Abstract
The synthesis of lithium-deficient $Li_xBC$ ($x > 0.5$) via high-temperature deintercalation is reported. Despite theoretical predictions of high-temperature superconductivity analogous to $MgB_2$, experimental characterization using neutron and X-ray diffraction, magnetometry, and electron microscopy reveals no superconducting transition above 2 K. The absence of superconductivity is attributed to structural distortions, specifically the buckling of boron-carbon layers, which alters the electronic density of states and suppresses the predicted high-$T_c$ state.
Introduction
The discovery of superconductivity at 39 K in magnesium diboride ($MgB_2$) in 2001 sparked intense interest in layered borides and carbides. $LiBC$, a structural analogue to $MgB_2$, was identified as a prime candidate for even higher transition temperatures ($T_c$). While stoichiometric $LiBC$ is an insulator, theoretical models suggested that removing lithium—thereby "hole-doping" the boron-carbon layers—would create an electronic environment similar to $MgB_2$, potentially yielding a $T_c$ exceeding 100 K. This study details the challenging synthesis of these $Li_xBC$ phases and the subsequent investigation into their physical properties to verify if theory holds up in the laboratory.
Experimental Methods
The synthesis of $Li_xBC$ was achieved through high-temperature lithium deintercalation. Stoichiometric $LiBC$ was heated to 1500 °C (1773 K) in sealed tantalum tubes for durations up to 24 hours. This aggressive thermal treatment was necessary to overcome the kinetic barriers to lithium removal from the stable lattice.
Structural characterization relied heavily on neutron powder diffraction performed at the ISIS spallation source. Because lithium is a light element, neutron diffraction provides significantly better contrast and precision for determining lithium occupancy compared to standard X-ray techniques. Rietveld refinement was used to extract lattice parameters and stoichiometric ratios.
Additionally, Selected Area Electron Diffraction (SAED) was performed to examine the long-range ordering of the crystal lattice and to detect any potential superstructures arising from lithium vacancy patterns.
Magnetic properties were measured using SQUID magnetometry down to 2 K in fields of 20 G and 40 G to search for the Meissner effect, the definitive signature of superconductivity.
Results and Discussion
The deintercalation process successfully produced crystalline $Li_xBC$ with lithium levels as low as $x \approx 0.5$. However, the samples exhibited phase separation rather than a continuous solid solution. For instance, a sample with an average composition of $Li_{0.78}BC$ was found to be a mixture of two distinct phases: $Li_{0.56}BC$ and $Li_{0.90}BC$.
As lithium is removed, the $c$-axis (interlayer spacing) increases due to reduced electrostatic attraction between the layers. Conversely, the $a$-axis (intralayer distance) contracts significantly. For the $Li_{0.56}BC$ phase, the refined lattice parameters were $a = 2.62631(7)$ Å and $c = 7.4390(4)$ Å. The B-C bond length shortened to 1.51496 Å in the most oxidized samples, a significant contraction from the parent material.
Despite achieving the hole-doping levels predicted to induce superconductivity, magnetometry measurements showed no evidence of a superconducting transition down to 2 K. Density Functional Theory (DFT) calculations provided a likely explanation for this discrepancy. The models revealed that lithium removal causes the B-C layers, which are perfectly flat in $LiBC$, to "buckle" by approximately 0.34 Å. This structural relaxation mixes the $\sigma$ and $\pi$ electronic bands, likely suppressing the pairing mechanism required for superconductivity.
Conclusion
While $Li_xBC$ can be synthesized with significant lithium deficiency, the material does not exhibit the high-$T_c$ superconductivity predicted by simple rigid-band models. The structural instability of the boron-carbon layers upon doping—specifically the buckling effect—appears to be the primary factor preventing the formation of a superconducting state. This work highlights the critical importance of accounting for lattice relaxation and local disorder in theoretical predictions of new superconducting materials.
Summary updated