Two-step growth of high-quality single crystals of the Kitaev magnet <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mi>α</mml:mi><mml:mtext>−</mml:mtext><mml:mi>RuCl</mml:mi></mml:math> <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:msub><mml:mrow/><mml:mn>3</mml:mn></mml:msub></mml:math>
Reiko Namba, Keigo Imamura, Ryoko Ishioka, K. Ishihara, Tatsuya Miyamoto, Hiroshi Okamoto, Yuuki Shimizu, Yohei Saito, Yassine Agarmani, Michael Lang, Hideki Murayama, Y. Xing, S. Suetsugu, Y. Kasahara, Y. Matsuda, K. Hashimoto, T. Shibauchi
Abstract
The layered honeycomb magnet $\ensuremath{\alpha}\text{\ensuremath{-}}{\mathrm{RuCl}}_{3}$ is the most promising candidate for a Kitaev quantum spin liquid (KQSL) that can host charge-neutral Majorana fermions. Recent studies have shown significant sample dependence of thermal transport properties, which are a key probe of Majorana quasiparticles in the KQSL state, highlighting the importance of preparing high-quality single crystals of $\ensuremath{\alpha}\text{\ensuremath{-}}{\mathrm{RuCl}}_{3}$. Here, we present a relatively simple and reliable method to grow high-quality single crystals of $\ensuremath{\alpha}\text{\ensuremath{-}}{\mathrm{RuCl}}_{3}$. We use a two-step crystal growth method consisting of a purification process by chemical vapor transport (CVT) and a main crystal growth process by sublimation. The obtained crystals exhibit a distinct first-order structural phase transition from the monoclinic $(C2/m)$ to the rhombohedral $(R\overline{3})$ structure at $\ensuremath{\sim}150$ K, which is confirmed by the nuclear quadrupole resonance spectra with much sharper widths than previously reported. The Raman spectra show the absence of defect-induced modes, supporting the good crystallinity of our samples. The jumps in the thermal expansion coefficient and specific heat at the antiferromagnetic (AFM) transition at 7.6--7.7 K are larger and sharper than those of previous samples grown by the CVT and Bridgman methods and do not show any additional AFM transitions at 10--14 K due to stacking faults. The longitudinal thermal conductivity in the AFM phase is significantly larger than previously reported, indicating a very long mean-free path of heat carriers. All the results indicate that our single crystals are of superior quality with good crystallinity and few stacking faults, which provides a suitable platform for studying the Kitaev physics.