ebook img

Upper critical field H_c2 in Bechgaard salts (TMTSF)_2PF_6 PDF

0.18 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Upper critical field H_c2 in Bechgaard salts (TMTSF)_2PF_6

Europhysics Letters PREPRINT Upper critical field H in Bechgaard salts (TMTSF) PF 6 c2 2 6 0 0 Ana Dom´ınguez Folgueras1,2,3 and Kazumi Maki3 2 1 Departamento de F´ısica, Universidad de Oviedo, 33007 Oviedo, Spain n a 2 Instituto de Ciencia de Materiales de Madrid, CSIC, Cantoblanco, 28049 Madrid, J Spain 4 3 Departmentof Physics &Astronomy, University of SouthernCalifornia, Los Angeles, CA 90089-0484, USA ] n o c PACS.74.70.Kn – Organic Superconductors. - r PACS.74.20.Rp – Pairing symmetries. p PACS.74.25.Op – Mixed states, critical fields, and surface sheaths. u s . t a m Abstract. – ThesymmetryofthesuperconductingorderparameterinBechgaardsaltsisstill - unknown, though the triplet pairing is well established by NMR data and large upper critical d field Hc2(0) ∼ 5 Tesla for H~ k a and H~ k b′. Here we examine the upper critical field of a n few candidate superconductors within the standard formalism. The present analysis suggests o strongly chiral f-wave superconductor somewhat similar to the one in Sr RuO is the most c 2 4 [ likely candidate. 1 v 5 6 Introduction. – The Bechgaard salts (TMTSF) PF is the first organic superconductor 0 2 6 1 discovered in 1980 [1]. For a long time the superconductivity is believed to be conventional 0 s-wave [2]. Recently the symmetry of the superconducting energy gap becomes the central 6 issue [3,4]. The upper critical field at T = 0K, H 5 Tesla for both H~ a and H~ b c2 ′ 0 ∼ k k for both (TMTSF) PF [5] and (TMTSF) ClO [6] are clearly beyond the Pauli limit [7,8] / 2 6 2 4 t indicating the triplet pairing. More recently the NMR data from (TMTSF) PF [9]indicates a 2 6 m clearly the triplet pairing. Therefore the candidates for the superconductivity in Bechgaard salts are more likely within p-wave and f-wave superconductors. In the following we shall ex- - d aminetheuppercriticalfieldofthesesuperconductorsfollowingthestandardmethodinitiated n by Gor’kov [10] and extended by Luk’yanchuk and Mineev [11] for unconventional supercon- o ductors. Also we take the quasiparticle energy in the normal state as in the standard model c : for Bechgaardsalts [2] v i X ξ(k)=v(ka kF) 2tbcosbk 2tccosck (1) | |− − − r a with v :v :v 1 :1/10:1/300 and v = √2t b and v = √2t c. There are earlier analysis b c b b c c ∼ of H of Bechgaard salts starting from the one dimensional models [12,13]. However, those c2 models predict diverging H (T) for T 0K or the reentrance behaviour, which have not c2 → beenobservedinthe experiments[5,6]. Also,thequasilinearTdependence ofH (T)inboth c2 (TMTSF) PF and (TMTSF) ClO is very unusual. Among the models we have considered, 2 6 2 4 (cid:13)c EDPSciences 2 EUROPHYSICSLETTERS Fig. 1 – |∆(~k)| of chiral f-wave and chiral f′-waveSC are sketched in a) and b) respectively. the chiral f-wave superconductor with ∆ ~k 1 sgn(k )+ı sinχ cosχ , looks most ′ (cid:16) (cid:17) ∼ (cid:16)√2 a 2(cid:17) 2 promising, where χ =~b~k and χ =~c~k where~b and~c are crystal vectors. 1 2 Also if the superconductor belongs to one of the nodal superconductors [14] and if nodes layparalleltok~ within the twosheets ofthe Fermisurface,the angledependent nuclearspin c relaxation rate T1−1 in a magnetic field rotated within the b′ −c∗ plane will tell the nodal directions. Beforeproceeding,weshow ∆(~k) ostwochiralf-wavesuperconductorsinFig.1a)andb). | | 1 1 where ∆(k) (1+cos2χ )(1 1cos2χ ) 2 and ∆(k) (1+cos2χ )(1 1cos2χ ) 2 | | ∼ 1 − 2 2 | | ∼ 2 − 2 2 for chiral f1 and(cid:2)chiral f2 respectively. (cid:3) (cid:2) (cid:3) Upper critical field for H~ b~. – Inthefollowingweneglectthe spincomponentof∆~(~k). k ′ Most likely the equal spin pairing is realised in Bechgaard salts as in Sr RuO [4]. In this 2 4 case the spin component is characterised by a unit vector dˆ. Also dˆis most likely oriented parallel to c~. Let’s assume dˆ c~, though H (T) is independent of dˆas long as the spin ∗ k ∗ c2 orbitinteractionis negligible. Experimentaldata frombothUPt andSr RuO indicate that 3 2 4 the spin-orbit interactions in these systems are not negligible but extremely small [15]. We consider a variety of triplet superconductors (see Fig. 1): A. Simple p-wave SC: ∆~(k) sgn(k ). Following [16] the upper critical field is deter- a ∼ mined by ∞ du lnt= (1 K ) (2) 1 − Z sinhu − 0 ∞ du C lnt= (C K ) (3) 2 − Z sinhu − 0 where K1 = e−ρu2|s|2 1+2Cρ2u4s4 (4) h i (cid:0) (cid:1) AnaDom´ınguezFolguerasandKazumiMaki:UppercriticalfieldHc2 inBechgaardsalts(TMTSF)2PF63 0.4 0 0.35 cchhiirraall fp’--wwaavvee -0.05 cph-wiraalv ep-wave pwave 0.3 -0.1 0.25 -0.15 ρ0 0.2 C-0.2 0.15 -0.25 0.1 -0.3 0.05 -0.35 00 0.2 0.4 0.6 0.8 1 -0.40 0.2 0.4 0.6 0.8 1 a) t b) t Fig.2–NormalisedHc2(t)andC(t)forH~ k~b′ areshownina)andb)respectively. Heresolid,dashed and dotted lines are chiral f′-wave,chiral p-waveand simple p-waverespectively. 1 16 2 K = e ρu2s2 ρ2u4s 4+C 1 8ρu2 s2+12ρ2u4 s4 ρ3u6 s6+ ρ4u8 s8 (5) 2 − | | ∗ h (cid:18)6 (cid:18) − | | | | − 3 | | 3 | | (cid:19)(cid:19)i and t = T , ρ = vavbeHc2(T), s = 1 sgn(k )+ısinχ , χ =~c~k and ... means average over Tc 2(2πT)2 √2 a 2 2 h i χ . Here v , v are the Fermi velocities parallel to the a axis and the c axis respectively. 2 a c Here we assumed that ∆(~r) is given by [16] ∆(~r) 1+C(a+)4 (6) ∼ i (cid:0) (cid:1) wherei= Cne−eBx2−nk(x+ız)−(n4ekB)2 istheAbrikosovstate[17]anda+ = √21eB (−ı∂z−∂x+2ıeHz) is the raising oPperator. Then in the vicinity of t 1 we find ρ = 2 ( lnt) = 0.237697( lnt) and C = → 7ζ(3) − − 93ζ(5) ρ. −647ζ(3) For t 0 on the other hand we obtain → v v eH (0) ρ = limρt2 = a c c2 =0.1583 (7) 0 t 0 2(2πT )2 → c and C = 0.031. From these we obtain − H (0) c2 h(0)= =0.6659 (8) ∂Hc2(t) ∂t |t=1 Both ρ (t) and C(t) are evaluatednumerically and shownin Fig.2 a)and b) respectively. 0 Here ρ (t)=t2ρ(t)=vv eH (t)/2(2πT )2. 0 c c2 c B.Chiralp-waveSC:∆~(k)=1/√2sgn(k )+isin(χ ). Here 1 sgn(k )+isin(χ )isthe a 2 √2 a 2 analogue of eıφ in the 3D systems in the quasi 1D system. For a chiral state the Abrikosov function is written as [18] ∆(~r,~k) (s+Cs∗(a†)2) (9) ∼ i 4 EUROPHYSICSLETTERS where s= 1 sgn(k )+ısin(χ ). Then we obtained eq. 3 with √2 a 2 K1 = e−ρu2|s|2 s2 2C s4 (10) h | | − | | i K = e ρu2s2 s4+C s2 1 4ρu2 s2(cid:0)+2ρ2u4 s4 (cid:1) (11) 2 − | | h −| | | | − | | | | i (cid:0) (cid:0) (cid:1)(cid:1) and the same expressions for t, ρ,... For t 1 we find C =1 √1.5= 0.2247 and ρ=0.3838( lnt). → − − − On the other hand, for t 0 we obtain C = 0.3660 and ρ =0.27343. 0 → − From these we obtain h(0) = 0.71324. We obtain ρ(t) and C(t) numerically. They are shown in Fig. 2 a) and b) respectively. C. Chiral f-wave SC: ∆ˆ(k) dˆscosχ . H (t) is determined from eq. 3 where now: 1 c2 ∼ K = (1+cos2χ )e ρu2s2 s2 2ρu2 s4 (12) 1 1 − | | h | | − | | i K = (1+cos2χ )e ρu2s2 ρu2 s4+C s2 1 4ρu2 s(cid:0)2+2ρ2u4 s4 (cid:1) (13) 2 1 − | | h − | | | | − | | | | i (cid:0) (cid:0) (cid:1)(cid:1) Here now ... means the average over both χ and χ . Then it is easy to see that the 1 2 h i chiral f-wave SC has the same H (t) and C(t) as the chiral p-wave SC, since the variable χ c2 1 is readily integrated out. D. Chiral f-wave SC: ∆ˆ(k) dˆscosχ . Now we have a set of equations similar to the ′ 2 ∼ chiral f-wave except 1+cos2χ in both eqs. 13 has to be replaced by 4(1+cos2χ ). Then 1 3 1 we obtain for t 1 C = 0.2247 and ρ = 0.5181( lnt). On the other hand, for t 0 we → − − → find C = 0.3660 and ρ =0.3734. 0 − We show ρ and C(t) of the chiral f-wave in Fig.2 a) and b) respectively. 0 ′ Note that C(t) is the same for three chiral states (chiral p-wave, chiral f-wave and chiral f-wave) as well as chiral p-wave studied in [18] ′ Therefore for the magnetic field H~ b, the chiral f-wave have the largest H (t) if we ′ ′ c2 k assume T and v, v are the same. Also H (t) of these states are closest to the observation. c c c2 Upper critical field for H~ ~a. – k A. Simple p-wave SC: ∆ ~k = sgn(k ). The equation for H (t) is given by [16] and a c2 (cid:16) (cid:17) can be written as in eq.3 with K1 = e−ρu2|s|2 1+2Cρ2u4 s4 (14) h | | i 16 (cid:0) 2 (cid:1) K = e ρu2s2 ρ2u4 s4+C 1 8ρu2 s2+12ρ2u4 s4 ρ3u6 s6+ ρ4u8 s8 (15) 2 − | | h (cid:18) | | (cid:18) − | | | | − 3 | | 3 | | (cid:19)(cid:19)i where t= T , ρ= vavbeHc2(t) and s=sinχ +ısinχ with χ =~b~k and χ =~c~k. Tc 2(2πT)2 1 2 1 2 Then for t 1, we find C = 93ζ(5) ρ and ρ = 2 ( lnt) = 0.2377( lnt). While for → −508ζ(3) 7ζ(3) − − 2 t 0 C = 3 3 + 1 = 0.0170129 and ρ = vavbeHc2(0) = 1 exp[α +2Cβ ] = → 2β0 −r(cid:16)2β0(cid:17) 12 − 0 2(2πTc)2 4γ 0 0 0.1751209, where α = ln s2 = 0.220051 and β = s4 = 4 1 = 0.0170. From these 0 −h | | i 0 −h s4 π − | | we obtain h(0)=0.73673. Both h(t) and C(t) are evaluated numerically and we show them in Fig. 3 a) and b) respectively. AnaDom´ınguezFolguerasandKazumiMaki:UppercriticalfieldHc2 inBechgaardsalts(TMTSF)2PF65 0.25 0 p-wave chiral p-wave 0.2 chiral f-wave -0.025 chiral f’-wave 0.15 -0.05 ρ0 C 0.1 -0.075 0.05 -0.1 p-wave chiral p-wave chiral f-wave chiral f’-wave 0 -0.125 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 a) t b) t Fig. 3 – Normalised Hc2(t) and C(t) for H~ k ~a are shown in a) and b) respectively. Here solid, dashed, dashed-dotted and dotted lines are chiral f′-wave, chiral f-wave, chiral p-wave and simple p-waverespectively. B. Chiral p-wave SC: ∆(k) 1 sgn(k )+ısinχ . Now H (t) is determined by ∼ (cid:16)√2 a 2(cid:17) c2 a similar set of equations as in sec. 1.B. In particular we find for t 1 C = 0.027735 → − and ρ = 0.212598(lnt) while for t 0 C = 0.067684 and ρ = 0.139672. We obtain 0 → − h(0)=0.6566. We show h(t) and C(t) in Fig. 3 a) and b) respectively. C. Chiral f-wave SC: ∆ˆ(k) dˆscosχ . Again we use a similar set of equations as those 1 ∼ discussed in sec. 1.C, we find for t 1 C = 0.0356236 and ρ = 0.2744495(lnt) while for → − t 0 C = 0.066 and ρ = 0.1920 and h(0) = 0.6997. Both h(t) and C(t) are evaluated 0 → numerically and shown in Fig. 3 a) and b). D. Chiral f-wave SC: ∆ˆ(k) dˆscosχ . Now we find for t 1 C = 0.05 and ρ = ′ 2 ∼ → − 0.2910(lnt), while for t 0 C = 0.1019 and ρ =0.2090. 0 − → − We have shown again h(t) and C(t) in Fig. 3 a) and b) respectively. Comparing these results with H (T) from (TMTSF) PF and (TMTSF) ClO [4,5], we c2 2 6 2 4 canconcludebothH~ b~ andH~ ~athechiralf-waveSCismostconsistentwithexperimental k ′ k ′ data. In particular these states have relatively large h(0) (see Table I).On the other hand almost the same H (0) for H~ b~ and H~ ~a has to be still accounted. c2 k ′ k Nodal lines in ∆(~k). – We have seen that from the temperature dependence of H (T), c2 we deduce the chiral f-wave and chiral f are the most favourable. They have nodal lines on ′ the Fermi surface (i.e. the χ χ plane), the chiral f-wave SC at χ = π, while chiral f-wave SC at χ = π. 1 − 2 1 ±2 ′ 2 ±2 These nodal lines may be detected if the nuclear spin relaxation rate is measured in a magnetic field rotated within the b c plane. ′ ∗ − Following the standard procedure given in [14] the quasiparticle density of states in the vortex state for T <<T and E =0 is given by c N 0,H~ = 2 v2√eH 1+cosθ2sinχ 2 21 (16) π2 10 (cid:16) (cid:17) (cid:0) (cid:1) 6 EUROPHYSICSLETTERS Table I – Summary of results. Here ρ (0)= vˆ2eHc2(0) and h(0)= Hc2(0) 0 2(2πTc)2 ∂H∂c2t(t)|t=1 symmetry C(0) C(1) −∂∂ρt|t =1 ρ0(0) h(0) p-wave -0.031 0 0.2377 0.1583 0.6659 H~ kb′ chiral p-wave -0.2247 -0.3660 0.3838 0.2734 0.71324 chiral f′-wave -0.2247 -0.3660 0.5181 0.3734 0.72073 p-wave -0.017 0 0.2377 0.1751 0,7366 H~ ka chiral p-wave -0.066 -0.028 0.2126 0.1396 0,6566 chiral f-wave -0.066 -0.035 0.2744 0.1920 0.6997 chiral f′-wave -0.1019 -0.05 0.2910 0.2090 0,7182 where χ is the position of the nodal line on the χ axis. So for the chiral f-wave SC we 10 10 findχ = π andN 0,H~ exhibits the simple angulardependence. Onthe other handwhen 10 2 (cid:16) (cid:17) nodal lines are on the χ axis, the θ dependence will be too small to see. Finally this gives 2 2 2 T1−1(cid:16)H~(cid:17)/T1−N1 =(cid:18)π2(cid:19) ~v2(eH)(cid:0)1+cosθ2(cid:1) (17) for the chiral f-wave SC. We show the θ dependence of T1−1 in Fig. for a few candidates. The chiralf-wave SC has thestrongestθ dependence(solidline)whilethechiralh-waveSC(dashedline)andthechiral p-wave SC (dotted line) have a similar θ dependence. Concluding remarks. – We have computes the upper critical field of Bechgaard salts for a variety of model superconductors with the standard microscopic theory. We find: a) Assuming all these superconductors have the same T , the chiral f-wave SC (∆~(k) c ′ ∼ 1 sgn(K )+ısinχ cosχ ) appears to be the most favourablewith largestH ’s for both √2 a 2 2 c2 (cid:16) (cid:17) H~ b and H~ a; b) however, non of these states exhibit the quasi T linear dependence of ′ k k Hc2(T)asobservedin [4]; c)Alsothe presenttheorypredictsHc2(0) (vvc)−1 and(vbvc)−1 ∼ for H~ b and H~ a respectively. This means H (0) for H~ a is about 5 time larger than ′ c2 k k k theoneforH~ b contrarytoobservation;d)fromH (0) 5TandT =1.5Kwecanextract ′ c2 c k ∼ V2 =√vvc 1.5 104cms−1, consistent with the known values of v, vc. ∼ ∗ ∗∗∗ WethankStuartBrownandPaulChaikinforusefuldiscussiononpossibledetectionofthe nodalstructureof∆(~k)sinBechgaardsaltsthroughNMR.Wehavebenefitedfromdiscussion with Stephan Haas and David Parker. ADF acknowledges gratefully the financial support of Ministeriode Educaci´onyCiencia(Spain)(AP2003-1383)andJaimeFerrerandPacoGuinea for useful discussion. REFERENCES [1] Jerome D., Mazard A., Ribault M. and Bechgaard K., J. Phys (France) Lett, 41 (1980) L95 [2] IshiguroT.,YamajiK.andSaitoG.,OrganicSuperconductors, Springer-Verlag(Berlin1999) [3] Sigrist M. and Ueda K., Rev. Mod. Phys., 63 (1991) 239 [4] Maki K., Haas S., Parker and Won H., Chinese J. Phys, 43 (2005) 532 AnaDom´ınguezFolguerasandKazumiMaki:UppercriticalfieldHc2 inBechgaardsalts(TMTSF)2PF67 2 1.8 1.6 1.4 1.2 1 0 0.5 1 1.5 θ Fig. 4 – The angle dependentnuclear spin relaxation rate for a few nodal superconductors is shown. (Chiral f-wave, chiral h-waveand chiral p-waveare represented in solid, dashed and dotted lines.) [5] Lee I.J., Chaikin P.M. and Naughton M.J., Phys. Rev. B, 63 (2002) R180502 [6] Oh J.I. and Naughton M.J., Phys. Rev. Lett., 92 (2004) 067001 [7] Clogston A.M., Phys. Rev. Lett., 9 (1967) 266 [8] Chandrasekhar B.S.,Appl. Phys. Lett., 1(1962) 7 [9] LeeI.J.,BrownS.E.,ClarkW.G.,StrouseM.J.,NaughtonM.J.,KangW.andChaikin P.M., Phys. Rev. Lett., 88 (2002) 017004 [10] Gor’kov L.P., Soviet Phys. JETP, 10 (1960) 59 [11] Luk’yanchuk I. and Mineev V.P., Soviet Phys. JETP, 66 (1987) 1168 [12] Lebed A.G.,JETP Lett, 44 (1986) 114 [13] Depuis N., Mantambaux G. and Sa de Melo C.A.R.,Phys. Rev. Lett., 70 (1993) 2613 [14] Won H., Haas S. Parker D., Telang S., Vanyolos A. and Maki K., Lectures on the Physics of Highly Correlated Electron Systems IX, AIP Conference Proceedings 789 (Melville 2005) [15] Maki K., Haas S., Parker D. and Won H.,cond-mat/0504635, (2005) [16] Won H. and Maki K., Europhys. Lett., 30 (1995) 421 Phys. Rev. B,53 (1996) 5927 [17] Abrikosov A.A., Soviet Phys. JETP, 5 (1957) 1174 [18] Wang G.F. and Maki K.,Europhys. Lett., 45 (1999) 71

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.