Relativistic Effects in LaBi 2 \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} Thin Films Reiley Dorrian, Sungmin Song, Jinwoong Kim, Mizuki Ohno, Seung-Hoon Jhi, Nicholas Kioussis, Joseph Falson Abstract Chemical substitution in crystalline quantum materials is a powerful way to explore the consequences of strong spin-orbit coupling on their structural and electronic properties. In this work, we present an investigation of thin films of the La Pn 2 \text{La}\textit{Pn}\text{}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} ( Pn = Sb, Bi) class of layered square-net intermetallics. We report the growth of LaBi 2 \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} with a pristine layer-by-layer growth mode, classifying it as a good metal displaying superconductivity at ∼ \sim 0.55 K. Compared to LaSb 2 \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , we attribute the enhanced metallic behavior and improved growth dynamics of LaBi 2 \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} to significant relativistic corrections to its electronic band structure and the resulting impact on both surface energy and intrinsic phonon scattering. I Introduction Spin-orbit coupling (SOC) – the relativistic coupling of an electron’s momentum to its spin – is a mechanism of great importance for discovering and engineering novel electronic phases in crystalline materials. Under certain symmetry constraints, SOC acts to gap out spin-degenerate crossings in the electronic band structure, leading to inverted bands and symmetry-protected topological states Hasan and Kane 2010 ; Qi and Zhang 2011 . In the superconducting state, strong SOC can impart significant resilience to applied magnetic fields through both scattering processes Klemm et al. 1975 and momentum-dependent spin stiffness in the band structure, also known as Ising pairing Zhou et al. 2016 , while potentially enabling unconventional pairing symmetries through the mixing of spin-singlet and -triplet components Youn et al. 2012 . Figure 1: (a) Yb-monoclinic structure observed in thin films of the light rare-earth dipnictides. Substitution of the Sb anion with Bi is expected to enhance SOC due to the larger atomic mass. (b) Degree of spacer-layer Pn dimerization in several reported compounds Wang and Steinfink 1967 , including the results presented in this work. As the strength of SOC is related to the atomic mass of constituent atoms, chemical substitution in crystalline compounds is a commonly employed method for engineering the impacts of SOC Hasan and Kane 2010 ; Qi and Zhang 2011 ; De la Barrera et al. 2018 . Exploring this degree of freedom has been fruitful in compounds containing the pnictogens Sb or Bi, where substitution of the former with the latter is found to fundamentally influence the topology of the electronic ground state Liu and Vanderbilt 2013 ; Liu et al. 2016 ; He et al. 2015 ; Glazkova et al. 2022 . However, investigations of the more subtle ways in which SOC influences a material’s basic characteristics – its growth kinetics, crystal structure, or metallic transport properties – have received less attention. In pursuit of this perspective, this work targets a class of compounds known for a chemistry-sensitive competition between structural and electronic phases – the \textit{LnPn}\text{}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} ( Ln = lanthanide element, Pn = Sb, Bi) family of layered square-net intermetallics. A schematic representation of this structural class is shown in Fig. 1 a. Variations in the oxidation state of the Ln cation influence the stacking configuration of the corrugated Ln-Pn and quasi-planar Pn layers, resulting in distinct orthorhombic or monoclinic structure types which can be parametrized via the dimerization of Pn atoms within the corrugated layer (Fig. 1 b) A. Papoian and Hoffmann 2000 . In addition to the chemistry of the Ln site, previous work on thin films of \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and \text{CeSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} discovered an instability with respect to Sb deficiency, establishing UHV thin-film growth techniques as a way to access new polymorphs with modified electronic properties Llanos et al. 2024a ; Dorrian et al. 2025 . With a desire to study the impacts of anion-tunable SOC in this subset of materials, this report builds upon the aforementioned work on (La, Ce) \text{Sb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} Llanos et al. 2024a ; Dorrian et al. 2025 and focuses on thin-films of \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} synthesized via molecular beam epitaxy (MBE). The substitution of Sb with Bi in this work acts to amplify SOC and significantly influences the thin-film growth mode and conductivity. We find \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} to exhibit superior metallic behavior compared to \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , although not due to an intuition based on augmented hopping amplitudes, but rather due to suppressed phonon scattering at elevated temperatures arising from the spin-orbit induced shifting of electronic states. II Methods II.1 Experimental Details We have synthesized \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} thin films using an MBE system with a base pressure of 10^{-10} mbar on MgO (001) substrates which are laser-annealed as in Refs. Llanos et al. 2024b ; Llanos et al. 2024a ; Dorrian et al. 2025 to optimize surface quality before growth. La and Bi fluxes were provided from conventional effusion cells with a large Bi/La beam-flux ratio (10 \sim 15) while the substrate was heated with a standard SiC heating coil. A growth rate-controlling La flux of about 7.5\times 10^{-9} mbar established a growth rate of approximately 0.034 Å/s. Due to the known air-sensitivity of the compound, all films were finalized with an amorphous Ge capping layer of approximately 5 nm thickness deposited in-situ at room temperature and then stored in a \text{N}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} -atmosphere glovebox. Even with these precautions, film degradation was observed within a few hours of air exposure (See Fig. S1 of the Supplemental Material Sup ). Structural properties of the films were characterized by X-ray diffraction (XRD) (SmartLab, Rigaku). Electrical data above 2 K were collected using a Quantum Design Dynacool PPMS with a 14 T superconducting magnet. The film was cut into an approximately square-shaped piece ( \approx 1 - 2 mm side lengths) and wire-bonded into a four-point van der Pauw geometry with Al wire. Magnetoresistance and Hall resistivity were extracted by symmetrizing the computed \rho_{\mathrm{xx}} or anti-symmetrizing the diagonal R_{yx} Van der Pauw channel, respectively, with respect to a magnetic field applied perpendicular to the film. Measurements of the superconducting transition below 1 K were gathered using a Leiden dilution refrigerator equipped with a two-axis vector magnet (9 and 3 T). The sample is fashioned into a rectangular ( \sim 1 \times 2 mm) piece and bonded with a linear four-point bond geometry. II.2 Theoretical Details Density functional theory (DFT) calculations were performed using the Quantum ESPRESSO package Giannozzi et al. 2009 . The exchange-correlation interaction was described by the PBEsol functional Perdew et al. 2008 , chosen for its accuracy in reproducing experimental lattice constants. Ion-electron interactions were treated using norm-conserving pseudopotentials Hamann 2013 . The electronic wavefunctions were expanded in a plane-wave basis set with a kinetic energy cutoff of 100 Ry. To account for the partial occupancies of electronic states near the chemical potential, we employed the Methfessel-Paxton smearing technique Methfessel and Paxton 1989 with a broadening width of 0.01 Ry. The Brillouin zone integration was performed using the Monkhorst-Pack scheme Monkhorst and Pack 1976 , where we used a \mathbf{k} -grid of 8\times 6\times 6 for the bulk calculations. For the slab models used in the surface energy calculations, an 8\times 8\times 1 \mathbf{k} -grid was adopted, and a vacuum layer of approximately 15\,\text{\AA} was inserted to avoid spurious interactions between periodic images. All atomic structures were fully relaxed until the residual forces on each atom converged to less than 1.0\times 10^{-6}\,\text{Ry}/a_{0} . The electronic structure and electron-phonon coupling calculations were performed using interface with Wannier90 Mostofi et al. 2008 and EPW code Poncé et al. 2016 . The elements of the electron-phonon matrix were initially computed on coarse grids of 8\times 6\times 6 \mathbf{k} -points and 4\times 2\times 2 \mathbf{q} -points, which were then interpolated onto dense \mathbf{k} - and \mathbf{q} -grid of 30\times 20\times 20 for electron and phonon wavevectors, respectively. The electron group velocities were calculated using Wannier90, while the conductivity and scattering rates were obtained using EPW based on the Boltzmann Transport Equation (BTE). Integration for conductivity and scattering rates was performed using adaptive smearing within an energy window of \pm 0.5 eV around the Fermi level, ensuring sufficient coverage of the active transport states. Figure 2: (a) Schematic view of growth approaches attempted for \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} synthesis, showing substrate temperature over time with the provided elemental fluxes represented by colored bars at the top. (b) Intensity of specular RHEED diffraction over the course of growth for each approach. Inset: RHEED patterns at selected stages of growth. (c) Snippet of the X-ray diffraction (XRD) \theta/2\theta diffraction pattern resulting from each growth approach. Figure 3: (a) \theta/2\theta diffraction pattern from a 75 nm film produced by the optimized two-step growth approach. (b) Rocking curve of the (006) peak. (c) Azimuthal \varphi -scan of an asymmetric diffraction peak showing a [100] {}_{\text{LaBi}_{2}}\parallel [110] {}_{\text{MgO}} register with the MgO substrate. (d) Reciprocal space map along the h=1 rod, showing finite splitting along Q_{z} consistent with the Yb-mono structure. III Properties of \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} Thin Films We begin by canvassing the history of \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} . Previous reports have provided conflicting characterizations of the crystal structure, ranging from trigonal to orthorhombic crystal classes, and the structure is absent in the Inorganic Crystal Structure Database (ICSD) Hulliger and Ott 1977 ; Nomura et al. 1977 ; Petrovic et al. 2002 . This is presumably due in part to the extreme air sensitivity of the compound. Nevertheless, a layered compound reminiscent of \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} is typically assumed to describe the structure. Despite being an excellent metal, \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} remains the only member of the La-Bi alloy system without reported ambient-pressure superconductivity besides the rock-salt structured LaBi Hulliger and Ott 1977 . Meanwhile, in addition to the modified dimerization illustrated in Fig. 1 b, thin films of both \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} Llanos et al. 2024a and \text{CeSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} Dorrian et al. 2025 display a unique structure-type characterized by a monoclinic shear, in contrast to bulk orthorhombic crystals. The unique insights that thin films have provided into the delicate structural properties of this class of materials have motivated our efforts to synthesize \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and characterize its structure. Figure 2 illustrates our entry point for producing high-quality, single-phase films. Four representative films correspond to distinct sequences of substrate temperatures (labeled T_{\mathrm{G1}}\approx 450-465^{\circ} C and T_{\mathrm{G2}}\approx 325-335^{\circ} C, Fig. 2 a) with fixed La and Bi beam fluxes. We focus first on the initial 15 minutes of growth. At the higher temperature T_{\mathrm{G1}} (black), the film quickly develops a streaky RHEED pattern with oscillating intensity (Fig. 2 b), indicating a flat crystal coating the MgO surface in a layer-by-layer fashion. Out-of-plane XRD shows reflections from the (001) \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} lattice planes without impurity phases, where we have assumed the same diffraction conditions as observed in bulk crystals (Fig. 2 c) Yoshihara et al. 1975 ; Zhou et al. 2018 . This higher T_{\mathrm{G}} , however, is found to be unsuitable for prolonged growth; the RHEED pattern gradually dims and degrades into a spottier series of streaks, as shown in Fig. 2 b. XRD confirms this to represent a complete decomposition of the \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} film into (111)-oriented LaBi, as shown in the green trace in Fig. 2 c. This can be understood in terms of the continuous thermal desorption of Bi from the substrate; due to its low melting point and high vapor pressure, lower growth temperatures are typically employed in the initial stages of growth for thin films of Bi-based compounds to ensure proper Bi adhesion, followed by a higher T_{\mathrm{G}} deposition to improve film crystallinity Li et al. 2010 ; Fan et al. 2020 ; Bansal et al. 2011 ; Ueda et al. 2020 ; Loke et al. 2023 . To test this approach, we begin with a lower temperature T_{\mathrm{G2}} as illustrated by the orange traces in Figs. 2 a- 2 c. Here, RHEED dims rapidly and displays diffuse features even at the 15 minute mark, with XRD reflecting a mixed-phase film composed of LaBi [002], \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , and peaks associated with \text{AuCu}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{3}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{3}}\text{}} -structured \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{3}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{3}}\text{}} Kinjo et al. 2016 . Some combination of limited surface migration energy and excessive Bi incorporation at low temperatures likely contributes to this behavior, driving the system towards multiple competing phases. The approach to realizing single-phase films is therefore a two-step process; starting with a high- T_{\mathrm{G}} buffer to obtain a phase pure template layer followed by an extended low- T_{\mathrm{G}} growth period (blue), the RHEED pattern persists and oscillates in intensity, and the resulting film is found to be phase-pure \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} with sharp (001) Bragg peaks and pronounced Laue fringes. This modified two-step procedure demonstrates the complex thermodynamics at play in the thin-film growth of intermetallic phases containing volatile species. Table 1: Calculated relative total energies, lattice constants, and angles of the Sm-type and Yb-mono bulk phases of \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} without (with) SOC. SG denotes the space group. Experimental values reported in the literature and obtained in this work are also included for comparison. System SG \Delta E (meV/La) a (Å) b (Å) c (Å) \alpha ( ∘ ) \beta ( ∘ ) \gamma ( ∘ ) Without SOC Sm-type 64 110.21 4.575 4.575 18.275 90.00 90.00 90.74 Yb-mono 12 0.000 4.649 4.588 17.457 90.00 87.11 90.00 With SOC Sm-type 64 70.06 4.598 4.598 18.121 90.00 90.00 90.69 Yb-mono 12 0.000 4.689 4.573 17.411 90.00 87.38 90.00 Experimental Bulk exp. Nomura et al. 1977 4.74 4.56 17.51 This work 4.73 4.57 17.50 90.0 86.85 90.0 Figure 3 summarizes the structural analysis of a 75 nm film grown via the optimized two-step procedure. Sharp, high-intensity Bragg peaks from the {001} \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} lattice planes in the XRD \theta/2\theta scan (Fig. 3 a) indicate both considerable crystal quality and the strong X-ray scattering expected from heavy Bi ions. The spacing of the Bragg peaks corresponds to an out-of-plane lattice constant c\approx 17.5 Å. Rocking curve analysis of the (006) diffraction condition (Fig. 3 b) gives a FWHM \approx 0.045^{\circ} , indicating a uniform orientation of lattice planes throughout the film. \varphi -scans of both the \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} film and MgO substrate in Fig. 3 c demonstrate a [100] {}_{\text{LaBi}_{2}}\parallel [110] {}_{\text{MgO}} epitaxial register without 45 ∘ -rotated domains. Figure 3 d shows asymmetric reciprocal space mapping (RSM) taken along the h=1 rod (See Fig. S4 of the Supplemental Material for extended RSM Sup ). As observed in epitaxial films of \textit{Ln}\text{Sb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , we observe a finite splitting of the ( \pm 1 0 12) peak along Q_{z} associated with an out-of-plane tilt of the \vec{a}^{*} reciprocal lattice vector. This implies a monoclinic structure for the \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} thin films Llanos et al. 2024a , in contrast to the orthorhombic space group previously assumed in bulk crystal studies. The diffraction observed in RSM can be accurately indexed with in-plane lattice constants a\approx 4.73 Å, b\approx 4.57 Å, and a monoclinic tilt of \beta\approx 86.9^{\circ} applied to one of three possible space groups (No. 12 A2/m , No. 5 A2 , No. 8 Am ) Flack 2016 . While a definitive space group assignment based on XRD alone is not possible, supporting density functional theory (DFT) calculations of the thermodynamic stability of potential stacking configurations point towards the Yb-monoclinic structure as lower in energy than the competing Sm-type structure reported in bulk \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} crystals Llanos et al. 2024a . The energy per La atom, the calculated lattice parameters both with and without SOC, and thin-film and bulk crystal experimental lattice parameters Nomura et al. 1977 ; Petrovic et al. 2002 are reported in Table 1 . The close agreement between these results suggests that the MBE-grown films stabilize in the same stacking configuration as bulk samples grown in Bi flux, which is in contrast to the case of \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} Llanos et al. 2024a . The lattice parameters appear consistent with the Yb-monoclinic phase to within \sim 1 \% . We therefore suspect that bulk studies have up to now misindexed monoclinic {}\mathrm{LaBi}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}} as an orthorhombic system. Due to the restrictions imposed by the samples’ air sensitivity, we have been unsuccessful in directly imaging the crystal structure using electron microscopy. Figure 4: (a) Temperature-dependent sheet resistivity \rho_{\mathrm{xx}} of a 75 nm film down to 2 K. (b) Hall resistivity \rho_{\mathrm{yx}} and (c) symmetrized magnetoresistance ratio (MRR) under an out-of-plane field at a series of temperatures. (d) Superconducting transitions below 1 K plotted as the temperature-dependent ratio of resistance to normal-state resistance R_{0} , for a series of perpendicular magnetic fields. (e) Upper critical field H_{c2}^{\perp} (defined at R=0.5R_{0} ) versus temperature. Fit to the single-gap Ginzberg-Landau formula shown as dashed line. Figure 5: Calculated surface energy \gamma as a function of the number of quintuple layers for \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} (blue) and \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} (green). The filled (open) circles represent the results obtained with (without) spin-orbit coupling (SOC). A summary of magnetotransport measurements on the 75 nm thick \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} film is presented in Fig. 4 . The temperature-dependent sheet resistivity (Fig. 4 a) shows metallic behavior with a RRR (\rho_{\mathrm{xx}}(\text{300 K})/\rho_{\mathrm{xx}}(\text{2 K}) ) of over 32. The anti-symmetrized Hall resistivity and symmetrized magneto-resistance ratio (MRR) are presented in Fig. 4 b-c. The non-linear \rho_{\mathrm{yx}} suggests at least two carrier types with a dominant hole-like species. The nearly linear, non-saturating MRR surpasses 100\% at 14 T. Previous reports of \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} transport properties were restricted to temperatures above the millikelvin regime. In Fig. 4 d, we present resistance measurements down to T = 20 mK where a sharp transition to a zero-resistance superconducting state is observed at about 550 mK at zero field. Suppression of this transition via a magnetic field applied perpendicular to the film confirms it as a transition into a superconducting ground state. Our phase-pure XRD analysis and UHV growth environment with persistent Bi desorption make it unlikely that this superconducting transition is a result of an extrinsic impurity phase; furthermore, reports of extrinsic superconductivity due to filamentary or thin-film Bi impurities in Bi-based compounds tend to occur at much higher temperature scales on the order of 1–2 K Xiang et al. 2019 . We therefore conclude that this represents the first observation of intrinsic superconductivity in \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} at ambient pressure. The superconducting coherence length is obtained by fitting the temperature-dependent critical field H_{c2}(T) to the single-gap Ginzburg-Landau model, H_{c2}^{\perp}(T)=\left(\frac{\Phi_{0}}{2\pi\xi_{ab}^{2}}\right)\left(1-\left(\frac{T}{T_{c}}\right)^{2}\right). (1) We find \xi_{ab}\approx 221 nm. Based upon an estimate of the mean free path of 300 nm using the multicarrier Drude framework, we can infer that the films are likely clean-limit superconductors at this thickness. Figure 6: Comparison of experimental transport characteristics between \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} for several film thicknesses. (a) RRR and (b) base-temperature residual resistivity, the latter plotted on a log scale. Approximate (c) carrier mobility and (d) mean free path as estimated by fits of magnetotransport data to the two-carrier Drude model. IV Comparison of La \text{Pn}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , Pn = Sb , Bi Despite the crystallographic similarities, there are notable differences between the growth dynamics and electronic properties of \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , which will be the focus of the remainder of this manuscript. One notable difference is the strong RHEED oscillations observed during growth, as shown in Fig. 2 , which are not observed in \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} for any experimental condition. As RHEED oscillations are typically associated with a layer-by-layer growth mode, this contrast points to a stark difference in the surface energies of the (001) lattice planes. To substantiate this analysis, density functional theory (DFT) calculations were performed to evaluate the surface energies of the LaSb- or LaBi-terminated corrugated layers as a function of quintuple layer (QL) thickness, where two QLs stacked along \vec{c} comprise one conventional unit cell (See Fig. S5 of the Supplemental Material Sup ). The results of these calculations are shown in Fig. 5 . The surface energies are determined from Sun and Ceder 2013 ; Boettger 1994 \gamma={{E_{\mathrm{s}}-nE_{\mathrm{bulk}}}\over{2A}}, (2) where E_{\mathrm{s}} is the total energy of the slab, E_{\mathrm{bulk}} the bulk energy per atom, n the number of atoms in the slab, and A the surface area. The surface energies are found to converge rapidly at two QL, and two features stand out; \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} has a smaller surface energy for all thicknesses than \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , and the inclusion of SOC significantly lowers these values, especially for \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} . Figure 7: (a) \mathrm{LaSb_{2}} and (b) \mathrm{LaBi_{2}} band structures with (solid) and without (dashed) SOC. In the case of SOC, colors denote the band- and k-dependent \Sigma_{n\textbf{k}}^{\prime\prime} which is proportional to the scattering rate. (c) Calculated scattering rate for different band indices n and k points, and (d) temperature-dependent in-plane conductivity, \sigma_{\textrm{in-plane}}=(\sigma_{xx}+\sigma_{yy})/2 for \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} (green) and \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} (blue), respectively. The transport properties of \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} are compared in Fig. 6 across several film thicknesses. The RRR (Fig. 6 a) and low- T residual resistivity (Fig. 6 b) values are directly obtained from electrical measurements, whereas the carrier mobilities (Fig. 6 c) and mean free paths (Fig. 6 d) are extracted from multicarrier Drude fits as described in the Fig. S5 of the Supplemental Material Sup . \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} is observed to be more conducting than \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} regardless of the metric used. While this may be intuitively attributed to an increased hopping integral for the enlarged Bi orbitals, our proceeding analysis of the band structure and phonon scattering rates provides an alternative insight into this behavior. In order to understand the underlying origin of the transport properties of the two compounds, we have carried out ab initio calculations of the electrical conductivity tensor elements, \sigma_{\alpha\beta} , for the bulk phase, employing the self-energy relaxation time approximation (SERTA) Ziman 2001 ; Poncé et al. 2016 ; Giustino 2017 ; Poncé et al. 2018 , \sigma_{\alpha\beta}=en_{e}\mu_{e,\alpha\beta}=-\frac{e^{2}}{\Omega}\sum_{n\in\mathrm{CB}}\int_{\Omega_{\mathrm{BZ}}\in\mathrm{CB}}d\mathbf{k}\frac{\partial f_{n\mathbf{k}}^{0}}{\partial\varepsilon_{n\mathbf{k}}}v_{n\mathbf{k},\alpha}v_{n\mathbf{k},\beta}\tau_{n\mathbf{k}}^{0}, (3) where \alpha , \beta denote Cartesian coordinates, n_{e} the electron carrier concentration, e the electron charge, \mu_{e,\alpha\beta} the electron mobility tensor component, \Omega and \Omega_{\mathrm{BZ}} the volumes of the unit cell and the first Brillouin zone, respectively, \varepsilon_{n\mathbf{k}} the single-particle electron eigenvalue of band index n and wave vector \mathbf{k} , f_{n\mathbf{k}}^{0} the equilibrium Fermi-Dirac distribution function, v_{n\mathbf{k},\alpha}=\hbar^{-1}\partial\varepsilon_{n\mathbf{k}}/\partial{k_{\alpha}} the \alpha component of the group velocity, the sum runs over band indices n within the conduction band (CB), and \tau_{n\mathbf{k}}^{0} is the band- and \mathbf{k} -dependent electron-phonon (el-ph) relaxation time. The scattering rate, 1/ \tau_{n\mathbf{k}}^{0} , defined as the inverse of the relaxation time can be directly calculated from the imaginary part of the el-ph self-energy Poncé et al. 2018 ; Poncé et al. 2016 ; Giustino 2017 , \begin{split}1/\tau_{n\mathbf{k}}^{0}(\omega,T)=2\Sigma^{\prime\prime}_{n\mathbf{k}}(\omega,T)=\frac{2\pi}{\hbar}\sum_{m\nu}\int_{BZ}\frac{d\mathbf{q}}{\Omega_{\mathrm{BZ}}}\left|g_{mn,\nu}(\mathbf{k},\mathbf{q})\right|^{2}\\ \hskip 9.24994pt\times\Big\{\left[n_{\mathbf{q}\nu}(T)+f_{m\mathbf{k}+\mathbf{q}}(T)\right]\delta\left(\omega-(\varepsilon_{m\mathbf{k}+\mathbf{q}}-\varepsilon_{F})+\omega_{\mathbf{q}\nu}\right)\\ \hskip 9.24994pt+\left[n_{\mathbf{q}\nu}(T)+1-f_{m\mathbf{k}+\mathbf{q}}(T)\right]\delta\left(\omega-(\varepsilon_{m\mathbf{k}+\mathbf{q}}-\varepsilon_{F})-\omega_{\mathbf{q}\nu}\right)\Big\}.\end{split} (4) where \omega is the energy of the electron, T the temperature, \omega_{\mathbf{q}\nu} the frequency, n_{\mathbf{q}\nu}(T) the occupation (using Bose-Einstein statistics) of the phonon mode \nu at wave vector \mathbf{q} , and g_{mn,\nu}(\mathbf{k},\mathbf{q}) the electron-phonon matrix element between Kohn-Sham states ( n,\mathbf{k} ) and ( m,\mathbf{{k+q}} ) interacting with a phonon mode with wave vector \mathbf{q} and band index \nu . The temperature dependence of the scattering rate arises from the electron and phonon occupations at finite temperatures, while electron-phonon matrix is calculated at T = 0. Figures 7 a and 7 b show the band structures of \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , respectively, calculated with and without SOC. In both panels, the band structures are color-coded according to the imaginary part of the self-energy (i.e., inverse of the relaxation time). While \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} exhibits relatively large scattering rates in some parts of the bands, \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} shows overall smaller scattering rates throughout the Brillouin zone. To quantitatively corroborate the different scattering rates (and hence electron-phonon interactions) between the two systems, the scattering rate, 1/\tau_{n\mathbf{k}}^{0}(\omega,T=0) , as a function of the chemical potential is collected over all band indices and sampled \mathbf{k} -points and displayed as a scatter plot, as shown in Fig. 7 c. This clearly demonstrates (i) substantially larger scattering rates in \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} compared to those in \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} over a wide range of chemical potential (and hence charge doping) and (ii) a wider distribution of scattering rates in \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , indicating the sensitivity of the relaxation time to both band index and the electron momentum. We attribute this difference in scattering rates between \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} to significant relativistic effects in \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} . The calculated temperature-dependent in-plane conductivity, \sigma_{\textrm{in-plane}}=(\sigma_{\textrm{xx}}+\sigma_{\textrm{yy}})/2 , is shown in Fig. 7 d for \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} . As these calculations neglect the impact of static disorder, they cannot be compared directly to our experimentally obtained values. Nevertheless, they show that the relatively larger conductivity of \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} over the entire temperature range considered arises from the lower scattering rates in \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} compared to \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , rather than from differences in their relative group velocities (See Fig. S8 of the Supplemental Material Sup ). In summary, we attribute both key differences between the isostructural compounds \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} to the enhanced SOC in \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and its impact on the electronic structure. Our calculations show that the relativistic correction fundamentally alters the Fermi surface by shifting bands relative to the Fermi energy at high-symmetry points in the Brillouin zone. Consequently, the resulting energy landscape exhibits weaker phonon scattering and lower surface energy, which lead to more metallic transport and enhanced layer-by-layer growth in \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} compared to \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} . V Conclusion We have reported the thin-film synthesis and the structural and electrical characterization of \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} . Our structural analysis agrees well with bulk-reported lattice parameters but reveals a monoclinic structure, similar to thin films of \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} Llanos et al. 2024a , suggesting that previous bulk studies misindexed the compound. We have also demonstrated ambient-pressure superconductivity with T_{\mathrm{c}}\sim 0.55 K. Beyond these findings, this work elucidates the subtle role of SOC in influencing both thin-film growth dynamics and metallicity. Compared to \text{LaSb}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , SOC in \text{LaBi}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} lowers the (001) surface energy and suppresses carrier scattering near the Fermi energy, resulting in layer-by-layer growth and superior electrical conductivity. VI