Search

We report thermal conductivity measurements on single crystals of herbertsmithite, over a wide range of temperatures (0.05-120 K) in magnetic fields up to 15 T. We also report measurements of the thermal Hall effect, found to be vanishingly small. At high temperatures, in the paramagnetic regime, the thermal conductivity has a negligible field dependence. Upon cooling and the development of correlations, the onset of a clear monotonic field dependence below about 20 K signals a new characteristic temperature scale that may reflect the subtle crossover towards the quantum spin liquid regime. Deconfined spinons, if present, are not detected and phonons, as the main carriers of heat, are strongly scattered by the intrinsic spin excitations and the magnetic defects. In view of the colossal fields required to affect the intrinsic spins, most of the field-induced evolution is attributed to the progressive polarization of some magnetic defects. By elaborating a phenomenological model, we extract the magnetization of these main scattering centers which does not resemble the Brillouin function for free spins 1/2, requiring to go beyond the paradigm of isolated paramagnetic spins. Besides, the onset of a nonmonotonic field dependence below about 2 K underlines the existence of another characteristic temperature scale, previously highlighted with other measurements, and sheds new light on the phase diagram of herbertsmithite down to the lowest temperatures., Comment: 14 pages, 7 figures

Electronic nematicity is the spontaneous loss of rotational symmetry in a metal, without breaking translational symmetry. In the cuprate superconductors, there is experimental evidence for nematicity, but its origin remains unclear. Here we investigate the onset of nematicity in the transport of charge by means of electric and thermoelectric measurements in underdoped YBa$_{\rm 2}$Cu$_{\rm 3}$O$_{\rm y}$, performed by passing the current (electrical or thermal) first along the $a$ axis then the $b$ axis of the orthorhombic structure in the same crystal, with a hole doping $p = 0.12$. Upon cooling, we observe no additional in-plane anisotropy -- beyond the background anisotropy due to the CuO chains -- in either the resistivity $\rho$ or the Seebeck coefficient $S$ as the temperature $T^{\star}$~for the onset of the pseudogap phase is crossed. We conclude that the pseudogap phase of cuprates is not nematic. However, at temperatures much lower than $T^{\star}$, a strong additional anisotropy is observed, most clearly in the Peltier coefficient $\alpha = S / \rho$. We interpret it as nematicity associated with the development of charge order., Comment: 7 pages, 4 figures

In various "strange" metals, electrons undergo Planckian dissipation, a strong and anomalous scattering that grows linearly with temperature, in contrast to the quadratic temperature dependence expected from the standard theory of metals. In some cuprates and pnictides, a linear dependence of the resistivity on magnetic field has also been considered anomalous - possibly an additional facet of Planckian dissipation. Here we show that the resistivity of the cuprate strange metals Nd-LSCO and LSCO is quantitatively consistent with the standard Boltzmann theory of electron motion in a magnetic field, in all aspects - field strength, field direction, temperature, and disorder level. The linear field dependence is found to be simply the consequence of scattering rate anisotropy. We conclude that Planckian dissipation is anomalous in its temperature dependence but not in its field dependence. The scattering rate in these cuprates does not depend on field, which means their Planckian dissipation is robust against fields up to at least 85 T.

The material $\alpha$-RuCl$_3$ has been the subject of intense scrutiny as a potential Kitaev quantum spin liquid, predicted to display Majorana fermions as low energy excitations. In practice, $\alpha$-RuCl$_3$ undergoes a transition to a state with antiferromagnetic order below a temperature $T_{\rm N}$ $\approx$ 7 K, but this order can be suppressed by applying an external in-plane magnetic field of $H_\parallel$ = 7 T. Whether a quantum spin liquid phase exists just above that field is still an open question, but the reported observation of a quantized thermal Hall conductivity at $H_\parallel$ $>$ 7 T by Kasahara and co-workers $\big[$Kasahara ${\it et \ al}$., Nature ${\bf 559}$, 227 (2018)$\big]$ has been interpreted as evidence of itinerant Majorana fermions in the Kitaev quantum spin liquid state. In this study, we re-examine the origin of the thermal Hall conductivity $\kappa_{\rm xy}$ in $\alpha$-RuCl$_3$. Our measurements of $\kappa_{\rm xy}$($T$) on several different crystals yield a temperature dependence very similar to that of the phonon-dominated longitudinal thermal conductivity $\kappa_{\rm xx}$($T$), for which the natural explanation is that $\kappa_{\rm xy}$ is also mostly carried by phonons. Upon cooling, $\kappa_{\rm xx}$ peaks at $T \simeq$ 20 K, then drops until $T_{\rm N}$, whereupon it suddenly increases again. The abrupt increase below $T_{\rm N}$ is attributed to a sudden reduction in the scattering of phonons by low-energy spin fluctuations as these become partially gapped when the system orders. The fact that $\kappa_{\rm xy}$ also increases suddenly below $T_{\rm N}$ is strong evidence that the thermal Hall effect in $\alpha$-RuCl$_3$ is also carried predominantly by phonons. This implies that any quantized signal from Majorana edge modes would have to come on top of a sizable -- and sample-dependent -- phonon background., Comment: 6 pages, 3 figures

We report measurements of the Seebeck effect in both the $ab$ plane ($S_{\rm a}$) and along the $c$ axis ($S_{\rm c}$) of the cuprate superconductor La$_{1.6-x}$Nd$_{0.4}$Sr$_{x}$CuO$_4$ (Nd-LSCO), performed in magnetic fields large enough to suppress superconductivity down to low temperature. We use the Seebeck coefficient as a probe of the particle-hole asymmetry of the electronic structure across the pseudogap critical doping $p^{\star} = 0.23$. Outside the pseudogap phase, at $p = 0.24 > p^{\star}$, we observe a positive and essentially isotropic Seebeck coefficient as $T \rightarrow 0$. That $S > 0$ at $p = 0.24$ is at odds with expectations given the electronic band structure of Nd-LSCO above $p^{\star}$ and its known electron-like Fermi surface. We can reconcile this observation by invoking an energy-dependent scattering rate with a particle-hole asymmetry, possibly rooted in the non-Fermi liquid nature of cuprates just above $p^{\star}$. Inside the pseudogap phase, for $ p < p^{\star}$, $S_{\rm a}$ is seen to rise at low temperature as previously reported, consistent with the drop in carrier density $n$ from $n \simeq 1 + p$ to $n \simeq p$ across $p^{\star}$ as inferred from other transport properties. In stark contrast, $S_{\rm c}$ at low temperature becomes negative below $p^{\star}$, a novel signature of the pseudogap phase. The sudden drop in $S_{\rm c}$ reveals a change in the electronic structure of Nd-LSCO upon crossing $p^{\star}$. We can exclude a profound change of the scattering across $p^{\star}$ and conclude that the change in the out-of-plane Seebeck coefficient originates from a transformation of the Fermi surface., Comment: 12 pages, 12 figures

Five transport coefficients of the cuprate superconductor Bi$_2$Sr$_{2-x}$La$_x$CuO$_{6+\delta}$ were measured in the normal state down to low temperature, reached by applying a magnetic field (up to 66T) large enough to suppress superconductivity. The electrical resistivity, Hall coefficient, thermal conductivity, Seebeck coefficient and thermal Hall conductivity were measured in two overdoped single crystals, with La concentration $x = 0.2$ ($T_{\rm c}=18$K) and $x = 0.0$ ($T_{\rm c}=10$K). The samples have dopings $p$ very close to the critical doping $p^{\star}$ where the pseudogap phase ends. The resistivity displays a linear dependence on temperature whose slope is consistent with Planckian dissipation. The Hall number $n_{\rm H}$ decreases with reduced $p$, consistent with a drop in carrier density from $n = 1+p$ above $p^{\star}$ to $n=p$ below $p^{\star}$. This drop in $n_{\rm H}$ is concomitant with a sharp drop in the density of states inferred from prior NMR Knight shift measurements. The thermal conductivity satisfies the Wiedemann-Franz law, showing that the pseudogap phase at $T = 0$ is a metal whose fermionic excitations carry heat and charge as do conventional electrons. The Seebeck coefficient diverges logarithmically at low temperature, a signature of quantum criticality. The thermal Hall conductivity becomes negative at low temperature, showing that phonons are chiral in the pseudogap phase. Given the observation of these same properties in other, very different cuprates, our study provides strong evidence for the universality of these five signatures of the pseudogap phase and its critical point., Comment: Added references

The nature of the pseudogap phase of cuprates remains a major puzzle. One of its new signatures is a large negative thermal Hall conductivity $\kappa_{\rm xy}$, which appears for dopings $p$ below the pseudogap critical doping $p^*$, but whose origin is as yet unknown. Because this large $\kappa_{\rm xy}$ is observed even in the undoped Mott insulator La$_2$CuO$_4$, it cannot come from charge carriers, these being localized at $p = 0$. Here we show that the thermal Hall conductivity of La$_2$CuO$_4$ is roughly isotropic, being nearly the same for heat transport parallel and normal to the CuO$_2$ planes, i.e. $\kappa_{\rm zy}(T) \approx \kappa_{\rm xy} (T)$. This shows that the Hall response must come from phonons, these being the only heat carriers able to move as easily normal and parallel to the planes . At $p > p^*$, in both La$_{\rm 1.6-x}$Nd$_{\rm 0.4}$Sr$_x$CuO$_4$ and La$_{\rm 1.8-x}$Eu$_{\rm 0.2}$Sr$_x$CuO$_4$ with $p = 0.24$, we observe no c-axis Hall signal, i.e. $\kappa_{\rm zy}(T) = 0$, showing that phonons have zero Hall response outside the pseudogap phase. The phonon Hall response appears immediately below $p^* = 0.23$, as confirmed by the large $\kappa_{\rm zy}(T)$ signal we find in La$_{1.6-x}$Nd$_{\rm 0.4}$Sr$_x$CuO$_4$ with $p = 0.21$. The microscopic mechanism by which phonons become chiral in cuprates remains to be identified. This mechanism must be intrinsic - from a coupling of phonons to their electronic environment - rather than extrinsic, from structural defects or impurities, as these are the same on both sides of $p^*$. This intrinsic phonon Hall effect provides a new window on quantum materials and it may explain the thermal Hall signal observed in other topologically nontrivial insulators.

The thermal conductivity $\kappa$ of the quasi-2D organic spin-liquid candidate EtMe$_3$Sb[Pd(dmit)$_2$]$_2$ (dmit-131) was measured at low temperatures, down to 0.07 K. We observe a vanishingly small residual linear term $\kappa_0/T$, in $\kappa/T$ vs $T$ as $T \to 0$. This shows that the low-energy excitations responsible for the sizeable residual linear term $\gamma$ in the specific heat $C$, seen in $C/T$ vs $T$ as $T \to 0,$ are localized. We conclude that there are no mobile gapless excitations in this spin liquid candidate, in contrast with a prior study of dmit-131 that reported a large $\kappa_0/T$ value [Yamashita et al., Science 328, 1246 (2010)]. Our study shows that dmit-131 is in fact similar to $\kappa$-(BEDT-TTF)$_2$Cu$_2$(CN)$_3$, another quasi-2D organic spin-liquid candidate where a vanishingly small $\kappa_0/T$ and a sizeable $\gamma$ are seen. We attribute heat conduction in these organic insulators without magnetic order to phonons undergoing strong spin-phonon scattering, as observed in several other spin-liquid materials., Comment: This revision includes a response to Yamashita's paper [M. Yamashita, Journal of the Physical Society of Japan 88, 083702 (2019)], data on a new sample, two new figures (Figs. 5 and 8), and a Supplementary Material (available upon request)

We present a detailed study of the temperature (T) and magnetic field (H) dependence of the electronic density of states (DOS) at the Fermi level, as deduced from specific heat and Knight shift measurements in underdoped YBa2Cu3Oy. We find that the DOS becomes field-independent above a characteristic field H_{DOS} and that the H_{DOS}(T) line displays an unusual inflection near the onset of the long range 3D charge-density wave order. The unusual S-shape of H_{DOS}(T) is suggestive of two mutually-exclusive orders that eventually establish a form of cooperation in order to coexist at low T. On theoretical grounds, such a collaboration could result from the stabilisation of a pair-density wave state, which calls for further investigations in this region of the phase diagram, Comment: 6 pages, 4 figures

The perfectly linear temperature dependence of the electrical resistivity observed as $T \rightarrow$ 0 in a variety of metals close to a quantum critical point is a major puzzle of condensed matter physics . Here we show that $T$-linear resistivity as $T \rightarrow$ 0 is a generic property of cuprates, associated with a universal scattering rate. We measured the low-temperature resistivity of the bi-layer cuprate Bi2212 and found that it exhibits a $T$-linear dependence with the same slope as in the single-layer cuprates Bi2201, Nd-LSCO and LSCO, despite their very different Fermi surfaces and structural, superconducting and magnetic properties. We then show that the $T$-linear coefficient (per CuO$_2$ plane), $A_1$, is given by the universal relation $A_1 T_F = h / 2e^2$, where $e$ is the electron charge, $h$ is the Planck constant and $T_F$ is the Fermi temperature. This relation, obtained by assuming that the scattering rate 1 / $\tau$ of charge carriers reaches the Planckian limit whereby $\hbar / \tau = k_B T$, works not only for hole-doped cuprates but also for electron-doped cuprates despite the different nature of their quantum critical point and strength of their electron correlations., Comment: main + SI

The three central phenomena of cuprate superconductors are linked by a common doping $p^{\star}$, where the enigmatic pseudogap phase ends, around which the superconducting phase forms a dome, and at which the resistivity exhibits an anomalous linear dependence on temperature as $T \to 0$. However, the fundamental nature of $p^{\star}$ remains unclear, in particular whether it marks a true quantum phase transition. We have measured the specific heat $C$ of the cuprates Eu-LSCO and Nd-LSCO at low temperature in magnetic fields large enough to suppress superconductivity, over a wide doping range across $p^{\star}$. As a function of doping, we find that the electronic term $C_{\rm el}$ is strongly peaked at $p^{\star}$, where it exhibits a $-T$log$T$ dependence as $T \to 0$. These are the classic signatures of a quantum critical point, as observed in heavy-fermion and iron-based superconductors where their antiferromagnetic phase ends. We conclude that the pseudogap phase of cuprates ends at a quantum critical point, whose associated fluctuations are most likely involved in the $d$-wave pairing and the anomalous scattering., Comment: Includes Supplementary Information

The properties of cuprate high-temperature superconductors are largely shaped by competing phases whose nature is often a mystery. Chiefly among them is the pseudogap phase, which sets in at a doping $p^*$ that is material-dependent. What determines $p^*$ is currently an open question. Here we show that the pseudogap cannot open on an electron-like Fermi surface, and can only exist below the doping $p_{FS}$ at which the large Fermi surface goes from hole-like to electron-like, so that $p^*$ $\leq$ $p_{FS}$. We derive this result from high-magnetic-field transport measurements in La$_{1.6-x}$Nd$_{0.4}$Sr$_x$CuO$_4$ under pressure, which reveal a large and unexpected shift of $p^*$ with pressure, driven by a corresponding shift in $p_{FS}$. This necessary condition for pseudogap formation, imposed by details of the Fermi surface, is a strong constraint for theories of the pseudogap phase. Our finding that $p^*$ can be tuned with a modest pressure opens a new route for experimental studies of the pseudogap., Comment: 15 pages, 5 figures, 7 supplemental figures

The thermal conductivity $\kappa$ of the Kondo insulator SmB$_6$ was measured at low temperature, down to 70 mK, in magnetic fields up to 15 T, on single crystals grown using both the floating-zone and the flux methods. The residual linear term $\kappa_0/T$ at $T \to 0$ is found to be zero in all samples, for all magnetic fields, in agreement with previous studies. There is therefore no clear evidence of fermionic heat carriers. In contrast to some prior data, we observe a large enhancement of $\kappa(T)$ with increasing field. The effect of field is anisotropic, depending on the relative orientation of field and heat current (parallel or perpendicular), and with respect to the cubic crystal structure. We interpret our data in terms of heat transport predominantly by phonons, which are scattered by magnetic impurities., Comment: publish version

The Seebeck coefficient $S$ of the cuprate YBa$_{2}$Cu$_{3}$O$_{y}$ was measured in magnetic fields large enough to suppress superconductivity, at hole dopings $p = 0.11$ and $p = 0.12$, for heat currents along the $a$ and $b$ directions of the orthorhombic crystal structure. For both directions, $S/T$ decreases and becomes negative at low temperature, a signature that the Fermi surface undergoes a reconstruction due to broken translational symmetry. Above a clear threshold field, a strong new feature appears in $S_{\rm b}$, for conduction along the $b$ axis only. We attribute this feature to the onset of 3D-coherent unidirectional charge-density-wave modulations seen by x-ray diffraction, also along the $b$ axis only. Because these modulations have a sharp onset temperature well below the temperature where $S/T$ starts to drop towards negative values, we infer that they are not the cause of Fermi-surface reconstruction. Instead, the reconstruction must be caused by the quasi-2D bidirectional modulations that develop at significantly higher temperature., Comment: 7 pages, 5 figures

We use the Nernst effect to delineate the boundary of the pseudogap phase in the temperature-doping phase diagram of cuprate superconductors. New data for the Nernst coefficient $\nu(T)$ of YBa$_{2}$Cu$_{3}$O$_{y}$ (YBCO), La$_{1.8-x}$Eu$_{0.2}$Sr$_x$CuO$_4$ (Eu-LSCO) and La$_{1.6-x}$Nd$_{0.4}$Sr$_x$CuO$_4$ (Nd-LSCO) are presented and compared with previous data including La$_{2-x}$Sr$_x$CuO$_4$ (LSCO). The temperature $T_\nu$ at which $\nu/T$ deviates from its high-temperature behaviour is found to coincide with the temperature at which the resistivity deviates from its linear-$T$ dependence, which we take as the definition of the pseudogap temperature $T^\star$- in agreement with gap opening detected in ARPES data. We track $T^\star$ as a function of doping and find that it decreases linearly vs $p$ in all four materials, having the same value in the three LSCO-based cuprates, irrespective of their different crystal structures. At low $p$, $T^\star$ is higher than the onset temperature of the various orders observed in underdoped cuprates, suggesting that these orders are secondary instabilities of the pseudogap phase. A linear extrapolation of $T^\star(p)$ to $p=0$ yields $T^\star(p\to 0)\simeq T_N(0)$, the N\'eel temperature for the onset of antiferromagnetic order at $p=0$, suggesting that there is a link between pseudogap and antiferromagnetism. With increasing $p$, $T^\star(p)$ extrapolates linearly to zero at $p\simeq p_{\rm c2}$, the critical doping below which superconductivity emerges at high doping, suggesting that the conditions which favour pseudogap formation also favour pairing. We also use the Nernst effect to investigate how far superconducting fluctuations extend above $T_{\rm c}$, as a function of doping, and find that a narrow fluctuation regime tracks $T_{\rm c}$, and not $T^\star$. This confirms that the pseudogap phase is not a form of precursor superconductivity., Comment: 24 pages and 26 figures including Appendix

The electrical resistivity $\rho$ and Hall coefficient R$_H$ of the tetragonal single-layer cuprate Nd-LSCO were measured in magnetic fields up to $H = 37.5$ T, large enough to access the normal state at $T \to 0$, for closely spaced dopings $p$ across the pseudogap critical point at $p^\star = 0.235$. Below $p^\star$, both coefficients exhibit an upturn at low temperature, which gets more pronounced with decreasing $p$. Taken together, these upturns show that the normal-state carrier density $n$ at $T = 0$ drops upon entering the pseudogap phase. Quantitatively, it goes from $n = 1 + p$ at $p = 0.24$ to $n = p$ at $p = 0.20$. By contrast, the mobility does not change appreciably, as revealed by the magneto-resistance. The transition has a width in doping and some internal structure, whereby R$_H$ responds more slowly than $\rho$ to the opening of the pseudogap. We attribute this difference to a Fermi surface that supports both hole-like and electron-like carriers in the interval $0.2 < p < p^\star$, with compensating contributions to R$_H$. Our data are in excellent agreement with recent high-field data on YBCO and LSCO. The quantitative consistency across three different cuprates shows that a drop in carrier density from $1 + p$ to $p$ is a universal signature of the pseudogap transition at $T=0$. We discuss the implication of these findings for the nature of the pseudogap phase., Comment: 11 pages, 12 figures

Superconductivity in cuprates peaks in the doping regime between a metal at high p and an insulator at low p. Understanding how the material evolves from metal to insulator is a fundamental and open question. Early studies in high magnetic fields revealed that below some critical doping an insulator-like upturn appears in the resistivity of cuprates at low temperature, but its origin has remained a puzzle. Here we propose that this 'metal-to-insulator crossover' is due to a drop in carrier density n associated with the onset of the pseudogap phase at a critical doping p*. We use high-field resistivity measurements on LSCO to show that the upturns are quantitatively consistent with a drop from n=1+p above p* to n=p below p*, in agreement with high-field Hall data in YBCO. We demonstrate how previously reported upturns in the resistivity of LSCO, YBCO and Nd-LSCO are explained by the same universal mechanism: a drop in carrier density by 1.0 hole per Cu atom., Comment: 16 pages, 4 figures, Supplementary material (12 pages, 7 figures)

The thermal conductivity of the iron-based superconductor FeSe was measured at temperatures down to 50 mK in magnetic fields up to 17 T. In zero magnetic field, the electronic residual linear term in the T = 0 limit, \kappa_0/T, is vanishingly small. Application of a magnetic field H causes no increase in \kappa_0/T initially. Those two observations show that there are no zero-energy quasiparticles that carry heat and therefore no nodes in the superconducting gap of FeSe. The full field dependence of \kappa_0/T has the classic shape of a two-band superconductor, such as MgB2: it rises exponentially at very low field, with a characteristic field H* << Hc2, and then more slowly up to the upper critical field Hc2. This shows that the superconducting gap is very small on one of the pockets in the Fermi surface of FeSe.

The thermal conductivity kappa of the iron-arsenide superconductor Ba1-xKxFe2As2 was measured for heat currents parallel and perpendicular to the tetragonal c axis at temperatures down to 50 mK and in magnetic fields up to 15 T. Measurements were performed on samples with compositions ranging from optimal doping (x = 0.34; Tc = 39 K) down to dopings deep into the region where antiferromagnetic order coexists with superconductivity (x = 0.16; Tc = 7 K). In zero field, there is no residual linear term in kappa(T) as T goes to 0 at any doping, whether for in-plane or inter-plane transport. This shows that there are no nodes in the superconducting gap. However, as x decreases into the range of coexistence with antiferromagnetism, the residual linear term grows more and more rapidly with applied magnetic field. This shows that the superconducting energy gap develops minima at certain locations on the Fermi surface and these minima deepen with decreasing x. We propose that the minima in the gap structure arise when the Fermi surface of Ba1-xKxFe2As2 is reconstructed by the antiferromagnetic order.

In the temperature-concentration phase diagram of most iron-based superconductors, antiferromagnetic order is gradually suppressed to zero at a critical point, and a dome of superconductivity forms around that point. The nature of the magnetic phase and its fluctuations is of fundamental importance for elucidating the pairing mechanism. In Ba{1-x}KxFe2As2 and Ba{1-x}NaxFe2As2, it has recently become clear that the usual stripe-like magnetic phase, of orthorhombic symmetry, gives way to a second magnetic phase, of tetragonal symmetry, near the critical point, between x = 0.24 and x = 0.28. Here we report measurements of the electrical resistivity of Ba{1-x}KxFe2As2 under applied hydrostatic pressures up to 2.75 GPa, for x = 0.22, 0.24 and 0.28. We track the onset of the tetragonal magnetic phase using the sharp anomaly it produces in the resistivity. In the temperature-concentration phase diagram of Ba{1-x}KxFe2As2, we find that pressure greatly expands the tetragonal magnetic phase, while the stripe-like phase shrinks. This raises the interesting possibility that the fluctuations of the former phase might be involved in the pairing mechanism responsible for the superconductivity.

The Seebeck coefficient $S$ of the cuprate superconductor La$ _{2-x} $Sr$_{x} $CuO$ _{4}$ (LSCO) was measured in magnetic fields large enough to access the normal state at low temperatures, for a range of Sr concentrations from $x = 0.07$ to $x = 0.15$. For $x = 0.11$, 0.12, 0.125 and 0.13, $S/T$ decreases upon cooling to become negative at low temperatures. The same behavior is observed in the Hall coefficient $R_{H}(T)$. In analogy with other hole-doped cuprates at similar hole concentrations $p$, the negative $S$ and $R_{H}$ show that the Fermi surface of LSCO undergoes a reconstruction caused by the onset of charge-density-wave modulations. Such modulations have indeed been detected in LSCO by X-ray diffraction in precisely the same doping range. Our data show that in LSCO this Fermi-surface reconstruction is confined to $0.085 < p < 0.15$. We argue that in the field-induced normal state of LSCO, charge-density-wave order ends at a critical doping $p_{\rm CDW} = 0.15 \pm 0.005$, well below the pseudogap critical doping $p^\star \simeq 0.19$.

The pseudogap is a central puzzle of cuprate superconductors. Its connection to the Mott insulator at low doping $p$ remains ambiguous and its relation to the charge order that reconstructs the Fermi surface at intermediate $p$ is still unclear. Here we use measurements of the Hall coefficient in magnetic fields up to 88 T to show that Fermi-surface reconstruction by charge order in YBa$_2$Cu$_3$O$_y$ ends sharply at a critical doping $p = 0.16$, distinctly lower than the pseudogap critical point at $p^* = 0.19$. This shows that pseudogap and charge order are separate phenomena. We then find that the change of carrier density from $n = 1 + p$ in the conventional metal at high p to $n = p$ at low $p$ - a signature of the lightly doped cuprates - starts at $p^*$. This shows that pseudogap and antiferromagnetic Mott insulator are linked.

Quantum oscillations and negative Hall and Seebeck coefficients at low temperature and high magnetic field have shown the Fermi surface of underdoped cuprates to contain a small closed electron pocket. It is thought to result from a reconstruction by charge order, but whether it is the order seen by NMR and ultrasound above a threshold field or the short-range modulations seen by X-ray diffraction in zero field is unclear. Here we use measurements of the thermal Hall conductivity in YBCO to show that Fermi-surface reconstruction occurs only above a sharply defined onset field, equal to the transition field seen in ultrasound. This reveals that electrons do not experience long-range broken translational symmetry in the zero-field ground state, and hence in zero field there is no quantum critical point for the onset of charge order as a function of doping., Comment: 20 pages and 5 figures in Main text + 9 pages and 6 figures in Supplementary material

Nematicity has emerged as a key feature of cuprate superconductors, but its link to other fundamental properties such as superconductivity, charge order and the pseudogap remains unclear. Here we use measurements of transport anisotropy in YBa$_2$Cu$_3$O$_y$ to distinguish two types of nematicity. The first is associated with short-range charge-density-wave modulations in a doping region near $p = 0.12$. It is detected in the Nernst coefficient, but not in the resistivity. The second type prevails at lower doping, where there are spin modulations but no charge modulations. In this case, the onset of in-plane anisotropy - detected in both the Nernst coefficient and the resistivity - follows a line in the temperature-doping phase diagram that tracks the pseudogap energy. We discuss two possible scenarios for the latter nematicity., Comment: 8 pages and 7 figures. Main text and supplementary material now combined into single article

The recent detection of charge-density modulations in YBa2Cu3Oy and other cuprate superconductors raises new questions about the normal state of underdoped cuprates. In one class of theories, the modulations are intertwined with pairing in a dual state, expected to persist up to high magnetic fields as a vortex liquid. In support of such a state, specific heat and magnetisation data on YBa2Cu3Oy have been interpreted in terms of a vortex liquid persisting above the vortex-melting field Hvs at T = 0. Here we report high-field measurements of the electrical and thermal Hall conductivities in YBa2Cu3O6.54 that allow us to probe the Wiedemann-Franz law, a sensitive test of the presence of superconductivity in a metal. In the T = 0 limit, we find that the law is satisfied for fields immediately above Hvs. This rules out the existence of a vortex liquid and it places strict constraints on the nature of the normal state in underdoped cuprates., Comment: 8 pages, includes main text and supplementary information

Cuprate superconductors have a universal tendency to form charge density-wave (CDW) order which competes with superconductivity and is strongest at a doping $p \simeq 0.12$. Here we show that in the archetypal cuprate YBa$_{2}$Cu$_{3}$O$_{y}$ (YBCO) pressure suppresses charge order, but does not affect the pseudogap phase. This is based on transport measurements under pressure, which reveal that the onset of the pseudogap at $T^*$ is independent of pressure, while the negative Hall effect, a clear signature of CDW order in YBCO, is suppressed by pressure. We also find that pressure and magnetic field shift the superconducting transition temperature $T_{\rm c}$ of YBCO in the same way as a function of doping - but in opposite directions - and most effectively at $p \simeq 0.12$. This shows that the competition between superconductivity and CDW order can be tuned in two ways, either by suppressing superconductivity with field or suppressing CDW order by pressure. Based on existing high-pressure data and our own work, we observe that when CDW order is fully suppressed at high pressure, the so-called "1/8 anomaly" in the superconducting dome vanishes, revealing a smooth $T_{\rm c}$ dome which now peaks at $p \simeq 0.13$. We propose that this $T_{\rm c}$ dome is shaped by the competing effects of the pseudogap phase below its critical point $p^{\star} \sim 0.19$ and spin order at low doping., Comment: New Hall data added. 11 pages, 9 figures and 2 tables

We report a sudden reversal in the pressure dependence of Tc in the iron-based superconductor RbFe2As2, at a critical pressure Pc = 11 kbar. Combined with our prior results on KFe2As2 and CsFe2As2, we find a universal V-shaped phase diagram for Tc vs P in these fully hole-doped 122 materials, when measured relative to the critical point (Pc, Tc). From measurements of the upper critical field Hc2(T) under pressure in KFe2As2 and RbFe2As2, we observe the same two-fold jump in (1/Tc)(-dHc2/dT) across Pc, compelling evidence for a sudden change in the structure of the superconducting gap. We argue that this change is due to a transition from one pairing state to another, with different symmetries on either side of Pc. We discuss a possible link between scattering and pairing, and a scenario where a d-wave state favored by high-Q scattering at low pressure changes to a state with s+- symmetry favored by low-Q scattering at high pressure., Comment: 9 pages, 11 figures, 1 table, Published

The Fermi surface of a metal is the fundamental basis from which its properties can be understood. In underdoped cuprate superconductors, the Fermi surface undergoes a reconstruction that produces a small electron pocket, but whether there is another, as yet undetected portion to the Fermi surface is unknown. Establishing the complete topology of the Fermi surface is key to identifying the mechanism responsible for its reconstruction. Here we report the discovery of a second Fermi pocket in underdoped YBa2Cu3Oy, detected as a small quantum oscillation frequency in the thermoelectric response and in the c-axis resistance. The field-angle dependence of the frequency demonstrates that it is a distinct Fermi surface and the normal-state thermopower requires it to be a hole pocket. A Fermi surface consisting of one electron pocket and two hole pockets with the measured areas and masses is consistent with a Fermi-surface reconstruction caused by the charge-density-wave order observed in YBa2Cu3Oy, provided other parts of the reconstructed Fermi surface are removed by a separate mechanism, possibly the pseudogap., Comment: 23 pages, 5 figures

Geometrical Berry phase is recognized as having profound implications for the properties of electronic systems. Over the last decade, Berry phase has been essential to our understanding of new materials, including graphene and topological insulators. The Berry phase can be accessed via its contribution to the phase mismatch in quantum oscillation experiments, where electrons accumulate a phase as they traverse closed cyclotron orbits in momentum space. The high-temperature cuprate superconductors are a class of materials where the Berry phase is thus far unknown despite the large body of existing quantum oscillations data. In this report we present a systematic Berry phase analysis of Shubnikov - de Haas measurements on the hole-doped cuprates YBa$_2$Cu$_3$O$_{y}$, YBa$_2$Cu$_4$O$_8$, HgBa$_2$CuO$_{4 + \delta}$, and the electron-doped cuprate Nd$_{2-x}$Ce$_x$CuO$_4$. For the hole-doped materials, a trivial Berry phase of 0 mod $2\pi$ is systematically observed whereas the electron-doped Nd$_{2-x}$Ce$_x$CuO$_4$ exhibits a significant non-zero Berry phase. These observations set constraints on the nature of the high-field normal state of the cuprates and points towards contrasting behaviour between hole-doped and electron-doped materials. We discuss this difference in light of recent developments related to charge density-wave and broken time-reversal symmetry states., Comment: new version with added supplementary information

We report a sudden reversal in the pressure dependence of Tc in the iron-based superconductor CsFe2As2, similar to that discovered recently in KFe2As2 [Tafti et al., Nat. Phys. 9, 349 (2013)]. As in KFe2As2, we observe no change in the Hall coefficient at the zero temperature limit, again ruling out a Lifshitz transition across the critical pressure Pc. We interpret the Tc reversal in the two materials as a phase transition from one pairing state to another, tuned by pressure, and investigate what parameters control this transition. Comparing samples of different residual resistivity, we find that a 6-fold increase in impurity scattering does not shift Pc. From a study of X-ray diffraction on KFe2As2 under pressure, we report the pressure dependence of lattice constants and As-Fe-As bond angle. The pressure dependence of these lattice parameters suggests that Pc should be significantly higher in CsFe2As2 than in KFe2As2, but we find on the contrary that Pc is lower in CsFe2As2. Resistivity measurements under pressure reveal a change of regime across Pc, suggesting a possible link between inelastic scattering and pairing symmetry., Comment: 7 figures, 7 pages, published

The upper critical field Hc2 is a fundamental measure of the pairing strength, yet there is no agreement on its magnitude and doping dependence in cuprate superconductors. We have used thermal conductivity as a direct probe of Hc2 in the cuprates YBa2Cu3Oy and YBa2Cu4O8 to show that there is no vortex liquid at T = 0, allowing us to use high-field resistivity measurements to map out the doping dependence of Hc2 across the phase diagram. Hc2(p) exhibits two peaks, each located at a critical point where the Fermi surface undergoes a transformation. The condensation energy obtained directly from Hc2, and previous Hc1 data, undergoes a 20-fold collapse below the higher critical point. These data provide quantitative information on the impact of competing phases in suppressing superconductivity in cuprates., Comment: to appear in Nature Communications; Supplementary Information file available upon request

The pairing mechanism in iron-based superconductors is the subject of ongoing debate. Proximity to an antiferromagnetic phase suggests that pairing is mediated by spin fluctuations, but orbital fluctuations have also been invoked. The former typically favour a pairing state of extended s-wave symmetry with a gap that changes sign between electron and hole Fermi surfaces (s+-), while the latter yield a standard s-wave state without sign change (s++). Here we show that applying pressure to KFe2As2 induces a change of pairing state. The critical temperature Tc decreases with pressure initially, and then suddenly increases, above a critical pressure Pc. The constancy of the Hall coefficient through Pc rules out a change in the Fermi surface. There is compelling evidence that the pairing state below Pc is d-wave, from bulk measurements at ambient pressure. Above Pc, the high sensitivity to disorder argues for a particular kind of s+- state. The change from d-wave to s-wave is likely to proceed via an unusual s + id state that breaks time-reversal symmetry. The proximity of two distinct pairing states found here experimentally is natural given the near degeneracy of d-wave and s+- states found theoretically. These findings make a compelling case for spin-fluctuation-mediated superconductivity in this key iron-arsenide material., Comment: 8 pages, 7 figures, including SI, published in Nature Physics

We report superconductivity in the ternary half-Heusler compound LuPtBi, with Tc = 1.0 K and Hc2 = 1.6 T. The crystal structure of LuPtBi lacks inversion symmetry, hence the material is a noncentrosymmetric superconductor. Magnetotransport data show semimetallic behavior in the normal state, which is evidence for the importance of spin-orbit interaction. Theoretical calculations indicate that the strong spin-orbit interaction in LuPtBi should cause strong band inversion, making this material a promising candidate for 3D topological superconductivity., Comment: 5 pages, 4 figures, Table I revised, discussion of the Pauli limit revised

Charge density-wave order has been observed in cuprate superconductors whose crystal structure breaks the square symmetry of the CuO2 planes, such as orthorhombic YBa2Cu3Oy (YBCO), but not so far in cuprates that preserve that symmetry, such as tetragonal HgBa2CuO4+d (Hg1201). We have measured the Hall (R_H), Seebeck (S), and Nernst coefficients of underdoped Hg1201 in magnetic fields large enough to suppress superconductivity. The high-field R_H(T) and S(T) are found to drop with decreasing temperature and become negative, as also observed in YBCO at comparable doping. In YBCO, the negative R_H and S are signatures of a small electron pocket caused by Fermi-surface reconstruction, attributed to charge density-wave modulations observed in the same range of doping and temperature. We deduce that a similar Fermi-surface reconstruction takes place in Hg1201, evidence that density-wave order exists in this material. A striking similarity is also found in the normal-state Nernst coefficient, further supporting this interpretation. Given the model nature of Hg1201, Fermi-surface reconstruction appears to be common to all hole-doped cuprates, suggesting that density-wave order is a fundamental property of these materials.

The transition temperature Tc of cuprate superconductors falls when the doping p is reduced below a certain optimal value. It is unclear whether this fall is due to strong phase fluctuations or to a decrease in the pairing gap. Different interpretations of photoemission data disagree on the evolution of the pairing gap and different estimates of the upper critical field Hc2 are in sharp contradiction. Here we resolve this contradiction by showing that superconducting fluctuations in the underdoped cuprate Eu-LSCO, measured via the Nernst effect, have a characteristic field scale that falls with underdoping. The critical field Hc2 dips at p = 0.11, showing that superconductivity is weak where stripe order is strong. In the archetypal cuprate superconductor YBCO, Hc2 extracted from other measurements has the same doping dependence, also with a minimum at p = 0.11, again where stripe order is present. We conclude that competing states such as stripe order weaken superconductivity and this, rather than phase fluctuations, causes Tc to fall as cuprates become underdoped., Comment: Supplementary Information file available upon request; Nature Physics (2012)

The nature of the pairing state in iron-based superconductors is the subject of much debate. Here we argue that in one material, the stoichiometric iron pnictide KFe2As2, there is overwhelming evidence for a d-wave pairing state, characterized by symmetry-imposed vertical line nodes in the superconducting gap. This evidence is reviewed, with a focus on thermal conductivity and the strong impact of impurity scattering on the critical temperature Tc. We then compare KFe2As2 to Ba0.6K0.4Fe2As2, obtained by Ba substitution, where the pairing symmetry is s-wave and the Tc is ten times higher. The transition from d-wave to s-wave within the same crystal structure provides a rare opportunity to investigate the connection between band structure and pairing mechanism. We also compare KFe2As2 to the nodal iron-based superconductor LaFePO, for which the pairing symmetry is probably not d-wave, but more likely s-wave with accidental line nodes.

The electrical resistivity rho of the iron-arsenide superconductor Ba1-xKxFe2As2 was measured in applied pressures up to 2.6 GPa for four underdoped samples, with x = 0.16, 0.18, 0.19 and 0.21. The antiferromagnetic ordering temperature T_N, detected as a sharp anomaly in rho(T), decreases linearly with pressure. At pressures above around 1.0 GPa, a second sharp anomaly is detected at a lower temperature T_0, which rises with pressure. We attribute this second anomaly to the onset of a phase that causes a reconstruction of the Fermi surface. This new phase expands with increasing x and it competes with superconductivity. We discuss the possibility that a second spin-density wave orders at T_0, with a Q vector distinct from that of the spin-density wave that sets in at T_N., Comment: Two higher K concentrations were added, revealing a steady expansion of the new phase in the T-P phase diagram

The thermal conductivity of the iron-arsenide superconductor KFe2As2 was measured down to 50 mK for a heat current parallel and perpendicular to the tetragonal c-axis. A residual linear term (RLT) at T=0 is observed for both current directions, confirming the presence of nodes in the superconducting gap. Our value of the RLT in the plane is equal to that reported by Dong et al. [Phys. Rev. Lett. 104, 087005 (2010)] for a sample whose residual resistivity was ten times larger. This independence of the RLT on impurity scattering is the signature of universal heat transport, a property of superconducting states with symmetry-imposed line nodes. This argues against an s-wave state with accidental nodes. It favors instead a d-wave state, an assignment consistent with five additional properties: the magnitude of the critical scattering rate for suppressing Tc to zero; the magnitude of the RLT, and its dependence on current direction and on magnetic field; the temperature dependence of the thermal conductivity., Comment: To appear in Physical Review Letters

The thermal conductivity k of the iron-arsenide superconductor K-Ba122 was measured down to 50 mK in a magnetic field up to 15 T, for a heat current parallel and perpendicular to the tetragonal c axis. In the range from optimal doping (x ~ 0.4) down to x = 0.16, there is no residual linear term in k(T) at T = 0, showing that there are no nodes in the superconducting gap anywhere on the Fermi surface. Upon crossing below x = 0.16, a large residual linear term suddenly appears, signaling the onset of nodes in the superconducting gap, most likely vertical line nodes running along the c axis. We discuss two scenarios: 1) accidental nodes in an s-wave gap, resulting from a strong modulation of the gap around the Fermi surface, in which minima deepen rapidly with underdoping; 2) a phase transition from a nodeless s-wave state to a d-wave state, in which nodes are imposed by symmetry., Comment: This paper has been replaced by arXiv: 1602.03914

The thermal conductivity k of the iron-arsenide superconductor LiFeAs (Tc ~ 18K) was measured in single crystals at temperatures down to T~50mK and in magnetic fields up to H=17T, very close to the upper critical field Hc2~18T. For both directions of the heat current, parallel and perpendicular to the tetragonal c-axis, a negligible residual linear term k/T is found as T ->0, revealing that there are no zero-energy quasiparticles in the superconducting state. The increase in k with magnetic field is the same for both current directions and it follows closely the dependence expected for an isotropic superconducting gap. There is no evidence of multi-band character, whereby the gap would be different on different Fermi-surface sheets. These findings show that the superconducting gap in LiFeAs is isotropic in 3D, without nodes or deep minima anywhere on the Fermi surface. Comparison with other iron-pnictide superconductors suggests that a nodeless isotropic gap is a common feature at optimal doping (maximal Tc)., Comment: 4 pages, 3 figures

Quantum oscillations have revealed the presence of a small pocket in the Fermi surface of the cuprate superconductor YBCO, whose nature and origin are the subject of much debate. Interpretations include electron and hole pockets; scenarios include Fermi-surface reconstruction by antiferromagnetism, d-density-wave order, and stripe order. Here we report quantum oscillations in the Seebeck and Nernst coefficients of YBCO and show, from the magnitude and sign of the Seebeck coefficient, that they come from an electron pocket. Using measurements of the Seebeck coefficient as a function of hole doping p, we show that the evolution of the Fermi surface in YBCO is the same as in Eu-LSCO, a cuprate where stripe order (a modulation of spin and charge densities) is well established. The electron pocket is most prominent where stripe order is strongest, at p = 1/8. This shows that Fermi-surface reconstruction is a generic mechanism of underdoped cuprates, intimately related to stripe order., Comment: 15 pages, 5 figures, Supplementary information now integrated into article

The thermal conductivity k of the iron-arsenide superconductor Ba(Fe_{1-x}Co_x)_2As_2 was measured down to 50 mK for a heat current parallel (k_c) and perpendicular (k_a) to the tetragonal c axis, for seven Co concentrations from underdoped to overdoped regions of the phase diagram (0.038 < x < 0.127). A residual linear term k_c0/T is observed in the T = 0 limit when the current is along the c axis, revealing the presence of nodes in the gap. Because the nodes appear as x moves away from the concentration of maximal T_c, they must be accidental, not imposed by symmetry, and are therefore compatible with an s_{+/-} state, for example. The fact that the in-plane residual linear term k_a0/T is negligible at all x implies that the nodes are located in regions of the Fermi surface that contribute strongly to c-axis conduction and very little to in-plane conduction. Application of a moderate magnetic field (e.g. H_c2/4) excites quasiparticles that conduct heat along the a axis just as well as the nodal quasiparticles conduct along the c axis. This shows that the gap must be very small (but non-zero) in regions of the Fermi surface which contribute significantly to in-plane conduction. These findings can be understood in terms of a strong k dependence of the gap Delta(k) which produces nodes on a Fermi surface sheet with pronounced c-axis dispersion and deep minima on the remaining, quasi-two-dimensional sheets., Comment: 12 pages, 13 figures.

The electrical resistivity of the quasi-1D organic superconductor (TMTSF)2PF6 was recently measured at low temperature from the critical pressure needed to suppress the spin-density-wave state up to a pressure where superconductivity has almost disappeared. This data revealed a direct correlation between the onset of superconductivity at Tc and the strength of a non-Fermi-liquid linear term in the normal-state resistivity, going as r(T) = r0 + AT + BT2 at low temperature, so that A goes to 0 as Tc goes to 0. Here we show that the contribution of low-frequency antiferromagnetic fluctuations to the spin-lattice relaxation rate is also correlated with this non-Fermi-liquid term AT in the resistivity. These correlations suggest that anomalous scattering and pairing have a common origin, both rooted in the low-frequency antiferromagnetic fluctuations measured by NMR. A similar situation may also prevail in the recently-discovered iron-pnictide superconductors., Comment: ISCOM'09 proceedings to be published in Physica B

Pulsed field measurements of the Hall resistivity and magnetoresistance of underdoped YBa2Cu4O8 are analyzed self-consistently using a simple model based on coexisting electron and hole carriers. The resultant mobilities and Hall numbers are found to vary markedly with temperature. The conductivity of the hole carriers drops by one order of magnitude below 30 K, explaining the absence of quantum oscillations from these particular pockets. Meanwhile the Hall coefficient of the electron carriers becomes strongly negative below 50 K. The overall quality of the fits not only provides strong evidence for Fermi-surface reconstruction in Y-based cuprates, it also strongly constrains the type of reconstruction that might be occurring., Comment: 5 pages, 4 figures, updated after publication in Physical Review B (Rapid Communication)

The temperature and magnetic field dependence of the in-plane thermal conductivity $\kappa$ of the iron-arsenide superconductor Ba(Fe$_{1-x}$Co$_x$)$_2$As$_2$ was measured down to $T \simeq 50$ mK and up to $H = 15$ T as a function of Co concentration $x$ in the range 0.048 $ \leq x \leq $ 0.114. In zero magnetic field, a negligible residual linear term in $\kappa/T$ as $T \to 0$ at all $x$ shows that there are no zero-energy quasiparticles and hence the superconducting gap has no nodes in the $ab$-plane anywhere in the phase diagram. However, the field dependence of $\kappa$ reveals a systematic evolution of the superconducting gap with doping $x$, from large everywhere on the Fermi surface in the underdoped regime, as evidenced by a flat $\kappa (H)$ at $T \to 0$, to strongly $k$-dependent in the overdoped regime, where a small magnetic field can induce a large residual linear term, indicative of a deep minimum in the gap magnitude somewhere on the Fermi surface. This shows that the superconducting gap structure has a strongly $k$-dependent amplitude around the Fermi surface only outside the antiferromagnetic/orthorhombic phase., Comment: version accepted for publication in Physical Review Letters; new title, minor revision, revised fig.1, and updated references

The thermal conductivity $\kappa$ of the iron-arsenide superconductor Ba$_{1-x}$K$_x$Fe$_2$As$_2$ ($T_c \simeq$ 30 K) was measured in single crystals at temperatures down to $T \simeq 50$ mK ($\simeq T_c$/600) and in magnetic fields up to $H = 15$ T ($\simeq H_{c2}$/4). A negligible residual linear term in $\kappa/T$ as $T \to 0$ shows that there are no zero-energy quasiparticles in the superconducting state. This rules out the existence of line and in-plane point nodes in the superconducting gap, imposing strong constraints on the symmetry of the order parameter. It excludes d-wave symmetry, drawing a clear distinction between these superconductors and the high-$T_c$ cuprates. However, the fact that a magnetic field much smaller than $H_{c2}$ can induce a residual linear term indicates that the gap must be very small on part of the Fermi surface, whether from strong anisotropy or band dependence, or both.

Combining old and new de Haas-van Alphen (dHvA) and magnetoresistance data, we arrive at a detailed picture of the Fermi surface of the heavy fermion superconductor UPt3. Our work was partially motivated by a new proposal that two 5f valence electrons per formula unit in UPt3 are localized by correlation effects -- agreement with previous dHvA measurements of the Fermi surface was invoked in its support. Comprehensive comparison with our new observations shows that this 'partially localized' model fails to predict the existence of a major sheet of the Fermi surface, and is therefore less compatible with experiment than the originally proposed 'fully itinerant' model of the electronic structure of UPt3. In support of this conclusion, we offer a more complete analysis of the fully itinerant band structure calculation, where we find a number of previously unrecognized extremal orbits on the Fermi surface., Comment: 23 pages, 12 figures, latex, iopart class

The de Haas-van Alphen effect was observed in the underdoped cuprate YBa$_2$Cu$_3$O$_{6.5}$ via a torque technique in pulsed magnetic fields up to 59 T. Above an irreversibility field of $\sim$30 T, the magnetization exhibits clear quantum oscillations with a single frequency of 540 T and a cyclotron mass of 1.76 times the free electron mass, in excellent agreement with previously observed Shubnikov-de Haas oscillations. The oscillations obey the standard Lifshitz-Kosevich formula of Fermi-liquid theory. This thermodynamic observation of quantum oscillations confirms the existence of a well-defined, close and coherent, Fermi surface in the pseudogap phase of cuprates., Comment: published version

We report the observation of Shubnikov-de Haas oscillations in the underdoped cuprate superconductor YBa$_2$Cu$_4$O$_8$ (Y124). For field aligned along the c-axis, the frequency of the oscillations is $660\pm 30$ T, which corresponds to $\sim 2.4$ % of the total area of the first Brillouin zone. The effective mass of the quasiparticles on this orbit is measured to be $2.7\pm0.3$ times the free electron mass. Both the frequency and mass are comparable to those recently observed for ortho-II YBa$_2$Cu$_3$O$_{6.5}$ (Y123-II). We show that although small Fermi surface pockets may be expected from band structure calculations in Y123-II, no such pockets are predicted for Y124. Our results therefore imply that these small pockets are a generic feature of the copper oxide plane in underdoped cuprates., Comment: v2: Version of paper accepted for publication in Physical Review Letters. Only minor changes to the text and references