Anomalous superfluid density in quantum critical superconductors

Anomalous superfluid density in quantum
critical superconductors

K. Hashimoto * \ Y. Mizukami * , R. Katsumata * , H. Shishido * ^ M. Yamashita * ^ H. Ikeda * , Y. Matsuda * ,
J. A. Schlueter ^, J. D. Fletcher A. Carrington ' , D. Gnida *% D. Kaczorowski " , and T. Shibauchi *

* Department of Physics, Kyoto University, Kyoto 606-8502, Japan, ^Materials Science Division, Argonne National Laboratory, Argonne, Illinois 60439, U.S.A., I'H.
H. Wills Physics Laboratory, University of Bristol, Tyndall Avenue, Bristol, BS8 1TL, United Kingdom,* 'Institute of Low Temperature and Structure Research,
Polish Academy of Sciences, P. O. Box 1410, 50-950 Wroclaw, Poland, * Present address: Institute for Materials Research, Tohoku University, Sendai 980-8577,
Japan,* Present address: Department of Physics and Electronics, Osaka Prefecture University, Sakai, Osaka 599-8531, Japan, § Present address: RIKEN, Wako,
Saitama 351-0198, Japan, and ttjo whom correspondence should be addressed. Email: shibauchi@scphys.kyoto-u.ac.jp

Submitted to Proceedings of the National Academy of Sciences of the United States of America

When a second-order magnetic ptiase transition is tuned to zero
temperature by a non-thermal parameter, quantum fluctuations
are critically enhanced, often leading to the emergence of un-
conventional superconductivity. In these 'quantum critical' su-
perconductors it has been widely reported that the normal-state
properties above the superconducting transition temperature Tc
often exhibit anomalous non-Fermi liquid behaviors and enhanced
electron correlations. However, the effect of these strong critical
fluctuations on the superconducting condensate below Tc is less
well established. Here we report measurements of the magnetic
penetration depth in heavy-fermion, iron-pnictide, and organic su-
perconductors located close to antiferromagnetic quantum critical
points showing that the superfluid density in these nodal super-
conductors universally exhibit, unlike the expected T-linear de-
pendence, an anomalous 3/2 power-law temperature dependence
over a wide temperature range. We propose that this non-integer
power-law can be explained if a strong renormalization of effec-
tive Fermi velocity due to quantum fluctuations occurs only for
momenta k close to the nodes in the superconducting energy gap
A{k). We suggest that such 'nodal criticality' may have an impact
on low-energy properties of quantum critical superconductors.

superfluid stiffness | d-wave superconductivity | spin fluctuations | mass
enhancement | quasiparticle scattering

The physics of materials located close to a quantum critical point
(QCP) is an important issue because the critical fluctuations as-
sociated with this point may produce unconventional high tempera-
ture superconductivity |1. 2|. Quantum oscillations |j3]|4] and spe-
cific heat measurements |5 | have shown that, in some systems, as the
material is tuned towards the QCP by controlling an external param-
eter such as doping, pressure, or magnetic field, the effective mass
strongly increases due to enhanced correlation effects. Along with
this the temperature dependence of the resistivity shows a strong de-
viation from the standard AT^ dependence in the Fermi liquid (FL)
theory of metals, and often shows an anomalous T-linear behavior
which corresponds to the A coefficient diverging as zero temperature
is approached.

Although there are many studies of non-FL behavior in the nor-
mal metallic state 1 1 , 2 1, relatively little is known about how the QCP
affects the superconducting properties below the critical temperature
Tc. The superconducting dome often develops around the putative
QCP so that when the temperature is lowered below T^, the super-
conducting order parameter starts to develop and the Fermi surface
becomes gapped. It is therefore natural to consider that the low-
energy quantum critical fluctuations are quenched by the formation of
the superconducting gap A, which means that the system avoids the
anomalous singularities associated with the QCP. Perhaps because of
this reasoning the superconducting properties are usually analyzed by
the conventional theory without including temperature/field depen-
dent renormalization effects resulting from the proximity to the QCP.
For example, in Refs. |6 7| the NMR relaxation rate l/Ti in the su-
perconducting state is fitted to the temperature dependence expected

for particular gap functions with the assumption that the normal-state
1/TiT is virtually temperature independent below Tc even when it
has strong temperature dependence above Tc due to the magnetic
fluctuations.

In superconductors near the QCP, the electron pairing is of-
ten unconventional with a superconducting energy gap A(fc) which
changes sign on different parts of the Fermi surface ||8] |9]- This
sign change stems from a repulsive pairing interaction for example
resulting from antiferromagnetic spin fluctuations. In many cases,
this leads to the presence of nodes in the gap A(fe) where the gap
changes sign. The low-energy excitations from the ground state in
these superconductors are governed by these nodal regions. In such
nodal superconductors, the effect of quantum critical fluctuations on
the excited quasiparticles should be k dependent. Then the question
arises as to how this effect modifies the low-energy properties in the
superconducting state.

The penetration depth A(r) is a fundamental property of the su-
perconducting state which parameterizes the ability of a supercon-
ductor to screen an applied field by the diamagnetic response of the
superconducting condensate. As fermionic quasiparticles are ther-
mally excited from the condensate a paramagnetic cuiTent is created
which reduces the screening and increases A. So measurements of
A(r) give direct information about density and Fermi velocity of
these quasiparticles |I0|. When the effective mass is enhanced by
the quantum critical fluctuations, the effective Fermi velocity is ex-
pected to be suppressed accordingly. In a one-component Galilean
invariant superfluid, electron correlation effects may not cause the
renormalization in the low-temperature penetration depth |11|. In
superconducting materials, however, strong electron correlations do
affect the renormalization resulting in an enhanced penetration depth,
which has been reported both theoretically |12 13] and experimen-
tally 1 141 ITsi . To discuss the energy-dependent effect of quantum
criticality on superconducting quasiparticles, the temperature depen-
dence of penetration depth at low temperatures is thus of particular
importance.

Reserved for Publication Footnotes

www.pnas.org/cgi/dol/1 0.1 073/pnas.1 221 9761 1

PNAS I Issue Date | Volume | Issue Number | 1-9

Results

Penetration Depth. Here we begin by presenting results for the
iieavy-fermion system Ce„nna,i+2 which is located close to a QCP
(T is the transition metal element, n is the number of Celns layers
alternating with the Tln2 blocks along the c axis). The most studied
member of this series is the n — 1 member CeCoIns (T^ = 2.3 K)
IT6I . in which clear evidence for non-FL behaviors in the normal
state I17l|18 !| and nodal superconductivity has been found ||6lll9ll20l .
The recent discovery of superconductivity at ambient pressure in
the n = 2 member Ce2PdIn8 (Tc = 0.68 K) (21], which exhibits
very similar non-FL properties [22, 23] |M] |25] [26, 27] to those in
CeCoIns, allows a detailed comparison of the superconducting prop-
erties to extract common features in these superconductors near the
antiferromagnetic QCP. The magnetic penetration depth measure-
ments have been made for high-quality single crystals of Ce2PdIn8
and CeCoIns down to ~ 60 mK by using a tunnel diode oscillator
with the resonant frequency of ~ 13 MHz (see 5/ Text: SI 1). The
weak ac field is applied along the c axis, which excites supercurrents
in the ab plane.

The temperature dependent change in the in-plane penetration
depth AA(r) = X{T) - A(T = 0) in both CeaPdIng and CeCoIng
(Fig. lA) exhibits strong temperature variation at low-temperatures,
much steeper than the flat exponential dependence expected for a
fully gapped superconductor. The AA(T) data for CeaPdIng are re-
producible in different crystals and the data for CeCoIns are fully
consistent with the previous studies |28. 29, 30 1. The strong temper-
ature dependence indicates substantial excitations of quasiparticles
at low energies, evidencing the presence of line nodes in the energy
gap. This is consistent with the residual density of states (DOS) in
the low temperature limit observed by thermal conductivity 1 19 23),
specific heat 1201 1241 . and NMR measurements |6 27 |. In particular,
a d^2_y2 order parameter with nodes along the (110) directions has
been established in CeCoIns by angle-resolved thermal conductiv-
ity 1 19 1 and specific heat measurements [20] . The strong similarity
between CeaPdIng and CeCoIns found in the low-temperature A(r)
points to common nodal structure in these superconductors.

In a pure d-wave superconductor with line nodes such as high-
Tc cuprates, it is well-established that AA(T) shows a linear T-
dependence at low temperatures, which stems from the linear energy
dependence of low-energy DOS of quasiparticles. However, a clear
deviation from the T-linear dependence is observed in the present
heavy-fermion superconductors (Fig. lA). The data also strongly de-
viate from the dependence expected for the dirty limit case

I3II (Fig. 16). We rather find that the power-law dependence T°'
with an unusual exponent a — 3/2 can describe the observed
low-temperature variation in a wide range in both superconductors
(Fig. IQ.

A few explanations for the super-linear temperature dependence
of AA(r) in d-wave superconductors have been put forth, including
the effect of impurity scattering 131], non-local effect near the nodes
|32l and phase fluctuations |33 |. Often an interpolation formula de-
scribing a crossover from T to dependence AA(T) oc T'^/{T +
T* ) is used to describe the experimental data. In particular, the im-
purity effect leads to the crossover temperature T* ~ 0.83\/rAo,
where F is the impurity scattering rate and Ao is the maximum gap.
This successfully accounts for the systematic change of AA(r) with
impurity scattering observed in Zn-doped YBaaCusOe.gs (Ref. 1 34 1).
In the present heavy-fermion case, however, fitting to this crossover
formula in these relatively clean superconductors yields T* /Tc val-
ues (~ 0.5 for CeaPdIng and ~ 0.2 for CeCoIns) substantially larger
than the estimates from these theories (see 5/ Text: SI2 and Ref. 1 30|).
Therefore, the T'^/^ dependence in a wide T/Tc range commonly ob-
served in these superconductors with quite different Tc, which is dis-
tinctly different from the T-linear dependence in e.g. YBaaCusOe.gs
(Ref. 1 34 1), rather suggests some inherent mechanism related to their
closeness to the antiferromagnetic QCP.

Superfluid Density. To discuss the precise temperature evolution of
quasiparticle excitations, we analyze our data in terms of the super-
fluid density 1/A^(T). We use the reported values of A(0) (280 nm
for CeCoIns |30| and 1010 nm for CeaPdlng |35|), from which the
normalized superfluid density paiT) = A^(0)/A^(T) has been ob-
tained (Fig. 2A). The factor of ~ 3.5 difference in the slope of
d\/d{T/Tc) (Fig. lA) is consistent with the difference in A(0) in
these two superconductors, which is also consistent with the larger 7
value in CeaPdIng |24|. This results in an almost collapse of the full
temperature dependence psiT) into a single curve (Fig. 2A, inset),
and the low-temperature variation shows the (T/Tc)^''^ dependence
with nearly identical slopes.

This (T/Tc)^/^ dependence can be also seen in other classes
of materials which are close to antiferromagnetic order. In
Fig. 2B we show data for the organic superconductor k-(BEDT-
TTF)2Cu[N(CN)2]Br which is consistent with but measured to lower
temperature than that in a previous report [|36 :: and the iron-pnictide
superconductor BaFe2(Aso,7Po,3)2 1151 . Both these materials are
known to have line nodes in their superconducting gap 1371 1381 .

Fig. 1. Temperature dependence of the magnetic penetration deptti in heavy-fermion superconductors near the antiferromagnetic QCP. (A): Low-temperature changes
in the magnetic penetration depth AA(T) = A(T) — A(T = 0) of single crystals of Ce2Pdlng and CeCoIns. The curves are shifted vertically for clarity and the
data for CeCoIns are multiplied by 3.5. The inset shows the ac susceptibility over the whole temperature range measured by the frequency shift of the tunnel diode
oscillator, showing sharp superconducting transitions. The dashed line is an exponential temperature dependence expected for a fully gapped s-wave superconductor.
(B): The same data plotted against (T/Tc)^. The solid line represents a T'^ dependence. (C): The same data plotted against (T/Tc)'^^^. The solid line represents
a T^/^ dependence.

2 I www.pnas.org/cgi/doi/10.1073/pnas.1221976110

Footline Author

B

1.00

0.95

FT

0.90

0.85

0.80

0.100

K-(ET)2Cu[N(CN)2]Br
BaFe2(Aso.7Po.3)2

0.025 0.050 0.075 0.100

Fig. 2. Universal T^^^ dependence of superfluid density in unconventional superconductors in the vicinity of the antiferromagnetic order. (A): The normalized
superfluld density ps as a function of (T/Tc)^^^ at low temperatures for Ce2Pdln8 and CeColns. The lines represent T^/^ dependence. The Inset shows
the overall temperature dependence up to (T/Tc) = 1. (B): A similar plot for Iron-pnictide superconductor BaFe2(Aso.7Po.3)2 and organic superconductor
K-(ET)2Cu[N(CN)2]Br. The solid lines are the fits to T^/^ dependence. The low-temperature data for BaFe2(Aso.7Po.3)2 are vertically shifted for clarity

In the BaFe2(Asi_i:Pa;)2 series, there is clear evidence for the an-
tiferromagnetic QCP being located at a:: ~ 0.30 |15|. In n-
(ET)2Cu[N(CN)2]Br, although the proposed phase diagram suggests
that the boundary between the superconducting and antifeiTomag-
netic states is a first-order phase transition |39 1, the anomalous criti-
cal exponent near the Mott critical end point |40 | suggests the pres-
ence of strong antiferromagnetic quantum fluctuations. The normal-
ized superfluid density ps(T) in these superconductors shows very
similar [T /Tc)'^^'^ dependence at low temperatures with slight devi-
ations at the lowest temperatures below T/Tc ~ 0.05 ((T/Tc)'*^^ <
0.01).

These results imply that the (T/Tc)^^'^ dependence of superfluid
density in a wide T/Tc range is a robust property in unconventional
superconductors, in which strong antiferromagnetic fluctuations are
present (for comparisons between iron-pnictides and cuprates, see 5/
Text: S13).

Discussion

If the quantum fluctuations survive the Fermi-surface gapping, the
effective mass is expected to diverge in the zero temperature limit.
In such a case, the strong enhancement of mass leads to a reduction
of superfluid density when approaching the zero-temperature limit.
It has been recently found in iron-pnictides that the zero-temperature
superfluid density 1/ (0) shows a strong reduction at the QCP | IS*],
indicating the strong quantum critical fluctuations directly affect the
superconducting condensate. However, the temperature dependence
of the normalized superfluid density ps (T) actually continues to rise
with decreasing temperature (Figs. 2A and B, insets). This suggests
that the effect of temperature dependent quantum fluctuations enters
as corrections in the temperature dependence of superfluid density at
low temperatures. The anomalous non-integer power-law tempera-
ture dependence of Ps{T) universally observed in quantum critical
superconductors thus calls for further theoretical understanding.

Nodal Quantum Critlcallty. Below we will propose a possible sce-
nario that this universal behavior can be naturally explained by in-
voking a strong momentum dependence of renormalization due to
the nodal gap structure. Suppose that the gap formation below Tc
quenches the low-energy quantum critical fluctuations, which pre-
vents the effective mass enhancement in the superconducting state.
Then in the nodal superconductors, because of the strong momentum
dependence of the gap magnitude we may consider that the degree of

quenching of the quantum fluctuations in the superconducting state
has a strong momentum dependence as well; the low-energy fluctu-
ations are expected to be strongest near the nodes where the FeiTni
surface is not gapped (Fig. 3A). To model this effect, we consider the
angle dependence of the effective Fermi velocity v*p ((j)) along the un-
derlying FeiTni surface. We consider a simple cylindrical Fermi sur-
face and assume that the renormalization in Vp{k)/vp is inversely
related to the enhancement in the effective mass m* (k) (Fig. 36).
(Here 1/vp is given by the dynamic effective mass which is different
from, but closely related to the thermodynamic mass II12I ). The ef-
fective mass enhancement on approaching the QCP can be described
by m*^ oc (p — Pqcp)~^, where p is a nonthermal parameter con-
trolling the distance from the QCP at pqcp- The critical exponent
P has been estimated experimentally by using magnetic fields as the
parameter p, and in several materials a value of /3 close to unity has
been reported |41) |42j. In the present case, we take the magnitude
of the superconducting gap | A| as the control parameter, because the
gap magnitude determines the degree of quenching of low-energy
fluctuations. We thus assume vp{k) oc j A(fc)j'^^^, which allows us
to calculate the temperature dependence of superfluid density by the
integral over the Fermi surface S as 1 10, 13]

where Vp (vp) is the effective velocity in the superconducting
(normal) state, the subscripts i,j denote the directions of the cur-
rent and vector potential (we take both along a), Y{k,T) —
-2 — = dE is Yosida function, and f(E) is the

Fermi-Dirac function for the quasiparticle energy E. By using the
d^2_y2 formula of A(fc) — Ao cos(2(^) and l3 — 1, the normal-
ized superfluid density ps is calculated with Vp{(l>) oc | cos(20)|^''^
(Fig. 3C), which results in the y^/^ dependence at low temperatures
(Fig. 3D).

As real materials will never be located exactly at the QCP there
will be a cutoff for the diverging m* near the nodes. This can be mod-
elled by introducing a minimum value for Vp. This leads to an up-
ward deviation from the t'-^^^ dependence of ps at very low tempera-
tures approaching T-linear behavior at sufficiently low temperatures
(Fig. 3D). This can explain the essential features of the experimental
observations. We note that the presence of the cutoff can be also ex-
pected even at the exact QCP, because the dynamical susceptibility in
the zero temperature limit should diverge only on certain regions of

Footllne Author

PNAS I Issue Date | Volume | Issue Number | 3

Fig. 3. Nodal quantum criticality in unconventional superconductors. (A): The momentum dependent gap A(fc) (whose magnitude is illustrated by thin lines with
grey shades) opens on the Fermi surface (thick line), and has nodes (red circles) at certain directions. In d^2_j,2-wave superconductors, for example, A(fc) has
strong in-plane anisotropy Ao cos(20) as a function of azimuthal angle (f. In quasi-two dimensional systems, the Fermi surface is approximated by a cylinder, and
thus the gap has nodal lines at perpendicular to the planes. At the nodes, the gap is zero and thus the quantum critical fluctuations may be present (red shades) on
the ungapped Fermi surface. (B): The nodal quantum fluctuations lead to the momentum dependence of the renormalization in l/vp{k) near the nodes (blue lines).
(C): The angle dependence of the renormalized Fermi velocity Vp{(f>) relative to the unrenormalized one vp along the Fermi surface, assumed for calculations of
the superfluid density in (D). Near the nodes, we illustrate different cutoff levels, which model finite distances from the QCP or disorder. {D): Calculated normalized
superfluid density as a function of (T/TcY/'^ with different cutoff levels, which explains the deviation from the T'^/'^ dependence at very low temperatures. The inset
is the full temperature dependence up to (T/Tc) = 1.

the Fermi surface (determined by the momentum dependence of the
spin fluctuations), which in general may be different from the nodal
points |9|. Of course, disorder would also produce some additional
changes to the temperature dependence but we have not included this
in the present model.

An additional factor could also come from the temperature de-
pendence of the renormalization of m* . As only the theiTnally ex-
cited quasiparticles will be renormalized, the angular range near the
nodes where this occurs (i.e., where A(fc) < ^bT) is quite limited
at low temperature and so this has a rather minor effect. In our tem-
perature range of interest, T < 0.2Tc, this region is actually limited
to a narrow angle range of ~ ±3° near the nodes. If we add a cut-
off constant of 0.3 covering this angle range (Fig. 3C), we found that
the T-dependence is affected only in the lowest temperature range
of [T/Tcf-^ < 0.01, above which the T^^^ dependence of ps still
holds (Fig. 3Z)). This exercise implies that the inclusion of the tem-
perature dependence of renormalization will not change ps (T) sig-
nificantly in the temperature range of interest (for more discussion,
see 5/ Text: SI4).

We also note that in iron-pnictides the gap symmetry is most
likely s-wave f43|, and the model based on the d-wave gap may
not be applicable. However, the fundamental physics that the low-
energy excitations are governed by the nodal regions should be es-

sentially the same. Although the detailed structure of the momentum-
dependent Fermi velocity (such as the precise value of the critical
exponent /?) will affect the detailed Ps{T) (see 5/ Text: SIS), it is
striking that such a simple model can capture the salient feature of
the unusual behavior of quasiparticle excitation in the superconduct-
ing state of these materials. In a FL theory, the renormalization of
the effective FeiTni velocity vp in Eq. [1] can be described by the in-
terplay between fc-dependent quasiparticle interaction and nodal gap
structure 1 13 1. More detailed theoretical calculations will be needed
to account for: strong energy dependence of critical fluctuations, the
effects of disorder (such as quasiparticle scattering interference |44 |)
and the inelastic quasiparticle scattering rate.

We suggest that the nodal quantum criticality proposed here is
an important aspect of unconventional superconductivity close to the
magnetic QCP. Besides the penetration depth analyzed here there will
be implications for most other superconducting properties such as
thermal conductivity and the NMR relaxation rate l/Ti{T) which
has been long known to have strong deviations from the T"^ law,
which in the usual analysis gives overestimates of the residual quasi-
particle DOS in quantum critical superconductors ||6]2]. It should be
straightforward to extend our analysis to these other properties.

4 I www.pnas.org/cgi/doi/10.1073/pnas.1221976110

Footline Author

ACKNOWLEDGMENTS. We thank A. Chubukov, H. Fukazawa, R. W. Giannetta,
K. Kanoda, S. Kasahara, H. Kontani, S. K. Goh, P.J. Hirschfeld, A. H. Nevidom-
skyy, T. Nomoto, R. Prozorov, I. Vekhter, Y. Yanase, and Y. F. Yang for discussions.
Tliis work is supported by KAKENHI from JSPS, Grant-in-Aid for GCOE program
"The Next Generation of Pfiysics, Spun from Universality and Emergence" from

IVIEXT, Japan, EPSRC in the UK, the National Science Centre (Poland) under
the research Grant No. 201 1/01 /B/ST3/04482 in Poland and by Argonne, a U.S.
Department of Energy Office of Science laboratory, operated under Contract No.
DE-AC02-06GH11357 in the USA.

1 . See, for example, Varma CM, Nussinov Z, & van Saarloos W (2002) Singular or
non-Fermi liquids. P/iys Rep 361:267-41 7.

2. Gegenwart P, SI Q, & Steglich F (2008) Quantum criticality in iieavy-fermion metais.
WafP/iys 4:186-197.

3. Shishido H, Settai R, Harlma H, & OnukI Y (2005) A drastic change of the Fermi
surface at a critical pressure in CeRhlns : dHvA study under pressure. J Phys Soc
Jpn 74:1103-1106.

4. Shishido H, etal. (2010) Evolution of the Fermi surface of BaFe2(Asi_a.Px)2 on
entering the superconducting dome. P/iysRevl.ett104:057008.

5. Stewart GR (2001) Non-Fermi-liquid Isehavior in d- and f-eiectron metals. Rev Mod
Phys 73:797-855.

6. Kohori Y, efa/. (2001) NIUIR and NQR studies of the heavy fermion superconductors
CeTlns (T = Co and Ir). Phys /?ev 664:1 34526.

7. Nakal Y, etal. (2010)-^^ Rand ^-^^As NMR evidence for a residual density of states at
zero energy In superconducting BaFe2(Aso.67Po.33)2- Phys BevB 81 :020503(R).

8. SIgrlst M, & Ueda, K (1991) Phenomenologlcal theory of unconventional super-
conductivity. Rev Mod Phys 63:239-31 1 .

9. Hirschfeld PJ, Korshunov MM, & Mazin II (2011) Gap symmetry and structure of
Fe-based superconductors. Rep Pmg P/iys 74:124508.

10. Chandrasekhar BS, EInzel D (1993) The superconducting penetration depth from
the semlclassical model. Ann Phys 2:535-546.

11. Leggett AJ (1965) Theory of a superfluid Fermi liquid. I. General formalism and
static properties. Phys Rev140:A1869-A1888.

12. Varma CM, Miyake K, & Schmitt-RInk S (1986) London penetration depth of heavy-
fermion superconductors. Phys Rev LeH 57:626-628.

13. Jujo T (2002) Effect of Umklapp scattering on magnetic field penetration depth In
high-Te cupiates. J Phys Soc Jpn 71 :888-904.

14. Gross F, el al. (1986) Anomalous temperature dependence of the magnetic field
penetration depth in superconducting UBCis. ZPhys 664:175-188.

15. Hashimoto K, etal. (2012) A sharp peak of the zero-temperature penetration depth
at optimal composition In the iron-based superconductor BaFe2(ASi_^P^)2. Sc/-
ence 336:1554-1557.

16. Petrovic C, etal. (2001) Heavy-fermlon superconductivity in CeColn^ at 2.3 K. J
Phys Condens /Wa(fer13:L337-L342.

17. Nakajima Y, et al. (2007) Non-Fermi liquid behavior In the magnetotransport of
CeMln5 (M : Co and Rh): Striking similarity between quasi two-dimensional heavy
fermion and high-T^ cupiates. J Phys Soc Jpn 76:024703.

18. Paglione J, ef a/. (2003) Field-Induced quantum critical point In CeColn -,. Phys Rev
LeH 91:246405.

19. Izawa K, et al. (2001) Angular position of nodes in the superconducting gap of
quasi-2D heavy-fermion superconductor CeColn5. Phys Rev Lett 87:057002.

20. An K. et al. (2010) Sign reversal of field-angle resolved heat capacity oscillations
In a heavy fermion superconductor CeColn5 and <i^2 _ 2 pairing symmetry. Phys
Rev Lett -104:037002.

21 . Kaczorowski D, Pikul AP, Gnida D, & Tran VH (201 0) Emergence of a superconduc^
ing state from an antiferromagnetic phase In single crystals of the heavy fermion
compound Ce2Pdlns. Phys Rev Lett103:027003(2009);/i»'(/. 104:059702.

22. Kaczorowski D, Gnida D, Pikul AP, & Tran VH (2010) Heavy-fermion superconduc-
tivity in Ce2Pdlng. Solid State Commun 150:411-414.

23. Dong JK, etaL (2011) Field-induced quantum critical point and nodal supercon-
ductivity in the heavy-fermion superconductor Ce2Pdlng. Phys RevXI :011010.

24. Tokiwa Y, Gegenwart P, Gnida D, & Kaczorowski D (201 1 ) Quantum criticality near
the upper critical field of Ce2Pdln8. Phys Rev B84:140507(R).

25. I\1atusiak M, Gnida D, & Kaczorowski D (201 1 ) Quantum criticality in Ce2 Pding : A
thermoelectric study. Phys RevS84:115110.

26. Gnida D, IVIatuslak M, & Kaczorowski D (201 2) Anomalous magnetotransport in the
heavy-fermlon superconductor Ce2 Pding . Phys Rev B 84:060508(R).

27. Fukazawa H, Nagashima R, Shimatani S, Kohori Y, & Kaczorowski D (2012) Nodal
superconductivity and non-Fermi-liquid behavior in Ce2Pdlng studied by ^^^in
nuclear quadrupole resonance. Phys Rev B 86:094508.

28. Ormeno RJ, Sibley A, Gough CE, Sebastian S, & Fisher IR (2002) Microwave con-
ductivity and penetration depth in the heavy fermion superconductor CeColng.
Phys Rev Lett 88:047005.

29. Chia EEIVI et al. (2003) Nonlocallty and strong coupling in the heavy fermion su-
perconductor CeColns: A penetration depth study. Phys Rev 6 67:014527.

30. dzcan S, e( al. (2003) London penetration depth measurements of the heavy-
fermion superconductor CeColn5 near a magnetic quantum critical point. Eum-
phys Leff 62:412-418.

31. Hirschfeld PJ & Goldenfeld N (1993) Effect of strong scattering on the low-
temperature penetration depth of a cf-wave superconductor. Phys Rev B 48:4219-
4222.

32. Kosztin I & Leggett AJ (1997) Nonlocal effects on the magnetic penetration depth
In d-wave superconductors. Phys Rev Le(f 79:135-138.

33. Chen 0, Kosztin I, & Levin K (2000) Unusual thermodynamlcal and transport sig-
natures of the BCS to Bose-Elnstein crossover scenario below t^. Phys Rev Lett
85:2801-2804.

34. Bonn DA, et aL (1994) Comparison of the influence of Ni and Zn impurities on the
electromagnetic properties of YBa2CU306.g5. Phys Rev 650:4051 -4063.

35. Tran VH, Hillier AD, Adroja DT, & Kaczorowski D (201 2) Antiferromagnetic spin fluc-
tuations In the heavy-fermion superconductor Ce2 Pding. Phys Rev 686:094525.

36. Carrlngton A, ef al. (1999) Low-temperature penetration depth of n-
(ET)2Cu[N(CN)2]Brand k-(ET)2Cu(NCS)2. Phys Rev LeH 83:4172-4175.

37. Hashimoto K, ef al. (2010) Line nodes In the energy gap of superconducting
BaFe2(Asi_a;Px)2 single crystals as seen via penetration depth and thermal
conductivity. Phys Rev 6 81 :220501 (R).

38. Taylor OJ, Carrlngton A, & Schlueter JA (2007) Specific-heat measurements of
the gap structure of the organic superconductors K-(ET)2Cu[N(CN)2]Br and n-
(ET)2Cu(NCS)2. Phys Rev Left 99:057001 .

39. Kanoda K (2006) Metal-insulator transition in k-(ET)2X and (DCNQI)2M: Two con-
trasting manifestation of electron correlation. J Phys Soc Jpn 75:051007.

40. Kagawa F, Miyagawa K, & Kanoda K (2009) IVIagnetic IVIott criticality in a ^-type
organic salt probed by NIVIR. A/at Phys 5:880-884.

41. Gegenwart P, ef al. (2002) IVIagnetic-Fleld Induced Quantum Critical Point in
YbRhjSis. Phys Rev Le/f 89:056402.

42. Matsumoto Y, et a/. (201 1 ) Quantum criticality without tuning in the mixed valence
compound /J-YbAIBj. Science 331:31 6-31 9.

43. Yamashita lU, et a/. (2011) Nodal gap structure of superconducting
BaFe2(Asi_xPx)2 from angle-resolved thermal conductivity in a magnetic field.
Phys Rev 6 84:060507(R).

44. Wang Q-H, & Lee D-H (2003) Quasiparticle scattering interference in high-
temperature superconductors. Phys Rev B 67:02051 1 (R).

Footline Author

PNAS I Issue Date | Volume | Issue Number | 5

Supporting Information

Hashimoto et al. 10.1073/pnas.l221976110
SI Text

SI1-Sample information and experimental techniques

High-quality single crystals of Ce2PdIn8 and CeCoIns were grown
by the self-flux method IS1IIS2I . The typical lateral size of the crys-
tals is 150 X 150 ^ra^ . For Ce2PdIn8 very thin crystals with a thick-
ness of less than 10 were prepared to avoid contamination of
Celns. We have checked that our crystals used in this study show
no antiferromagnetic signals. The magnetic penetration depth mea-
surements down to very low temperatures (~ 60 mK) have been per-
formed by using a tunnel diode oscillator with the resonant frequency
of 13 MHz [S3jlS4], which is mounted on a '^He-'^He dilution refrig-
erator. The sample is placed on a sapphire rod, the other end of which
is glued to a copper block on which a Ru02 thermometer is mounted.
The sample is placed at the center of a solenoid which forms part of
the resonant tank circuit. A weak ac field is applied along the c-axis
so that the supercurrent flows within the afe-plane. Changes in the
resonant frequency are directly proportional to changes in the mag-
netic penetration depth, AA(T) = GAf{T). The calibration factor
G is determined from the geometry of the sample |S5 1.

Samples of K-(BEDT-TTF)2Cu[N(CN)2]Br were grown via the
standard electrocrystalisation method |S6I. Measurements were also
made using a tunnel diode oscillator. Special care was taken to cool
the samples slowly across the temperature region 70 - 90 K where a
structural ordering transition occurs. The cooling rate was restricted
to 0.02 K/min, which is sufficiently slow enough to avoid suppression
in Tc and reduction in electronic contribution to the heat capacity

(S3.

The data in Fig. 26 was taken for a sample with dimensions
0.8 X 0.7 X 0.4 mm^. In these organic samples even weak radio
frequency (RF) fields can produce self-heating effects at the lowest
temperatures (T < 300 mK). To eliminate this effect, measurements
were made with the sample at different positions on the axis of the
RF coil, effectively reducing the applied field. Typically, a reduction
in field of a factor three was sufficient to eliminate all signs of self-
heating. This procedure was tested in three samples and consistent
results were obtained.

SI2-Comparisons with existing theories

The origins of the deviation from the T-linear temperature depen-
dence of A(r) in a d-wave superconductor are discussed in the frame-
work of two different theories. Hirschfeld and Goldenfeld |S8 | have
shown that a small amount of impurity causes a residual density of
states near Ep, which can change the linear-T dependence (pure
regime) to a quadratic T-dependence (impurity-dominated regime)
at a crossover temperature T*^^. This characteristic temperature
is given by T-*-^p ~ 0.83\/rAo, where F is the impurity scatter-
ing rate and Ao is the maximum gap. On the other hand, Kosztin
and Leggett | S9 1 have pointed out that since the coherence length ^
diverges near the nodes in a d-wave superconductor, Cooper pairs
near the nodal directions, where ^ 3> A, cannot participate in the
field screening. Therefore, the effective penetration depth should be
larger than the local limit value below a characteristic temperature
T^i^ ~ Ao^(0)/A(0), which leads to a quadratic T-dependence.
Thus, for both theories the interpolation formula between two regions
is given by AA(T) oc T"^ /{T + T*). Here we note an important dis-
tinction between the temperature dependence of AA(T), which is
directly measured, and the normalised superfluid density, Ps(T) =
A^(0)/A^(r). The d-wave form for the superfluid density, ps{T) =
1 - ^lT/T,, gives AA(T)/A(0) = \KT/T,) + Ip^'iT/T^f + ■ ■ ■,

1.00

— «===J 1 1

1

0.98

-

0.94

0.92

• CeCoIng

• CesPdIng

0.90

0.88

1 1 1

1

0.05 0.10 0.15 0.20 0.25

T/T,

Fig. S1. Expanded view of normalized superfluid density plotted against T/Tc
for Ce2Pdlng and CeColns. The solid lines are fit to the crossover behavior in a
d-wave superconductor described in the text.

which lead to a small quadratic component to A(r) that depends on
^, where /j, is proportional to the inverse of the slope of the gap A((^)
at the nodes, fj, = (4 In 2)fcsrc/(dA(0)/d0|nodc) 1S10|. Then,
even in the pure d-wave model we have slightly concave tempera-
ture dependence for A(T). Therefore, we take the following formula
to describe the crossover behavior expected in these theories;

p.(T) = i-M(rVTe)/(r + T*). [s-i]

Figure SI shows the low-temperature superfluid density Ps(T)
both for Ce2PdIn8 and CeCoIns. We carefully examine the possibil-
ity of the two theories mentioned above by evaluating T*^^ from the
present data. The obtained values of T^xp/Tc from the fits are 0.48
and 0.22 for Ce2PdIn8 and CeCoIns, respectively. If we assume that
the origin of nonlinearity of A(r) comes from the residual density of
states induced by impurities, we can estimate the impurity scattering
rate F via T* ^ 0.83\/rA^. The obtained value of F for Ce2PdIn8
is O.I 1 K. By using this and Abrikosov-Gor'kov theory|Sl 1 1, we es-
timated that the Tc reduction is 11% for Ce2PdIn8. However, recent
specific heat measurements have suggested that the impurity level of
Ce2PdIn8 is comparable to 0.22% Cd-doped CeCoIns 1S12|. In this
case, Tc reduction is less than 5 %. Similarly, the obtained value
of T* for CeCoIns using Eq. [S-1] is much larger than the value
estimated from thermal transport measurements (< 30 mK) |S13|.
Therefore, impurity effect are unlikely to explain the nonlinearity of
A(r) in these systems. As for the nonlocal effects, the reported val-
ues of ^0 ~ 8nm |S2 S14 S15| for both compounds give T^j, ~
14 mK and 140 mK for Ce2PdIn8 and CeCoIns, respectively, which
indicates that the non-locality effect should be important at much
lower temperatures than our T^^p- Thus, we can exclude these ef-
fects as primary origins of the deviation from the T-linear behavior
mps{T).

Reserved for Publication Footnotes

6 I www.pnas.org/cgi/doi/10.1073/pnas.1221976110

Footline Author

We also note that the T^/"^ dependence of low-temperature super-
fluid density has been derived from the Bardeen-Cooper-Schrieffer
(BCS) to Bose-Einstein condensation (BEC) crossover below Tc
|S16|. However, this assumes an s-wave state, which is not appli-
cable to the nodal superconductors focused in the present study.

SI3-Comparisons between iron-pnicteides and cuprates

We comment on the difference in the high- Tc cuprate superconduc-
tors. In clean single crystals of optimally doped YBa2Cu307_4, the
superfluid density exhibits a rather wide range of T-linear behav-
ior and the crossover temperature is as low as T*^^/Tc ~ 0.01. In
hole-doped cuprates, the superconducting dome is separated from
the antiferromagnetic order, and a possible QCP of the pseudogap
phase has been discussed near the optimum doping of the supercon-
ducting dome. The absolute value of the superfluid density shows a
broad maximum near the optimal doping IIS17I . which is opposite to
the iron-pnictide BaFe2(Asi_a;Pi:)2 case, where the sharp minimum
is observed at the antiferromagnetic QCP IS181 . The latter case is
consistent with the enhanced mass (which enters inversely to the su-
perfluid density) associated with the QCP. This difference in doping
dependence seems consistent with the difference in the temperature
dependence of superfluid density; the y^/^ dependence is a common
property associated with the mass enhancement due to the antiferro-
magnetic QCP.

To see the doping dependence of superfluid density in the iron-
pnictide BaFe2(Asi_2,Pa;)2, we plot the low-temperature ps as a
function of {T/TaY'^ in a wide doping range (Fig. S2). The anti-
ferromagnetic QCP composition in this system has been located at
X = 0.30 ISI8I . In high composition samples (x > 0.38) which are
relatively far from the QCP, the low-temperature deviations from the
T^^^ dependence is more pronounced than those in samples closer
to the QCP (0.27 < x < 0.33). These results further support that
the non-integer power law is related to the closeness to the magnetic
QCP. In hole-doped cuprate superconductors, the antiferromagnetic
QCP is actually located outside the low-doping end point of the su-
perconducting dome. It deserves further studies to understand how
the phase diagrams relate to these differences of low-energy quasi-
particle excitations in cuprates and iron-based high- Tc superconduc-
tors.

0.05 0.10

{T/T,)'-'

Fig. S2. The normalized superfluid density ps as a function of (T/Tc)^-^ at
low temperatures for BaFe2(Asi_£Pi:)2 single crystals. The original penetration
depth data are taken from Ret. IS181 .

SI4-Temperature dependence of the renormalization

Near the QCP, the temperature dependence of the renormalization
is an important factor. In the normal state above the upper critical
field, the specific heat is largely enhanced at low temperatures |S19 1,
indicating the strong temperature dependence of m* which may be
approximated by 1/Vt at low temperatures (Fig. S3A). (Note that
due to the field-induced QCP, this dependence is the strongest near
the upper critical field and the actual temperature dependence at zero
field may be weaker.) In the superconducting state, owing to the mo-
mentum dependence of the renormalization we consider, l/vp{T)
will level off at momentum dependent temperature. Near the nodes,
the temperature dependence of renormalization will continue to low
temperatures, whereas in the antinodal directions it will become in-
dependent of temperature near Tc.

If we assume v*p((j),T)/vF = m&x{C^T/Tc, \ cos(2<j!>)j^''^),
we can calculate the temperature dependence of averaged mass en-
hancement in the superconducting state (Fig. S3A). Here we separate

A 1.15

LlOh

o

c^"^ 1.05

1.00

• m*(7)/m*(0.2TJ
Fit

A/(7)/A/(0.2r,)

X

0.05

0.10

0.15

1.15

1.00

0.20

B

1.0

- ' 1 ' 1 ' W

— V((>)/''f =

C0S(2^)|"'

- - - 1/a/

--

- -

= max(C(7■/7J"^|cos(2^)|'") /

1/a/

—

r'' , 1 , 1

, 1 , \

0.2

Fig. S3. (A): Temperature dependence of the effective mass normalized by its 0.2Tc value In CeColns (black circles) determined from the specific heat
measured at 5 T in the normal state (above the upper critical field) S19 can be fitted to a l/VT dependence (black line). The inverse of the effective
Fermi velocity averaged over the Fermi surface In the superconducting state (corresponding to the first term in Eq. [1] In the main text) calculated assuming
Vp{4>, T)/vp = max(Ci/T/Tc, | cos(2</))|-'/^) with a proper constant C = Vb\ cos 2(45° it 3°)|-'-''^, has a much weaker temperature dependence (red
line). (B): Temperature dependence of the normalized superfluid density with (solid lines) and without (dashed lines) assuming the temperature dependence of
renormalization. The total superfluid density (black) can be separated to the diamagnetic (red) and paramagnetic (blue) components.

Footline Author

PNAS I issue Date | Volume | Issue Number | 7

the two terms in Eq. [1] in the main text as A = ^ — Ap ^,
where the first term A^^ = |^|^ / "i^^f '^^ corresponds to the dia-
magnetic current and the second term comes from the paramagnetic
current carried by the thermally excited quasiparticles. In Fig. S3S
the temperature dependence of each contribution is presented with
and without the temperature dependence of renormalization. These
results indicate that in the superconducting state the temperature de-
pendence of the averaged renormalization is weak, which seems to be
consistent with the recent high-frequency conductivity results IS20I .
Moreover, the temperature dependence of the total superfluid density
A^^ (T) remains almost unchanged even we consider the temperature
dependence of the renormalization in the superconducting state.

SI5-Effect of quantum critical exponent

To see how the power-law of superfluid density varies with the quan-
tum critical exponent /3, we have computed pa (T) with several values
of /3 up to ~ 2. The low-temperature 1 - Ps{T) is fitted to the T"
dependence up to T = Q.2Tc and the obtained power a is plotted
against the exponent /3 in Fig. S4. We find that the power a varies
linearly with /3, which can be approximated by a ~ 1 + /3/2.

The fact that we universally obtained a close to 3/2 in the present
quantum critical superconductors suggests that the exponent /? in
these materials is not far from unity. Indeed, in /3-YbAlB4 in which
zero-field quantum criticality has been suggested IIS2II . thermody-
namic measurements shows /? ~ 1, which leads to a « 1.54
(Fig. S4). In the study of YbRh2Si2, where the quantum critical field
is 0.66 T, /? ~ 1 has also been reported |S22|. We note that in a sim-
ilar system YbRh2(Sio.95Geo.o5)2 where the critical field is reduced
to 0.027 T, smaller value of /3 ~ 2/3 has been suggested |S23|, but
this value gives a ~ 1.36, which is in practise difficult to distinguish
from the 3/2 power-law dependence. We also note that in CeCoIns

and Ce2PdIn8 superconductors the estimated values for (3 from trans-
port coefficient A near the upper critical field Hc2 are rather scattered
(~ 1.37 for CeCoIns tS24i| and ~ 0.57 for Ce2PdIn8 | S14|), but the
first-order transition of the low-temperature Hc2 due to the strong
Pauli paramagnetism P S25I IS 121 may provide some complexity in
evaluating the precise value of critical exponent.

0.5 1.0 1.5 2.0

Fig. S4. Low-temperature power a of the superfluid density obtained from ttie
fitting up to T/Tc = 0.2 for the calculated ps{T/Tc) with different values of
the quantum critical exponent /3. The line is a fit to a linear dependence.

51. Shishido H, etal. (2002) Fermi surface, magnetic and superconducting properties
of LaRhlns and CeTlnj (T: Co, Rh and Ir). J P/iys Soe Jpn 71:162-173.

52. Kaczorowski D, Pikul AP, Gnida D, & Tran VH (2009) Emergence of a superconduct-
ing state from an antiferromagnetic phase in single crystals of tlie tieavy fermion
compound C e2Pdln8. Phys Rev LefM 03:027003; ibid. Phys Rev Lett 104:059702
(2010).

53. Carrington A, Giannetta RW, Kim JT, & Giapintzakis J (1999) Absence of nonlinear
Meissner effect in YBa2Cu306 95. Ptiys /?evS59:R14173-R14176.

54. Hashimoto K, et al. (2010) Line nodes in the energy gap of superconducting
BaFe2(ASi_^P:c)2 single crystals as seen via penetration depth and thermal con-
ductivity Pljys Rev B 81 :220501 (R).

55. Prozorov R, Giannetta RW, Carrington A, & Araujo-Moreira FM (2000) Meissner-
London state in superconductors of rectangular cross section in a perpendicular
magnetic field. Phys Rev 6 62:1 1 5-1 1 8.

56. Kini M, et al. (1990) A new ambient-pressure organic superconductor, k-
(ET)2Cu[N(CN)2]Br, with the highest transition temperature yet observed (inductive
onset T„ = 11.6K, resistive onset = 12.5 K). Inorg Chem 29:2555.

57. Taylor OJ, Carrington A, & Schlueter JA (2008) Superconductor-insulator phase
separation induced by rapid cooling of K-(BEDT-TrF)2Cu[N{CN)2]Br. Pliys Rev B
77:060503.

58. Hirschfeld PJ & Goldenfeld N (1993) Effect of strong scattering on the low-
temperature penetration depth of a cf-wave superconductor. Pliys Rev 6 48:421 9-
4222.

59. Kosztin I & Leggett AJ (1997) Nonlocal effects on the magnetic penetration depth
in d-wave superconductors. Phys Rev Le» 79:135.

510. Xu D, Yip SK, & Sauls JA (1995) Nonlinear Meissner effect In unconventional
superconductors. Phys Be/S 51:1 6233-1 6253.

511. Abrikosov AA & Gor~fkov LP (1960) Contribution to the theory of superconduct-
ing alloys with paramagnetic impurities Zh E/csp Teor F/z 39:1 781. [Sov. Phys. JETP
/.ett.12, 1243(1961)].

SI 2. Toklwa Y, Gegenwart P, Gnida D, & Kaczorowski D (2011) Quantum criticality near
the upper critical field of Ce2Pdln8. Pliys RevB84:140507(R).

513. Movshovich R, et ai. (2001) Unconventional superconductivity in Celrlns and
CeColhs : Specific heat and thermal conductivity studies. Piiys Rev Lett 86:51 52-
5155.

514. Dong JK, et ai. (2011) Field-Induced quantum critical point and nodal supercon-
ductivity in the heavy-fermion superconductor Ce2Pdln8. Phys Rev X 1 :01 1 01 0.

SI 5. Settal R, et al. (2001) Quasi-two-dimensional Fermi surfaces and the de Haas-
van Alphen oscillation in both the normal and superconducting mixed states of
CeColng. J Phys Condens /Wa(fer13:L627-L634.

SI 6. Chen Q, Kosztin I, & Levin K (2000) Unusual thermodynamical and transport sig-
natures of the BCS to Bose-Einstein crossover scenario below t,.. Phys Rev Lett
85:2801-2804.

517. Tallon JL, Loram JW, Cooper JR, Panagopoulos C, & Bernhard C (2003) Super-
fluid density in cuprate high-T,. superconductors: A new paradigm. Phys Rev B
68:180501.

518. Hashimoto K, et ai. (201 2) A sharp peak of the zero-temperature penetration depth
at optimal composition in the Iron-based superconductor BaFe2(ASi_ j.P^)2. Sci-
ence 334: 1554-1 557.

519. Bianchi A, Movshovich R, Vekhter I, Paglluso PG, & Sarrao JL (2003) Avoided
antiferromagnetic order and quantum critical point In CeColng. Phys Rev Lett
91:257001.

520. Trunclk CJS et ai. (2012) Nodal quasiparticle dynamics In the heavy
fermion superconductor CeColng revealed by precision microwave spectroscopy.
arXiv:1 21 0.5571

521 . Matsumoto Y, et ai. (201 1 ) Quantum criticality without tuning in the mixed valence
compound 8-YbAIB4. Science 331:31 6-31 9.

522. Gegenwart P, et al. (2002) Magnetic-field induced quantum critical point in
YbRh2Sl2. PhysRev ieff 89:056402.

523. Custers J, ef ai. (2003) The break-up of heavy electrons at a quantum critical point.
Wafure 424:524-527.

524. Paglione J, ef ai. (2003) Field-induced quantum critical point in CeColng . Phys Rev
Le(f 91:246405.

525. Izawa K, ef a/. (2001) Angular position of nodes In the superconducting gap of
quasi-2D heavy-fermion superconductor CeColng . Phys Rev Lett 87:057002.

8 I www.pnas.org/cgi/doi/10.1073/pnas.1221976110 Footline Author