase T sf, but keep optimized T c /T sf - 3d spins in the active layer e , etc

58 Generalized c/a plot 160 Hg-Pb-1:2:2:1:2 Superconducting Transition T C (K) MgB 2 CuO-based YBCO FeAs-based Pu1-1-5 YNi 2 B 2 C Ce ,8 1,0 1,2 1,4 1,6 1,8 2,0 2,2 2,4 2,6 2,8 3,0 3,2 3,4 3,6 3,8 c/a

59 Implications for designing materials 1-1-5: c/a ~ : c/a ~ 3.0 3d spins as the active layer. Examples: A 2 MB 8 M = Cu, Fe, Co, Ni, Mn,Ru,Re,Mo A=La,Y,Ca,Sr,Ba, Mg, K B = Bi, Sb, Ge, Sn, In, As, etc e , etc

60 Implications for designing materials c/a ~ dspinsasthe active layer. Examples: AM 2 B 2 M = Cu, Fe, Co, Ni, Mn,Ru,Re,Mo A=La,Y,Ca,Sr,Ba, Mg,K,Li,Mg B = Bi, Sb, Ge, Sn, In, etc e , etc

61 Implications for designing materials 2-1-4: c/a ~ 3/4-3d spins astheactivelayer. Examples: A 2 MB 4 M = Cu, Fe, Co, Ni, Mn,Ru,Re,Mo A=La,Y,Ca,Sr,Ba, Mg, K B = Bi, Sb, Ge, Sn, In, Pb, etc e , etc

62 Thank you for your attention!

ESR signal in â-ybalb 4 : cond-mat 0908.0044 Abs. Der. (arb. units) T=4.2K 9.5 GHz Zn-flux In-flux In-flux YbRh 2 Si 2 Sichelschmidt et al. PRL91, 156401. 0 1 2 3 4 H (koe) Duque et al.

63 ESR signal in â-ybalb 4 : cond-mat Abs. Der. (arb. units) T=4.2K 9.5 GHz Zn-flux In-flux In-flux YbRh 2 Si 2 Sichelschmidt et al. PRL91, H (koe) Duque et al. PRB (2009) Abrahams, E., Wolfle,P.Phys. Rev. B (2008). Schlottmann, P.Phys. Rev. B (2009). Kochelaev, B. I. arxiv: (2009). 171 A ~ 1300 Oe which is of the order of the typical values found for 171 Yb in low symmetry systems. Irrefutable indication that the ESR spectra found for â-ybalb4 acquires the characteristic of the Yb 3+ ions at low-t.

64 ESR temperature dependence Both phases show T-independent ESR intensity typical of CESR g-value for the â-phase shifts to larger value at low-t (g ~ 3): typical for Yb 3+ Krammers doublets) For â-ybalb4 ΔH(T) shows a weak nonmonotonic increase as a function of temperature which results in an average linewidth broadening of ~ 0.4 Oe/K in the whole temperature range. This rate is much smaller than the linear Korringa-rate found for the ESR line observed at low-t for YbRh2Si2. In contrast, for α-ybalb4 ΔH(T) increases dramatically with decreasing-t which avoid the observation of the resonance for T <40K.

65 g-anisotropy This remarkable and unprecedented ESR signal found in the HFS â-ybalb4 behaves as a CESR at high temperatures and acquires characteristics of Yb3+ LM ESR at low temperature. This dual behavior in same ESR spectra strikes as an in situ unique observation of the Kondo localization of the f-electron at the quantum critical point (QCP) of â- YbAlB4.

Phase diagrams of pressure-tuned Heavy Fermion Systems

Phase diagrams of pressure-tuned Heavy Fermion Systems G. Knebel, D. Aoki, R. Boursier, D. Braithwaite, J. Derr, Y. Haga, E. Hassinger, G. Lapertot, M.-A. Méasson, P.G. Niklowitz, A. Pourret, B. Salce,