Access the full text.
Sign up today, get DeepDyve free for 14 days.
(2019)
arXiv e-prints
E. Fradkin (1991)
Field theories of condensed matter systems
(2015)
New J
(2009)
Annals of Physics 324
(2018)
Reports on Progress in Physics 81
(2000)
The European Physical Journal B -Condensed Matter and Complex Systems
O. Castro-Orgaz, W. Hager (2019)
and sShallow Water Hydraulics
D. Scalapino (2006)
Numerical Studies of the 2D Hubbard ModelarXiv: Strongly Correlated Electrons
(2011)
Nuclear Physics B 853
(1987)
Science 235
(2017)
Scientific Reports 7
W. Marsden (2012)
I and J
(2000)
Physical Review B 62
(2016)
EPL (Europhysics Letters) 113
(2014)
Comptes Rendus Physique 15
1 1 1 1 C. P epin, D. Chakraborty, M. Grandadam, and S. Sarkar Institut de Physique Th eorique, Universit e Paris-Saclay, CEA, CNRS, F-91191 Gif-sur-Yvette, France. The physics of the pseudo-gap phase of high spectives to the PG state of cuprate superconductors. temperature cuprate superconductors has been We then focus on one speci c theoretical approach where an enduring mystery in the past thirty years. uctuations protected by a speci c Higgs mechanism are The ubiquitous presence of the pseudo-gap phase held responsible for the unusual properties of the pseudo- in under-doped cuprates suggests that its under- gap phase. standing holds a key in unraveling the origin of high temperature superconductivity. In this pa- per, we review various theoretical approaches to 1. MOTT PHYSICS AND STRONGLY this problem, with a special emphasis on the con- CORRELATED ELECTRONS cept of emergent symmetries in the under-doped region of those compounds. We dierentiate the An overall glance at the phase diagram of the cuprate theories by considering a few fundamental ques- superconductors shows that superconductivity sets up tions related to the rich phenomenology of these close to an anti-ferromagnetic (AF) phase transition, but materials. Lastly we discuss a recent idea of two also close to a Mott insulating phase (Fig. 2). The pres- kinds of entangled preformed pairs which open ence of the metal-insulating transition so close to super- a gap at the pseudo-gap onset temperature T conductivity is unusual and led many theoreticians to through a speci c Higgs mechanism. We give a attribute the PG phase to a precursor of the Mott tran- review of the experimental consequences of this sition (a few review papers [15{24]). The Coulomb inter- line of thoughts. action has a high energy scale U = 1eV which prohibits The Pseudo-Gap (PG) state of the cuprates was dis- the double occupancy on each site and induces strong covered in 1989 [1], three years after the discovery of high correlations between electrons. An eective Hamiltonian temperature superconductivity in those compounds. It can be derived by integrating out formally the Coulomb was rst observed in nuclear magnetic resonance (NMR) interaction, which leads to AF super-exchange interac- experiments, in an intermediate doping regime 0:06 < tion associated with a constraint prohibiting double oc- p < 0:20, as a loss of the density of states at the Fermi cupancy on each site for conduction electrons (see e.g. level [1{3] at temperatures above the superconducting [16, 25, 26]) and is given by transition temperature T . Subsequently, angle resolved photo emission spectroscopy (ARPES) established that a X X H = t c c + J S S ; (1) ij j i j part of the Fermi surface is gapped in the Anti-Nodal Re- i i;j hi;ji gion (ANR) (regions close to (0;) and (; 0) points) of the Brillouin zone, leading to the formation of Fermi `arcs'. Though the PG state shows behaviors of a metal, where c (c ) is a creation (annihilation) operator for i; i; the appearance of Fermi `arcs' instead of a full Fermi an electron at site i with spin , S = c c is the i ; i; i; surface results into the violation of the conventional Lut- spin operator at site i ( is the vector of Pauli matri- tinger theorem of Fermi liquid theory. Furthermore, sur- ces), t = t and J is the spin-exchange interaction. No ij ji face spectroscopies like ARPES (see e.g. [4{8]) and scan- double occupancy constraint is ensured by taking n 1 ning tunneling spectroscopy [9{11] show that the mag- with n = c c being the number operator. The- i i; i; nitude of the anti-nodal (AN) gap is unchanged when ories then associate the formation of the PG phase as entering the superconducting (SC) phase below T (see a precursor of the Mott transition, involving some frac- Fig. 1a). The PG state persists up to a temperature T tionalization of the electron due to the constraint n 1. which decreases linearly with doping. The AN gap is also The typical and simplest realization of such a program is visible in two-body spectroscopy, for example in the B 1g to consider that the electron fractionalizes into \spinons" channel of Raman scattering[12{14], which shows that (f ) and \holons" (b ) subject to a local U(1) or gauge below T , the pair breaking gap of superconductivity fol- symmetry group, lows the T line with doping, in good agreement with the AN gap observed in ARPES. In contrast to the SC y y c = f b ; (2) phase, the PG state is found to be independent of dis- i i i order or magnetic eld. Despite of numerous invaluable experimental and theoretical investigations over the last where f is the creation operator for a fermion carry- three decades, various puzzles of the PG state remains to ing the spin of the electron, b is the annihilation oper- be solved. ation for a boson carrying the charge. The no double In this paper, we review three dierent theoretical per- occupancy constraint n 1 is replaced by an equality arXiv:1906.10146v1 [cond-mat.supr-con] 24 Jun 2019 2 PG SC+PG (π,π) UD50 T=10K (0,0) -0.2 -0.1 0.0 0.1 0.2 (eV) E-E (a) Figure 2. Symmetrized EDC near the antinode for underdoped Bi-2212 with a T of 50 K. Two features are seen in the spectrum: a low-energy peak associated (c) UD75 (b) UD65 with superconductivity and a broader feature at higher energy associated with the pseudogap. For such a deeply underdoped system, the intensity and energy position of the superconducting feature are strongly inﬂuenced FIG. by the 2:underlying The schematic phase diagram of cuprates pseudogap. superconductors as a function of hole doping [35]. On the left hand side, for 0 < p < 0:06, the system has an AF phase. The Green region below T denotes the coupling [9, 21], so it is not obvious a priori why we should characterize the broad hump in superconducting state. Below T the short range charge co ﬁgure 2 with the pseudogap. In principle, multiple components—superconductivity, pseudogap, order is observed, whereas the purple region is the mode-coupling and maybe others—contribute to the antinodal lineshape. The characterization charge order observed with application of a magnetic done in ﬁgure 2 is reasonable because of the increasing inﬂuence of the pseudogap in the eld. T is represented with a dashed line which underdoped regime, the doping dependence of the broadly peaked featureencloses at highertener he PG gy [phase 22], marked with yellow. In Region 3 and the proximity of this larger energy scale to the pseudogap energy scale and Region above T2. wThis e have additional source of damping, respectively due to the approach of the Mott transition picture will become clear upon studying more experiments, such as the momentum dependence and the proximity to the PG quantum critical point. of the superconducting gap. While a d-wave superconducting gap is a hallmark of cuprate high-temperature superconductivity, recent experiments have shown that the gap functions of underdoped systems constraint deviate from a simple d-wave form, 1(k) =|cos(k ) cos(k )|/2, near the antinode [23]–[27]. x y This is exempliﬁed in ﬁgure 3, which shows the low-temperature gap function of two y y f f + b b = 1; (3) i i i i lanthanum-based cuprates: La Ba CuO (LBCO) x = 0.083 and La Sr CuO (LSCO) 2 x x 4 2 x x 4 x = 0.11, adapted from [26]. The gap is extracted from the leading-edge midpoint, a model- which is enforced using a Lagrange multiplier [27]. In independent measure. Close to the node (the momentum position where the superconducting gap is zero), the gap follows a simple d-wave form, but in the antinodal this formalism, region (momentaf f gives the total fermion density i i 0.05 -0.10 -0.05 0.00 0.05 -0.10 -0.05 0.00 0.05 near the Brillouin zone axis), the gap increases more rapidly, deviating strongly from the near- and b b gives the total holon density. The invariance E-E (eV) E-E (eV) i F F y y nodal momentum dependence. This deviation is more pronounced for the more underdoped i 50 under the charge conjugation, f ! e f and b ! (f) UD65 i i (g) UD75 sample. Thus, the Fermi surface can be divided into two general regions with distinct i momentum 40 e b is implemented through the constraint in Eq. 3. dependences of the gap, although we note that the crossover may not be abrupt. 30 Other ways of fractionalizing the electrons of course exist, The same phenomenon is observed in underdoped Bi-2212. Figure 4 shows the gap 20 with the celebrated Z gauge theory leading to \visons" functions at 10 K for four dopings, partially adapted from [23] and [24]. At each momentum, 10 excitations [28, 29], or with extensions to higher groups gaps were extracted by ﬁtting symmetrized EDCs to a minimum model [28]. The sample with 0 0 like the pseudo-spin group SU(2)[30, 31] or the spin group the largest hole 1.0 concentration in ﬁgure 4 (UD92) 0.0 0.2 0.4follo 0.6 ws 0.8 a 1.0simple d-wave form all around the 0.0 0.2 0.4 0.6 0.8 1.0 SU(2) [32{34]. 0.5*|cos(kx)-cos(ky)| 0.5*|cos(kx)-cos(ky)| In all these cases the fractionalization of the electron (b) into various entities is eective in the ANR of the Bril- New Journal of Physics 12 (2010) 105008 (http://www.njp.org/) louin zone, opening a PG through the formation of small FIG. 1: a) The anti-nodal gap measured by ARPES hole pockets. In this framework, the global understand- spectroscopy, which shows two gaps, the rst energy ing of the phase diagram (as a function of doping) goes scale correspond to the PG scale E whereas the second in the following way (see Fig. 3a). A rst line increas- energy scale corresponds to the spectroscopic gap E . ing with doping, describes the Bose condensation of the spec The data have been symmetrized with respect to the holons. Implicit is the assumption that the system is energy E . b) Two sets of ARPES lines scanning the F three dimensional so that the Bose condensation T oc- Fermi surface of Bi1221 in the under-doped region. One curs at nite temperature. A second line T decreasing sees that the quasi-particle peak at E = 0 present in the with doping describes the PG phase and is typically as- nodal region (the upper curves) evolves into a Bogoliubov quasi-particle peak at nite E in the anti-nodal region (the lower curves). One can see how the Bogoliubov peak in the anti-nodal region is still visible, whereas somewhat broadened. Data taken from Ref. [7] Gap (meV) Intensity (A.U.) Gap (meV) Gap (meV) 46 NAGAOSA PATRICK A. LEE AND NAOTO It would seem a task to determine tivity. hopeless (2.5) solu- from first which of the principles many proposed t-J In this pa- tions is realized as a solution to the model. be- boson-boson has been as A interaction term dropped we set for ourselves a rather more modest goal. We per order in the of the J term. small (of x rewriting ing ) mean-field solution want to ask the question: Given a the mean-field decou- Equation (2.5) leads naturally to and including fluctuations around can we obtain a it, by pling' of which are consistent description physical quantities with the rather severe constraints set experiments? by (2.6) Anderson that fluctua- Baskaran and recognized tions about the RVB mean-field solution are naturally the same term can be written as Alternatively, theories. This was elaborated in a point gauge paper by re- Ioffe and and we shall draw from the Larkin, heavily We the sults of this work. find that, even though spin and of freedom are separated on the charge degrees mean-field are the level, they strongly coupled by gauge which would lead to the decoupling the we field. In order to reproduce photoemission data, mean-field Fermi surface need a theory with a spinon D (2. 7) (f & ifJ iffy Luttinger's con- which theorem. This leads us to obeys sider the uniform RVB state. A short version of this ab- At half-filling, there exists an symmetry' (an SU(2) earlier. of our main results is work was published One sence of a down is to an when spin equivalent spin up that a linear T resistivity emerges due to scattering there is a fermion so that the two by exactly single site), per field fluctuations. A mathematically related decou- are equivalent. For finite the two gauge decouplings x, behav- distinct mechanism for linear T though are distinct. For the bulk of this we shall physically plings paper and Ioffe and Kotliar =0. ior is given Ioffe Wiegmann. treat the while that decoupling assuming by D; g; related have also published a work which is closely very There are it convenient two rationales for this. First, is the one. to present formally to extend the sum to a sum over N spin degrees of freedom, and a large-1V In this perform expansion. MODEL LAGRANGIAN II. THE case it is clear that scales and there is no as N simple y; of . Grilli and Kotliar' have defining Indeed, way D; t-J with the model defined on a We shall begin square 'WO, found in the large-X the mean field that, limit, yI lattice =0 concentra- is stable for some intermediate doping D, = =2 = in the if we treat the E case 'n;n tion x J/t. Secondly, H t c +J S (2.1) —, ), c, g(S; o. I& we for intermediate i, saddle-point approximation, expect j, J 'WO of order while that y', below a temperature J, doping 'c, the sum over where tr n—;=g and ij c; c, S; , ttc;&, d-state lower tempera- at a develop symmetry may D; is to is over the nearest neighbor. Equation (2.1) subject where that there should be a temperature range ture, so occu- constraint that a site cannot be the important given 'WO =0. y' mean-field but Schematically the phase D, con- The constraint is more than one electron. pied by that shown in 1. There are look like Fig. diagram may slave-boson the veniently implemented using '%0 Below the solid line and, four different regimes. yI. method have uniform RVB when is d-state symmetric, we a D, (2.2) we have a Fermi-liquid state. In (b)%0 and =f;b; region I, c, in the heavy- similar to that which quite appears phase, carries the label that spin is a fermion operator where, treated Grilli This has been fermion problem. region by interpreted as ' that can be is a boson operator and b, =0. shall but b We Kotliar. In WO and region II, ( ) D, occu- The constraint of no double a vacancy. creating is now replaced pancy by =l, (2. 3) +b, b, f; gf; t, formu- in functional which can be implemented a integral T IV field and a Grassmann field with la over a complex b II over an additional field A. on each the integration site, III — P Z d A, db;*df; df;* (Xo+H)dr db; exp f f of FIG. 1. Schematic slave-boson mean-field phase diagram where t-J of the uniform the model. The solid line denotes the onset Bose- mean-field The dotted line denotes the RVB state. while the of the boson Einstein condensation temperature fermion opera- denotes the onset of of the dashed line pairing are Fermi (II) spin-gap phase, tors. The four regions (I) liquid, *b, +i A. (2.4) 1), -(f, +b, f, and strange metal phase. , (III) superconductor, (IV) sociated to a precursor of pairing. For example in the U(1) theory, the T line corresponds to the condensation D E y y of spinon pairs f f on a bond. Strong coupling i j theories of the PG with electron fractionalization can be considered as various examples of the celebrated Reso- nant Valence Bond (RVB) theory [36] introduced in the ⇤ early days after the discovery of cuprates, where some sort of spin liquid associated with uctuating bond states is considered as the key ingredient for the formation of the PG [37, 38]. When the two lines cross, below T and T the electron re-con nes so that the SC transition is also a con ning transition from the gauge theory perspec- tive. Above is the strange metal (SM) phase (Region IV) whereas on the right we nd the typical metallic- Landau Fermi liquid phase (Region I). (a) Although this line of approach has been supported by a huge body of theoretical work, we are skeptical that this is the nal solution for the PG phase. First, fractional- ization of the electron produces, when it happens, very diag !7 spectacular eects like the quantization of the resistivity, observed for example in the theory of fractional Quan- tum Hall Eect (QHE). Here no spectroscopic signature of \recon nement" of the electron has been observed at nite energy. Moreover, in most of the under-doped re- gion of cuprates, the ground state is a superconductor. AF If the electron fractionalize at T and that the ground state is a superconductor, it means that at T the elec- tron shall re-con ne, form the Cooper pairs and globally freeze the phase of the Cooper pairs. Of course it is possi- ble but quite unlikely. A key experiment which illustrates this feature is maybe ARPES measurements which shows the presence of Bogoliubov quasi-particle in the ANR of the Brillouin zone, in the under-doped region inside the SC phase (see e.g. [7]). Taken at its face value this ex- (b) perience clearly hints at the formation of the SC state on the whole Fermi surface, including the ANR, rather than on small hole pockets centered around the nodes (see Fig.1b). In this paper, we will now focus on the two other avenues of investigations which are the uctuations and the hidden phase transition at T . 2. PHASE FLUCTUATIONS The realization that phase uctuations are important in the under-doped region of the cuprates stems back from the seminal paper by Emery and Kivelson, which (c) pointed out that close to a Mott transition, the local- FIG. 3: a) The phase diagram from strong coupling ization of the electrons induces strong phase uctuations theories. The rst line, below which the Bose of all the elds present in the system [40]. The origin condensations (\holons") occurs, and the second line, of these uctuations is deep and comes from the Heisen- below which sees the formation of \spinon-spinon" berg uncertainty principle, in which the particle/wave pairing, cross each other resulting into formation of a duality ensures that the localization of particles in space spin liquid or \Resonating Valence Bond" (RVB) state. will generate phase uctuations. Careful study of the b) Schematic picture of the eight hot spots (EHS) penetration depth as a function of doping shows that model. The dots in red are the eight hot spots. The cuprates are in the class of superconductors where the diagonal modulation wave vector is shown in blue (from phase of the Cooper pairs strongly uctuates at T lead- Ref. [39]) c) Schematic depiction of bond variables on a square lattice. 4 ing to a Berezinsky Kosterlitz Thouless transition typical uctuations is present in many theories for cuprates. A of strong 2D uctuations (see e.g. [41, 42]). Early exper- quick glance at the phase diagram of the cuprates would iments showed a linear dependence between the T and convince anyone that the most prominent features are the penetration depth, giving a lot of impetus to the uc- the AF phase and the SC phase. Hence as a natural rst tuations scenario [43, 44]. guess, the rotation from the SC to the AF state has been Another phenomenology which can be successfully ex- tried, leading to an enlarged group of SO(5) symmetry [60{62]. The ten generators of the SO(5) group corre- plained by phase uctuations of the Cooper pairs, lies in spectroscopic studies of the spectral gap associated to the spond to transitions between the various states inside the quintuplet (three magnetic states, and two SC states transition at T . Surface spectroscopies like STM and ARPES tell us that the magnitude of the spectral gap which are complex conjugate of one another). These theories are based on an underlying principle that the associated to the PG, which lies in the ANR of the Bril- louin zone, is unchanged when one goes through the SC `partners' are nearly degenerate in energy. The `part- ners' posses an exact symmetry in the ground state in T [45, 46], whereas the gap broadens more and more un- some nearby parameter regime, for example the SC and til T . On the other hand, in the nodal region, the spec- the AF states show exact degeneracy at half lling. At tral gap vanishes at T as it should within the standard other parameters, the symmetry is realized only approxi- BCS theory. Moreover, we learn from Raman scattering spectroscopies that in the B channel, which scans the mately in the ground state. The `hidden' exact symmetry 1g emerges when the system is perturbed from the ground ANR, the spectral gap follows the T line as a function of doping, but is at the same time associated to the \pair state by increasing the temperature or the magnetic eld. In the framework of emergent symmetries, the PG is cre- breaking" below T [13]. ated by uctuations over a wide range of temperature. These considerations lead to a very strong phe- Later on, theories based on a SU(2) gauge structure in nomenology based on the phase uctuations of the Coop- the pseudo spin space led the formation of ux phase [25], ers pairs, opening a channel for damping between T and and rotation from the SC state to ux phases was envi- T and giving a remarkable explanation for the lling of sioned. This theory had an emergent SU(2) symmetry the AN spectral gap with temperature, up to T [46{ with uctuations acting as well up to temperatures T . 52]. This theory was corroborated by the observation of an enormous Nernst eect going up to T in the Lan- thanum and Bismuth compounds which was interpreted as the presence of vortices up to a very high temperature reaching very close to T [53{55]. A long controversy followed, in order to determine the extend of the temperature regime of phase uctuations. Further experimental investigations, including Nernst ef- fect on YBCO compounds [56]-where, contrarily to the LSCO compounds, the contribution of the quasi-particles Recently, new developments have shown observation of has opposite sign from the contribution of the vortices, charge modulations in the underdoped region of cuprate transport studies [57] and direct probing with Josephson superconductors, in most of the compounds through var- SQUID experiments [58], led to the conclusion that, the ious probes, starting with STM [9, 10], NMR [63, 64] and phase uctuations extend only a few tens of degrees above X-ray scattering [65{68]. At high applied magnetic eld, T but do not reach up to T . This issue can be resolved the charge modulations reconstruct the Fermi surface, if we consider a competition of the SC phase uctuations forming small electron pockets [69{76]. In YBCO, the with uctuations of a partner-competitor like particle- charge modulations are stabilized as long range uniaxial hole pairs, then it has been shown, through the study of Q = (Q ; 0) 3D order above a certain magnetic thresh- a non linear -model (NLM) [59], that the true region old [77{79] and the thermodynamic lines can be deter- of the SC phase uctuations is reduced to a temperature mined with ultra-sound experiment [80, 81]. With the window close to T . In the same line of thoughts, we ex- ubiquitous observation of charge modulations inside the amine here the possibility of extending uctuations to a PG phase, a rotation of the SC state towards a Charge bigger symmetry group, where not only the phase of the Density Wave (CDW) state has been proposed [82{86]. Cooper pairs uctuates but also there is a quantum su- This rotation has also two copies of an SU(2) symme- perposition of Cooper pairs with a \partner-competitor" try group where one rotates the SC state to the real eld. and imaginary parts of the particle-hole pair where (i; j) are sites on a bond (see Fig.3c) with r = r a j i x;y and r = (r + r ) =2 and Q is the modulation wave vec- i j 2.1. Extended symmetry groups and the case for tor corresponding to the CDW state. In the case of the SU(2) symmetry eight hot spots (EHS) model (de ned after Eq. 6), Q is the the diagonal wave vector Q = (Q ; Q ) (shown in 0 0 The idea to \rotate" the d-wave superconductor to an- Fig.3b) but experimentally, it is axial with Q = (0; Q ) other `partner', hence enlarging the group allowed for or Q = (Q ; 0). We have the two l = 1 representations 0 5 ( ; ; ) with 1 0 1 y y i(i+ ) ^ j = p d c c e ; i j h i a y iQr+i(i ) y iQri(i ) j j = d c c e + c c e ; j i 0 i j h i y y b iQr+i(i ) iQri(i ) ^ j j = d c c e + c c e ; j i i j i(i+ ) ^ j = d c c e (4) 1 j i and the corresponding - operators satisfy the SU(2) algebra, (a) [ ; ] = l (l + 1) m (m 1) ; m m1 [ ; ] = m ; (5) z m m with (note that is identical for both representations a and b) y y y y a iQr 2i iQr 2i i j = [c c e e + c c e e ]; i i j j a iQr 2i iQr 2i i j = [c c e e + c c e e ]; i i j j (b) b y y iQr 2i y y iQr 2i i j FIG. 4: a) Charge modulations observed in the vortex = [c c e e + c c e e ]; + i i j j core, from the seminar paper [10]. The data were taken at magnetic eld B = 0T and B = 8T and then b iQr 2i iQr 2i i j = [c c e e c c e e ]; subtracted. Charge modulations \appreared" inside the i i j j vortex core. b) Schematic picture of a Skyrmion in the 1 1 -space corresponding to the O(3) NLM in Eq.(15). = [ ; ] = (n ^ + n ^ 1) : (6) z + i j 2 2 The axes are (m ; m ; m ). x y z An exact realization of the emergent symmetry has been found, where the Fermi surface is reduced to eight where is the gap in the particle-hole channel whereas \hot spots" (crossing of the Fermi surface and the AF is the gap in the particle-particle channel. zone boundary as shown in Fig.3b) and the electrons in- Below T , a great amount of uctuations are present teract with an AF critical mode in d = 2 [82{84, 87]. which are described by the O(4) NLM [82, 88, 89]. For In this simple model, the gap equations could be solved the four- elds n , = 1 : : : 4, with n = ( + ) =2, 1 1 1 a b showing the exact SU(2) symmetry between the Cooper n = ( ) =2, n = , n = the eective 2 1 1 3 4 0 0 pairing and particle-hole channels. We observed some or- action writes dering of a composite order parameter, which is a super- 4 4 X X position of gaps in the particle-particle and particle-hole 2 2 S = 1=2 d x (@ n ) ; with jn j = 1: (9) channels =1 =1 y y y j0i = c c + c c j0i ; (7) k+Q k k k 2.2. What can the model explain at this stage? where the modulations in the charge sector occur at a nite diagonal wave vector Q = (Q ; Q ), where Q and x y x 2.2.1. Charge modulations inside the vortex core Q are the distance between two hot spots, as pictured in Fig.3b. A composite gap is formed at T with the mean At this stage, the model can already explain a few square of the gaps in each channel properties of cuprate superconductors. Since the model treats very seriously the competition between CDW and 2 2 E = jj +jj ; (8) SC pairing order parameters, it predicts charge modu- Vortex liquid LETTER doi:10.1038/nature10345 Magnetic-field-induced charge-stripe order in the high-temperature superconductor YBa Cu O 2 3 y 1 1 1 1 1 2,3 2,3 2,3 ¨ ´ Tao Wu , Hadrien Mayaffre , Steffen Kramer , Mladen Horvatic , Claude Berthier , W. N. Hardy , Ruixing Liang , D. A. Bonn NATURE 1 PHYSICS DOI: 10.1038/NPHYS2502 LETTERS & Marc-Henri Julien LETTER RESEARCH 11–13 Electronic charges introduced in copper-oxide (CuO ) planes reconstruction . Thus, whatever the precise profile of the static 6 a co generate high-transition-temperature (Tc) superconductivity but, charge modulation is, the reconstruction must be related to the trans- 4.2 K under special circumstances, they can also order into filaments lational symmetry breaking by the charge ordered state. Leboeuf et al, Nature (2013) 10 1 a b called stripes . Whether an underlying tendency towards charge The absence of any splitting or broadening of Cu2E lines implies a 9.9 K 120 order is present in all copper oxides and whether this has any one-dimensional character of the modulation within the planes and lations inside the 14.9 K vortex core (Fig.4a)[9{11, 63, 90]. In- relationship with superconductivity are, however, two highly con- imposes strong constraints on the charge pattern. Actually, only two 28.5 T 2,3 Static 30 troversial issues . To uncover 33.5 T underlying electronic order, mag- types of modulation are compatible with a Cu2F splitting (Fig. 2). The 19.6 K deed, since the SC order parameter vanishes there, the netic fields strong enough to destabilize superconductivity can be –1 charge order first is a commensurate short-range (2a or 4a period) modulation NMR 4–6 24.9 K 15 T used. Such experiments, including quantum oscillations in running along the (chain) b axis. However, this hypothesis is highly competing order emerges at the core. The special struc- co YBa Cu O (an extremely clean copper oxide in which charge unlikely: to the best of our knowledge, no such modulation has ever 2 3 y 15 T order has not until now been observed) have suggested that super- been observed in the CuO planes of any copper oxide; it would there- ture associated to this feature is called a meron, or half 7–9 conductivity competes with spin, rather than charge, order . Here fore have to be triggered by a charge modulation pre-existing in the –2 we report nuclear magnetic resonance measurements showing that filled chains. A charge-density wave is unlikely because the finite-size skyrmion, in the pseudo-spin space. It can be noted Superconducting high magnetic fields actually induce charge order, without spin chains are at best poorly conducting in the temperature and doping 11,14 order, in the CuO2 planes of YBa2Cu3Oy. The observed static, uni- range discussed here . Any inhomogeneous charge distribution that it is a generic prediction of the theories of emer- B co directional, modulation of the charge density breaks translational such as Friedel oscillations around chain defects would broaden rather gent symmetries, that the competing order shows up 15 symmetry, thus explaining quantum oscillation results, and we than split the lines. Furthermore, we can conclude that charge order c d argue that it is most probably the same 4a-periodic modulation occurs only for high fields perpendicular to the planes because the ∆ 0.06 1 inside the vortex core. For example, the SO(5) theory as in stripe-ordered copper oxides . That it develops only when NMR lines neither split at 15 T nor split in a field of 28.5 T parallel ¬2 28.5 T superconductivity fades away and near the same 1/8 hole doping to the CuO planes (along either a or b), two situations in which predicts AF correlations inside the vortex core [91, 92], as in La Ba CuO (ref. 1) suggests that charge order, although 22x x 4 superconductivity remains robust (Fig. 1). This clear competition 33.5 T B visibly pinned by CuO chains in YBa Cu O , is an intrinsic pro- between charge order and superconductivity is also a strong indication 2 3 y BQ13456 PRB May 28, 2018 13:44 CO 0.04 whereas the SU(2) symmetry which rotates superconduc- Field-induced pensity of the superconducting planes of high-T copper oxides. that the charge ordering instability arises from the planes. YBCO The Vortex solid ortho II structure of YBa Cu O (p5 0.108, where p is the The only other pattern compatible with NMR data is an alte char rnatge or ion of der 2 3 6.54 tivity to the - ux phase predicts that the - ux orbital p = 0.108 m hole concentration per planar Cu) leads to two distinct planar Cu more and less charged Cu2F rows defining a modulation with a period Spin NMR sites: Cu2F are those Cu atoms located below oxygen-filled 0.02 of four lattice spacings along the a axis (Fig. 2). Strikingly, this corre- state [16, 93] is present inside the vortex core. Although order ¬4 chains, and Cu2E are those below oxygen-empty chains . The main sponds to the (site-centred) charge stripes found in La Ba CuO at 0 22x x 4 15 T 0 discovery of our work 20 is that, on cooling in a field 40 H of 28.5 T along the d60 oping levels near p5 x5 0.125 (ref. 1). Being a proven electronic AF correlations were observed in the vortex core in the 15 T 0 0.04 0.08 0.12 0.16 c axis (that is, in the conditions for which quantum oscillations are instability of the planes, which is detrimental to superconductivity , Temperature (K) La-compounds [94], for YBCO, BSCCO and Hg-series, resolved; see Supplementary Materials), the Cu2F lines undergo a stripe order not only provides a simple explanation of the NMR splitting p (hole/Cu) profound change, whereas the Cu2E lines do not (Fig. 1). To first order, but also rationalizes the striking effect of the field. Stripe order is also 0 10 20 30 e f STM experiments [9, 10] and NMR [90] gave evidence 2.0 this change can be described as a splitting of Cu2F into two sites having fully consistent with the remarkable similarity of transport data in (a) Figure 4 | Phase diagram of underdoped YBa Cu O . The charge ordering CHAKRABORTY, MORICE, AND PÉPIN PHYSICAL REVIEW B 00,004500(2018) 2 3 y Figure 2 | Thermodynamic phase diagram. Magnetic ﬁeld–temperature B (T) both different hyperfine shifts K5 Æhzæ/H0 (where Æhzæ is the hyperfine YBa2Cu3Oy and in stripe-ordered copper oxides (particularly the for charge modulations. This ubiquitous observation of temperature T (defined as the 11–1ons 3 et of the Cu2F line splitting; blue open field due to electronic spins) and quadrupole frequencies n (related to dome-shaped dependenc cha e of rgeT around p5 0.12) . However, stripes 1.5 Q 0 phase diagram of underdoped YBCO (p = 0.108) obtained from the the electric field gradient). Additional effects might be present (Fig. 1), mustcir becle pars) allcoi el frncid om pla esnewit to h plaTne in (br Yown Ba Cuplu O ,swh siegns reas),ththe ey are temperature at which the Hall 2 3 y charge modulations inside the vortex core is a nice test for c anomalies seen in butthe theyelastic are minorc inonstant comparisoncwith(F theig. obs1). erveBlack d splittinsquar g. Change ess indicat perpende icuthe lar in, for example, La Ba CuO . We speculate that this 29.5 K 11 22x x 4 87 measurements suggest the presence 11 of an incommensurate 1.0 constant R changes its sign. T is considered as the onset of the Fermi surface H 0 0 in 50field-dependent an 100 d temper 0 ature-dependen 50t orbital occupancy 100 (for explains why the charge transport along the c axis in YBa Cu O 2 3 y 11–13 the theory of emergent SU(2) symmetry, but the presence 34.8 K transition from a vortex lattice to a vortex liquid at B , which cannot be m reconstruction . The continuous line represents the superconducting 88 unidirectional three dimensional (3D) CO with out-of-plane example d 2 2 versus d 2 2) without on-site change in electronic becomes coherent in high fields below T (ref. 15). If so, stripe fluctua- T (K) x {y z {r T (K) 0 density are implausible, and any change in out-of-plane charge density tions must be involved in the incoherence along c above T . 40.1 K transition temperature T . The dashed lin 0 e indicates the speculative nature of resolved above 40 K. Red circles correspond to the phase transition c of charge modulations inside the vortex core could also 89 correlation length ∼ 10 lattice spacings. The 3D CO has the or lattice would affect Cu2E sites as well. Thus, the change in n can Once we know the doping dependence of n (ref. 16), the difference Q Q the extrapolation of the field-induced charge order. The magnetic transition 42.2 K towards static charge order at B , as observed in c . The error bars on the g 8 only arise from a differco entiation in the charge dens11ity between Cu2F Dn 5 3206 50 kHz for p5 0.108 implies a charge density variation be explained 90 samebin-plane y a strong incommensuration competitionasbthe etw 2D een counterpart. SC andThis Q temperatures (T ) are from muon-spin-rotation (mSR) data (green stars) . T sites (or at the oxygen sites bridging them). A change in the asymmetry as small as Dp5 0.036spi 0.n01 hole between Cu2Fa and Cu2Fb. A 0 43.5 K ﬁeld scale B (B ) correspond to the width of the transition in the m co 91 indicates that the appearance of the 3D CO is somehow related ¬2 parameter and/or in the direction of the principal axis of the electric canon and icalTstripevan descis ripti h clo on (se Dpto5the 0.5 hosam le) ise the criref tica orel con inadeq cen uate tration p5 0.08. A scenario of CDW without invoking any emergent symmetry [95, 96]. spin 44.8 K field gradient could also be associated with this charge differentiation, at the NMR timescale of ,10 s, at which most (below T ) or all derivative (raw data) of c (B). The charge-order transition is almost 0 92 to the 2D long-range CO. At high magnetic ﬁelds, all these field-induced spin order has been predicted for p. 0.08 (ref. 8) by analogy with but these are relatively small effects. CO 4 (above T ) of the charge differentiation is averaged out by fluctuations 50 K 0 5 21 temperature independent up to ⇡ 40 K. Above 40 K the ﬁeld scale BLaat Sr CuO , for which the non-magnetic ground state switches to co 1.855 0.145 4 93 experiments thus demonstrate the appearance of a long-range The charge differentiation occurs below T 5 506 10 K for faster than 10 s . This should not be a surprise: the metallic nature of charge p5 0.108 (Fig. 1 and Supplementary Figs 9 and 10) and 676 5 K for the com antpoun iferro d atma allgne fieldstic isor incom derpati inble fiewith lds ful grleat chaer rgetha ordn er,aevfe enw teslas (ref. 7 and references which charge order sets in rises. In the Supplementary Information, we 94 CO irrespective of its structure. 17 p5 0.12 (Supplementary Figs 7 and 8). Within error bars, for each of if this order is restricted to the direction perpendicular to the stripes . therein). Our work, however, shows that spin order does not occur up to,30 T. argue that the ovthe erall sampl beha es Tviour coinci ofdes the with char T , g thee-or tempe der rature phase at which boundary the Actualin ly,this there is compelling evidence of stripe fluctuations down to 0 charge 0 95 Acompetition between theCO and the superconductivity 2.2.2. B-T phase diagram In contrast, the field-induced charge or 18der reported here raises the question of 0 20 40 60 80 100 Hall constant R becomes negative, an indication of the Fermi surface very low temperatures in stripe-ordered copper oxides , and indirect B–T diagram is consistent with a theoretical model of superconductivity in 96 ¬4 (SC) is already noticeable at zero or moderate magnetic ﬁelds. T (K) whether a similar field-dependent charge order actually underlies the field 1 2 ∆ Laboratoire National des Champs Magne´tiques Intenses, UPR 3228, CNRS-UJF-UPS-INSA, 38042 Grenoble, France. Department of Physics and Astronomy, University of British Columbia, Vancouver, competition withBritish a densit Columbia V6T y-1Z1, wa Canada. ve stat Canadian e Institute . The for Advanced green Research, diamond Toronto, Ontario isM5G the 1Z8, Canada. dependence of the spin order in La Sr CuO and YBa Cu O . Error bars 22x x 4 2 3 6.45 97 First evidence of this competition can be viewed from the Figure 3 | Slow spin fluctuations instead of spin order. a, b, Temperature NMR represent the uncertainty in defining the onset of the NMR line splitting (Fig. 1f This leads us to a second set of experimental evidence, temperature T = 6350±10 K at which NMR experiments detect the onset 8SEPTEMBER 2011 | VOL 477 | NATURE | 191 98 suppression of the zero ﬁeld T in the same doping range where co c dependence of the planar Cu spin-lattice relaxation rate 1/T for p5 0.108 ©2011 Macmillan Publishers Limited. All rights reserved and Supplementary Figs 8–10). explicitly the phase diagram in the presence of an applied of a charge modulation at a ﬁeld B = 28.5 T in YBCO at doping p = 0.11 (a) and p5 0.12 (b). The absence of any peak/enhancement on cooling rules 99 the short-range CO is observed [23]. Second, the intensity out the occurrence of a magnetic transition. c, d, Increase in the Cu spin–spin (ref. 4). Within the error bars, this onset temperature agrees with our magnetic 100 eld of the aszero-ﬁeld described x-ray in Figs. scattering 5a, 5bCO [35peaks , 80]. decreases For the for fluctuations strongly enhances the spin–lattice (1/T ) and spin–spin relaxation rate 1/T on cooling below,T , obtained from a fit of the spin- 2 charge ﬁndings. Dashed lines are guides to the eye. 101 T< T [23,36]. The presence of a moderate magnetic ﬁeld compound ¬6YBCO, a phase diagram could be derived as a a (1/T ) relaxation rates between T and T for La nuclei. For echo decay to a stretched form s(t) / exp(2(t/T ) ), for p5 0.108 (c) and 2 charge spin 102 reduces this decrease [23,26,36]. The competition can be p5 0.12 (d). e, f, Stretching exponent a for p5 0.108 (e) and p5 0.12 (f). The the more strongly hyperfine-coupled Cu, the relaxation rates become function of an applied magnetic eld up to roughly 20T . with a density-wave order (see discussion in the Supplementary 103 further substantiated by scanning tunneling microscopy, which deviation from a5 2 on cooling arises mostly from an intrinsic combination of so large that the Cu signal is gradually ‘wiped out’ on cooling below FIG. 1. A schematic B-T phase (b) diagram of underdoped YBCO For this speci c compound, one observes at H = 17T c 63 Gauss Information). ian and exponentFor ial decT ays,below combine40 d with K or some so,spatia static l distrcharge ibution oforder sets 0 10 20 30 T (refs 18, 23, 24). In contrast, the Cu(2) signal here in 104 detects the short-range CO in regions of space where the charge summarizing various experiments. Type-II superconductors have two [80], a second order phase transition towards a 3D charge T values (Supplementary Information). The grey areas define the crossover 2 in only above a threshold field of 18 T, akin to the situation in YBa Cu O does not experience any intensity loss and 1/T does not B (T) 2 3 y 1 105 amplitude of the superconductivity is reduced (both near the critical ﬁelds: a lower critical ﬁeld (B )andauppercriticalﬁeld(B ). FIG. 5: a) Experimental B c 1 T phase diagram from c2 temperature T below which slow spin fluctuations cause 1/T to increase slow 2 order (CO) state with one uniaxial vector of modulations La Sr CuO (x = 0.145) in which a magnetic field is necessary show any peak or enhancement as a function of temperature (Fig. 3). 2 x x 4 106 vortex core [54]at low ﬁelds andsurroundingZn impurities The system completely expels magnetic ﬁelds for B< B showing and to become field dependent; note that the change of shape of the spin-echo c1 Ref.[80]. The shape of the lines is done with ultrasound Moreover, the anisotropy of the linewidth (Supplementary Figure 1 | Field dependence of the sound velocity in underdoped to destabilize superconductivity and to drive the system to a [77, 97]. The shape of this transition is very at in tem- 107 [32]atzero magnetic ﬁeld). decay occurs at a slightly higher (,115 K) temperature than T . T is the Meissner effect and allows magnetic ﬁeld ﬂux slowlines slowto penetrate Information) indicates that the spins, although staggered, align mostly experiments. b) Schematic phase diagram described YBa Cu O . a,b, Field dependence of the longitudinal mode c magnetically ordered state . Close to the onset temperature of slightly lower than T , which is consistent with the slow fluctuations being a perature 2 3[80 y , 81], a fact that cannot be accoun11 ted by a charge 108 In this paper, we will focus on the competition between at various locations (called vortices) for B <B <B .Cuprates c1 c2 along the field (that is, c axis) direction, and the typical width of the consequence of charge-stripe order. The increase of a at the lowest through the SU(2) theory for competition between (propagation q and polarization u of the sound wave along a axis) in static charge order, T , the threshold field B sharply increases co co simple mo 109 del the SC of comp and the etition long-range betwCO eenatthe high twmagnetic o orders, ﬁelds. butThis have very low B and so form vortices with 2 1/2the application of a very c1 central lines at base temperature sets an upper magnitude for the static temperatures probably signifies that the condition cÆh æ t = 1, where t is z c c underdoped YBCO (p = 0.108) at different temperatures from T = 4.2K to and the phase boundary tends to become vertical. This is in charge modulations and Cooper pairing [35]. 23 110 competition is prominent in the B-T phase diagram of under- small magnetic ﬁeld. B varies with temperature as shown in the c2 can be explained by a pseudo-spin op transition, where spin polarization as small as gÆS æ# 23 10 m for both samples in the correlation time, is no longer fulfilled, so that the associated decay is no z B T = 24.9K (a), and from T = 29.5K to T = 50 K (b). The curves are shifted agreement with the theoretical phase of competing order with ﬁgure and its exact proﬁle depends on the speciﬁcs of the sample. 111 doped cuprates. In Fig. 1,weshow a schematic B-T phase longer a pure exponential. We note that the upturn of 1/T is already present at fields of,30 T. These consistent observations rule out the presence of the system suddenly goes from the SC state to the CO. for clarity. The measurements were performed in static magnetic ﬁeld up to superconductivity that predicts that superconducting fluctuations 15 T, whereas no line splitting is detected at this field. The field therefore affects At low temperatures, the vortices form periodic arrays called vorte magxnetic order, in agreement with an earlier suggestion based on the 112 diagram summarizing various experiments on underdoped The model of pseudo-spin op comes directly from the ex- the spin fluctuations quantitatively but not qualitatively. g, Plot of NMR signal 28 T. Black arrows indicate the ﬁeld B corresponding to the vortex lattice have no significant effect on charge order in this part of m solid with local short-range charge modulations inside the vortex cpre ore. sence of free-electron-like Zeeman splitting . 113 (0.11 ! p ! 0.13) YBCO. Our endeavor in this work will be CDW states are the generators of an O(3) Lie algebra intensity (corrected for a temperature factor 1/T and for the T decay) against pression of the NLM where the constraint plays the role 2 melting. At low temperature, the loss of the vortex lattice compression the phase diagram. Increasing the temperature, this solid melts for B> B .Though BIn stripe-ordered copper oxides, the strong increase of 1/T on m m 2 114 to understand the following salient features of the long-range temperature. Open circles, p5 0.108 (28.5 T); filled circles, p5 0.12 (33.5 T). of the modulus value canof be the estimat spin ed and in a is in magnetic agreementspin- op with previous transition. studies (see We now0 turn to the analysis of the symmetry 1 of the coo charge ling below T is accompanied by a crossover of the time decay has a different temperature dependence than B , it intersects the B charge c2 c2 115 CO in the B-T phase diagram: The absence of any intensity loss at low temperatures also rules out the presence of the spin-echo from the high-temperature Gaussian form Supplementary Information). For T > 40 K, B cannot be resolved. Red modulation. line in the tw Ino dthe ifferframework ent limits of zof ero the tempLandau erature antheory d zero mof agnephase tic The pseudo-spin op transition accoun m ts for the atness of magnetic order with any sig nificant moment. Error bars represent the added 116 (1) The long-range CO phase is associated with an onset 2 18,23 a a @ A exp(2K(t/T ) ) to an exponential form exp(2t/T ) . A similar 2G 2E arrows indicate the ﬁeld B where the charge-order phase transition uncert transitions, ainﬁtieesldL in. sig A =tnal an hiana gh anomaly lys mis, agexp neteri ic in men ﬁ ethe ltal d, con selastic ysdit teion m ssconstant and h0owTs mea a looccurs sur ngeme -ranntg;s.at e cahaphase rg(10) e co 2 [98] in temp 117 magnetic erature ﬁeld of (the B ),transition, which is found buttoibe t do surprisingly es not ex- insen- + co m n crossover occurs here, albeit in a less extreme manner because of the All measurements are with H || c. 1 a a occurs. This transition is not related to vortex physics because it is also transition densityonly waveiforder a coupling with the in transition the free energy ﬁeld BF .=Bg isQinsensiti " (where ve co co + c 0mn plain the 118 phase sitive toof theco-existence temperature and [80remains , 90, 99 ﬂat ].up Ac tocoun the scale tingof T z c + 2 absence of magnetic order: 1/T sharply increases below T and the 2 charge seen in acoustic modes c and c (Fig. 3 and Supplementary Fig. S3), m and to tenmare peraintegers ture at low and temgperaisturaescoupling and markconstant) s a sudden between rise. T isthe co 44 55 mn 119 [41,48,55]. evidence (explaining the rotational symmetry breaking) over a broad decay actually becomes a combination of exponential and Gaussian for this phase requires to break the exact SU(2) symme- the temperature at high magnetic ﬁelds where the charge order marks which are insensitive to the ﬂux line lattice because those modes involve order parameter Q and the strain " is symmetry allowed, that is, temperature range in YBa Cu O (refs 14, 19–22). Therefore, instead 120 (2) At high magnetic ﬁelds, the long-range CO transi- decays (Fig. 3). In Supplementary Information we provide evidence where only the op 2 erator 3 y in Eq.(6) contribute (since try underlying the NLM by an amount of roughly 5%. m n a transition to the pseudogap phase. T is effectively insensitive to co atomic motions parallel to the vortex ﬂux lines (uk Hk c). only if Q and " transform according to the same irreducible of being a defining property of the ordered state, the small amplitude of that the typical values of the 1/T below T imply that antiferro- 121 tion temperature (T )isnearly independent of themagnetic 2E charge co we rotate to the real part of the particle-hole pair), and As a conclusion, an SU(2) emergent symmetry and its the magnetic ﬁeld. The green region is the pseudogap phase where the representation charge differentiation . Inis Fig. more3likwe ely compare to be a conseq theuenc field e ofdependence stripe at magnetic (or ‘spin-density-wave’) fluctuations are slow enough to 122 ﬁeld [41,55]. Remarkably, T is very close to zero-ﬁeld co where the notation stands for a hermitian 17 matrix. The 212 one looses any coherence of either charge or superconducting order. NLM can account for the whole phase diagram owing phase stabilized by the magnetic field above B is straightforward. 0 order T (th = 20 e sm Kecti ofcfour phasedifferent of an electr modes onic liqcuid , ccry,sta c l and ) remcainithat ng display appear frozen on the timescale of a cyclotron orbit 1/v < 10 s. co 11 44 55 66 c 123 T [27,28,41,56]and signiﬁcantlylower than T (transition co The pseudogap phase persists for temperatures below T , which is structure of L in Eq.(10) give rise to collective modes. partly fluctuating (that is, nematic). In principle, such slow fluctuations could reconstruct the Fermi sur- to High-field the fact that NMRitmeasurements is broken byin5% YBCO [35].at similar doping have an anomaly at B . To explain the presence of such coupling for co 124 temperature of the short-range CO) in the doping range 0.11 ! 25,26 very large compared to T or T . At low temperatures, the magenta In stripe copper oxides, charge or c der at coT 5 T is always accom- face, provided that spins are correlated over large enough distances These collective modes are spin char zero, ge charge two, and re- shown that charge order develops above a threshold field B > 15 T all these modes, we rely on group theory arguments. YBCO is co 125 p ! 0.13. In fact, T is bound by zero-ﬁeld T for all doping. co c charge order region merges with the blue superconducting order RMN panied by spin order at T , T . Slowing down of the spin (see also ref. 9). It is unclear whether this condition is fulfilled here. The spin charge and below T = 50 ± 10 K (ref. 4). Given the similar field and an orthorhombic system (point group D ), and given the even ect the structure of the SU(2) symmetry 2h . They can be co 126 (3) Sound velocity measurements [41,55], NMR [27,28], region showing the coexistence of both the orders. The short-range temperature scales, it is natural to attribute the anomaly seen in character of the strains we have only to consider the character table considered as Pair Density Wave (PDW) excitations since 8SEPTEMBER 2011 | VOL477 | NATURE | 193 127 and x-ray scattering [51–53]measurements all indicatethat charge order is present even for low magnetic ﬁelds, but is not shown the elastic constant 2.2.3. at B to Col the lective thermodynamic modes transition towards of point group D shown in Table 1. co 2 ©2011 Macmillan Publishers Limited. All rights reserved 128 there exists a coexisting phase in the B-T phase diagram at low they have non zero center of mass wave vector. They in this schematic. the static charge order. To represent the different symmetric charge modulations that 129 temperatures. This is further illustrated by the identiﬁcation could be responsible for the mode observed in the A 1g The phase diagram in Fig. 2 shares common features with the transform according to each irreducible representation of the point 130 of the upper critical ﬁeld of the superconductor [57–59]to where the SC is suppressed partially. This CO inside each 147 An emergent symmetry is characterized by a set of channel in Raman Scattering [12] and can also be seen by theoretical phase diagram of superconductivity in competition group D and to determine to which acoustic mode they couple, we 131 be higher than the onset ﬁeld of the long-range CO at low halo ﬂuctuates enormously with no long-range CO. It was 148 collective modes [100]. Let us assume that the CDW is spectroscopy experiments, like X ray, MEELS [101, 102] 132 temperatures. postulated that only an interlayer coupling [61]oraﬁnite 149 real, as in the rst set of Eq.(4). Here the operators or soft X-rays [103], where the resolution in q-space can 80 NATURE PHYSICS | VOL 9 | FEBRUARY 2013 | www.nature.com/naturephysics 133 In particular, we argue that the ﬂatness of B is a signature magnetic ﬁeld [27,61](inducingvortex-vortexinteraction)can 150 co and which enable the \rotation" from the SC to the be traced. The theory predicts that the mode occurs 134 of an SU(2) symmetry between the SC and the CO. The stabilize a true long-range CO. 151 © 2013 Macmillan Publishers Limited. All rights reserved 135 coexisting phase in the phase diagram is a result of a weak In Sec. II,we focus on asimilar Ginzburg-Landau theory of 152 136 biquadratic symmetry breaking between the two, caused by the competing SC and 2D CO, but in a different perspective. 153 137 interaction terms in the free energy. We will treat a Ginzburg-Landau free energy with effective 154 138 Theoretically, competing orders [60–77]are studied enor- homogeneous order parameters (averaging the vortex induced 155 139 mously in the context of the underdoped cuprates. In the inhomogeneities) near the upper critical magnetic ﬁeld of the 156 140 presence of a magnetic ﬁeld, the SC is suppressed near a vortex superconductor. In this approach, the magnetic ﬁeld renormal- 157 141 core. As a result, any competing order like charge density izes the effective mass (coefﬁcient of the quadratic term of 158 142 wave [61,64,78], spin density wave [62–64,79], or pair density order parameters in the free energy) of the SC order parameter. 159 143 wave [80]becomes recognizable near the vortex cores. The We couple this effective free energy of the SC with the free 160 144 competing orders are often treated within a Ginzburg-Landau energy of the CO and study the competition. Note that we 161 145 theory. In particular, it was shown in Ref. [61]that theCO only consider the long-range CO. We phenomenologically 162 146 can coexist with the SC in a halo surrounding the vortex core construct the B-T phase diagram with increasing coupling 163 004500-2 ¬4 ¬4 v /v (10 ) v /v (10 ) s s s s –1 –1 1/T ( s ) 1/T (ms ) 2 1 Intensity (arb. units) B (T) T (K) 7 around the same wave vector as the charge modulations neighbor (n.n.) Coulomb interaction is retained and has a typical linear shape. Using the EHS model, we could t the slope of the mode to a recent X-ray experi- H = t (c c + h:c) ij j; i; ments on BSCCO compounds [103]. i;j; + J S S + V n n ; (11) i;j i j i;j i j i;j 2.2.4. Symmetry of the spectral gap with respect to zero energy where c (c ) is a creation (annihilation) operator i; i; for an electron at site i with spin , n = c c i i; i; The PG state is identi ed in various spectroscopies is the number operator and S = c c is the i ; i; i; with either the pair-breaking peak as seen in the B 1g spin operator at site i ( is the vector of Pauli matri- channel in Raman [104, 105], or the gap seen in STM ces). J is an eective AF coupling which comes for i;j [106] or ARPES [7, 107, 108]. The spectral \peak" or example from the Anderson super-exchange mechanism the gap concern the ANR of the Brillouin zone. One [120]. The constraint of no double occupancy typical most noticeable feature about this gap is that it goes \un- of the strong Coulomb onsite interaction is implemented changed" when one decreases the temperature to reach through the Gutzwiller approximation by renormalizing inside the SC phase [7, 107, 108]. Said in other words, the hoping parameter and the spin-spin interaction with the SC gap in the ANR below T remains the same when 2p 4 t (p) = g t = t and J (p) = g J = J , where p is t J 2 the temperature is raised up to T . This very unusual 1+p (1+p) the hole doping while the density-density interaction is feature contrasts with the spectroscopic behavior around left invariant. We also assume that the AF correlations the node, as seen for example in the B channel in Ra- 2g are dynamic, strongly renormalized, and short ranged, as man spectroscopy [104, 105], or in ARPES resolved in given by the phenomenology of Neutron scattering stud- k-space [108], where the gap as a function of tempera- ies for cuprates [121] and V is a residual Coulomb in- ture vanishes at T which is typical of a standard BCS i;j teraction term. behavior. The behavior in the ANR naturally calls for an The theory of the SU(2) emergent symmetry, although interpretation in terms of the phase uctuations of the producing some strong phenomenology suers from a few SC pairing order parameter whereas the phase coherence weaknesses. Maybe the most prominent one is the sta- sets up below T . This led to the development of a pow- bility of the symmetry with respect to a small change of erful phenomenology of the PG [109, 110], in which one parameters in the model [122]. Let's consider again the distinguished feature is that the spectral gap is symmet- EHS model which has an exact realization of the sym- ric around E = 0 [20]. This feature is easily reproduced metry. If we try to generalize to a real Fermi surface, we by any theory of preformed pairs and emergent symme- notice that the symmetry is valid only at the hot spots, tries. On the other hand, all other theories will produce but anywhere away in the Brillouin zone it is broken. In an asymmetry of the PG around E = 0 (see e.g. Ref. order to resolve this issue, we rst enlarged the space [111{113]). of the charge modulations by considering multiple wave A somewhat related issue is the evolution of the num- vectors depending on the position in k-space. Although ber of carriers with doping, studied in a recent Hall mea- a bit arti cial, this trick was developed in order to max- surement [114]. It is shown that the number of carriers imize the area in k-space where the SU(2) symmetry is evolves rst linearly as the doping p, and then goes quite realized. The object thus created in the charge channel abruptly to 1 + p, forming a large hole Fermi surface has the form of charge excitons and forms droplets in real around the critical doping p . This behavior can be ac- space. This enabled us to have a theory for the prolif- counted by all the theories which open a gap around the eration of droplets in the under-doped region of cuprate eight hot spots (see Fig.3b), including theories coming superconductors [120, 123]. Formation of droplets, or from strong coupling scenarios [20, 115{117] or theories phase separation in real space, is a very interesting idea which open a gap on weaker coupling scenarios with stan- to explain the PG region because it has been observed dard symmetry breaking [20, 117, 118]. through NMR experiments that the PG phase is very robust with respect to induced disorder [22]. When a va- cancy is induced in the compound, for example through 2.3. Microscopic models for emergent symmetries Zn doping [2], or irradiation with electrons [124], most of the energy scales get powerfully reduced, like the SC A rst generalization of the EHS model, was to start T or the charge ordering temperature T , but the PG c co with a model for short range AF correlations, rather than line remains unaected. More precisely, NMR studies a model of an AF QCP with critical uctuations in 2D. tell us that around the Zn- impurity a small region of A possible model consists of a simpli cation of the t- AF correlations is forming, as if the close vicinity of the J model of Eq.(17), where the constraint of no double impurity was revealing the underlying presence of short occupancy in the Gutzwiller projection operators [119] is range AF correlations. But in-between the Zn- impu- treated at the mean- eld level and where an extra nearest rities, we recover the unaected PG. This tells us that 8 3. PHASE TRANSITION AND HIGGS an anti-bounding with the optical mode which pushes it 34 MECHANISM AT T at a higher energy than experimentally observed . The most commonly accepted explanation within the frame- work of itinerant magnetism, is that the INS resonance To summarize, in the last section, we gave an attempt is a particle-hole bound state below the spin gap (a spin- of the generalization of the SU(2) symmetry for realis- triplet exciton) which is stabilized by repulsive interac- tic Fermi surfaces and more realistic starting point of 34–36,38–40,43,44 tion left within the d-wave SC state. . This a model of short range AF interactions. The price we scenario well reproduces the structure of the spin exci- have to pay is to consider multiple wave vectors for the tation in the SC state in the optimally and overdoped charge modulations. Although such a scenario is valid in regime. In the underdoped regime, the observation of the theory, one could ask whether it is eectively realized in INS resonance in the PG state above T leads to a more the experiments and whether we could get a more robust complex situation. The shape of the resonance changes from “X” to “Y” with the presence of some additional mechanism to get uctuations at the T line. spectral weight in the vicinity of Q, whereas, in Hg-1201, the energy of the collective mode remains unchanged compared to the SC phase (see Fig. 1). This observation 3.1. Chiral model and Hopf mapping is very diﬃcult to account for theoretically.Recently, an incommensurate spiral spin order stabilized by quantum A new idea came by noticing that the SU(2) symme- ﬂuctuations upon doping the AF Mott insulator has been FIG. 1. (Color online) Schematic Temperature- doping (T,p) FIG. 6: Phase diagram for the Inelastic Neutron try that rotates the SC phase to the CDW modulations proposed to explain the evolution of the energy ﬂuctua- phase diagram of hole-doped cuprate compounds. The anti- Scattering response at (; ) from Ref.[126]. ferromagnetic (AF) phase develops close to half ﬁlling (p =0). tion spectrum around Q with doping in YBCO . The comes with two copies of SU(2) in the case (like in the The SC phase appears at intermediate hole doping and T is main diﬃculties lies on a correct modelization of the PG c EHS model) where the charge sector is a \bond exciton"- maximum at optimal doping p .Above T ,in the under- o c phase which,-if we believe the excitonic explanation in with a real and imaginary part [127]. As in the O(4) doped regime, p< po,the system exhibits a large pseudo-gap the SC phase, has to retain a certain amount of coher- ⇤ NLM Eq.(9), we have two complex operator elds (PG) phase until the temperature T = T .The RES SU(2) ence if the collective mode is to be observed at all in this model explains the PG phase by the proliferation of local ex- regime. citonic patches above the temperature T (orange line) . ^ prolif z = d c c ; 1 j i ij In parallel, ERS measurments in Hg-1201 provides very The proliferation temperature vanishes below the doping p . the PG is made of some short range \droplets" ( in real Below p , the preformed particle hole pairs in the RES are interesting and complementary information for the study c X space), which don't \communicate" with one another, so iQ(r +r )=2 ^ i j more stable than the Cooper pairs. Consequently, the Fermi of collective modes in the underdoped regime. A no- z = d c c e ; (12) 2 j ij that when a strong local perturbation is present ( like the surface is fractionalized : the Cooper pairs develop in the ticeable change of behavior is observed in Raman data Zn- impurity) the PG is only aected locally [2, 125]. Nregion onlywhilethepreformedparticle-holepairspopu- around 0.12 hole doping. Raman scattering is a dy- late in the AN region. The system exhibits a two-gap regime namical response, which probes the charge channel at ^ with d being the d-wave form factor and the constraint (green area). For doping above p ,the particle-hole pairs are q =0.Moreover, speciﬁc structure factors enable to writes W less e iden stable tify tha tw n othregimes e Cooperaspaairfunction s and the of SCdoping gaps ou (Fig. t the 6) scan the Brillouin zone with respect to respective sym- whole Fermi surface. The system exhibits a one-gap regime [126]. At lower doping in the under-doped region (0:06 < q metries : the A response is isotropic, the B symmetry 1g 1g (blue area).The two black arrows represent the two hole dop- 2 2 x < 0:12) the droplets proliferate, which translates itself E = jz j +jz j ; (13) 1 2 scans the anti-nodal (AN) regions (0,±⇡ ) and (±⇡, 0), ing where further calculations are performed. left arrow) Be- as a mixed character of the PG in the AN region with while the B symmetry selects the nodal (N) region 2g low p ,we have T =0 then RES is strong compared to c prolif superposition of charge and SC gaps. This induces a spe- where E is a high energy scale which is interpreted as (±⇡/ 2,±⇡/ 2) .For doping p< 0.12,the Raman data SC and the AN region is massively gapped by RES mecha- cialniresp sm and onse in the > Inelastic in the AN Neutron region o Scatterin f the ﬁrstgBZ (INS), . It exhibits a large SC coherence peak in the B symme- RES SC the PG energy scale in the case of the cuprate supercon- 2g is a two-gaps regime and the energy ﬂuctuation spectrum of try, while its intensity is very low in the B symmetry. with the spin one mode observed at the AF wave vector z 1g ductors. Considering the spinor eld = , the the spin susceptibility exhibits the same Y-shape in both the For higher doping, p 0.12,the SC coherent peak has (; ) having a \Y"shape typical of the PG phase. The PG and the SC phase. right arrow) Close to optimal doping ahuge intensity in the B symmetry and decreases in 1g INS mode is treated in this theory as coming from spin constraint in Eq.(13) is invariant with respect to factor- (for p <p), we have T 6=0.The RES is weaker than 47,48 c prolif the B symmetry . This change of behavior around 2g i excitons inside the SC phase or in the PG phase. On SC state and < in the AN region of the ﬁrst BZ. izing a global phase ! e . This gauge invariance RES SC the same doping in both Raman and INS probes suggests theConsequen other hand, tly inat thelarger SC stat doping e, the AN (0:12 zone< isx nearly < 0:20), com- the is visible in the SU(2) chiral model where the elds are that the coherence eﬀect that are getting lost around T pletely gapped out by Cooper pairs. It is a one-gap regime droplets become rarer and the gap in the ANR below T is c allowed to uctuate within an SU(2) matrix. are a key in the explanation of the feature 3). To the best where the energy ﬂuctuation spectrum exhibit a X-shape in a Cooper pair gap. In this region, the INS mode changes of our knowledge, the feature 3) has only been observed Z the SC phase which transforms itself in a Y-shape above T . form below T with now an \hourglass" or typical \X" c ab in Hg-1201 compound. d y S = d xTr[@ ' @ ']; with ' = z z ; ab a shape especially visible in the YBCO compound, where 2 2 Here, we calculate the two-particle responses in both the splitting between the lower branches of the \X" com- charge and spin sectors and compare them with exper- addressed before and a comprehensive study of the re- ing from the border of the particle-hole continuum inside imental observations reported by ERS and INS in the lations between neutron and Raman susceptibilities in and z z = 1: (14) the SC phase (see [121] and reference therein). underdoped regime, within a new theoretical explana- this region are given here for the ﬁrst time. There have a=1 tion for the of the PG phase : the Resonant Excitonic been many proposals for the PG phase of the cuprates, 51–53 54–56 State (RES) which can be described as preformed exci- based on AF ﬂuctuations ,strong correlations , Using the Hopf mapping of the sphere S represented 49,50 The formulation 57,58 of the SU(2) symmetry using 42m ,59ultiple tonic (particle-hole) pairs .Although diﬀerent the- loop current or emergent symmetry models .A by Eq. (13) to the sphere S , the model in Eq.(14) can wave vectors was intended to preserve the symmetry in oretical approaches have been developed to explain the recent study proposes explain the PG phase with a SU(2) be reduced to and O(3) NLM as follows (see a generic the ANR of the Brillouin zone. Although the possibility a a PG phase, as stated above, the issue of the change of emergent symmetry model where the SU(2) symmetry proof in [128]). We de ne m = z z where are 60,61 shape of the INS resonance across T has never been relates the SC state to the charge sector . The PG c of multiple wave vectors is interesting, it is not very likely Pauli matrices and the indices a = x; y; z, the eective that a physical system does not choose one wave vector at action is now least at lower temperature. Is there another route to get 3 3 an organizing principle to treat the competition between X X 2 2 d a a S = 1=2 d x (@ m ) ; with jm j = 1: (15) the charge modulations and the SC without resorting to a=1 a=1 the multiple wave vectors trick? 9 We see that the form in Eq.(15) is similar to the one where the global U(1) phase of the spinor gets of Eq.(9) but with three elds (O(3) NLM) rather than frozen. As a result, one gauge eld acquires a mass four (O(4) NLM). In the case of the two elds of Eq.(12) 2 2 E = j j +j j opening a gap in the ANR of the ij ij which represent the case of cuprates, the operators m , Brillouin zone and E characterizes the PG energy scale. m and m correspond to PDW uctuations. y z Owing to this Higgs mechanism, the PG line will show universal features independent of the material speci cs or disorder. Thus at T , the PP and PH pairs get 3.2. U (1) U (1) gauge theory and Higgs phase entangled and compete strongly with each other. Due transition at T to this competition, both the amplitudes jz j and jz j 1 2 uctuate wildly along with uctuations in the relative The two complex elds and in Eq. (12) are de- ij ij phase. The eective theory just below T corresponds ned on bonds (i; j) with r = r a (see Fig.3c). j i x;y ij to the O(3) NLM or, equivalently, to the SU(2) chi- and represent preformed pairs in the particle-particle ij ral model, and the typical excitations are made of PDW (PP) channel and the particle-hole (PH) channel. In or- -modes. The freezing of the global phase at T has der to construct a continuum eld theory, these elds are removed one copy of the O(3) NLM leaving the other de ned on the midpoint of the bond, r = (r + r ) =2 i j one untouched. It is to be noted that the O(3) NLM such that z (r) = and z (r) = . The eective 1 ij 2 ij has topological excitations which are skyrmions in the eld theory will then have a U(1) U(1) gauge struc- pseudo-spin space and may account for the recently ob- ture. One U(1) corresponds to the usual charge sym- served anomalous thermal Hall eect [129]. To explain metry (usually broken by superconducting ground state) the anomalous Hall eect, there are other recent pro- and the other is a consequence of the fact that we have posals based on proximity to a quantum critical point pairs on bonds. By writing z and z on bonds, we have 1 2 of a `semion' topological ordered state [130], presence of doubled the gauge structure to U(1) U(1) with two spin-dependent next-nearest neighbor hopping in the - gauge elds introduced in order to accommodate two in- ux phase [131] or presence of large loops of currents dependent phases. Without any loss of generality, the [132]. If we continue to lower the temperature in the U(1) U(1) gauge theory can be formulated in terms under-doped region, dierent temperature lines emerge of a global and a relative phase of the two kinds of pre- (see Fig. 2). First, there are two cross-over mean- eld formed pairs. In terms of the spinor , the global and lines corresponding to the condensation of the amplitudes relative phases can be expressed as follows of each eld (T ) = jz j and (T ) = jz j , leading 0 1 0 2 0 0 to a uniform d-wave Cooper pairing contribution below z ~ i i ' = e e ; (16) T and modulated d-wave charge contribution below fluc z ~ T . Below each line, the preformed pairs have a non zero co precursor gap in each channel, and these two channels where z ~ and z ~ are two elds which can carry d-wave 1 2 compete. The two elds z and z still satisfy Eq. (13) symmetry and modulations. Typically here, z ~ = djz j 1 2 1 1 iQr with a condensed part ( and ) and a uctuation part 0 0 and z ~ = djz j e . Two gauge elds a and b are 2 2 such that jz j = (T ) + jz j and jz j = (T ) + jz j. 1 0 1 2 0 2 introduced in the theory to enforce the gauge invariance. i i ' These lines do not have to be identical. In Fig.2 they The transformation ! e e , a ! a +@ , b ! b + @ ' leaves the following action are represented, with the rst one T corresponding to co a quasi long range charge order (experimentally this or- 1 1 1 d 2 ~ ~ der is bi-dimensional, so that it is impossible to have it S = d x jD j + V ( ) + F F + F F ; a;b 2g 4 4 fully long range at nite temperature) and the second (17) one T corresponding to the phase uctuation regime fluc with D = @ ia i b ; of the Cooper pairs. Note than when T < T , a fluc co PDW composite order is induced below T , since the F = @ a @ a ; fluc eld z z = has the symmetry of a SC order with 1 2 and F = @ b @ b ; non-zero center of mass and its phase was frozen at T . invariant, where is the Pauli matrix in the spinorial These lines are determined by solving the gap equations space and a and b are gauge elds corresponding in each channel, corresponding to the microscopic model respectively to the spinor's global phase and relative in Eq.(11). Below T , the second freezing of the relative phase '. A linear combination of the two gauge elds, phase occurs and we are fully in the SC phase. One con- a + b = 2A gives twice the electro-magnetic gauge eld sequence of our theory, is that the SC ground state is and another combination, a b = a ~ is identi ed as a actually a super-solid (or PDW), a fact that is supported dipolar eld on a bond. by X-Ray experiments, which show that the charge or- The concept of two kinds of preformed pairs and der signal does not go to zero as T ! 0 in the SC phase (see e.g. [65]). Since our theory of preformed pairs is the resultant U(1) U(1) gauge structure opens up unique possibilities with hierarchy of phenomenon occur- based on two complex elds, the phase in the particle- hole channel produces currents of d-density wave (dDW) ring with reduction in temperature from a high value. At T , the system encounters a rst Higgs mechanism [133] type, as well as a series of pre-emptive transitions 10 breaking C symmetry (nematicity), time reversal sym- time to the question of whether charge modulations metry (TR) and creating loop currents. The reasoning are \strong" or \weak" phenomenon in under-doped follows the steps of earlier works [134{136]. Another con- cuprates. Charge modulations have been seen in most sequence of this double stage freezing of the phase is that of the compounds in the under-doped region, and the phase slip associated to the charge ordering is now although they are bi-dimensional, in theory they are sti and related to the phase of the superconductor. This \long range enough" to lead to a re-con guration of the very unusual situation promises to open space for future Fermi surface observed through Quantum Oscillations experimental veri cations. (QO) [72]. But in this speci c Raman experiment we see for the rst time that the magnitude of the charge precursor gap in the B channel is comparable to 2g 3.3. What can this work explain? the magnitude of the Pair breaking gap in the B 1g channel, comforting the idea that the charge modulation is a \strong" eect in the physics of underdoped cuprates. The main idea of this work is to have two sets of pre- formed pairs, one in the particle-hole channel, that we At this point, it is important to note that historically called \bond excitons", and one in the particle-particle in the physics of cuprates, it is argued that there exists channel-the Cooper pairs that become entangled at T . two energy scales corresponding to the PG phase (see The freezing of the entropy down to T = 0 then follows e.g. the review [16]). First there is a higher energy scale its own route showing cascade of phenomenon occurring associated with the depletion in the NMR Knight shift, at dierent temperatures. The notion of preformed pairs referred to as the \Alloul-Warren" gap. This scale is shall be distinguished from the typical scenario of phase also seen in Raman spectroscopies as a hump rather than uctuations [42, 48] around a mean eld amplitude. In- a peak. Second scale corresponds to the spectroscopy deed, in the phase uctuations scenario, the focus is put peaks at a lower energy scale seen in STM, ARPES or on the uctuations only with no restrictions on the size Raman spectroscopy. Both the \Alloul-Warren" and the of the pairs, whereas the preformed pair scenarios [50{ spectroscopic gaps seem to be related to each other and 52] put the accent on the very short size of the Cooper behave similarly with doping, decreasing in a quasi-linear pair which opens a wide region in temperature where the fashion as doping increases[13]. The argument of two pairs behave like a hard core boson and undergo Bose PG energy scales is typical of strong coupling theories Einstein condensation. In view of the strong correlations where the higher energy scale is characterized with spin in cuprates and the very short size of the Cooper pairs, singlet formation and the lower energy scale is associated the concept of preformed pairs is very natural. Moreover, with superconducting uctuations. In this work, the PG recent experiments have revived the issue of uctuating state is accompanied by a single energy scale E which is preformed pairs with their observation up to T in pump observed as spectroscopy peaks. The higher energy hump probe measurements [137{139] and in the over-doped re- in Raman spectroscopy or the \Alloul-Warren" gap is not gion of the phase diagram in time domain spectroscopy an independent energy scale and might be related to the [140]. We review below a few experiments which could coupling of fermions to a collective mode [142{144]. be explained by such a scenario. Using a BCS-like scaling argument for the dierent en- ergy scales, 3.3.1. A precursor in the charge channel 1=J 1 0 = 2~! e ; 1=J 2 0 = 2~! e ; A recent Raman scattering experiment performed on 1=J the compound Hg1223 shows for the rst time that there E = 2~! e (18) is a precursor gap in the charge channel (see Fig.7a) [13]. It is seen as a spectral peak in the B channel, which is where the density of state at the Fermi level is taken to 2g 0 visible below the ordering temperature T but follows be constant, ! is the cut-o frequency of the interaction co c T rather than T as a function of oxygen doping. and J = (J + J )=2, we can obtain their approximate co 1 2 This peak is attributed to charge modulations. The doping dependence. The evolution of the energy scales situation is very similar at what is observed in the AN with doping is shown on the right panel of Fig.7b where region of the Brillouin zone. In the B channel (which J and J have been chosen to be linearly decreasing with 1g 1 2 is scanning the AN region) a pair breaking peak is doping and vanish at two dierent doping p and p . In 1 2 observed below T , but which follows T rather than T this simpli ed approach, E goes to zero at an interme- c c with doping. This spectral peak is also seen in ARPES diate doping p . We obtain the linear dependence of [107, 108], where it is shown that it remains unchanged and in an extended range of doping with as above T and just broadens through an additional source observed in the Raman experiment Fig.7a. We also show of damping [49, 108, 141], up to the PG temperature that there exists only one PG energy scale E . scale T . It is also seen in STM [106], at similar energy The concept of preformed pairs can then account, with- scales. A new feature of this very spectacular Raman out resorting to the idea of a topological order, for the scattering experiment is that it answers for the rst observation that upon application of a strong magnetic T (K) CDW Nodal region eld, the PG survives without a true T = 0 long range (a) order [114]. Indeed around p application of a magnetic eld up to 100T will certainly destroy SC phase coher- ence, but to destroy the pairs a much bigger eld will be YBCO 2Δ CDW Hg-1223 called for. YBCO CDW Our theory [127] addresses directly the formation of Hg-1223 200 the spectral gaps and which are visible in the B 2g 200 and B channels respectively 1g (b) X 2000 1 J (q; k+q = ; (19) k;! 2 1600 2 2 (! + q; k+q k+q 1 J (q; + k+q = : k;! (! + ) (! + k+q k+Q+q 400 k+q 2Δ 2ΔCDW 2Δ q; PG SC 0.10 0.12 0.14 0.16 with J (q; ) being related to the original model param- Doping p eter as J (q; ) 3J (p)V and is the inverse temper- (a) ature. Solving Eq.(19) while keeping the momentum de- pendence of the gaps and allowing only one of them to be non-zero at each k-point we obtain the repartition shown in the left panel of Fig.7b. We see that the particle- particle pairing gap (yellow) is favored in the ANR while the particle-hole pairing gap (violet) prevails in the nodal region. This result agrees well with the Raman experi- ment where the SC pair-breaking peak appears in the B symmetry which is probing the anti-nodal part of the 1g Brillouin zone, while the precursor in the charge channel appears in the B symmetry probing the nodal region. 2g (b) The PG energy scale is identi ed in the present theory as the scale at which the two preformed pairs start to FIG. 7: a) Raman Scattering experiment [13] in Hg1223 2 2 get entangled and compete, E = jj +jj , as rst where the two spectroscopic gaps in the B (Cooper 1g pairing called here) and B (particle-hole pairing described in Eq.(13) for the case of the emergent SU(2) SC 2g called here) channels are shown to be of the symmetry. Theories which can account for a precursor CDW same order or magnitude and to decrease linearly with gap in the charge channel, which does not follow T but co doping like T . The higher energy hump also has T with doping, and which is of the same size as the pre- PG the same doping dependence and is thus possibly not an cursor gap in the Cooper channel, are very rare. This independent energy scale. b) Theoretical solution of the set of experiments, if con rmed, thus can be considered gap equations Eq.(19) from the toy model Eq.(11). In as a signature of a mechanism for two kinds of entangled the left hand side, a quarter of the Brillouin zone is preformed pairs. It has to be noted that the precursor represented with the Cooper pairing gap in yellow in the charge channel has been seen, so far, only in the and the preformed particle-hole gap in blue. Hg1223 compound, where the three layers enhance the Calculations are done for T T . On the right hand visibility of the charge channel. Similar ndings were re- side, the variation of the magnitude of the gap with ported for YBCO but the analysis is more dicult due doping is given, using a simpli ed BCS-like scaling to the persistent noise in the B channel. 2g argument. 3.3.2. Phase locking at T halo [145{148]. But a stranger feature can be inferred The locking of the global phase of the two kinds of by our theory. Since the phases of the Cooper pairs and preformed pairs at T has serious experimental conse- the charge modulations have been locked at T , inside quences. The PP and PH pairs become entangled at the vortices, the \phase-slip" of the charge modulations T and at lower temperatures T , the remaining relative cos (Q r + ) is locked over a very long range h i 6= 0, c r r phase is xed. Now, if one applies an external magnetic longer than the typical size of each modulation patch, and eld, it will create vortices where the SC order parame- thus much longer than the size of each vortex. This situ- ter will be suppressed. Inside each vortex the competing ation was observed in STM, where one sees that although order parameter (in this case a bond excitonic order) is there is a bit of spreading of the phase-slips around the present, which typically occurs in theories with compet- mean value, the mean value itself is typically long range ing orders. A PDW order will be present in the vortex over the whole sample [149]. A concrete and strong ex- -1 -1 2Δ (cm ) Energy gap (cm ) CDW perimental prediction would then naturally be to check 4. CONCLUSION the link between the patches of charge modulations and the SC phase. A Josephson-type of setup, where the SC We have presented in this paper a review of our un- phase is monitored through an applied current, would in- derstanding of the PG phase of cuprate superconductors. evitably produce a correlation in the \phase-slips" of the This is an old problem, but which has generated a consid- charge modulations, a situation so unusual that, if veri- erable amount of creativity in the last thirty years, both ed, it would probably pin this theory to be the correct with theoretical concepts and with new experiments. We one. have put forward a scenario for the PG where a phase transition occurs at T where two kinds of preformed pairs, in the particle-hole and particle-particle channels, 3.3.3. Q = 0 orders at T get entangled and start to compete. The eective model below T still retains a large amount of uctuations, as One of the great experimental complexity of the PG is described as an O(3) NLM, or equivalently the SU(2) chiral model, which is reminiscent of theories of emer- line, is that Q = 0 orders have been observed around this line which break discrete symmetries. Loop currents gent symmetries with non abelian groups like the SU(2) group. We stress that although the uctuations below T observed through elastic neutron scattering [150], ne- maticity revealed through magneto-torque measurement belong to the same universality class as emergent sym- metries, the new mechanism does not rely on the pres- [151, 152] and parity breaking observed via a second har- ence of an exact symmetry in the Lagrangian over the monic generation [153], all show a thermodynamic signa- whole under-doped region. The gap opens in the ANR ture at T , the caveat being that Q = 0 orders cannot of the Brillouin zone due to the freezing of the global open a gap in the electronic density of states. We are phase of our two kinds of preformed pairs. The model thus in a complex situation where the depletion of the has a U (1)U (1) gauge structure which enables to iden- density of states in the ANR cannot be explained by the tify two phase transitions, one at T and another one at Q = 0 orders whereas any theory which gives an under- lower temperatures, at T . It has to be noted that models standing of the PG will have to account for the presence of these intra unit cell orders at T . with SU(2) gauge structure, both in the spin sector and the pseudo-spin sectors have been put forward recently to The phase transition at T is of a very peculiar na- explain the PG though corresponding Higgs transitions ture, with the freezing of the global phase of the spinor at T . in Eq.(16) but also the opening of a gap it looks like a The main dierence between our approach and those usual Higgs phenomenon. But the dierence with the Meisner eect in a superconductor for example, is that models is that we do not \fractionalize" the electron at T into \spinon" and \holons" of some kinds. Instead, the opening of the gap is made of a composite order rather than corresponding to the condensation of a sim- the electron keeps its full integrity, and the PG is due in our scenario to the entanglement of two kinds of pre- ple eld. The entanglement of two kinds of preformed pairs induces multiple orders in the PG phase. One can formed pairs in the charge and Cooper channels. The two consider a composite eld = as a direct product approaches being antinomic, the future will tell whether of the two entangled elds. As the eld correspond one of the two is the right one (or whether maybe a third to particle-particle pairs with a nite center of mass mo- line of idea is needed). An argument in our favor lie in mentum, this eld has the same symmetries as that of the recent observation of a precursor gap in the charge a PDW eld. Since has only a global phase and that channel whose energy scale is related to the PG energy scale [13]. In consideration of the fractionalized scenario, it is frozen at T a global phase coherence sets up. But still, the eld has huge uctuations at T around a it has been argued that there is some continuity between iQr the PG at half lling and the PG in the under-doped re- mean average hi = jj e in which the uctua- tions of the amplitude might \wash out" the modulation gion where the ground state is a superconductor, which would push the balance towards a fractionalized scenario term. The PDW eld will condense to a long range order for temperatures below T where both the SC and the for T (see e.g. [113] and references therein). We argue fluc that \something" must happen at a doping of the order bond-excitonic orders obtain uniform components. One feature of this PDW is that its modulation wave vector of p 0:05 where numerous of \bond orders" sitting on will be same as that of the bond-excitonic order. Since the Oxygen atoms start to show up. Our intuition is that we work with complex elds and , the theory can it is maybe the doping at which we lose the Zhang-Rice potentially open auxiliary orders around T especially singlet [156], which thus liberates some degrees of free- dom on the Oxygen atoms, leading to the formation of those, like loop currents or nematic order, which occur at Q = 0 but break a discrete symmetry Z or C . This d-wave \bond" charge orders. At this stage it is just a 2 4 speculation. situation has been described in previous works [135, 154], in particular in the case where T is identi ed with the The scenario of entangled preformed pairs being simple formation of a long range PDW order [154] and where the enough, we hope that it will be possible to produce a 2 2 strong predictive experiment in the near future. second order magneto-electric tensor l = j j j j Q Q acquires a non zero value [155]. This work has received nancial support from 13 the ERC, under grant agreement AdG-694651- IIP (Natal, Brazil), for very inspirational visits, and for CHAMPAGNE. The authors also like to thank the wonderful hospitality. [1] H. Alloul, T. Ohno, and P. Mendels, Phys. Rev. Lett. [33] A. Ferraz and E. Kochetov, Nuclear Physics B 853, 710 63, 1700 (1989). (2011). [2] H. Alloul, P. Mendels, H. Casalta, J. F. Marucco, and [34] I. Ivantsov, A. Ferraz, and E. Kochetov, Phys. Rev. B J. Arabski, Phys. Rev. Lett. 67, 3140 (1991). 98, 214511 (2018). [3] W. W. Warren et al., Phys. Rev. Lett. 62, 1193 (1989). [35] D. Chakraborty, C. Morice, and C. P epin, Phys. Rev. [4] J. C. Campuzano et al., Nature 392, 157 (1998). B 97, 214501 (2018). [5] J. C. Campuzano et al., Phys. Rev. Lett. 83, 3709 [36] G. Kotliar, Phys. Rev. B 37, 3664 (1988). (1999). [37] P. W. Anderson, Science 235, 1196 (1987). [6] K. M. Shen et al., Science 307, 901 (2005). [38] P. W. Anderson et al., J. Phys. Condens. Matter 16, [7] I. M. Vishik et al., New Journal of Physics 12, 105008 R755 (2004). (2010). [39] T. Kloss, X. Montiel, V. S. de Carvalho, H. Freire, and [8] I. M. Vishik et al., Phys. Rev. Lett. 104, 207002 (2010). C. P epin, Rep. Prog. Phys. 79, 084507 (2016). [9] W. D. Wise et al., Nat. Phys. 4, 696 (2008). [40] Emery, V. J. and S. A. Kivelson, Nature 374, 434 [10] J. E. Homan et al., Science 295, 466 (2002). (1995). [11] M. H. Hamidian et al., Nat. Phys. 12, 150 (2015). [41] L. Benfatto, S. Caprara, and C. D. Castro, The Euro- [12] S. Benhabib et al., Phys. Rev. Lett. 114, 147001 (2015). pean Physical Journal B - Condensed Matter and Com- [13] B. Loret et al., Nature Physics , 1 (2019). plex Systems 17, 95 (2000). [14] B. Loret et al., Physical Review B 96, 094525 (2017). [42] L. Benfatto, C. Castellani, and T. Giamarchi, Phys. [15] M. R. Norman and C. P epin, Rep. Prog. Phys. 66, 1547 Rev. Lett. 98, 117008 (2007). (2003). [43] Y. J. Uemura et al., Phys. Rev. Lett. 62, 2317 (1989). [16] P. A. Lee, N. Nagaosa, and X.-G. Wen, Rev. Mod. Phys. [44] C. C. Homes et al., Nature (London) 430, 539 (2004). 78, 17 (2006). [45] M. R. Norman, H. Ding, and M. Randeria, J. Phys. [17] A. Chubukov, D. Pines, and J. Schmalian, J 2003 The Chem. Solids 59, 1902 (1998). Physics of Conventional and Unconventional Supercon- [46] M. R. Norman, A. Kanigel, M. Randeria, U. Chatterjee, ductors vol 1ed K.H. Bennemann and J. B. Ketterson, and J. C. Campuzano, Phys. Rev. B 76, 174501 (2007). Berlin:Springer, 2003. [47] M. R. Norman, M. Randeria, H. Ding, and J. C. Cam- [18] T. M. Rice, K.-Y. Yang, and F. C. Zhang, Reports on puzano, Phys. Rev. B 52, 615 (1995). Progress in Physics 75 (2012). [48] S. Banerjee, T. V. Ramakrishnan, and C. Dasgupta, [19] T. Senthil and P. A. Lee, Phys. Rev. B 79, 245116 Phys. Rev. B 83, 024510 (2011). (2009). [49] S. Banerjee, T. V. Ramakrishnan, and C. Dasgupta, [20] S. Verret, O. Simard, M. Charlebois, D. S en echal, and Phys. Rev. B 84, 144525 (2011). A. M. S. Tremblay, Physical Review B 96, 125139 [50] R. Boyack, C.-T. Wu, P. Scherpelz, and K. Levin, Phys- (2017). ical Review B 90, 220513 (2014). [21] E. Fradkin, S. A. Kivelson, and J. M. Tranquada, Rev. [51] R. Boyack, Q. Chen, A. A. Varlamov, and K. Levin, Mod. Phys. 87, 457 (2015). arXiv.org , 321 (2017). [22] H. Alloul, Comptes Rendus Physique 15, 519 (2014). [52] C.-C. Chien, Y. He, Q. Chen, and K. Levin, Physical [23] D. J. Scalapino, Numerical Studies of the 2D Hubbard Review B 79, 214527 (2009). Model, in Handbook of High-Temperature Superconduc- [53] L. Li et al., Phys. Rev. B 81, 054510 (2010). tivity, pages 495{526, Springer New York, New York, [54] L. Li, N. Alidoust, J. M. Tranquada, G. D. Gu, and NY, 2007. N. P. Ong, Phys. Rev. Lett. 107, 277001 (2011). [24] K. Le Hur and T. M. Rice, Annals of Physics 324, 1452 [55] L. Li, Y. Wang, and N. P. Ong, Phys. Rev. B , 056502 (2009). (2013). [25] P. A. Lee, N. Nagaosa, T.-K. Ng, and X.-G. Wen, Phys. [56] O. Cyr-Choiniere et al., Physical Review B 97, 064502 Rev. B 57, 6003 (1998). (2018). [26] P. A. Lee and N. Nagaosa, Phys. Rev. B 46, 5621 [57] F. Rullier-Albenque, H. Alloul, and G. Rikken, Phys. (1992). Rev. B 84, 014522 (2011). [27] G. Kotliar and J. Liu, Phys. Rev. B 38, 5142 (1988). [58] N. Bergeal et al., Nat. Phys. 4, 608 (2008). [28] T. Senthil and M. P. A. Fisher, Physical Review B 62, [59] G. Wachtel and D. Orgad, Physical Review B 90, 7850 (2000). 224506 (2014). [29] T. Senthil and M. P. A. Fisher, Phys.Rev.Lett. 86, 292 [60] E. Demler and S.-C. Zhang, Phys. Rev. Lett. 75, 4126 (2001). (1995). [30] P. Lee and X.-G. Wen, Phys. Rev. Lett. 78, 4111 [61] E. Demler, W. Hanke, and S.-C. Zhang, Rev. Mod. (1997). Phys. 76, 909 (2004). [31] P. A. Lee, N. Nagaosa, T.-K. Ng, and X.-G. Wen, Phys. [62] S.-C. Zhang, Science 275, 1089 (1997). Rev. B 57, 6003 (1998). [63] T. Wu et al., arXiv:1404.1617 [cond-mat.supr-con] [32] S. Sachdev, H. D. Scammell, M. S. Scheurer, and (2014). G. Tarnopolsky, Phys. Rev. B 99, 054516 (2019). [64] T. Wu et al., Nature 477, 191 (2011). 14 [65] E. Blackburn et al., Phys. Rev. Lett. 110, 137004 [106] M. Kugler, O. Fischer, C. Renner, S. Ono, and Y. Ando, (2013). Phys. Rev. Lett. 86, 4911 (2001). [66] S. Blanco-Canosa et al., Phys. Rev. Lett. 110, 187001 [107] A. Damascelli, Z. Hussain, and Z.-X. Shen, Rev. Mod. (2013). Phys. 75, 473 (2003). [67] S. Blanco-Canosa et al., Phys. Rev. B 90, 054513 [108] I. M. Vishik, Reports on Progress in Physics 81, 062501 (2014). (2018). [68] G. Ghiringhelli et al., Science 337, 821 (2012). [109] P. A. Lee, Phys. Rev. X 4, 031017 (2014). [69] N. Doiron-Leyraud et al., Nature 447, 565 (2007). [110] X. Montiel, T. Kloss, and C. P epin, EPL (Europhysics [70] D. LeBoeuf et al., Nature 450, 533 (2007). Letters) 115, 57001 (2016). [71] S. E. Sebastian et al., Phys. Rev. Lett. 108, 196403 [111] W. Wu et al., Phys. Rev. X 8, 021048 (2018). (2012). [112] S. Sakai, M. Civelli, and M. Imada, Phys. Rev. Lett. [72] W. Tabis et al., Nat. Commun. 5, 5875 (2014). 116, 057003 (2016). [73] N. Bari si c et al., Nature Physics 9, 761 (2013). [113] S. Sakai, M. Civelli, and M. Imada, Physical Review B [74] G. Grissonnanche and F. Laliberte, (2015). 98, 195109 (2018). [75] F. Lalibert e et al., Nat. Commun. 2, 432 (2011). [114] S. Badoux et al., Nature (London) 531, 210 (2016). [76] J. Chang et al., Phys. Rev. Lett. 104, 057005 (2010). [115] K.-Y. Yang, T. M. Rice, and F.-C. Zhang, Phys. Rev. [77] S. Gerber et al., Science 350, 949 (2015). B 73 (2006). [78] J. Chang et al., Nat. Commun. 7 (2016). [116] S. Chatterjee, S. Sachdev, and A. Eberlein, Phys. Rev. [79] H. Jang et al., Proc. Natl. Acad. Sci. 113, 14645 (2016). B 96, 075103 (2017). [80] D. LeBoeuf et al., Nat. Phys. 9, 79 (2013). [117] J. G. Storey, EPL (Europhysics Letters) 113, 27003 [81] F. Lalibert e et al., npj Quantum Materials 3, 11 (2018). (2016). [82] K. B. Efetov, H. Meier, and C. P epin, Nat. Phys. 9, [118] C. Morice, X. Montiel, and C. P epin, Physical Review 442 (2013). B 96, 134511 (2017). [83] C. P epin, V. S. de Carvalho, T. Kloss, and X. Montiel, [119] A. Paramekanti, M. Randeria, and N. Trivedi, Phys. Phys. Rev. B 90, 195207 (2014). Rev. Lett. 87, 217002 (2001). [84] M. A. Metlitski and S. Sachdev, Phys. Rev. B 82, [120] X. Montiel, T. Kloss, and C. P epin, Phys. Rev. B 95, 075128 (2010). 104510 (2017). [85] W. A. Atkinson, A. P. Kampf, and S. Bulut, New J. [121] V. Hinkov et al., Nat. Phys. 3, 780 (2007). Phys. 17, 013025 (2015). [122] X. Wang, Y. Wang, Y. Schattner, E. Berg, and R. M. [86] H. Freire, V. S. de Carvalho, and C. P epin, Phys. Rev. Fernandes, Phys. Rev. Lett. 120, 247002 (2018). B 92, 045132 (2015). [123] X. Montiel, T. Kloss, and C. P epin, Scienti c Reports [87] A. Abanov, A. V. Chubukov, and J. Schmalian, Adv. 7, 3477 (2017). Phys. 52, 119 (2003). [124] F. Rullier-Albenque et al., EuroPhys. Lett. 50, 81 [88] M. Einenkel, H. Meier, C. P epin, and K. B. Efetov, (2000). Phys. Rev. B 90, 054511 (2014). [125] H. Alloul, J. Bobro, M. Gabay, and P. J. Hirschfeld, [89] L. E. Hayward, D. G. Hawthorn, R. G. Melko, and Rev. Mod. Phys. 81, 45 (2009). S. Sachdev, Science 343, 1336 (2014). [126] X. Montiel and C. P epin, Phys. Rev. B 96, 094529 [90] T. Wu et al., Nat. Commun. 4, 2113 (2013). (2017). [91] D. P. Arovas, A. J. Berlinsky, C. Kallin, and S.-C. [127] D. Chakraborty et al., arXiv e-prints , arXiv:1906.01633 Zhang, Phys. Rev. Lett. 79, 2871 (1997). (2019). [92] A. Ghosal, C. Kallin, and A. J. Berlinsky, Phys. Rev. [128] E. Fradkin, Field theories of condensed matter B 66, 214502 (2002). systems , in Field theories of condensed matter sys- [93] P. A. Lee and X.-G. Wen, Phys. Rev. B 63, 224517 tems, page see chap.7.9, Sarat Book House, 2007. (2001). [129] G. Grissonnanche et al., arXiv e-prints , [94] B. Lake et al., Science 291, 1759 (2001). arXiv:1901.03104 (2019). [95] S. A. Kivelson, D.-H. Lee, E. Fradkin, and [130] S. Chatterjee et al., arXiv e-prints , arXiv:1903.01992 V. Oganesyan, Phys. Rev. B 66, 144516 (2002). (2019). [96] Y. Zhang, E. Demler, and S. Sachdev, Phys.l Rev. B [131] J. H. Han, J.-H. Park, and P. A. Lee, arXiv e-prints , 66, 094501 (2002). arXiv:1903.01125 (2019). [97] J. Chang et al., Nat. Commun. 7, 11494 (2016). [132] C. M. Varma, arXiv e-prints , arXiv:1903.04699 (2019). [98] H. Meier, M. Einenkel, C. P epin, and K. B. Efetov, [133] S. Chakravarty, R. B. Laughlin, D. K. Morr, and Phys. Rev. B 88, 020506 (2013). C. Nayak, Phys. Rev. B 63, 094503 (2001). [99] J. Ka cmar c k et al., Phys. Rev. Lett. 121, 167002 [134] Y. Wang and A. Chubukov, Phys. Rev. B 90, 035149 (2018). (2014). [100] C. Morice, D. Chakraborty, and C. P epin, Phys. Rev. [135] Y. Wang, D. F. Agterberg, and A. Chubukov, Phys. B 98, 224514 (2018). Rev. B 91, 115103 (2015). [101] S. Vig et al., SciPost Physics 3, 026 (2017). [136] Y. Wang, D. F. Agterberg, and A. Chubukov, Phys. [102] M. Mitrano et al., Proceedings of the National Academy Rev. Lett. 114, 197001 (2015). of Sciences 115, 5392 (2018). [137] S. Rajasekaran et al., Science 359, 575 (2018). [103] L. Chaix et al., Nature Physics 13, 952 (2017). [138] W. Hu et al., Nature Materials 13, 705 (2014). [104] A. Sacuto et al., Reports on Progress in Physics 76, [139] D. Fausti et al., Science 331, 189 (2011). 022502 (2013). [140] F. Mahmood, X. He, I. Bo zovi c, and N. P. Armitage, [105] T. P. Devereaux and R. Hackl, Rev. Mod. Phys. 79, Phys.Rev.Lett. 122, 027003 (2019). 175 (2007). [141] M. R. Norman, M. Randeria, H. Ding, and J. C. Cam- 15 puzano, Phys. Rev. B 57, R11093 (1998). (2015). [142] M. Eschrig and M. R. Norman, Phys. Rev. Lett. 85, [150] B. Fauqu e et al., Phys. Rev. Lett. 96, 197001 (2006). 3261 (2000). [151] Y. Sato et al., Nature Physics 13, 1074 (2017). [143] M. Eschrig, Advances in Physics 55, 47 (2006). [152] H. Murayama et al., arXiv e-prints , arXiv:1805.00276 [144] A. Chubukov, T. Devereaux, and M. Klein, Phys. Rev. (2018). B 73, 094512 (2006). [153] L. Zhao et al., Nat. Phys. 13, 250 (2016). [145] M. H. Hamidian et al., Nature 532, 343 (2016). [154] D. F. Agterberg, D. S. Melchert, and M. K. Kashyap, [146] S. D. Edkins et al., arXiv e-prints , arXiv:1802.04673 Phys. Rev. B 91, 054502 (2015). (2018). [155] S. Sarkar, D. Chakraborty, and C. P epin, arXiv e-prints [147] Y. Wang et al., Phys. Rev. B 97, 174510 (2018). , arXiv:1906.08280 (2019). [148] Z. Dai, Y.-H. Zhang, T. Senthil, and P. A. Lee, Phys. [156] F. C. Zhang and T. M. Rice, Phys. Rev. B 37, 3759 Rev. B 97, 174511 (2018). (1988). [149] M. H. Hamidian et al., arXiv e-prints , arXiv:1508.00620
Condensed Matter – arXiv (Cornell University)
Published: Jun 24, 2019
You can share this free article with as many people as you like with the url below! We hope you enjoy this feature!
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.