betaFeedback

WorksPhysics and Astronomy

Divalent Yb in the Ca3_3Pd4_4Bi8_8 Structure Type: Crystal Growth and Physical Properties of Yb3_3Pd4_4Bi8_8

Abstract

We report on the single-crystal growth, crystal structure, and physical properties of Yb3_3Pd4_4Bi8_8. Single crystal x-ray diffraction shows that the material crystallizes in the Ca3_3Pd4_4Bi8_8 structure-type at ambient conditions, described by orthorhombic symmetry PbamPbam. 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 6p6p orbitals, with a density of states within 1% of that of the isoelectronic compound Ca3_3Pd4_4Bi8_8, consistent with a filled and chemically inert 4f4f 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, Yb3_3Pd4_4Bi8_8 is realized as a rare-earth analog of the Ca3_3Pd4_4Bi8_8 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 Yb2+^{2+} (4f144f^{14}) configuration and a magnetic Yb3+^{3+} (4f134f^{13}) 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, YbRh2_2Si2_2 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]; YbInCu4_4 shows a spectacular first order valence transition from the Yb3+^{3+} to an intermediate Yb valence as a function of temperature [8]; or YbB12_{12} is a prime candidate for a topological Kondo Insulator [9, 10]. Despite having an insulating bulk gap driven by 4f4f 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 β\beta-PdBi2_2 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 (γ≈1.2\gamma\approx1.2 J/mol·K2^2) with a possible antiferromagnetic order near 1 K [18, 19, 20]. Due to its non-centrosymmetric crystal structure (space group F4ˉ3mF\bar{4}3m) 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 Ca3_3Pd4_4Bi8_8 [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 4f144f^{14} configuration. The Ca3_3Pd4_4Bi8_8 crystal structure, PbamPbam, 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 Ca3_3Pd4_4Bi8_8 framework a natural setting for symmetry-enforced Dirac degeneracies, a question that the two existing scalar-relativistic studies of Ca3_3Pd4_4Bi8_8 could not address exhaustively [21, 22].

Here we report on the synthesis, crystal structure and physical properties of single-crystalline Yb3_3Pd4_4Bi8_8. It crystallizes in the Ca3_3Pd4_4Bi8_8 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 6p6p band complex essentially identical to that of Ca3_3Pd4_4Bi8_8, 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 Yb3_3Pd4_4Bi8_8 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 Yb3_3Pd4_4Bi8_8 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 Yb3_3Pd4_4Bi8_8 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 (λ\lambda) of 0.5701 Å. The detector used is a Dectris PILATUS3 1M area detector (pixel size 172 × 172 μ\mum) 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 systemOrthorhombic
Space groupPbamPbam
Space group no.55
aa (Å)10.8109(4)
bb (Å)17.0493(7)
cc (Å)4.1428(1)
Volume (Å3^3)763.6(3)
ZZ2
Wavelength (Å)0.5701
Detector distance (mm)170
Rotation per image (ϕ\phi) (°)0.25
χ\chi offset (°)-45
2θ2\theta offset (°)-19
(sin⁡(θ)/λ)max⁡(\sin(\theta)/\lambda)_{\max} (Å−1^{-1})0.611578
Absorption, μ\mu (mm−1^{-1})64.806
No. of reflections measured1816
No. of unique reflections (obs/all)741/741
RintR_{\mathrm{int}} (obs/all)0.0250/0.0250
No. of parameters47
RFR_F (obs/all)0.0448/0.0448
wRFwR_F (all)0.0645
GoF (obs/all)5.09/5.09
Δρmin⁡,Δρmax⁡\Delta\rho_{\min}, \Delta\rho_{\max} (e Å−3^{-3})-5.5, 4.36

Table I. TABLE I. Crystallographic data of Yb3_3Pd4_4Bi8_8 at 293 K. The criterion of observability is I>3σ(I)I > 3\sigma(I).

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 μ\mum 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 4f4f 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, z=10z = 10, 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 10−310^{-3} 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 N(EF)N(E_{\mathrm{F}}). Band structures were computed along Γ\Gamma–X–S–Y–Γ\Gamma–Z–U–R–T–Z, Y–T, U–X and S–R with 348 kk points. Two further cuts of 41 kk points each, one along a∗a^{*} and one along b∗b^{*} through the point (12,12,0.123)\left(\frac{1}{2},\frac{1}{2},0.123\right) on the S–R line, were computed to resolve the dispersion transverse to that line.

The isostructural, isoelectronic analogue Ca3_3Pd4_4Bi8_8 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 4f4f 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 a=10.8109(4)a = 10.8109(4) Å, b=17.0493(7)b = 17.0493(7) Å, c=4.1428(1)c = 4.1428(1) Å. The structure of Yb3_3Pd4_4Bi8_8 is isostructural to Ca3_3Pd4_4Bi8_8 [22]. As depicted in Table I, the fit to the data employing orthorhombic PbamPbam symmetry is excellent (RF=4.5%R_{\mathrm{F}} = 4.5\%). 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 cc axis, with Yb atoms occupying the voids. Each Pd atom is coordinated to six Bi atoms, forming distorted PdBi6_6 trigonal prisms. Pd(1) lies at the center of Pd(1)Bi6_6 wedged-type distorted trigonal prisms that share their rectangular bases to form Pd(1)2_2Bi8_8 dimeric units. These units are connected through their common edges to form the chains along the cc axis. The Pd-Pd distance within the dimeric unit is very short, at 2.864(5) Å in the (ab)(ab) plane, suggesting metal-metal bonding. The second chain consists of more regular Pd(2)Bi6_6 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 Yb3_3Pd4_4Bi8_8 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 Ca2+^{2+} and Yb2+^{2+} [38]. This is expected to promote the retention of the same structure type while introducing additional 4f4f 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 Yb3_3Pd4_4Bi8_8 should therefore be regarded as a qualitative guide rather than a quantitative predictor of structural stability.

AtomxxyyzzU11U_{11}U22U_{22}U33U_{33}U12U_{12}U13U_{13}U23U_{23}UisoeqU^{eq}_{iso}
Orthorhombic structure of Yb3_3Pd4_4Bi8_8 at T=293T = 293 K, SG: PbamPbam
Yb10000.039(1)0.00042(9)0.0269(9)-0.0089(8)000.0235(6)
Yb20.3572(1)0.19978(8)00.0202(8)0.0051(7)0.0233(7)-0.0048(5)000.0162(4)
Pd10.4540(2)0.0788(1)0.50.0177(14)0.0014(12)0.0306(12)0.0005(9)000.0166(7)
Pd20.0718(2)0.17623(12)00.0229(14)0.0036(12)0.020(10)0.0017(10)000.0156(7)
Bi10.1100(1)0.28681(6)0.50.0172(7)0.00010.0227(6)0.0012(4)000.0133(3)
Bi20.1958(1)0.08565(6)0.50.0148(7)0.0030(6)0.0230(6)0.0019(4)000.0136(4)
Bi30.3907(1)0.35729(6)0.50.0162(7)0.0043(6)0.0254(7)-0.0004(4)000.0153(4)

Table II. Room temperature atomic coordinates xx, yy, zz and atomic displacement parameters (ADPs) UijU_{ij} of Yb3_3Pd4_4Bi8_8 with space group PbamPbam. During refinement, the U22U_{22} 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.

Image not extracted

Figure 1. The crystal structure of Yb3_3Pd4_4Bi8_8 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 Eu2+^{2+} 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 XXAl4 compounds (XX = Ca, Sr, Ba, Eu) [40, 41, 42, 43] or the AMg2_2Zintl family of thermoelectrics (AA = Ca, Sr, Eu, Yb) [44] and several others. Moreover, in comparison with the non-magnetic Yb2+^{2+} (4f144f^{14}) configuration, Eu2+^{2+} (4f74f^{7}) carries a large localized effective magnetic moment μeff≈7.94μB\mu_{\mathrm{eff}} \approx7.94\mu_{\mathrm{B}}. 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 Yb3_3Pd4_4Bi8_8, with a field of 7 T applied parallel and perpendicular to the crystallographic cc axis, is presented in Figure 2(a). The magnetization isotherms M(H)M(H) collected at 300 K are shown in Figure 2(b). The magnetic susceptibility is nearly isotropic. The negative susceptibility indicates that Yb3_3Pd4_4Bi8_8 is diamagnetic, confirming that Yb is in the non-magnetic Yb2+^{2+} 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 −4×10−4-4 \times10^{-4} e.m.u./mol, which is comparable to that of Ca3_3Pd4_4Bi8_8 (∼−2×10−4\sim-2 \times10^{-4} e.m.u./mol) and Ca3_3Pt4_4Bi8_8 (∼−3×10−4\sim-3 \times10^{-4} e.m.u./mol) [22], with the larger magnitude in Yb3_3Pd4_4Bi8_8 reflecting the larger number of core electrons.

Image not extracted

Figure 2. (a) Temperature dependence of the magnetic susceptibility under an applied magnetic field of 7 T parallel and perpendicular to the cc axis. (b) Magnetic field dependence of the magnetization at 300 K for the field parallel and perpendicular to the cc axis.

Resistivity

Fig. 3 shows the temperature dependence of the electrical resistivity of Yb3_3Pd4_4Bi8_8 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 RRR=ρ(300K)/ρ(0K)≈4RRR = \rho(300\mathrm{K})/\rho(0\mathrm{K}) \approx4 is rather low. The temperature dependence can be well fitted using the Bloch-Grüneisen formula of the resistivity,

ρ(T)=ρ0+A(TθR)5∫0θR/Tx5(ex−1)(1−e−x) dx\rho(T) = \rho_0 + A\left(\frac{T}{\theta_R}\right)^5 \int_0^{\theta_R/T} \frac{x^5}{(e^x-1)(1-e^{-x})}\,\mathrm{d}x

Where AA is a prefactor and θR\theta_R is the characteristic temperature associated with phonon-mediated electrical transport. The fit yields ρ0=6.4 μΩ cm\rho_0 = 6.4\,\mu\Omega\,\mathrm{cm}, A=25.5 μΩ cmA = 25.5\,\mu\Omega\,\mathrm{cm} and θR≈105\theta_R \approx105 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 θR\theta_R is comparable to those reported for Ca3_3Pd4_4Bi8_8 and Ca3_3Pt4_4Bi8_8, 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 ([Pd4_4Bi8_8]∞16−_\infty^{16-} 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. β\beta − PdBi2_2 (tetragonal Phase) has TcT_\mathrm{c} = 5.4 K and α\alpha−PdBi2_2 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 Yb3_3Pd4_4Bi8_8 (not shown).

Image not extracted

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 (C/TC/T) 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 θD\theta_\mathrm{D} = 107.5 K is in excellent agreement with θR\theta_\mathrm{R} = 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 C/T≈0.035C/T \approx0.035 J mol−1^{-1}K−2^{-2} to C/T≈0.06C/T \approx0.06 J mol−1^{-1}K−2^{-2}; however, the origin of this jump in the specific heat is still unclear.

Image not extracted

Figure 4. Temperature dependence of the specific heat divided by temperature (C/TC/T) 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 PdBi6_6 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 MM–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.

ContactExp. (Å)PBE (Å)Δ\DeltaPBEsol (Å)Δ\Delta
Pd–Pd2.8642.923+2.05%2.910+1.61%
Bi–Pd2.7942.801+0.25%2.798+0.12%
Bi–Bi3.1233.178+1.75%3.111−0.39%
MM–Pd3.1033.047−1.83%3.032−2.31%
MM–Bi3.3023.270−0.97%3.284−0.54%
Ca3_3Pd4_4Bi8_8
Pd–Pd2.8512.859+0.30%——

Table III. Shortest interatomic contacts in Yb3_3Pd4_4Bi8_8 (MM = Yb): experiment at 293 K, and calculated values after relaxing the ionic positions at the experimental cell with the PBE and PBEsol functionals. Δ\Delta is relative to experiment. The isostructural Ca3_3Pd4_4Bi8_8 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 aa, against a computed displacement of 0.032 Å.

The calculations were performed with the 4f4f 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 Yb2+^{2+} assignment is correct and the 4f4f electrons are spectators at the Fermi energy. With Yb2+^{2+}, the compound is isoelectronic with Ca3_3Pd4_4Bi8_8, 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 (N(EF)N(E_{\mathrm{F}})) (Figure 5). Our N(EF)N(E_{\mathrm{F}}) for Ca3_3Pd4_4Bi8_8, 5.41 states eV−1^{-1} 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, χPauli+χLandau=23μB2NAN(EF)=+1.3×10−4\chi_{\mathrm{Pauli}}+\chi_{\mathrm{Landau}}=\frac{2}{3}\mu_{\mathrm{B}}^{2}N_{\mathrm{A}}N(E_{\mathrm{F}})=+1.3\times10^{-4} 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 −5.3×10−4-5.3\times10^{-4} emu/mol, consistent with a response dominated by the core diamagnetism of a closed 4f144f^{14} shell.

Image not extracted

Figure 5. Comparison of Yb3_3Pd4_4Bi8_8 with the isoelectronic, isostructural analogue Ca3_3Pd4_4Bi8_8, computed under an identical protocol, with scalar relativistic pseudopotentials, and aligned on E−EFE-E_{\mathrm{F}}; the two cells differ only by 0.18% in volume. (a) Total densities of states, per unit cell, which contains two formula units (Z=2Z=2). (b) Their difference: spectral weight is redistributed within the Pd 4d4d block below −2-2 eV, while above −2-2 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 EFE_{\mathrm{F}}, giving N(EF)=5.86N(E_{\mathrm{F}})=5.86 states eV−1^{-1} per formula unit, a modest increase over the 5.46 states eV−1^{-1} per formula unit computed without spin-orbit coupling. The corresponding band value of the Sommerfeld coefficient is γband=13.8\gamma_{\mathrm{band}}=13.8 mJ mol−1^{-1}K−2^{-2} per formula unit. The states at the Fermi level are dominated by Bi, which contributes 61% of N(EF)N(E_{\mathrm{F}}) against 20% for Pd and 17% for Yb (Figure 6c), and within the Bi contribution it is the 6p3/26p_{3/2} manifold that dominates. The Pd 4d4d shell is filled and lies 2–4 eV below EFE_{\mathrm{F}}, so the metallic conduction is carried by the Bi–Pd framework. The origin of the increase of N(EF)N(E_{\mathrm{F}}) lies in the Bi 6p6p manifold, which spin-orbit coupling splits into j=1/2j=1/2 and j=3/2j=3/2 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 6p3/26p_{3/2} states are over-represented at EFE_{\mathrm{F}} by 44%. The effect is not confined to Bi, however, as the Yb, Pd and Bi contributions increase comparably.

Image not extracted

Figure 6. Electronic structure of Yb3Pd4Bi8\mathrm{Yb}_{3}\mathrm{Pd}_{4}\mathrm{Bi}_{8} 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 EFE_{\mathrm{F}} 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 EFE_{\mathrm{F}} the Bi, Pd and Yb contributions are 61%, 20% and 17%. (d) Bi 6p6p resolved by total angular momentum. At EFE_{\mathrm{F}}, the j=3/2j = 3/2 to j=1/2j = 1/2 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 (Z=2Z = 2); every N(EF)N(E_{\mathrm{F}}) 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 Yb3_3Pd4_4Bi8_8. As the space group PbamPbam 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 kk points (Figure 7a). PbamPbam is nonsymmorphic: The bb- and aa-glides carry non-primitive translations in the abab plane, and their phases force additional band degeneracy on the zone boundary. Without spin–orbit coupling the whole kx=1/2k_x=1/2 and ky=1/2k_y=1/2 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 kz=0k_z=0 or kz=1/2k_z=1/2 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 Yb3_3Pd4_4Bi8_8 hosts symmetry-enforced Dirac points at EFE_{\mathrm{F}}. We note that the compound is metallic, so that a Z2\mathbb{Z}_2 invariant is not defined. The parity products at Γ\Gamma and Z are both −1-1, indicating no band inversion between them.

Image not extracted

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 kx=1/2k_x = 1/2 and ky=1/2k_y = 1/2 nodal surfaces that carry no glide protection. The Kramers degeneracy is never lifted. (b) The fourfold branch nearest EFE_{\mathrm{F}} 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 EFE_{\mathrm{F}}. The dotted line marks where the perpendicular cuts of (c) were taken, kz=0.123k_z = 0.123, which is 0.0035 reciprocal lattice units short of the exact crossing at kz=0.1265k_z = 0.1265. (c) The two perpendicular cuts, plotted back to back on a common momentum scale: left half along b∗b^*, right half along a∗a^*. Each cut is exactly even in δk\delta k, 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 EFE_{\mathrm{F}} 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 EFE_{\mathrm{F}} 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 EFE_{\mathrm{F}}, one along a∗a^* and one along b∗b^* 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 ∣δk∣|\delta k| over the first 0.02 Å−1^{-1}. The slopes give transverse velocities ℏva=0.50\hbar v_a=0.50 and ℏvb=0.54\hbar v_b=0.54 eV Å, that is 0.76 and 0.83×105 m s−10.83 \times10^{5}\ \mathrm{m\,s^{-1}}. These differ by only 8%, so the cone is nearly isotropic in the transverse plane even though the lattice parameters ∣a∗∣|a^{*}| and ∣b∗∣|b^{*}| 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 6p6p and Pd 4d4d states. The degeneracy at EFE_{\mathrm{F}} is a genuine fourfold Dirac crossing, carried by a nodal line that runs along c∗c^{*} and disperses linearly in the two perpendicular directions. Along the line itself, the same quartet disperses with ℏv=2.2 eV A˚\hbar v = 2.2\,\mathrm{eV\,\mathring{A}}, 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 EFE_{\mathrm{F}}: 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 ℏv\hbar v between 2.1 and 4.6 eV Å. Only the crossing at kz=0.1265k_z = 0.1265 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 b∗b^{*} and c∗c^{*} rather than by a∗a^{*} and b∗b^{*}.

Conclusion

We have successfully grown single crystals of Yb3Pd4Bi8\mathrm{Yb}_{3}\mathrm{Pd}_{4}\mathrm{Bi}_{8}. It is isostructural to Ca3Pd4Bi8\mathrm{Ca}_{3}\mathrm{Pd}_{4}\mathrm{Bi}_{8} and crystallizes in orthorhombic space group 55, Pbam. The magnetic response is diamagnetic, indicating that the Yb has a non-magnetic Yb2+\mathrm{Yb}^{2+} 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 R3Pd4Bi8\mathrm{R}_{3}\mathrm{Pd}_{4}\mathrm{Bi}_{8} 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 PbamP bam crystal structure enforces fourfold-degenerate nodal lines on the zone boundary, four of which cross EFE_{\mathrm{F}} and give symmetry-enforced Dirac crossings. As a non-magnetic analog, Yb3_3Pd4_4Bi8_8 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 Eu2+^{2+} 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. [1]C. M. Varma, Mixed-valence compounds, Rev. Mod. Phys. 48, 219 (1976).DOI
  2. [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. [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. [4]P. Gegenwart, Q. Si, and F. Steglich, Quantum criticality in heavy-fermion metals, Nat. Phys. 4, 186 (2008).DOI
  5. [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. [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. [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. [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. [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. [10]T. Kasuya, Gap state in YbB12 and SmB6: Real kondo insulators, Europhysics Letters 26, 277 (1994).
  11. [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. [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. [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. [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. [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. [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. [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. [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. [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. [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. [21]D. Johredt and A. Mewis, Ca3Pd4Bi8: Kristall- und elektronische Struktur, Z. anorg. allg. Chem. 628, 2671 (2002).
  22. [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. [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. [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. [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. [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. [27]Rigaku, CrysAlis Pro Version 171.43.90, Rigaku Oxford Diffraction. (2023).
  28. [28]V. Petříček, M. Dušek, and L. Palatinus, Crystallographic computing system JANA2006: general features, Z. Kristallogr. 229, 345 (2014).DOI
  29. [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. [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. [31]P. E. Blöchl, Projector augmented-wave method, Phys. Rev. B 50, 17953 (1994).pubmed.ncbi.nlm.nih.gov/9976227
  32. [32]A. Dal Corso, Pseudopotentials periodic table: From H to Pu, Comput. Mater. Sci. 95, 337 (2014).DOI
  33. [33]H. J. Monkhorst and J. D. Pack, Special points for brillouin-zone integrations, Phys. Rev. B 13, 5188 (1976).DOI
  34. [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. [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. [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. [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. [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. [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. [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. [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. [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. [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. [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).

Paper details

vlnc:W_srtvy55w
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents