The Kitaev model has become a source of much excitement in the field of condensed matter. It is a two dimensional model of spins ½ on a honeycomb lattice with bond-dependent interactions. Its interesting properties include a quantum spin liquid ground state and anyonic excitations. These properties could lead to exciting applications in quantum computing if materials were found to behave similarly to the Kitaev model. Such materials have been found, however the Kitaev model is too simple to describe these materials and additional interactions must be considered. The Heisenberg interaction is one such additional interaction. As such, we can define the Kitaev-Heisenberg model by combining the Kitaev and Heisenberg interactions. We can now ask ourselves if the quantum spin liquid ground state and anyonic excitations still exist in the Kitaev-Heisenberg model. To answer this question, a non-local string order parameter has been defined which is non-zero inside the quantum spin liquid phase and zero outside of it. This string order parameter was shown to exist and survive the Heisenberg interaction on the 2-leg ladder. In this thesis, we look to expand this result to multileg ladders such as the 3-leg, 4-leg, and 5-leg ladders to see if the string order parameter survives in the Kitaev-Heisenberg model in 2 dimensions. Our results show that the string order parameter does exist in multileg ladders, however the phase space window in which it survives the Heisenberg interaction is narrower than in the 2-leg ladder. / Thesis / Master of Science (MSc)

This work presents macroscopic and microscopic experiments of the disordered hexagonal double perovskite BaTi1/2Mn1/2O3, in order to characterize its electronic, magnetic and structural properties to support the possibility that this system hosts a spin liquid phase. Such assumption is based on the absence of a transition to a magnetically ordered phase in the magnetic and thermodynamic measurements, which points to a strong magnetic frustration in this material. In addition, it is observed the formation of a correlated spin state. To characterize this correlation, we resorted to Muon Spin Resonance (µSR) experiments to measure the low temperature spin dynamics. The zero field µSR relaxation regime displays dynamic magnetism down to T = 0.019 K and longitudinal field experiments support as well that dynamic magnetism persists at low temperatures, a behavior expected for a spin liquid system. The magnetic behavior of BaTi1/2Mn1/2O3 consists in the high temperature physics being dominated by the presence of magnetic trimers, magnetic dimers, and orphan spins. At lower temperatures, the effective magnetic degrees of freedom, composed by orphan spins and magnetic trimers, are correlated but no phase transition is detected down to T = 0.1 K, despite the effective exchange couplings between magnetic trimers and orphan spins being -8.5 K, resulting in a magnetic frustration parameter of at least 85. The possibility that disorder is responsible for the spin liquid ground state is discussed, however, other scenarios are not totally discarded. For example, the possibility that the measured state is not the true ground state, which could lie at even lowers temperatures or the possible formation of a spin glass state. This work raises questions that are not easy to answer. Ultimately, the growth of a single crystal is necessary to continue the characterization of BaTi1/2Mn1/2O3. Besides, theoretical and experimental developments in this field of research are needed to find a more direct and conclusive way to characterize the magnetic phases in this complex material. / Neste trabalho apresento dados experimentais macroscópicos e microscópicos da peroviskita hexagonal dupla BaTi1/2Mn1/2O3, a fim de caracterizar sua estrutura eletrônica, magnética e cristalina para embasar a possibilidade deste sistema apresentar uma fase de líquido de spin. Esta hipótese está baseada na ausência de transição para uma fase magneticamente ordenada nas medidas magnéticas e termodinâmicas, que apontam para uma forte frustração magnética neste material. Além disso, é observada a formação de um estado de spins correlacionados. Para caracterizar esta correlação, recorremos para experimentos de ressonância de múons (µSR) para medir a dinâmica de spins em baixas temperaturas. Dados de µSR para campo magnético nulo mostram em seu regime de relaxamento um magnetismo dinâmico para temperaturas tão baixas quanto T = 0.019 K, adicionalmente, experimentos com campo magnético longitudinal aplicado apontam também que o magnetismo dinâmico persiste em baixas temperaturas, um comportamento esperado para um sistema de líquido de spin. O comportamento magnético do BaTi1/2Mn1/2O3 consiste na física de altas temperaturas sendo dominada pela presença de trimers magnéticos, dimers magnéticos e spins órfãos. Para temperaturas mais baixas, os graus de liberdade magnéticos são efetivamente compostos por spins órfãos e trimers magnéticos, que estão correlacionados mas nenhuma transição de fase é detectada para temperaturas tão baixas quanto T = 0.1 K, mesmo que a constante de interação efetiva entre os spins órfãos e os trimers magnéticos seja -8.5 K, resultando num fator de frustração magnética de ao menos 85. A possibilidade da desordem ser responsável pelo estado fundamental de líquido de spin é discutida, no entanto, outros cenários não estão totalmente descartados, por exemplo, a possibilidade de que o estado medido não seja o verdadeiro estado fundamental, e que este estaria em temperaturas ainda mais baixas ou a possível formação de um estado de vidro de spin. Este trabalho levanta questões que não são fáceis de responder. Por fim, o crescimento de uma amostra monocristalina é necessário para continuar a caracterização do BaTi1/2Mn1/2O3. Ademais, desenvolvimentos de cunho teórico e experimental neste campo de pesquisa são necessários para encontrar um método mais direto e conclusivo para caracterizar a fase magnética neste material complexo.

This work presents macroscopic and microscopic experiments of the disordered hexagonal double perovskite BaTi1/2Mn1/2O3, in order to characterize its electronic, magnetic and structural properties to support the possibility that this system hosts a spin liquid phase. Such assumption is based on the absence of a transition to a magnetically ordered phase in the magnetic and thermodynamic measurements, which points to a strong magnetic frustration in this material. In addition, it is observed the formation of a correlated spin state. To characterize this correlation, we resorted to Muon Spin Resonance (µSR) experiments to measure the low temperature spin dynamics. The zero field µSR relaxation regime displays dynamic magnetism down to T = 0.019 K and longitudinal field experiments support as well that dynamic magnetism persists at low temperatures, a behavior expected for a spin liquid system. The magnetic behavior of BaTi1/2Mn1/2O3 consists in the high temperature physics being dominated by the presence of magnetic trimers, magnetic dimers, and orphan spins. At lower temperatures, the effective magnetic degrees of freedom, composed by orphan spins and magnetic trimers, are correlated but no phase transition is detected down to T = 0.1 K, despite the effective exchange couplings between magnetic trimers and orphan spins being -8.5 K, resulting in a magnetic frustration parameter of at least 85. The possibility that disorder is responsible for the spin liquid ground state is discussed, however, other scenarios are not totally discarded. For example, the possibility that the measured state is not the true ground state, which could lie at even lowers temperatures or the possible formation of a spin glass state. This work raises questions that are not easy to answer. Ultimately, the growth of a single crystal is necessary to continue the characterization of BaTi1/2Mn1/2O3. Besides, theoretical and experimental developments in this field of research are needed to find a more direct and conclusive way to characterize the magnetic phases in this complex material. / Neste trabalho apresento dados experimentais macroscópicos e microscópicos da peroviskita hexagonal dupla BaTi1/2Mn1/2O3, a fim de caracterizar sua estrutura eletrônica, magnética e cristalina para embasar a possibilidade deste sistema apresentar uma fase de líquido de spin. Esta hipótese está baseada na ausência de transição para uma fase magneticamente ordenada nas medidas magnéticas e termodinâmicas, que apontam para uma forte frustração magnética neste material. Além disso, é observada a formação de um estado de spins correlacionados. Para caracterizar esta correlação, recorremos para experimentos de ressonância de múons (µSR) para medir a dinâmica de spins em baixas temperaturas. Dados de µSR para campo magnético nulo mostram em seu regime de relaxamento um magnetismo dinâmico para temperaturas tão baixas quanto T = 0.019 K, adicionalmente, experimentos com campo magnético longitudinal aplicado apontam também que o magnetismo dinâmico persiste em baixas temperaturas, um comportamento esperado para um sistema de líquido de spin. O comportamento magnético do BaTi1/2Mn1/2O3 consiste na física de altas temperaturas sendo dominada pela presença de trimers magnéticos, dimers magnéticos e spins órfãos. Para temperaturas mais baixas, os graus de liberdade magnéticos são efetivamente compostos por spins órfãos e trimers magnéticos, que estão correlacionados mas nenhuma transição de fase é detectada para temperaturas tão baixas quanto T = 0.1 K, mesmo que a constante de interação efetiva entre os spins órfãos e os trimers magnéticos seja -8.5 K, resultando num fator de frustração magnética de ao menos 85. A possibilidade da desordem ser responsável pelo estado fundamental de líquido de spin é discutida, no entanto, outros cenários não estão totalmente descartados, por exemplo, a possibilidade de que o estado medido não seja o verdadeiro estado fundamental, e que este estaria em temperaturas ainda mais baixas ou a possível formação de um estado de vidro de spin. Este trabalho levanta questões que não são fáceis de responder. Por fim, o crescimento de uma amostra monocristalina é necessário para continuar a caracterização do BaTi1/2Mn1/2O3. Ademais, desenvolvimentos de cunho teórico e experimental neste campo de pesquisa são necessários para encontrar um método mais direto e conclusivo para caracterizar a fase magnética neste material complexo.

In this thesis we present results from quantum Monte Carlo for the fully-frustrated honeycomb lattice. The XXZ model is of interest in the classical limit, as there is a mapping between the classical fully-frustrated honeycomb Ising model groundstates and the classical hard-core dimer model groundstate. The aim of this work is to explore the effect of quantum fluctuations on the fully-frustrated honeycomb model to see what sort of interesting physics arises. One might expect unusual physics due to the quantum hard-core dimer model, where interesting physics are known to exist. This is because there is a duality mapping between the classical dimer model and the classical fully-frustrated honeycomb Ising model. Indeed, by studying the fully-frustrated honeycomb XXZ model we find that in some cases the system orders into crystal-like structures, a case of order-by-disorder. The most interesting case, when the frustrating bonds are chosen randomly, reveals to us a novel state without any discernible order while at the same time avoiding the freezing one would expect of a glass. This state is a featureless system lacking low temperature magnetic susceptibility---a candidate ``quantum spin liquid''. Future work that might more easily measure quantum spin liquid criteria is suggested.

In this thesis we present results from quantum Monte Carlo for the fully-frustrated honeycomb lattice. The XXZ model is of interest in the classical limit, as there is a mapping between the classical fully-frustrated honeycomb Ising model groundstates and the classical hard-core dimer model groundstate. The aim of this work is to explore the effect of quantum fluctuations on the fully-frustrated honeycomb model to see what sort of interesting physics arises. One might expect unusual physics due to the quantum hard-core dimer model, where interesting physics are known to exist. This is because there is a duality mapping between the classical dimer model and the classical fully-frustrated honeycomb Ising model. Indeed, by studying the fully-frustrated honeycomb XXZ model we find that in some cases the system orders into crystal-like structures, a case of order-by-disorder. The most interesting case, when the frustrating bonds are chosen randomly, reveals to us a novel state without any discernible order while at the same time avoiding the freezing one would expect of a glass. This state is a featureless system lacking low temperature magnetic susceptibility---a candidate ``quantum spin liquid''. Future work that might more easily measure quantum spin liquid criteria is suggested.

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第22685号 / 理博第4626号 / 新制||理||1665(附属図書館) / 京都大学大学院理学研究科化学専攻 / (主査)准教授 前里 光彦, 教授 吉村 一良, 教授 有賀 哲也 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DGAM

L’étude de l’état fondamental liquide de spins est un des domaines très actif de la recherche en matière condensée. Le réseau le plus à même de stabiliser un tel état fondamental semble être, à deux dimensions, le réseau kagomé de spins antiferromagnétiques 1/2. Il y a à présent un consensus théorique sur le fait que ce modèle stabilise un état fondamental liquide de spin. Cependant, la nature de cet état est encore inconnue, notamment la nature des corrélations. Nous ne savons toujours pas si ces dernières sont à courte portée avec un gap dans le spectre d’excitations, ou si elles sont à plus longue portée avec un spectre d’excitations sans gap. D’un point de vue expérimental il n’existe que très peu de matériaux et seul l’Herbertsmithite présente un réseau kagomé de spins 1/2 géométriquement parfait. Les différentes études réalisées sur ce composé pointent toutes vers un état liquide de spin sans gap mais révèlent aussi des déviations à l’hamiltonien de Heisenberg qui pourraient être responsables de la fermeture de ce gap.Cette thèse traite de l’étude expérimentale principalement par RMN et µSR de nouveaux composés kagomé à base de vanadium faisant partie d’une famille récemment synthétisée, les vanadates fluorés à géométrie kagomé. Le matériau que nous avons le plus étudié est un composé à réseau kagomé de spins 1/2 à base de V4+, (NH4)2[C7H14N][V7O6F18] (DQVOF). Le modèle magnétique de ce composé peut être décomposé en deux sous systèmes presque indépendants, des plans kagomé trimérisés isolés et des ions V3+ quasi paramagnétiques. Les études de µSR démontrent une absence de gel magnétique jusqu’à 20 mK donc un état liquide de spins dans DQVOF. Les études de chaleur spécifique et de RMN dévoilent un comportement liquide de spin sans gap malgré la trimérisation du réseau et la faible valeur supposée de l’interaction Dzyaloshinskii Moriya. Nos résultats montrent finalement que l’absence de gap, intrinsèque ou due à des déviations à l’hamiltonien idéal, est une caractéristique robuste des matériaux kagomé. Nous avons de plus étudié un second matériau de cette famille, (NH4)2[C2H8N][V3F12] (DDVF), dont le réseau magnétique est formé par des plans kagomé découplés entre eux à base de V3+ (S = 1). Ce réseau présente de fortes distorsions par rapport au réseau idéal et les expériences thermodynamiques et de µSR mettent en évidence une transition magnétique vers un état gelé à 10 K avec une mise en ordre à longue distance qui s’effectue à 6 K uniquement. / The search for quantum liquid state is a very active field in condensed matter research. In two dimensions, the antiferromagnetic spin 1/2 kagome lattice seems to be the most able to stabilize such a ground state. Indeed, from recent theoretical investigations, we are now quite sure that this model has a quantum spin liquid ground state. However, we still do not know its nature, in particular the nature of its correlations. They could be short ranged with a gap in the excitation spectrum, or long ranged with a gapless excitation spectrum. On the experimental side, only few materials exist and only one possesses a geometrically perfect lattice, the Herbertsmithite. All the experiments that have been done on this compound reveal a gapless spin liquid state along with deviations to the spin 1/2 Heisenberg hamiltonian which could be responsible of the gap closure.This thesis deals with the experimental study, mainly by NMR and µSR, of new vanadium based kagomé compounds which are part of a newly synthesized family, the kagome fluoride vanadates. The material that we studied the most is a spin 1/2 kagomé compound based on V4+, (NH4)2[C7H14N][V7O6F18] (DQVOF). The magnetic model of this compound can be decomposed in two rather independent parts, trimerized kagome planes and quasi paramagnetic V3+ ions. The µSR studies, showing the absence of frozen moment down to 20 mK, reveal a spin liquid ground state in DQVOF. The heat capacity and the NMR experiments point out a gapless behavior despite trimerization and likely weak Dzyaloshinskii Moriya interactions. Our results demonstrate that the gapless ground state, whether intrinsic or due to deviation to the ideal hamiltonian, is a rather robust characteristic of kagome materials.Furthermore, we studied another compound of this family, (NH4)2[C2H8N][V3F12] (DDVF), which magnetic lattice is made of uncoupled kagomé planes based on V3+ (S = 1). The lattice shows large deviations to the ideal kagomé and the thermodynamic experiments and the µSR studies reveal a magnetic transition to a frozen state at 10 K with a long distance order which is effective only below 6 K.

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第21548号 / 理博第4455号 / 新制||理||1640(附属図書館) / 京都大学大学院理学研究科物理学・宇宙物理学専攻 / (主査)教授 川上 則雄, 教授 田中 耕一郎, 教授 前野 悦輝 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

Thesis advisor: Fazel Tafti / The search for an ideal quantum spin-liquid (QSL) material which can host a QSL ground state as well as exotic excitations has been one of the leading research topics in condensed matter physics over the past few decades. Out of all the proposals to realize the physics of a QSL, the Kitaev model is the most promising proposal with a QSL ground state. The Kitaev Hamiltonian is exactly solvable via fractionalization of its spin degrees of freedom into Majorana excitations, and it can be engineered in real materials. Among all the proposed Kitaev candidates, α-Li2IrO3, Na2IrO3, Li2RhO3, and α-RuCl3 are the most promising candidates. During my Ph.D. research I explored new physics related to Kitaev materials via modification of the symmetry and structural properties of these known Kitaev candidates. First, I studied how modification of the inter-layer chemistry can alter the thermodynamic properties of Kitaev candidate α-Li2IrO3 via an enhancement of the spin-orbit coupling (SOC) effect. The light, octahedrally-coordinated inter-layer Li atoms are replaced with heavier, linearly-coordinated Ag atoms to synthesize Ag3LiIr2O6. In addition to these structural modifications to the parent compound α-Li2IrO3, having heavier elements between the honeycomb layers in the Ag compound increased the effect of SOC in the honeycomb layers and led to a decrease in the long-range ordering temperature in Ag3LiIr2O6 compared to its parent compound. Second, I studied the effect of local crystal distortion in the presence of a weak SOC effect to explore a new spin-orbital state different from the Jeff=1/2 state. Based on theoretical predictions, the ground states of Kitaev materials can be tuned to other exotic spin-orbital states such as an Ising spin-1/2 state. To provide the proper conditions for a competition between the trigonal crystal distortion and the SOC effect, I modified the crystal environment around the magnetic elements in the parent compound Li2RhO3 via a topo-chemical method and synthesized Ag3LiRh2O6. An increase in the strength of trigonal distortion in Ag3LiRh2O6, in the presence of weak SOC, led to a transition from the Jeff=1/2 ground state (Kitaev limit) in the parent compound to an Ising spin-1/2 ground state (Ising limit) in the product. This change in spin-orbital state resulted in a dramatic change in magnetic behavior. Whereas Li2RhO3 shows a spin-freezing transition at 6 K, Ag3LiRh2O6 reveals a robust long-range antiferromagnetic transition at 94 K. This is the first realization of a change of ground state between the Kitaev and Ising limits in the same structural family. Lastly, I studied how the crystal symmetry can be an important factor in the physics of Kitaev materials. Honeycomb layered materials can be crystallized in space groups C2/m, C2/c, and P6_322. However, the crystal symmetry of most Kitaev candidates is described by the C2/m space group. We successfully synthesized a polymorph of a 3d Kitaev candidate, hexagonal Na2Co2TeO6 (P6_322 space group) in space group C2/m. The change in crystal symmetry of this cobalt tellurate replaced three anti-ferromagnetic (AFM) orders at 27, 15, 7 K in the hexagonal polymorph by a single AFM peak at 9.6 K in the monoclinic Na2Co2TeO6. / Thesis (PhD) — Boston College, 2023. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Physics.

The experimental quest for a quantum spin-liquid state (QSL) in frustrated magnetic systems addresses fundamental scientific interests, as this intriguing quantum phase provides excellent grounds for discovering exotic collective phenomena. ZnCu3(OH)6Cl2 (herbertsmithite), an S=1/2 kagomé-lattice Heisenberg antiferromagnet, is the most promising candidate for experimentally realizing a QSL. However, despite years of intense research, the nature of its paramagnetic ground state remains highly debated. The root cause of the controversy lies in the difficulty in distinguishing the effects of defects from the intrinsic properties of the kagomé lattice. In this thesis, we present 17-O nuclear magnetic resonance (NMR) measurements of an isotope-enriched ZnCu3(OH)6Cl2 single crystal. We succeeded in distinguishing the intrinsic magnetic behavior of the kagomé lattice from the defect-induced phenomena down to T~0.01J, where J~200K is the Cu-Cu super-exchange interaction. We identify NMR signals arising from the nearest-neighbor 17-O sites of Cu2+ defects occupying the Zn2+ interlayer sites. From the 17-O Knight shift measurements, we show that these Cu2+ defects induce a large Curie-Weiss contribution to the bulk-averaged susceptibility at low temperatures. Moreover, our 17-O single-crystal lineshapes show no signature of nonmagnetic Zn2+ defects within the kagomé lattice, and therefore, we rule out “anti-site disorder” as a cause of the paramagnetic ground state in ZnCu3(OH)6Cl2. Most importantly, we demonstrate that the intrinsic spin susceptibility of the kagome lattice asymptotically tends to zero below T~0.03J, indicating the presence of a finite gap Δ = 0.03~ 0.07J in the spin excitation spectrum; this gap is completely suppressed under the application of a high magnetic field of ~ 9T. The behavior of low-energy spin fluctuations probed by the 17-O nuclear spin-lattice relaxation rate is consistent with the gap signature observed for the 17-O Knight shift. In short, our 17-O NMR results provide the first experimental evidence for a gapped QSL realized in ZnCu3(OH)6Cl2. / Thesis / Doctor of Philosophy (PhD)