WorksPhysics and Astronomy
Divalent Yb in the CaPdBi Structure Type: Crystal Growth and Physical Properties of YbPdBi
Abstract
We report on the single-crystal growth, crystal structure, and physical properties of YbPdBi. Single crystal x-ray diffraction shows that the material crystallizes in the CaPdBi structure-type at ambient conditions, described by orthorhombic symmetry . Magnetization measurements reveal a nearly isotropic diamagnetic response, indicating that Yb is stabilized in a non-magnetic divalent configuration. Electrical resistivity measurements show a metallic behavior. First-principles calculations including spin–orbit coupling describe a metal whose states at the Fermi level are dominated by Bi orbitals, with a density of states within 1% of that of the isoelectronic compound CaPdBi, consistent with a filled and chemically inert shell. They further show that the nonsymmorphic symmetry of the crystal enforces fourfold-degenerate nodal lines on the Brillouin-zone boundary, four of which cross the Fermi level and give symmetry-enforced Dirac points with a linear transverse dispersion. Despite strong metallic and covalent bonding within the Pd-Bi network, YbPdBi is realized as a rare-earth analog of the CaPdBi structure type with divalent Yb ions. This result offers a potential avenue to discover other rare-earth analogs within this structural family.
Introduction
In condensed matter physics and materials science, the drive to synthesize entirely new materials with novel crystal structures is the primary engine for discovering unprecedented quantum phenomena. New Yb-based intermetallic compounds have played an important role in understanding strongly correlated electron physics. This is mainly due to the fact that the valence of the Yb ion can fluctuate between a non-magnetic Yb () configuration and a magnetic Yb () configuration [1]. Compounds with either of the two configurations exist depending on the local environment of the Yb and the hybridization strength with the conduction electrons. Several Yb-based heavy fermion systems have been at the forefront of research. For example, YbRhSi is a prime example for heavy fermion quantum criticality [2, 3, 4]; half-Heusler compounds like YPTBi are famous for hosting topological semimetal and superconductivity with a very high angular momentum Cooper pairing [5, 6, 7]; YbInCu shows a spectacular first order valence transition from the Yb to an intermediate Yb valence as a function of temperature [8]; or YbB is a prime candidate for a topological Kondo Insulator [9, 10]. Despite having an insulating bulk gap driven by correlations, quantum oscillations have been observed at high fields, pointing to exotic, topologically protected metallic surface states [11, 12, 13, 14].
Recently, Pd-Bi compounds have been intensively studied, as they are hallmark systems for physics with strong spin-orbit coupling and superconductivity. The extremely strong spin-orbit coupling in Bi often leads to topological band properties with band inversion and topological surface states. Pd-Bi systems like PdBi and -PdBi are intrinsic bulk superconductors with topological states on the surface [15, 16, 17]. This means that the topological surface states and the superconducting gap coexist in the same system. Introducing Yb into the Pd-Bi system can add strong electronic correlations and open the way to heavy-fermion physics. As an example, YbPdBi is a half-Heusler heavy fermion compound with a strongly enhanced Sommerfeld coefficient of the specific heat ( J/mol·K) with a possible antiferromagnetic order near 1 K [18, 19, 20]. Due to its non-centrosymmetric crystal structure (space group ) it is a candidate Weyl semimetal, similar to YbPtBi.
Beyond the known half-Heusler phase, the Yb-Pd-Bi system also offers an opportunity for discovering new rare-earth intermetallics. In this respect, the reported alkaline-earth compound CaPdBi [21, 22] is a particularly useful structural prototype. Its Pd-Bi framework accommodates divalent alkaline-earth cations, suggesting that substitution by Yb may be possible if ytterbium adopts its divalent, non-magnetic configuration. The CaPdBi crystal structure, , is centrosymmetric, so that in a non-magnetic compound the combination of inversion and time reversal makes every band 2-fold degenerate throughout the zone and forbids Weyl nodes. At the same time, it is nonsymmorphic: its two glide planes carry half-lattice translations, and force bands to be 4-fold degenerate on parts of the zone boundary [23]. Together with the strong spin–orbit coupling of Bi, this makes the CaPdBi framework a natural setting for symmetry-enforced Dirac degeneracies, a question that the two existing scalar-relativistic studies of CaPdBi could not address exhaustively [21, 22].
Here we report on the synthesis, crystal structure and physical properties of single-crystalline YbPdBi. It crystallizes in the CaPdBi structure type. Magnetization, transport and heat-capacity measurements indicate metallic behavior with divalent Yb. These measurements are complemented by fully relativistic density-functional calculations, which place the Fermi level in a Bi band complex essentially identical to that of CaPdBi, and which locate four symmetry-enforced Dirac points at the Fermi level on the 4-fold nodal lines of the zone boundary.
Methods
Crystal growth and experimental probes
Single crystals of YbPdBi were grown using a self-flux method. The clean metals, Yb (≥ 99%, G. Holdron), Pd (Alfa Aesar, 99.98+%) and Bi (Puratronic(R), 99.98+%), were introduced in a Canfield crucible set [24, 25] in the 1:1:7, Yb:Pd:Bi molar ratio and sealed in an evacuated silica ampoule. The ampoule was heated to 1150 °C in 48 h, held there for 8 h, and slowly cooled to 500 °C at 1.6 °C per hour. At 500 °C, the formed crystals were filtered by removing the ampoule from the furnace, inverting it, and immediately spinning it in a centrifuge [26]. The as-synthesized YbPdBi single crystals exhibit a metallic gray color and needle-like morphology.
Energy dispersive X-ray analysis performed in a FE-SEM Zeiss Ultra+ scanning electron microscope with a Bruker SDD detector confirmed the YbPdBi stoichiometry.
Single-crystal X-ray diffraction data were acquired at the side-station of ID28 beamline at European Synchrotron Radiation Facility (ESRF), Grenoble, France, employing a wavelength () of 0.5701 Å. The detector used is a Dectris PILATUS3 1M area detector (pixel size 172 × 172 m) on a four-circle Euler diffractometer with angular step rotation of 0.25° and exposure of 0.25 seconds. Further crystallographic information can be inferred from Table I. Integration and subsequent refinement of the crystal structure was done using CRYSAL-ISPRO [27] and JANA 2006 [28].
| Crystal system | Orthorhombic |
| Space group | |
| Space group no. | 55 |
| (Å) | 10.8109(4) |
| (Å) | 17.0493(7) |
| (Å) | 4.1428(1) |
| Volume (Å) | 763.6(3) |
| 2 | |
| Wavelength (Å) | 0.5701 |
| Detector distance (mm) | 170 |
| Rotation per image () (°) | 0.25 |
| offset (°) | -45 |
| offset (°) | -19 |
| (Å) | 0.611578 |
| Absorption, (mm) | 64.806 |
| No. of reflections measured | 1816 |
| No. of unique reflections (obs/all) | 741/741 |
| (obs/all) | 0.0250/0.0250 |
| No. of parameters | 47 |
| (obs/all) | 0.0448/0.0448 |
| (all) | 0.0645 |
| GoF (obs/all) | 5.09/5.09 |
| (e Å) | -5.5, 4.36 |
Table I. TABLE I. Crystallographic data of YbPdBi at 293 K. The criterion of observability is .
Magnetization measurements were performed using an MPMS3 SQUID magnetometer (Quantum Design) in VSM scan mode down to 2 K and up to 7 T.
Resistivity measurements were performed in a physical property measurement system (PPMS, Quantum Design) using a four-point lock-in technique; 15 m Au wires were spot-welded to the sample for electrical contacts. The specific heat was measured using the PPMS with the heat pulse relaxation method.
First-principles calculations
Electronic-structure calculations were performed with the plane-wave pseudopotential code QUANTUM ESPRESSO (v7.2) [29] using the PBE generalized-gradient functional [30] and projector augmented-wave pseudopotentials [31] from PSLIBRARY 1.0.0 [32]. For Yb the shell is treated as core states, so that Yb is divalent by construction, as established by the diamagnetic susceptibility measurements. This choice makes Yb and Ca isovalent, , with 444 valence electrons in the 30-atom cell. Wavefunction and charge-density cutoffs of 70 and 700 Ry were used, well above the values suggested by the dataset generation tests (56 and 539 Ry for Pd, the most demanding species). These values were chosen by running convergence tests on total energies for cutoffs between 40 and 90 Ry. Spin–orbit coupling was included non-perturbatively through fully relativistic pseudopotentials in a noncollinear calculation; the scalar-relativistic results are reported alongside for comparison.
Ionic positions were relaxed at the experimental 293 K cell with a 4 × 3 × 10 shifted Monkhorst–Pack mesh [33] and Marzari–Vanderbilt smearing [34], until residual forces fell below Ry/bohr. Densities of states were obtained from a 12 × 8 × 32 shifted mesh, 384 points in the irreducible wedge, using the optimized tetrahedron method [35]; the integrated density of states recovers 444.0 electrons in every calculation reported. Orbital characters were obtained by projection onto Löwdin-orthogonalized atomic orbitals [36], which recover 98% of the density of states at the Fermi level . Band structures were computed along –X–S–Y––Z–U–R–T–Z, Y–T, U–X and S–R with 348 points. Two further cuts of 41 points each, one along and one along through the point on the S–R line, were computed to resolve the dispersion transverse to that line.
The isostructural, isoelectronic analogue CaPdBi was computed under an identical protocol but using only scalar relativistic pseudopotentials at its own experimental cell, which differs in volume by only 0.18%, to isolate the role of the Yb shell. The sensitivity of the relaxed geometry to the exchange–correlation functional was tested by repeating the relaxation with PBEsol [37].
Results and discussion
Crystal structure
All diffraction maxima were indexed on a primitive orthorhombic lattice corresponding to Å, Å, Å. The structure of YbPdBi is isostructural to CaPdBi [22]. As depicted in Table I, the fit to the data employing orthorhombic symmetry is excellent (). The refined atomic coordinates and atomic displacement parameters are listed in Table II. As illustrated in Figure 1, the structure is a 3D network of Pd-Bi chains running along the axis, with Yb atoms occupying the voids. Each Pd atom is coordinated to six Bi atoms, forming distorted PdBi trigonal prisms. Pd(1) lies at the center of Pd(1)Bi wedged-type distorted trigonal prisms that share their rectangular bases to form Pd(1)Bi dimeric units. These units are connected through their common edges to form the chains along the axis. The Pd-Pd distance within the dimeric unit is very short, at 2.864(5) Å in the plane, suggesting metal-metal bonding. The second chain consists of more regular Pd(2)Bi triangular prisms sharing their triangular faces, with a Pd-Pd distance of 4.1428(11) Å. The two chains are connected by the vertices of the prisms as seen in Figure 1b. The isostructural relationship between YbPdBi and its Ca-based analogs can be rationalized on crystal-chemical grounds. Although these compounds are far from the ionic limit and instead exhibit substantial metallic and covalent bonding contributions, the substitution of Ca by Yb is facilitated by the close match in the effective ionic radii of Ca and Yb [38]. This is expected to promote the retention of the same structure type while introducing additional electronic degrees of freedom. However, it should be emphasized that Shannon ionic radii are formulated for predominantly ionic environments with well-defined coordination geometries. Their application to complex intermetallic phases such as YbPdBi should therefore be regarded as a qualitative guide rather than a quantitative predictor of structural stability.
| Atom | ||||||||||
| Orthorhombic structure of YbPdBi at K, SG: | ||||||||||
| Yb1 | 0 | 0 | 0 | 0.039(1) | 0.00042(9) | 0.0269(9) | -0.0089(8) | 0 | 0 | 0.0235(6) |
| Yb2 | 0.3572(1) | 0.19978(8) | 0 | 0.0202(8) | 0.0051(7) | 0.0233(7) | -0.0048(5) | 0 | 0 | 0.0162(4) |
| Pd1 | 0.4540(2) | 0.0788(1) | 0.5 | 0.0177(14) | 0.0014(12) | 0.0306(12) | 0.0005(9) | 0 | 0 | 0.0166(7) |
| Pd2 | 0.0718(2) | 0.17623(12) | 0 | 0.0229(14) | 0.0036(12) | 0.020(10) | 0.0017(10) | 0 | 0 | 0.0156(7) |
| Bi1 | 0.1100(1) | 0.28681(6) | 0.5 | 0.0172(7) | 0.0001 | 0.0227(6) | 0.0012(4) | 0 | 0 | 0.0133(3) |
| Bi2 | 0.1958(1) | 0.08565(6) | 0.5 | 0.0148(7) | 0.0030(6) | 0.0230(6) | 0.0019(4) | 0 | 0 | 0.0136(4) |
| Bi3 | 0.3907(1) | 0.35729(6) | 0.5 | 0.0162(7) | 0.0043(6) | 0.0254(7) | -0.0004(4) | 0 | 0 | 0.0153(4) |
Table II. Room temperature atomic coordinates , , and atomic displacement parameters (ADPs) of YbPdBi with space group . During refinement, the parameter of the Bi1 site was constrained to a small positive value. When refined freely, this parameter became negative, resulting in a non-positive definite ADP tensor. This behavior is most likely attributable to minor imperfections in the absorption correction, as the compound is a strong X-ray absorber, and even small inaccuracies in absorption correction can disproportionately affect anisotropic displacement parameters. However, this constraint has a negligible effect on the overall structural model.
Figure 1. The crystal structure of YbPdBi is shown in (a) and (b). A picture of a single crystal on a millimetre grid is shown in (c).
Nevertheless, this cation substitution approach naturally extends to rare earths for a plethora of materials [39]. Eu-based analogs can be considered, as Eu also exhibits an ionic size comparable to alkaline-earth ions such as Ca, Sr, and Ba. The transition from alkaline-earth compounds to Eu-containing counterparts are widely realized across other intermetallic families like the Al4 compounds ( = Ca, Sr, Ba, Eu) [40, 41, 42, 43] or the AMgZintl family of thermoelectrics ( = Ca, Sr, Eu, Yb) [44] and several others. Moreover, in comparison with the non-magnetic Yb () configuration, Eu () carries a large localized effective magnetic moment . In metallic systems, these local moments are often coupled through the Ruderman–Kittel–Kasuya–Yosida (RKKY) interaction mediated by conduction electrons, giving rise to a rich variety of magnetic ground states. Thus, Eu-based compounds provide an ideal testbed for exploring the interplay between magnetism, RKKY exchange, strong spin–orbit coupling and emergent electronic phases.
Magnetization
The magnetization data collected from a single crystal of YbPdBi, with a field of 7 T applied parallel and perpendicular to the crystallographic axis, is presented in Figure 2(a). The magnetization isotherms collected at 300 K are shown in Figure 2(b). The magnetic susceptibility is nearly isotropic. The negative susceptibility indicates that YbPdBi is diamagnetic, confirming that Yb is in the non-magnetic Yb configuration. The upturn at low temperature indicates a small amount of paramagnetic impurity, which is not significant enough to overcome the diamagnetic response. The diamagnetic susceptibility is around e.m.u./mol, which is comparable to that of CaPdBi ( e.m.u./mol) and CaPtBi ( e.m.u./mol) [22], with the larger magnitude in YbPdBi reflecting the larger number of core electrons.
Figure 2. (a) Temperature dependence of the magnetic susceptibility under an applied magnetic field of 7 T parallel and perpendicular to the axis. (b) Magnetic field dependence of the magnetization at 300 K for the field parallel and perpendicular to the axis.
Resistivity
Fig. 3 shows the temperature dependence of the electrical resistivity of YbPdBi measured at zero magnetic field with the current applied along the crystallographic direction parallel to the needle axis, [1]. The resistivity decreases upon cooling, indicating metallic behavior. The residual resistivity ratio is rather low. The temperature dependence can be well fitted using the Bloch-Grüneisen formula of the resistivity,
Where is a prefactor and is the characteristic temperature associated with phonon-mediated electrical transport. The fit yields , and K. This agreement is consistent with electron–phonon scattering accounting for the temperature-dependent resistivity, without requiring an
additional magnetic scattering contribution. The value of is comparable to those reported for CaPdBi and CaPtBi, approximately 128 and 117 K, respectively [22]. These rather low Debye temperatures may originate from soft phonon modes associated with the heavy bismuth polyanionic framework ([PdBi] chains). The inset in Figure 3 shows the temperature dependence below 6 K. The resistivity starts to decrease near 5.4 K and 1.8 K it decreases strongly. This may be due to the onset of superconductivity, but we cannot exclude that this superconductivity is due to some residual flux on the surface. E.g. − PdBi (tetragonal Phase) has = 5.4 K and −PdBi which crystallize in a monoclinic phase and becomes superconducting below 1.7 K [15]. A magnetic field of 0.1 T is enough to completely suppress this extrinsic superconducting transition in YbPdBi (not shown).
Figure 3. Temperature dependence of the resistivity at zero magnetic field. The red solid line is a fit to the resistivity with the Bloch-Grüneisen formula. The inset shows the resistivity below 6 K. Below 1.8 K, the resistivity shows the onset of superconductivity, but its origin may be extrinsic due to Pd-Bi flux on the surface.
Heat capacity
The temperature dependence of the specific heat divided by the temperature () below 20 K is shown in Figure 4. The specific heat decreases monotonously to low temperatures without any anomalies above 1 K, supporting that the transitions seen in the resistivity are not bulk transitions. The temperature dependence of the specific heat can be fitted with the Debye model and the Debye temperature = 107.5 K is in excellent agreement with = 105 K from the temperature dependence of the resistivity. At low temperature, we observed a jump of the specific heat on cooling at 0.78 K from J molK to J molK; however, the origin of this jump in the specific heat is still unclear.
Figure 4. Temperature dependence of the specific heat divided by temperature () at zero magnetic field below 20 K. The red line is a fit with a Debye model for the specific heat.
Electronic structure
Relaxing the ionic positions at the experimental 293 K cell reproduces the refined structure closely (Table III). The Bi–Pd contact that defines the PdBi prisms is within 0.25% of experiment and the cation–anion distances within 2%. The least well reproduced contact is the short Pd–Pd bond, computed 2% too long. Repeating the relaxation with PBEsol, with everything else held fixed, shortens the bond, reducing the discrepancy to +1.6%, while improving Bi–Bi and leaving –Pd slightly worse. These discrepancies, stemming from approximations in the GGA semilocal density functionals, are small in absolute terms. In fact, the calculated 0 K geometry agrees with the 293 K refinement to within the thermal amplitude. For example, the computed displacement of Pd1 is 0.032 Å against a refined root-mean-square thermal amplitude of 0.133 Å for that atom.
| Contact | Exp. (Å) | PBE (Å) | PBEsol (Å) | ||
| Pd–Pd | 2.864 | 2.923 | +2.05% | 2.910 | +1.61% |
| Bi–Pd | 2.794 | 2.801 | +0.25% | 2.798 | +0.12% |
| Bi–Bi | 3.123 | 3.178 | +1.75% | 3.111 | −0.39% |
| –Pd | 3.103 | 3.047 | −1.83% | 3.032 | −2.31% |
| –Bi | 3.302 | 3.270 | −0.97% | 3.284 | −0.54% |
| CaPdBi | |||||
| Pd–Pd | 2.851 | 2.859 | +0.30% | — | — |
Table III. Shortest interatomic contacts in YbPdBi ( = Yb): experiment at 293 K, and calculated values after relaxing the ionic positions at the experimental cell with the PBE and PBEsol functionals. is relative to experiment. The isostructural CaPdBi is given for the Pd–Pd contact only. Differences between theory and experiments are well within thermal displacements. For example, the refined root-mean-square thermal displacement of Pd1 is 0.133 Å along , against a computed displacement of 0.032 Å.
The calculations were performed with the shell frozen in the core, so that Yb is divalent by construction, as established by the diamagnetic susceptibility reported above. The good agreement with the experimental geometry suggests that the Yb assignment is correct and the electrons are spectators at the Fermi energy. With Yb, the compound is isoelectronic with CaPdBi, and both the relaxed geometry (Table III) and the electronic structure (Figure 5) of the two compounds are almost indistinguishable. Specifically, both compounds have very similar densities of states at the Fermi energy () (Figure 5). Our for CaPdBi, 5.41 states eV per formula unit, agrees well with the value reported in Refs. 21 and 22. The Pauli and Landau contributions implied by the calculated density of states, emu/mol, are positive and about three times smaller in magnitude than the measured magnetic susceptibility. Subtracting them from the measured susceptibility leaves a core contribution of emu/mol, consistent with a response dominated by the core diamagnetism of a closed shell.
Figure 5. Comparison of YbPdBi with the isoelectronic, isostructural analogue CaPdBi, computed under an identical protocol, with scalar relativistic pseudopotentials, and aligned on ; the two cells differ only by 0.18% in volume. (a) Total densities of states, per unit cell, which contains two formula units (). (b) Their difference: spectral weight is redistributed within the Pd block below eV, while above eV the two curves are superimposable, with a Pearson correlation of 0.979.
The fully relativistic spin-orbit-coupling calculation confirms that the compound is a metal with a modest density of states at the Fermi level (Figure 6). Ten bands cross , giving states eV per formula unit, a modest increase over the 5.46 states eV per formula unit computed without spin-orbit coupling. The corresponding band value of the Sommerfeld coefficient is mJ molK per formula unit. The states at the Fermi level are dominated by Bi, which contributes 61% of against 20% for Pd and 17% for Yb (Figure 6c), and within the Bi contribution it is the manifold that dominates. The Pd shell is filled and lies 2–4 eV below , so the metallic conduction is carried by the Bi–Pd framework. The origin of the increase of lies in the Bi manifold, which spin-orbit coupling splits into and components separated by 0.66 eV. At the Fermi level the two are populated in the ratio 2.88, whereas their degeneracies alone would give 2.00 (Figure 6d): the states are over-represented at by 44%. The effect is not confined to Bi, however, as the Yb, Pd and Bi contributions increase comparably.
Figure 6. Electronic structure of including spin–orbit coupling. (a) Kohn-Sham bands along the Pbam high-symmetry path. Each band is occupied by an electron pair: the ten pairs that cross are highlighted. (b) Total density of states, scalar-relativistic and with spin–orbit coupling, on the same energy axis. (c) Element-resolved density of states with spin–orbit coupling; at the Bi, Pd and Yb contributions are 61%, 20% and 17%. (d) Bi resolved by total angular momentum. At , the to ratio is 2.9, against the value 2.0 expected from their degeneracies. The density-of-states axes in (b)–(d) are per unit cell, which contains two formula units (); every quoted on the figure, in the text, and in Figure 5 is per formula unit.
Finally, we discuss the symmetry and the topology of the electronic structure of YbPdBi. As the space group is centrosymmetric and the compound is non-magnetic, the product of inversion and time reversal makes every band twofold degenerate throughout the zone, which excludes Weyl nodes by symmetry. The spin–orbit calculation, in which 560 spinor states are obtained independently, find the degeneracy realised to better than 0.3 meV over all the bands and the 348 points (Figure 7a). is nonsymmorphic: The - and -glides carry non-primitive translations in the plane, and their phases force additional band degeneracy on the zone boundary. Without spin–orbit coupling the whole and planes are fourfold degenerate, forming two nodal surfaces (Figure 7a). Spin–orbit coupling removes the surfaces but leaves five fourfold nodal lines, X–S, S–Y, U–R, R–T and S–R: those that either lie in the or mirror plane or lie where the two glide planes intersect. Three fourfold branches cross the Fermi level along S–R and one along R–T (Figure 7b), so that YbPdBi hosts symmetry-enforced Dirac points at . We note that the compound is metallic, so that a invariant is not defined. The parity products at and Z are both , indicating no band inversion between them.
Figure 7. Spin–orbit coupling and the symmetry-enforced band degeneracies. (a) Scalar-relativistic and spin–orbit band structures overlaid on the identical 348-point path, each referred to its own Fermi level. Shading marks the two segments where spin–orbit coupling destroys the degeneracy. The two shaded segments are the parts of the and nodal surfaces that carry no glide protection. The Kramers degeneracy is never lifted. (b) The fourfold branch nearest along the S–R nodal line (heavy line); thin lines are the other fourfold branches in the window and open circles mark every crossing of . The dotted line marks where the perpendicular cuts of (c) were taken, , which is 0.0035 reciprocal lattice units short of the exact crossing at . (c) The two perpendicular cuts, plotted back to back on a common momentum scale: left half along , right half along . Each cut is exactly even in , so one half of each is shown; each branch drawn is a Kramers doublet, so four states meet at the node. The node sits 12 meV above because the cut misses the exact crossing by the amount given above; the nearest other bands are 0.25 eV away and lie outside the panel.
A nodal line is fourfold degenerate along its entire length, so a path running along the specific S–R high-symmetry path cannot by itself establish that the crossing at is a Dirac point: the dispersion has to be examined in the plane perpendicular to the line. We therefore computed two further cuts through the crossing nearest , one along and one along at $k_z=0.123 (Figure 7c). On the line, the four states are degenerate to better than 0.1 meV, the precision to which the eigenvalues are reported; away from it they separate into two Kramers doublets whose splitting grows linearly in over the first 0.02 Å. The slopes give transverse velocities and eV Å, that is 0.76 and . These differ by only 8%, so the cone is nearly isotropic in the transverse plane even though the lattice parameters and are very different. These group velocities are about an order of magnitude smaller than the Dirac velocity of graphene, as expected for a crossing built from Bi and Pd states. The degeneracy at is a genuine fourfold Dirac crossing, carried by a nodal line that runs along and disperses linearly in the two perpendicular directions. Along the line itself, the same quartet disperses with , about four times the transverse velocities, so the cone is strongly anisotropic overall even though it is nearly isotropic in the plane perpendicular to the line. The same argument applies to the other three crossings of : the fourfold degeneracy is enforced along the whole of S–R and R–T, each crossing is isolated from the rest of the spectrum by 180 to 600 meV, and each disperses linearly along its own line with between 2.1 and 4.6 eV Å. Only the crossing at was cut perpendicular, however, so the transverse cone is verified there alone; the
transverse velocities at the other three will differ, and the crossing on R–T has its transverse plane spanned by and rather than by and .
Conclusion
We have successfully grown single crystals of . It is isostructural to and crystallizes in orthorhombic space group 55, Pbam. The magnetic response is diamagnetic, indicating that the Yb has a non-magnetic configuration. The electrical resistivity is metallic and can be explained by electron-phonon scattering over the whole temperature range. These findings are supported by DFT calculations and demonstrate that the Pd–Bi atomic network of the structure type accommodates a stable divalent Yb ion while preserving its characteristic network of interconnected PdBi₆ trigonal prisms and metallic character. Electronic structure calculations with fully relativistic spin–orbit coupling show that the nonsymmorphic crystal structure enforces fourfold-degenerate nodal lines on the zone boundary, four of which cross and give symmetry-enforced Dirac crossings. As a non-magnetic analog, YbPdBi provides an important baseline for disentangling electronic and magnetic contributions, enabling comparison with future rare-earth members of this series. The successful stabilization of this compound thus motivates the search for Eu-based analogs, where the large localized Eu moment can be incorporated into the same Bi-rich metallic framework. Such compounds would offer a platform to explore the interplay between local-moment magnetism and strong spin–orbit coupling, which is expected to be mediated by conduction electrons via the RKKY interaction in an isostructural setting.
Acknowledgements
We thank Matthieu Quénot for the EDS analyses and Johnathan Bulled for support at the ID28 beamline. We acknowledge financial support from the French National Research Agency (ANR) under the project KON-WEY (ANR-24-CE97-0006). M.A.P. and V.T. acknowledge funding from the Laboratoire d’excellence LANEF in Grenoble (ANR-10-LABX-51-01). S.R. thanks the support from the ANR under the project SUPER-NICKEL (Grant No. ANR-21-CE30-0041-04) and also the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Grant Agreement No. 865826). The research at UdeM received support from the Natural Sciences and Engineering Research Council of Canada (Canada), as well as the RQMP through the grant https://doi.org/10.69777/309032. The theory work was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Science and Engineering, grant DE-SC0022288.
References
- [1]C. M. Varma, Mixed-valence compounds, Rev. Mod. Phys. 48, 219 (1976).DOI
- [2]O. Trovarelli, C. Geibel, S. Mederle, C. Langhammer, F. M. Grosche, P. Gegenwart, M. Lang, G. Sparn, and F. Steglich, YbRh₂Si₂: Pronounced non-fermi-liquid effects above a low-lying magnetic phase transition, Phys. Rev. Lett. 85, 626 (2000).
- [3]G. Knebel, R. Boursier, E. Hassinger, G. Lapertot, P. G. Niklowitz, A. Pourret, B. Salce, J. P. Sanchez, I. Sheikin, P. Bonville, H. Harima, and J. Flouquet, Localization of 4f State in YbRh₂Si₂ under Magnetic Field and High Pressure: Comparison with CeRh₂Si₂, J. Phys. Soc. Jpn. 75, 114709 (2006).DOI
- [4]P. Gegenwart, Q. Si, and F. Steglich, Quantum criticality in heavy-fermion metals, Nat. Phys. 4, 186 (2008).DOI
- [5]S. Chadov, X. Qi, J. Kübler, G. H. Fecher, C. Felser, and C. Zhang, Tunable multifunctional topological insulators in ternary heusler compounds, Nature Materials 9, 541 (2010).DOI
- [6]C. Y. Guo, F. Wu, Z. Z. Wu, M. Smidman, C. Cao, A. Bostwick, C. Jozwiak, E. Rotenberg, Y. Liu, F. Steglich, and H. Q. Yuan, Evidence for weyl fermions in a canonical heavy-fermion semimetal yptbi, Nature Communications 9, 4622 (2018).
- [7]H. Kim, K. Wang, Y. Nakajima, R. Hu, S. Ziemak, P. Syers, L. Wang, H. Hodovanets, J. D. Denlinger, P. M. Brydon, D. F. Agterberg, M. A. Tanatar, R. Prozorov, and J. Paglione, Beyond triplet: Unconventional superconductivity in a spin-3/2 topological semimetal, Science Advances 4, 1 (2018), arXiv:1603.03375.DOI
- [8]J. L. Sarrao, A. P. Ramirez, T. W. Darling, F. Freibert, A. Migliori, C. D. Immer, Z. Fisk, and Y. Uwatoko, Thermodynamics of the first-order valence transition in YbInCu4, Phys. Rev. B 58, 409 (1998).DOI
- [9]M. Kasaya, F. Iga, K. Negishi, S. Nakai, and T. Kasuya, A new and typical valence fluctuating system, YbBi2, Journal of Magnetism and Magnetic Materials 31-34, 437 (1983).
- [10]T. Kasuya, Gap state in YbB12 and SmB6: Real kondo insulators, Europhysics Letters 26, 277 (1994).
- [11]K. Hagiwara, Y. Ohtsubo, M. Matsunami, S.-i. Ideta, K. Tanaka, H. Miyazaki, J. E. Rault, P. L. Fèvre, F. Bertran, A. Taleb-Ibrahimi, R. Yukawa, M. Kobayashi, K. Horiba, H. Kumigashira, K. Sumida, T. Okuda, F. Iga, and S.-i. Kimura, Surface kondo effect and non-trivial metallic state of the kondo insulator YbB12, Nature Communications 7, 12690 (2016).DOI
- [12]Z. Xiang, Y. Kasahara, T. Asaba, B. Lawson, C. Tinsman, L. Chen, K. Sugimoto, S. Kawaguchi, Y. Sato, G. Li, S. Yao, Y. L. Chen, F. Iga, J. Singleton, Y. Matsuda, and L. Li, Quantum oscillations of electrical resistivity in an insulator, Science 362, 65 (2018), arXiv:1905.05140.DOI
- [13]C. A. Mizzi, S. K. Kushwaha, P. F. S. Rosa, W. A. Phelan, D. C. Arellano, L. A. Pressley, T. M. McQueen, M. K. Chan, and N. Harrison, The reverse quantum limit and its implications for unconventional quantum oscillations in YbB12, Nature Communications 15, 1607 (2024).
- [14]Z. Yang, C. Marcenat, S. Kim, S. Imajo, M. Kimata, T. Nomura, A. De Muer, D. K. Maude, F. Iga, T. Klein, D. Chowdhury, and Y. Kohama, Evidence for large thermodynamic signatures of in-gap fermionic quasiparticle states in a kondo insulator, Nature Communications 15, 7801 (2024).DOI
- [15]Y. Imai, F. Nabeshima, T. Yoshinaka, K. Miyatani, R. Kondo, S. Komiya, I. Tsukada, and A. Maeda, Superconductivity at 5.4 K in β-Bi2Pd, J. Phys. Soc. Jpn. 81, 113708 (2012).
- [16]M. Sakano, K. Okawa, M. Kanou, H. Sanjo, T. Okuda, T. Sasagawa, and K. Ishizaka, Topologically protected surface states in a centrosymmetric superconductor β-PdBi2, Nature Communications 6, 8595 (2015).DOI
- [17]L. Powell, W. Kuang, G. Hawkins-Potter, R. Jalil, J. Birkbeck, Z. Jiang, M. Kim, Y. Zou, S. Komrakova, S. Haigh, I. Timokhin, G. Balakrishnan, A. K. Geim, N. Walet, A. Principi, and I. V. Grigorieva, Multiphase superconductivity in PdBi2, Nature Communications 16, 291 (2025).DOI
- [18]G. LeBras, P. Bonville, J. A. Hodges, J. Hammann, M. J. Besnus, G. Schmerber, S. K. Dhar, F. G. Aliev, and G. Andre, Local symmetry lowering in the cubic intermetallics YbPdBi and YbNiSb, J. Phys. Condens. Matter 7, 5665 (1995).DOI
- [19]D. Kaczorowski, A. Leithe-Jasper, P. Rogl, H. Flandorfer, T. Cichorek, R. Pietri, and B. Andraka, Magnetic, thermodynamic, and electrical transport properties of ternary equiatomic ytterbium compounds YbTM(T=transition metal, M = Sn and Bi), Phys. Rev. B 60, 422 (1999).DOI
- [20]R. Pietri, B. Andraka, D. Kaczorowski, A. Leithe-Jasper, and P. Rogl, Magnetoresistance and low-temperature specific heat of the Yb compounds YbRhSn, YbPdBi, and YbPtSn, Phys. Rev. B 61, 12169 (2000).DOI
- [21]D. Johredt and A. Mewis, Ca3Pd4Bi8: Kristall- und elektronische Struktur, Z. anorg. allg. Chem. 628, 2671 (2002).
- [22]A. Ovchinnikov and A.-V. Mudring, Flux Growth, Crystal Structures, and Electronic Properties of the Ternary Intermetallic Compounds Ca3Pd4Bi8 and Ca3Pt4Bi8, Inorg. Chem. 61, 9756 (2022).DOI
- [23]S. M. Young, S. Zaheer, J. C. Y. Teo, C. L. Kane, E. J. Mele, and A. M. Rappe, Dirac semimetal in three dimensions, Phys. Rev. Lett. 108, 140405 (2012).
- [24]P. C. Canfield, T. Kong, U. S. Kaluarachchi, and N. H. Jo, Use of frit-disc crucibles for routine and exploratory solution growth of single crystalline samples, Philos. Mag. 96, 84 (2016).
- [25]P. C. Canfield and T. J. Slade, Use of Frit-Disc Crucible Sets to Make Solution Growth More Quantitative and Versatile, Z. Anorg. Allg. Chem. , 2500007 (2025).arxiv.org/abs/2503.10374
- [26]J. Paglione, N. P. Butch, and E. E. Rodriguez, Fundamentals of Quantum Materials A Practical Guide to Synthesis and Exploration (World Scientific Publishing, 2021).DOI
- [27]Rigaku, CrysAlis Pro Version 171.43.90, Rigaku Oxford Diffraction. (2023).
- [28]V. Petříček, M. Dušek, and L. Palatinus, Crystallographic computing system JANA2006: general features, Z. Kristallogr. 229, 345 (2014).DOI
- [29]P. Giannozzi, O. Andreussi, T. Brumme, O. Bunau, M. Buongiorno Nardelli, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, M. Cococcioni, N. Colonna, I. Carnimeo, A. Dal Corso, S. de Gironcoli, P. Delugas, R. A. DiStasio, A. Ferretti, A. Floris, G. Fratesi, G. Fugallo, R. Gebauer, U. Gerstmann, F. Giustino, T. Gorni, J. Jia, M. Kawamura, H.-Y. Ko, A. Kokalj, E. Küçükbenli, M. Lazzeri, M. Marsili, N. Marzari, F. Mauri, N. L. Nguyen, H.-V. Nguyen, A. Otero-de-la-Roza, L. Paulatto, S. Poncé, D. Rocca, R. S. Santini, B. S. S. Sharma, M. Schlipa, A. P. Seitsonen, A. Smogunov, I. Timrov, T. Thonhauser, P. Umari, N. Vast, X. Wu, and S. Baroni, Advanced capabilities for materials modelling with Quantum ESPRESSO, J. Phys.: Condens. Matter 29, 465901 (2017).
- [30]J. P. Perdew, K. Burke, and M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett. 77, 3865 (1996).pubmed.ncbi.nlm.nih.gov/10062328
- [31]P. E. Blöchl, Projector augmented-wave method, Phys. Rev. B 50, 17953 (1994).pubmed.ncbi.nlm.nih.gov/9976227
- [32]A. Dal Corso, Pseudopotentials periodic table: From H to Pu, Comput. Mater. Sci. 95, 337 (2014).DOI
- [33]H. J. Monkhorst and J. D. Pack, Special points for brillouin-zone integrations, Phys. Rev. B 13, 5188 (1976).DOI
- [34]N. Marzari, D. Vanderbilt, A. De Vita, and M. C. Payne, Thermal contraction and disordering of the Al(110) surface, Phys. Rev. Lett. 82, 3296 (1999).DOI
- [35]M. Kawamura, Y. Gohda, and S. Tsuneyuki, Improved tetrahedron method for the Brillouin-zone integration applicable to response functions, Phys. Rev. B 89, 094415 (2014).DOI
- [36]P.-O. Löwdin, On the non-orthogonality problem connected with the use of atomic wave functions in the theory of molecules and crystals, J. Chem. Phys. 18, 365 (1950).DOI
- [37]J. P. Perdew, A. Ruzsinszky, G. I. Csonka, O. A. Vydrov, G. E. Scuseria, L. A. Constantin, X. Zhou, and K. Burke, Restoring the density-gradient expansion for exchange in solids and surfaces, Phys. Rev. Lett. 100, 136406 (2008).pubmed.ncbi.nlm.nih.gov/18517979
- [38]R. D. Shannon, Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides, Acta Crystallographica Section A 32, 751 (1976).DOI
- [39]M. L. Fornasini and F. Merlo, Binary intermetallic phases formed by Ca, Sr, Ba, Eu and Yb: similarities and differences, Zeitschrift für Kristallographie - Crystalline Materials 221, 382 (2006).DOI
- [40]G. J. Miller, F. L. Li, and H. F. Franzen, The structural phase transition in calcium-aluminum compound (CaAl4): a concerted application of Landau theory and energy band theory, J. Am. Chem. Soc. 115, 3739 (1993).DOI
- [41]E. Parthé, B. Chabot, H. F. Braun, and N. Engel, Ternary BaAl4-type derivative structures, Acta Crystallographica Section B 39, 588 (1983).DOI
- [42]S. Ramakrishnan, S. R. Kotla, H. Pi, B. B. Maity, J. Chen, J.-K. Bao, Z. Guo, M. Kado, H. Agarwal, C. Eisele, M. Nohara, L. Noohinejad, H. Weng, S. Ramakrishnan, A. Thamizhavel, and S. van Smaalen, Non-centrosymmetric, transverse structural modulation in SrAl4, and elucidation of its origin in the BaAl4 family of compounds, Phys. Rev. Res. 6, 023277 (2024).
- [43]S. Ramakrishnan, S. R. Kotla, T. Rekis, J.-K. Bao, C. Eisele, L. Noohinejad, M. Tolkiehn, C. Paulmann, B. Singh, R. Verma, B. Bag, R. Kulkarni, A. Thamizhavel, B. Singh, S. Ramakrishnan, and S. van Smaalen, Orthorhombic charge density wave on the tetragonal lattice of EuAl4, IUCrJ 9, 378 (2022).DOI
- [44]Z. Yuan, K. Liu, Q. Liu, S.-Q. Xia, T. Zhou, and X.-F. Hou, Low-density AMgGe (A = Ca, Sr, Eu, Yb) thermoelectric Zintl phases with high power factors, J. Alloy. Compd. 1030, 180911 (2025).