Abstract
A prototypical quasi2D metallic compound, 1TTaS_{2} has been extensively studied due to an intricate interplay between a Mottinsulating ground state and a chargedensitywave order. In the lowtemperature phase, 12 out of 13 Ta^{4+} 5delectrons form molecular orbitals in hexagonal starofDavid patterns, leaving one 5delectron with S = ½ spin free. This orphan quantum spin with a large spinorbit interaction is expected to form a highly correlated phase of its own. And it is most likely that they will form some kind of a shortrange order out of a strongly spinorbit coupled Hilbert space. In order to investigate the lowtemperature magnetic properties, we performed a series of measurements including neutron scattering and muon experiments. The obtained data clearly indicate the presence of the shortranged phase and put the upper bound on ~0.4 µ_{B} for the size of the magnetic moment, consistent with the orphanspin scenario.
Similar content being viewed by others
Introduction
The instability of charge density waves (CDW) found in lowdimensional electron systems of layered materials has attracted enormous attention recently. 1TTaS_{2} is a prototypical quasi2D metallic compound with a strong electronphonon coupling responsible often for CDW instabilities. Upon cooling, it undergoes a series of firstorder phase transitions to CDW, Mott and superconducting phases (see e.g., ref. 1 and references therein).
1TTaS_{2} bulk crystal has a lamellar structure, with each layer composed of a triangular lattice of Ta atoms which is then sandwiched by S atoms in an octahedral TaS_{6} coordination with a weak Van der Waals bonding between the layers. Above ~540 K, the structure is unmodulated trigonal with P3m1 symmetry; below which the triangular lattice exhibits a series of structural modulations.^{2, 3} First, an incommensurate CDW phase sets in at T ~ 540 K. Upon further cooling, the structure changes to a nearly commensurate phase at T _{nCDW} ~ 350 K. Finally, the material turns into a Mott insulating phase with inplane √13 × √13 superlattice distortion, which coexists with a commensurate CDW phase below T _{CDW} ~180 K. Superconductivity emerges in 1TTaS_{2} below ~2 K by introducing disorders.^{4} Recent angleresolved photoemission spectroscopy experiments suggest that the melted Mott state and the superconductivity coexist in real space^{5} providing better understanding of the interplay between electron correlation, charge order, and superconductivity. Unlike other Mott insulators, the CDW superlattices play the role of localization centers in the ground state of 1TTaS_{2}.
According to the current understanding, the CDW phase is composed of molecular orbitals of 13 Ta atoms, forming a hexagonal pattern of socalled Davidstar clusters. As proposed more than three decades ago,^{6} each Ta^{4+} of 1TTaS_{2} provides one 5d electron, and thus there are 13 5delectrons per each David star. Out of these 13 electrons, 12 electrons form 6 covalent bonds and so they become inactive both electrically and magnetically, leaving one 5delectron localized well inside the David star. This orphan electron provides S = ½ per each David star, and becomes localized at low temperatures (T < 60 K) in the center of the David star.^{7, 8} We note that in the 2D quantum triangular lattice systems, a combination of reduced dimensionality, geometric frustration, and a small spin value significantly enhances quantum fluctuations, which can easily induce exotic quantum states. In contrast to other S = ½ triangular systems showing rich phase diagrams such as CuRE _{2}Ge_{2}O_{8},^{9} Ba_{3}CoSb_{2}O_{9},^{10} and others,^{11, 12} there is no apparent magnetic ordering in the insulating ground state of 1TTaS_{2} down to the lowest measured temperatures.
Instead, a commensurate CDW phase forms most likely through Mott–Hubbard physics with possibly intricate electron–phonon and electronelectron interactions.^{13} This then leaves the orphan spin (S = 1/2) with a large spinorbit interaction in the lowtemperature CDW phase. How this orphan quantum spin behaves at low temperatures has been at the focus of current attention and hence extensively studied both theoretically^{7, 8, 14} and experimentally.^{13, 15,16,17} There is even a suggestion that it may form a highly correlated, “hidden” phase of its own and just recently,^{18} 1TTaS_{2} was suggested to be considered as a quantum spin liquid. Yet, none of the experimental works focused directly on the investigation of this lowtemperature phase with a very recent exception.^{19} It is thus interesting to ask whether some kind of a shortrange ordering would be formed out of the strongly spinorbit coupled Hilbert space, in which a magnetic moment µ = g _{ s }√(S(S + 1))/13 µ_{B} ~ 0.13 µ_{B} for S = ½ per each David star is expected.
In this work, we tried to address the question of a possible lowtemperature correlated phase. For this, we have performed extensive bulk measurements, polarized neutron diffraction and µSR studies covering a wide range of time scales and spatial resolution. µSR and diffuse neutron scattering are ideal experimental techniques for investigating the weak magnetic signals and providing information on a correlation length of magnetic order, respectively.
Results
We measured the magnetic susceptibility in the temperature range of 2–400 K with magnetic field applied along the hexagonal basal plane using several samples from different batches (see Fig. 1a, b). Diamagnetic signal dominates the data from high temperatures down to ~50 K, except for intermediate interruptions by two jumps at T _{CDW} and T _{nCDW}, which signal a transition from the commensurate to the nearly commensurate phase and then to another incommensurate phase, respectively. While the transition to the incommensurate phase takes place at T _{nCDW} ~355 K, the transition at T _{CDW} shows a remarkable hysteresis as shown in the inset of Fig. 1b and reaches values of ~220 K upon warming and ~160 K upon cooling. Interestingly, all the χ(T) curves exhibit a strong paramagnetic Curie–Weiss tail below ~50 K, suggesting that some kind of a shortrange correlation might exist at low temperatures. Magnetic impurities were excluded as a source of the lowtemperature Curie contribution, considering the fact that we consistently observed a comparable lowtemperature tail of almost similar magnitude for all the five samples prepared from raw materials of different purity. We also note that the Curie constant is too high to be explained purely by an extrinsic effect.^{20} Our estimate of ~8 μ_{B}/f.u. for the impurity moments is a too big a value if we adopt the impurity based explanation. All the known structure phases of the 1TTaS_{2} compound are summarized in Table 1.
The magnetic susceptibility was fitted using the following equation with a diamagnetic/Pauli paramagnetic contribution χ _{0} and a Curietype paramagnetic contribution in the temperature region from 2 to 200 K;
where C is the Curie constant and θ is the paramagnetic temperature. The temperature dependence of the inverse magnetic susceptibility and the respective fits are shown in Fig. 1d. Parameters of the Curie–Weiss fit are also summarized in Table S1 in the Supplemental Material. The single crystals #1 and #2 are not diamagnetic down to ~50 K, probably due to a small presence of the 2Hpolymorph in the samples (discussed in the Supplemental Material) which reveals itself by a tiny bump at ~75 K corresponding to the CDW transition^{21} (marked by the arrow in Fig. 1b). We note that the size of the effective magnetic moment and the Curie–Weiss temperature (θ _{CW}) is similar for all the measured samples with μ _{eff} ~ 0.08 μ_{B}/f.u. and θ _{CW} ~ 0.02 K, respectively. Therefore, we observe ~50% reduction of the observed μ _{eff} value from the expected spinonly value of ~0.13 μ_{B} for one magnetic moment per each David star. On the other hand, the small value of θ indicates weak magnetic correlations present at low temperatures.
In order to investigate the thermodynamic nature of the transitions, we also measured the specific heat. The specific heat collected upon cooling from 400 down to 0.7 K is shown in Fig. 1a in comparison to the magnetic susceptibility data. In accordance with the magnetic susceptibility, the specific heat reveals a typical firstorder transition from the incommensurate to the nearly commensurate phase at T _{nCDW} ~355 K. However, we failed to observe any lambdalike peak signifying the firstorder transition into the commensurate phase in contrast to the behavior of χ(T) measured on the identical sample. Instead, the C _{p}/T (T) curve shows a moderate decrease followed by a clear change of the slope at T _{CDW} ~150 K. Similar discontinuous decrease/increase (depending on the rise/decrease of temperature) of the specific heat in the vicinity of T _{CDW} was observed in the case of incommensuratetocommensurate transition in 2 HTaSe_{2}.^{22} Moreover, a discontinuity in the specific heat of 1TTaS_{2} was predicted by the calculation based on the Born von Karman model.^{23} Upon further cooling, the C _{p}/T (T) curve reveals a sharp drop at ~50 K, which may be related to the upturn in the magnetic susceptibility at a similar temperature range.
The hightemperature specific heat data shown in Fig. 1c were modeled by the Debye equation. The phonon contribution can be written as
Our analysis produces the following parameters for the best fit: the Debye temperature of θ _{D} ~450 K, the Sommerfeld coefficient of γ = 1.84 mJ mol^{−1} K^{−2}, and the coefficient ß = 0.31 mJ mol^{−1} K^{−4} as shown in the upper inset of Fig. 1c. We note that these values correspond rather well to those obtained from the previous experiment carried on powder sample.^{24} Interestingly enough, the value of γ is close to the theoretical value of 3.56 mJ mol^{−1} K^{−2} from DFT calculations.^{25} We note that C _{p}(T) shows a modest increase below ~1 K, indicating some kind of residual entropy still remaining at very low temperatures.
After subtracting off the estimated lattice and electron contributions, we can get a broad hump centered at ~50 K in the magnetic heat capacity C _{mag}, indicative of some kind of shortranged order. Integrating the peak area yields an entropy released by this correlated phase, which is ≈40% of the theoretical maximum entropy Rln(2) = 5.76 J mol^{−1} K^{−1}.
When taken together, the analysis of the Curie–Weiss fits, the lowtemperature specific heat and the magnetic susceptibility provided clear evidence in support of the presence of a lowtemperature correlated phase. In order to get a more direct insight into the lowtemperature ground state of 1TTaS_{2}, we performed neutron scattering and muon spin relaxation experiments. Figure 2 shows the magnetic scattering data (dσ/dΩ)_{mag} of the 1TTaS_{2} compound. Although the signals are relatively weak, closer inspection of the data reveals a bump emerging in the Q–range from 0.7 to 1.2 Å^{−1} (for more information about the signal quality, see the Supplemental Material). To emphasize the possible presence of a broad peak marked by the red arrow in the inset of Fig. 2, the original data were shifted upwards by a factor of 0.0075 barn st^{−1} f.u.^{−1}.
We can model the data by the expression for the magnetic neutronscattering differential cross section,
which is valid in the case of a disordered magnetic configuration. The background term \({\left( {\frac{{d\sigma }}{{d\Omega }}} \right)_{bkg}}\) = 0.0075 barn st^{−1} f.u.^{−1} was added into the Eq. (3) to compensate the systematic errors which play an important role in our data with the low signaltonoise ratio. Details are discussed in the Supplemental Material. Using the formula, we can calculate the average spinspin correlation functions \(\left\langle {{S_0}.{S_n}} \right\rangle \) for the n ^{th} nearest neighbor radial shell around a central Ta atom at the origin. Here, γ _{n} is the neutron gyromagnetic ratio, r _{0} is the classical electron radius and f(Q) is the magnetic form factor of the magnetic species used considering Ta^{4+} in our case,^{26} g ^{2} S(S + 1) is the squared magnetic moment and Z _{n} and R _{n} are the coordination number and radial distance of the nth nearestneighbor shell, respectively. For a paramagnetic system without any spinspin correlations, the Qdependence of magnetic cross section corresponds straightforwardly to the evolution of the absolute squared magnetic moment per atom.
To define reasonable limits for the fitting parameters, we compared our data to those measured on the Lu_{2}Mo_{2}O_{7} oxide and the Lu_{2}Mo_{2}O_{5}N_{2} oxynitride.^{27} We recall that in our orphan spin scenario the spin moment is expected to be localized in the respective David star consistent with what one would expect in the lowtemperature CDW phase. The bump in the oxide data taken after ref. 27 was previously modeled by taking into account the nearest and the nextnearestneighbor correlations \(\left\langle {{S_0} \cdot {S_1}} \right\rangle \) = −0.029 and \(\left\langle {{S_0} \cdot {S_2}} \right\rangle \) = −0.056, respectively, proving the presence of the static shortrange molybdenum spin correlations (green dashed line in Fig. 2). On the other hand, the oxynitride data follow the f ^{2}(Q) dependence with zero \(\left\langle {{S_0} \cdot {S_i}} \right\rangle \) correlations (red dashed line in Fig. 2). Similar fit of our neutron data on 1T TaS_{2} produces the effective magnetic moment of 0.11 µ_{B}/f.u., representing an upper bound for the magnetic moment of the orphan spins.
As shown in ref. 28, double layers are connected by the cstacking vector across the hexagonal sheets. As a result, the lowtemperature phase is commensurate only in the basal plane and not along the caxis because the stacking order established along the caxis is not periodic. Related to this issue, two types of stacking patterns–screw and repetitive–have been previously observed in the ratio ~1:2.7 for 1TTaS_{2}. Due to the large spatial extent of the David star, the nearest neighbors that are expected to contribute to the magnetic cross section are located across the hexagonal layers and not within them. Indeed, assuming that the crucial correlations take place only within the CDW layer, the fit cannot describe satisfactorily the data (shown in Fig. S2 in the Supplemental Material). Therefore, the threedimensional character of the electronic correlations at low temperatures has to be considered. This result is indeed in good agreement with the recent experimental and theoretical studies, which suggested the importance of interlayer coupling.^{14, 29,30,31} In order to describe the broad peak well by the expression (3), we had to include interactions among the nearest (r _{1} = 5.90 Å, Z _{1} = 1) and the nextnearestneighbor (r _{2} = 8.29 Å, Z _{2} = 1) in the repetitive stacking and the nextnearestneighbor (\(r_2^\prime \) = 6.79 Å, Z’ _{2} = 1) in the screw stacking with respect to their ratio, as depicted in Fig. 3.
The best fit to the data by the Eq. (3) was obtained for the effective magnetic moment of 0.4 ± 0.05 μ_{B} (≈ 300% of the expected magnetic moment per each David star) and for the average correlations \(\left\langle {{S_0} \cdot {S_1}} \right\rangle \) = −0.07 ± 0.05, \(\left\langle {{S_0} \cdot {S_2}} \right\rangle \) = 0.17 ± 0.10, \({\left\langle {{S_0} \cdot {S_2}} \right\rangle ^\prime }\) = 0.39 ± 0.16, respectively (solid blue line in Fig. 2). The solid magenta line represents the fit without the correction on the background term. The magnetic moment is higher compared to the expected µ ~ 0.13 µ_{B} value of the magnetic moment per each David star; however, it is necessary to keep in mind that the values of the fitted parameters represent the upper limit. Also, the values and error bars of the average correlations need to be understood in terms of very low signaltonoise ratio. The dominant correlation is weakly antiferromagnetic, while the secondnearest neighbor correlations are more robust and ferromagnetic; further correlations among more distant Davidstar patterns can be neglected. Here we note that there are considerable nearest and nextnearest \(\left\langle {{S_0} \cdot {S_i}} \right\rangle \) correlations with opposite signs. Interestingly, the ferromagnetic nature of \(\left\langle {{S_0} \cdot {S_i}} \right\rangle \) for i = 2 corresponds to the positive value of the Curie–Weiss temperatures θ _{CW} obtained from the magnetization measurement, hinting at subtle ferromagnetic correlations.
In order to study further the shortrange magnetic order, we carried out the zerofield (ZF) muonspin relaxation study from 5 up to 250 K. The ZF spectra can be modeled using the following expressions:
where a _{i} and a _{bg} are the initial and background asymmetries, respectively. Eq. (5) describes the static Gaussian Kubo–Toyabe function, whereas σ is the nuclear depolarization rate. G _{mag}(t) from Eq. (6) is the magnetic part of the relaxation function, which represents the muonspin relaxation rate λ from the dynamic magnetic fields associated with fluctuating atomic spins. Consistent with the neutron experiment, we observe a sign of the presence of a lowtemperature shortranged magnetic ordering in the asymmetry spectra compared to high temperatures as shown in Fig. 4a, b. The asymmetry spectra can be characterized by the typical Kubo–Toyabe evolution and no pronounced qualitative differences among the spectra are observed for various temperatures (the spectra measured at 5 and 250 K are shown for comparison).
To suppress the nuclear contribution to the muon asymmetry function, small LF field of 5 mT was applied in a direction parallel to the muon polarization; which is sufficient to fully decouple the muons from the relaxation channel. The effect of the applied longitudinal field also indicates that the magnetism is dynamic in the investigated temperature range (see Fig. S3 in the Supplemental Material). Fig. 4c–e summarizes the results of the Kubo–Toyabe fit over the whole temperature range for the fitting parameters a _{i}, λ and σ. The ZF run of the initial asymmetry parameter shown in Fig. 4c remains temperatureindependent down to ~50 K, where the asymmetry starts to decrease slightly and at the lowest measured temperature of 5 K. This change points to the onset of a shortrange magnetic order. The relaxation rate λ in Fig. 4d shows a slight increase below ~50 K, implying that this extra relaxation might indicate additional internal magnetic fields. This result by itself is not conclusive as the effect is very weak, in agreement with the recent µSR measurements performed by Klanjšek et al.^{19}; however, together with the bulk and neutron data it forms a comprehensive picture about the lowtemperature state in 1TTaS_{2}. The depolarization rate σ in Fig. 4e reveals an interesting temperature evolution. Upon warming, it starts to deviate from the constant value at ~100 K, which is probably a sign of the muons diffusion, and a minimum of σ is observed at ~210 K. This value coincides well with the temperature T _{CDW} ~210 K (upon warming) of the commensurate CDW transition obtained from the magnetic susceptibility measurement.
Discussion
With those data from the specific heat and the magnetic susceptibility, we can calculate the SommerfeldWilson ratio. The ratio is given as
With χ _{0} = 5·10^{−4} emu mol^{−1} as obtained at the lowest measured temperature T = 1.8 K and for γ = 1.84 mJ mol^{−1} K^{−2} using the identical sample #4, we get R _{W} ~2 (R _{W} = 1 for the free electron gas) indicating the presence of considerable electronic correlations at low temperatures.
Another remarkable feature in χ(T) is the absolute size of the jumps Δχ _{1} and Δχ _{2} that is consistently seen for all the measured samples (see Table S1 in the Supplemental Material) although the absolute values of χ(T) are strongly sampledependent due to different susceptibility term χ _{0}. From the ratio of the jumps it is possible for us to estimate the relative ratio of the CDWinduced energy gaps as
where ΔN is the electron density of states. The early tunneling experiments on 1TTaS_{2} ^{32, 33} were carried out to investigate the size of the charge gaps directly and reported somewhat a larger ratio of ~2.5 for charge gap of ~10 eV at T _{CDW} and 0.4 eV at T _{nCDW}. Later, it was shown that the gap opening at the Fermi level is related to the Mott localization in the commensurate CDW phase.^{34} Turning to the question on the lowtemperature phase, both the magnetic susceptibility and the specific heat suggest the presence of a correlated phase; another support for this scenario can be found in the neutron data that show a clear bump at a finite Q value. The position of the bump in the diffuse magnetic scattering of 1TTaS_{2} is Q ~ 0.9 Å^{−1}, close to 2π/\(r_2^\prime \), where \(r_2^\prime \)= 6.79 Å is the next nearestneighbor distance across the screw stacking layers. As the respective correlation for this neighbor distance is positive, the Qposition implies ferromagnetic correlations between pairs of spins across the hexagonal layers. The correlation length extracted from the full width at half maximum (~0.285 ± 0.060 Å^{−1}) of the bump (see the Lorentz fit shown in the inset of Fig. 2) is L ~22 Å, which roughly corresponds to 4r _{1}. Thus, we conclude that correlated domains show a large spatial extent almost over two Davidstar patterns and across four hexagonal sheets.
Finally, the temperature evolution of the parameters a _{i} and λ obtained by µSR measurement reveals a change of slope below ~50 K. This temperature scale seems to have a physical significance as it corresponds to the abrupt changes in the temperature evolution of specific heat and magnetic susceptibility. Interestingly, a spontaneous ordering into the hidden state in 1TTaS_{2} was observed below 100 K after an application of the sub35fs laser pulse under nonequilibrium conditions.^{35} Our µSR experiment shows a subtle change in the temperature evolution of the relaxation rate and initial asymmetry value below 50 K, which can be taken as a signature of a lowtemperature correlated phase in accordance with the neutron and bulk measurements.
To sum up, we have performed complementary specific heat, magnetization, neutron diffraction, and µSR measurements of 1TTaS_{2} to study the nature of the lowtemperature hidden order within the commensurate CDW phase. All the results presented here consistently offer evidences of the “orphan spin” scenario with S = ½ moments localized in the centers of the David stars, albeit with a somewhat higher value of the effective magnetic moment as compared to the proposed ~0.13 µ_{B} per David star. Our results support also the important role of the interlayer correlations in the electronic structure of 1TTaS_{2} in line with theoretical studies.^{30}
Methods
Polycrystals and single crystals of 1TTaS_{2} was prepared by the solidstate hightemperature reaction of initial elements and by the chemical vapor transport method, respectively (for details of the synthesis, see the Supplemental Material). The samples were characterized by Xray diffraction (Bruker XRD D8 Discover for powder and Rigaku XtaLAB P200 for single crystals) and confirmed to have the trigonal crystal structure crystallizing in the P3m1 space group (see Fig. S1 in the Supplemental Material). The magnetization and heat capacity measurements were carried out using commercial setups: MPMS3 EverCool (Quantum Design, USA) and PPMS9 equipped with the lowtemperature He3 option (Quantum Design, USA).
Polarized diffuse neutron scattering was performed on D7 at Institut LaueLangevin using 4.2 g of sample placed in a doublewall Al cylinder can and incident energy E _{i} = 3.5 meV (λ = 4.8 Å). The temperature was kept at 1.5 K during the whole experimental time. The instrument background was estimated by adding the scattering from an empty and a cadmiumfilled sample holder, weighted by the sample transmission. This was subtracted from the data. The data were then subsequently corrected for polarization efficiency using the scattering from amorphous quartz. Vanadium was measured to estimate detector efficiency and to put the data on an absolute scale. Equal time was spent on measuring the scattering along the X, Y, and Z directions. The nuclear, magnetic, and nuclear spinincoherent scattering cross sections were separated as a function of both momentum and energy transfer using the 6point ‘xyz’polarization analysis.^{36} The nonspinflip and spinflip scattering along each direction was measured with a time ratio of 1:8.
µSR spectra were measured using the EMU spectrometer at the ISIS Facility, Rutherford Appleton Laboratory, U.K. in the temperature range from 5 to 250 K using a Close Cycle Refrigerator. The sample of the mass of ~20 g was enclosed in an Al sample holder. Any exposed Al to the beam was covered by a silver mask (dia. 30 mm) with a 30 μm mylar window for the muons to pass through before entering the sample. The decaypositron asymmetry function was measured as a function of time, in the zerofield (ZF) regime to examine the unperturbed magnetic behavior and in the longitudinal field (LF) of 5 mT.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
Sipos, B. et al. From Mott state to superconductivity in 1TTaS_{2}. Nat. Mat. 7, 960–965 (2008).
Wilson, J. A. et al. Chargedensity waves and superlattices in the metallic layered transition metal dichalcogenides. Adv. Phys 24, 117–201 (1975).
Spijkerman, A. et al. Xray crystalstructure refinement of the nearly commensurate phase of 1T−TaS_{2} in (3 + 2)dimensional superspace. Phys. Rev. B 56, 13757–13767 (1997).
Xu, P. et al. Superconducting phase in the layered dichalcogenide 1TTaS_{2} upon inhibition of the metalinsulator transition. Phys. Rev. B 81, 172503 (2010).
Ang, R. et al. RealSpace Coexistence of the Melted Mott State and Superconductivity in FeSubstituted 1TTaS_{2}. Phys. Rev. Lett. 109, 176403 (2012).
Fazekas, P. & Tosatti, E. Electrical, structural and magnetic properties of pure and doped 1TTaS_{2}. Philos. Mag. B 39, 229–244 (1979).
Rossnagel, K. & Smith, N. V. Spinorbit coupling in the band structure of reconstructed 1T−TaS_{2}. Phys. Rev. B 73, 073106 (2006).
Gan, L.Y. et al. Strain tuning of the charge density wave in monolayer and bilayer 1TTaS_{2}. Phys. Chem. Chem. Phys. 18, 3080–3085 (2016).
Cho, H. et al. Properties of spin1/2 triangularlattice antiferromagnets CuY_{2}Ge_{2}O_{8} and CuLa_{2}Ge_{2}O_{8}. Phys. Rev. B 95, 144404 (2017).
Koutroulakis, G. et al. Quantum phase diagram of the S = 1/2 triangularlattice antiferromagnet Ba_{3}CoSb_{2}O_{9}. Phys. Rev. B 91, 024410 (2015).
Rawl, R. et al. Ba_{8}CoNb_{6}O_{24}: A spin1/2 triangularlattice Heisenberg antiferromagnet in the twodimensional limit. Phys. Rev. B 95, 060412(R) (2017).
Zvyagin, S. A. et al. Magnetic excitations in the spin1/2 triangularlattice antiferromagnet Cs_{2}CuBr_{4}. New. J. Phys. 17, 113059 (2015).
Zhu, X. et al. Classification of charge density waves based on their nature. PNAS 112, 2367–2371 (2015).
Bovet, M. et al. Interplane coupling in the quasitwodimensional 1T−TaS_{2}. Phys. Rev. B 67, 125105 (2003).
Yoshida, M. et al. Memristive phase switching in twodimensional 1TTaS_{2} crystals. Sci. Adv 1, 1–6 (2015).
Ritschel, T. et al. Orbital textures and charge density waves in transition metal dichalcogenides. Nat. Phys 11, 328–331 (2015).
Cho, D. et al. Interplay of electronelectron and electronphonon interactions in the lowtemperature phase of 1TTaS_{2}. Nat. Comm 7, 10453 (2015).
Law, K. T. & Lee, P. A. Is 1TTaS_{2} a quantum spin liquid? Preprint at https://arxiv.org/abs/1704.06157 (2017).
Klanjsek, M. et al. A new hightemperature quantum spin liquid with polaron spins. Preprint at https://arxiv.org/abs/1704.06450 (2017).
DiSalvo, F. J. et al. Paramagnetic moments and localization in 1TTaS_{2}. Phys. Rev. B 22, 4241–4246 (1980).
Tidman, J. P. et al. The phase transition in 2HTaS_{2} at 75 K. Phil. Mag. 30, 1191–1194 (1974).
Craven, R. A. & Meyer, S. F. Specific heat and resistivity near the chargedensitywave phase transitions in 2H−TaSe_{2} and 2H−TaS_{2}. Phys. Rev. B 16, 4583–4593 (1977).
Balaguru Rayappan, J. B. et al. Thermal properties of 1TTaS_{2} at the onset of charge density wave states. Phys. B 405, 3172–3175 (2010).
Benda, J. A. Optical, electricaltransport, and heatcapacity studies of solidsolutions Ti_{ x }Ta_{1−x }S_{2}, Zr_{ x }Ta_{1−x }S_{2}, and Ti_{ x }Nb_{1−x }Se_{2}. Phys. Rev. B 10, 1409–1420 (1974).
YanBin, Q. et al. Anisotropic properties of TaS_{2}. Chin. Phys. 16, 3809–3814 (2007).
Kobayashi, K. et al. Radial integrals for the magnetic form factor of 5d transition elements. Acta Cryst. A 67, 473–480 (2011).
Clark, L. et al. From spin glass to quantum spin Liquid ground states in molybdate pyrochlores. Phys. Rev. Lett. 113, 117201 (2014).
Ishiguro, T. & Sato, H. Electron microscopy of phase transformations in 1TTaS_{2}. Phys. Rev. B 44, 2046–2060 (1991).
Svetin, D. et al. Threedimensional resistivity switching between correlated electronic states in 1TTaS_{2}. Preprint at https://arxiv.org/abs/1603.06214 (2016).
Darancet, P. et al. Threedimensional metallic and twodimensional insulating behavior in octahedral tantalum dichalcogenides. Phys. Rev. B 90, 045134 (2014).
Ma, L. et al. A metallic mosaic phase and the origin of Mottinsulating state in 1TTaS_{2}. Nat. Comm 7, 10956 (2016).
Ozaki, H. et al. Tunneling investigation of charge density wave energy gap in ITTaS_{2}. Phys. B 117118, 590–592 (1983).
Noutomi, S. et al. Tunneling study of the chargedensitywave state in 1TTaS_{2} and 1TTaSe_{2}. Solid State Commun. 50, 181–184 (1984).
Kim, J.J. et al. Observation of Mott Localization Gap Using Low Temperature Scanning Tunneling Spectroscopy in Commensurate 1TTaS_{2}. Phys. Rev. Lett. 73, 2103 (1994).
Stojchevska, L. et al. Ultrafast switching to a stable hidden quantum state in an electronic crystal. Science 344, 177–180 (2014).
Stewart, J. R. et al. Disordered materials studied using neutron polarization analysis on the multidetector spectrometer, D7. J. Appl. Cryst. 42, 69–84 (2009).
Acknowledgements
We acknowledge Y.W. Son for useful comments. This work was supported by the Institute for Basic Science (IBS) in Korea (IBSR009G1). The experiments were partially performed in MGML (https://mgml.eu/) through the program of Czech Research Infrastructures (LM2011025). We acknowledge the support and beam time at Institut Laue Langevin and ISIS Neutron and Muon Source, in providing neutron and muon research facilities used in this work. The work at Rutgers was funded by the Gordon and Betty Moore Foundation’s EPiQS Initiative through Grant GBMF4413 to the Rutgers Center for Emergent Materials.
Author information
Authors and Affiliations
Contributions
J.G.P. conceived the idea and supervised the project. M.K. carried out all the experiments: the neutron scattering experiments were done together with A.W. while muon experiments were done with A.H. L.W. and S.W.C. prepared the sample. M.K. and J.G.P. wrote the manuscript with contributions from all the other authors.
Corresponding authors
Ethics declarations
Competing interests
The authors declare that they have no competing financial interests.
Additional information
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Kratochvilova, M., Hillier, A.D., Wildes, A.R. et al. The lowtemperature highly correlated quantum phase in the chargedensitywave 1TTaS_{2} compound. npj Quant Mater 2, 42 (2017). https://doi.org/10.1038/s4153501700481
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1038/s4153501700481
This article is cited by

Dualistic insulator states in 1TTaS2 crystals
Nature Communications (2024)

Twocomponent nematic superconductivity in 4HbTaS2
Nature Communications (2024)

Stacking and spin order in a van der Waals Mott insulator 1TTaS2
Communications Materials (2023)

Strong correlations in twodimensional transition metal dichalcogenides
Science China Physics, Mechanics & Astronomy (2023)

Visualizing the evolution from Mott insulator to Anderson insulator in Tidoped 1TTaS2
npj Quantum Materials (2022)