⇦Chinese Physics Letters, 2020, Vol. 37, No. 4, Article code 047102Express Letter Pressure-Induced Metallization and Structural Phase Transition in the Quasi-One-Dimensional TlFeSe$_{2}$ * Zi-Yi Liu (刘子儀)1,2†, Qing-Xin Dong (董庆新)1,3†, Peng-Fei Shan (单鹏飞)1,3†, Yi-Yan Wang (王义炎)1,3, Jian-Hong Dai (戴建洪)1,3, Rajesh Jana1,3, Ke-Yu Chen (陈科宇)1,3, Jian-Ping Sun (孙建平)1,3, Bo-Sen Wang (王铂森)1,3,4, Xiao-Hui Yu (于晓辉)1,3,4, Guang-Tong Liu (刘广同)1,3,4, Yoshiya Uwatoko5, Yu Sui (隋郁)2, Huai-Xin Yang (杨槐馨)1,3,4, Gen-Fu Chen (陈根富)1,3,4**, Jin-Guang Cheng (程金光)1,3,4** Affiliations 1Beijing National Laboratory for Condensed Matter Physics and Institute of Physics, Chinese Academy of Sciences, Beijing 100190 2School of Physics, Harbin Institute of Technology, Harbin 150001 3School of Physical Sciences, University of Chinese Academy of Sciences, Beijing 100190 4Songshan Lake Materials Laboratory, Dongguan 523808 5Institute for Solid State Physics, University of Tokyo, Kashiwa, Chiba 277-8581, Japan Received 6 March 2020, online 13 March 2020 *Supported by the National Key R&D Program of China (2018YFA0305700), the National Natural Science Foundation of China (11904391, 11834016, 11874400, 11888101, 11921004), the Beijing Natural Science Foundation (Z190008), the Strategic Priority Research Program and the Key Research Program of Frontier Sciences of the Chinese Academy of Sciences (XDB25000000, QYZDB-SSW-SLH013), and the CAS Interdisciplinary Innovation Team (JCTD-2019-01). JPS acknowledges the support from the China Postdoctoral Science Foundation and the Postdoctoral Innovative Talent Program.
†These authors contributed equally to this work.
**Corresponding authors. Email: gfchen@iphy.ac.cn; jgcheng@iphy.ac.cn Citation Text: Liu Z Y, Dong Q X, Dan P F, Wang Y Y and Dai J H et al 2020 Chin. Phys. Lett. 37 047102    Abstract We report a comprehensive high-pressure study on the monoclinic TlFeSe$_{2}$ single crystal, which is an antiferromagnetic insulator with quasi-one-dimensional crystal structure at ambient pressure. It is found that TlFeSe$_{2}$ undergoes a pressure-induced structural transformation from the monoclinic phase to an orthorhombic structure above $P_{\rm c} \approx 13$ GPa, accompanied with a large volume collapse of $\Delta V/V_{0}=8.3{\%}$. In the low-pressure monoclinic phase, the insulating state is easily metallized at pressures above 2 GPa; while possible superconductivity with $T_{\rm c}^{\rm onset} \sim 2$ K is found to emerge above 30 GPa in the high-pressure phase. Such a great tunability of TlFeSe$_{2}$ under pressure indicates that the ternary $A$FeSe$_{2}$ system ($A$ = Tl, K, Cs, Rb) should be taken as an important platform for explorations of interesting phenomena such as insulator-metal transition, dimensionality crossover, and superconductivity. DOI:10.1088/0256-307X/37/4/047102 PACS:71.27.+a, 71.30.+h, 74.62.Fj, 74.72.Cj © 2020 Chinese Physics Society Article Text Since the discovery of high-$T_{\rm c}$ superconductivity in cuprates,[1] great efforts have been devoted to find new superconducting materials with high critical temperature and/or unconventional paring mechanism. The iron-based superconductors discovered in 2008 represent one of the celebrated examples in this field.[2–4] Although the microscopic mechanism of these two classes of unconventional superconductors remains elusive,[5] several rational designing principles and some candidate materials have been put forward to facilitate the discovery of new high-$T_{\rm c}$ superconductors.[6–9] Based on the first-principles density functional calculations, Zhao and coworkers[10] recently proposed that the ternary $A$FeSe$_{2}$ ($A$ = Tl, K, Rb, or Cs) with tetragonal $I\bar{4}m2$ symmetry are promising parent compounds of high-$T_{\rm c}$ superconductors. In their calculations, the structural, magnetic, and electronic properties of tetragonal $A$FeSe$_{2}$ were found to be quite similar to the parent compound of cuprate superconductors. In particular, these compounds are featured by an antiferromagnetically ordered ground state. It was further suggested that high-$T_{\rm c}$ superconductivity can be achieved in the $A$FeSe$_{2}$ via electron- or hole-doping. Motivated by this theoretical study, we are interested in the ternary $A$FeSe$_{2}$ and related compounds. A literature survey shown that the layered tetragonal structure used in the above theoretical study[10] is thermodynamically unstable for these $A$FeSe$_{2}$ compounds, which actually crystallize in the quasi-one-dimensional (Q1D) monoclinic structures with space group $C2/c$ for $A$ = K and Rb[11] and $C2/m$ for $A$ = Tl,[12] respectively. In the present study, we focus on TlFeSe$_{2}$ because it is the only one among these $A$FeSe$_{2}$ compounds that has been reported to adopt the layered tetragonal structure[13] in addition to the stable monoclinic structure. The physical properties of the tetragonal TlFeSe$_{2}$ remain largely unknown to date presumably due to the difficulty in stabilizing this structure. In contrast, most previous studies on TlFeSe$_{2}$ were concentrated on the monoclinic phase, which exhibits interesting structural and physical properties.[14–17] As shown in the inset of Fig. 1, the crystal structure of monoclinic TlFeSe$_{2}$ consists of infinite chains of edge-shared FeSe$_{4}$ tetrahedra along the $b$ axis, and the lattice parameters were reported to be $a=12.02$ Å, $b= 5.50$ Å, $c=7.13$ Å, and $\beta=118.52 ^{\circ} $.[14] It was reported that the monoclinic TlFeSe$_{2}$ undergoes a paramagnetic-to-antiferromagnetic transition at $T_{\rm N} \approx 290$ K, below which the magnetic moments of Fe ions form an antiparallel arrangement along the $b$ axis. The ordered magnetic moment of Fe is 2.1$\mu_{_{\rm B}}$ at 10 K as determined from neutron diffraction.[12] However, there exist some controversies regarding the nature of the electronic state in the Q1D TlFeSe$_{2}$. Experimentally, the resistivity of TlFeSe$_{2}$ exhibits an insulating behavior below room temperature.[18] This is consistent with a theoretical study by Luo et al.,[19] who found a robust gap in the density of states along the chain direction with staggered spin order. On the other hand, through detailed analysis of the magnetic properties, Seidov et al.[15] argued that TlFeSe$_{2}$ should possess a 1D metallic behavior and the experimentally observed insulating behavior should originate from the weak interchain interactions and/or extrinsic factors such as mechanical breaks or defects. In the course of our study on TlFeSe$_{2}$, we indeed confirm that the tetragonal phase is hard to synthesize but needle-like single crystals with the monoclinic structure can be readily obtained with the Bridgman method. As a Q1D material, the magnetic and transport properties of monoclinic TlFeSe$_{2}$ should be sensitive to interchain and intrachain interactions, which can be effectively tuned by applying external pressure.[20,21] In this work, we thus perform a comprehensive high-pressure (HP) study on the monoclinic TlFeSe$_{2}$ single crystal. We find that TlFeSe$_{2}$ undergoes a pressure-induced structural transformation from the monoclinic structure to a new orthorhombic structure above $P_{\rm c} \approx 13$ GPa, accompanying a large volume collapse of $\Delta V/V_{0}=8.3{\%}$. In the low-pressure monoclinic $C2/m$ phase, the insulating state can be easily converted into a metallic state at $P > 2$ GPa; while in the HP phase, possible superconductivity with $T_{\rm c}^{\rm onset} \sim 2$ K is found to emerge at $P > 30$ GPa. Our results demonstrate that the structural and physical properties of monoclinic TlFeSe$_{2}$ are greatly tunable so that the ternary $A$FeSe$_{2}$ should be regarded as an ideal platform for exploring more interesting physical phenomena.
cpl-37-4-047102-fig1.png
Fig. 1. Characterizations of TlFeSe$_{2}$ at ambient pressure. (a) Rietveld analysis of the powder XRD pattern of TlFeSe$_{2}$. A minor secondary phase TlSe is marked by the asterisk. The inset shows the picture of needle-shaped TlFeSe$_{2}$ single crystals and the schematic view of the crystal structural along the $b$ axis, highlighting the edge-shared FeSe$_{4}$ tetrahedra along the $b$ direction. (b) Temperature-dependent magnetic susceptibility $\chi (T)$ of TlFeSe$_{2}$ measured in zero-field-cooled mode under an external magnetic field of 0.1 T applied parallel and perpendicular to the chain direction ($b$-axis). (c) Temperature-dependent resistivity $\rho (T)$ and its temperature derivative $d\log \rho /dT$ along the chain direction under 0 T and 7 T.
Single crystals of TlFeSe$_{2}$ were grown by the Bridgman method. The starting materials, Tl, Fe and Se, were mixed in a ratio of 1 : 1 : 2 and put into a quartz tube. The tube was transferred from the colder zone (550$^{\circ}\!$C) to the hotter zone (700$^{\circ}\!$C) at a rate of 500 nm/s. After holding for 20 hours, the tube was cooled to room temperature slowly (10 K/h), and then needle-like single crystals can be obtained. The chemical composition of the resultant crystals was checked with Oxford X-Max energy dispersive x-ray (EDX) spectroscopy analysis in a Hitachi S-4800 scanning electron microscope. The resistivity and magnetic properties at ambient pressure were measured with a Quantum Design physical property measurement system (PPMS-9T) and a magnetic property measurement system (MPMS-7T), respectively. We have measured temperature-dependent resistivity of TlFeSe$_{2}$ along the $b$-axis over a wide pressure range with different HP techniques, i.e., the palm cubic anvil cell (CAC) from 2 to 12 GPa,[22] and the diamond anvil cell (DAC) from 15 to 45 GPa. For measurements in CAC, we first evaporated four gold pads on the sample surface and then attached 20-µm-diameter gold wires on the pads with silver paste so as to improve the electrical contact. Glycerol was used as the pressure transmitting medium (PTM). The pressure in CAC was estimated from the force-pressure calibration curve at room temperature. Details about the sample assembly and pressure calibration of CAC can be found elsewhere.[22] We have used DAC made of nonmagnetic CuBe alloy for resistance measurements up to 45.2 GPa by using the diamond culet of 300 µm in diameter. A plate of Rhenium covered with c-BN powder was used as gasket, which was first pre-indented to $\sim 30$ µm in thickness and then a hole of 130 µm in diameter drilled. Here c-BN powder was pressed into the gasket hole and further drilled a hole of 100 µm in diameter serving as the sample chamber. Then, KBr was loaded into the chamber as PTM, on which the TlFeSe$_{2}$ sample with dimensions of $\sim$$90 \times 30 \times 20\,µ$m$^{3}$ and a ruby chip were placed. Four pieces of thin platinum foil were utilized as the electrical contacts. The pressure in DAC was calibrated at room temperature using the ruby florescent method. All temperature-dependent measurements in DAC were performed in a liquid-helium cryostat with a slow cooling rate of $\sim $0.5 K/min in order to avoid cracks and to ensure a good thermal equilibrium. The HP angular dispersive x-ray diffraction (AD-XRD) experiments were conducted on TlFeSe$_{2}$ powder samples at room temperature in the 4W2 beamline of Beijing Synchrotron Radiation Facility (BSRF) using monochromatic x-ray radiation of wavelength $\lambda=0.6199$ Å. High pressure was generated using a symmetric DAC with Boehler-type diamond culet of 300 µm. A T301 stainless steel gasket was first pre-indented to a thickness of 30 µm and then a 100 µm-diameter hole was drilled as the sample chamber. The TlFeSe$_{2}$ powder ground from single crystals was prepressed to a pellet with a thickness of approximately 10 µm and then was loaded into the sample chamber with small ruby balls for pressure calibration. Silicon oil was used as the PTM to ensure the quasi-hydrostatic pressure condition. The diffracted patterns were collected on a Pilatus-CCD detector. The CeO$_{2}$ standard was used to calibrate the sample to detector distance and the orientation of the detector. The recorded two-dimensional (2D) images were integrated by using Fit2D program to acquire diffraction intensity versus 2$\theta$ profile. As shown in the inset of Fig. 1(a), the obtained TlFeSe$_{2}$ crystals are black in color with metallic luster and have a needle shape with typical dimensions of $\sim$$ 0.5\times 0.5\times 3$ mm$^{3}$. The longest dimension is directed along the $b$-axis. As checked under the microscope, these crystals actually are composed of a bundle of thin fibers, consistent with the Q1D character of the crystal structure. We first characterized the structural and physical properties of these TlFeSe$_{2}$ crystals at ambient pressure. Figure 1(a) displays the room-temperature powder XRD pattern, which confirms that we have obtained a nearly single-phase sample with a trace amount of TlSe secondary phase as marked by the asterisk. The XRD pattern has been analyzed with the Rietveld method in the monoclinic space group $C2/m$ (No. 14).[14] The refinement converged excellently with small reliability factors as illustrated in Fig. 1(a). The obtained lattice parameters $a=11.977(2)$ Å, $b=5.489(2)$ Å, $c=7.111(1)$ Å and $\beta=118.138(1)^{\circ}$ are consistent with those reported previously.[14,23] Figure 1(b) shows the temperature-dependent magnetic susceptibility $\chi (T)$ of TlFeSe$_{2}$ crystals measured in zero-field-cooled mode under an external magnetic field of $\mu_{0}H=0.1$ T applied parallel and perpendicular to the $b$ axis. As can be seen, the $\chi (T)$ curves along both directions exhibit a clear anomaly around $T_{\rm N} = 290$ K, signaling the occurrence of long-range antiferromagnetic order. The obtained $T_{\rm N}$ agrees well with the result reported previously.[15] It is noteworthy that the $\chi (T)$ curves for $H || b$ and $H \bot b$ are not overlapped with each other in the paramagnetic region at $T > T_{\rm N}$ due to the difference between intrachain and interchain interactions. Seidov et al. also observed such a magnetic anisotropy from the angular dependence of magnetic susceptibility at room temperature.[15] Because the thin-fiber crystals are weekly connected, it is difficult to measure resistivity $\rho (T)$ normal to the $b$-axis. We only measured $\rho (T)$ of TlFeSe$_{2}$ with the current applied along the $b$-axis, as shown in Fig. 1(c), which clearly reveals an insulating behavior. The resistance becomes too large to measure below $\sim $150 K and the application of 7 T external magnetic field has negligible influence on $\rho (T)$. Although no obvious anomaly can be discerned in $\rho (T)$ around $T_{\rm N}$, the $d\log \rho /dT$ curve exhibits a slope change around $T_{\rm N}$, as seen in Fig. 1(c). The above-mentioned characterizations at ambient pressure confirm that we have obtained high-quality TlFeSe$_{2}$ crystals with the Q1D monoclinic structure. It is an antiferromagnetic insulator with $T_{\rm N}=290$ K, exhibiting anisotropic magnetic and transport properties. Before studying the pressure effects on the electrical transport properties of TlFeSe$_{2}$, we first examined its structural stability against compression at room temperature. Asgerov et al.[23] have performed HP neutron diffraction measurements on TlFeSe$_{2}$ and found that the chain-like monoclinic structure with space group $C2/m$ is stable up to 5.2 GPa, the highest pressure in their study. In the present study, we have extended the pressure range to over 20 GPa by performing HP XRD with DAC. The XRD patterns shown in Fig. 2(a) clearly illustrate a pressure-induced structural transformation at $P_{\rm c} \approx 13$ GPa. For $P < P_{\rm c}$, the Q1D monoclinic structure is stable. The lattice parameters extracted from these XRD patterns with the LeBail method are displayed in Fig. 2(b) as a function of pressure. The lattice parameters decrease continuously with increasing pressure, but exhibit anisotropic compression behaviors; i.e., up to 12.76 GPa, the $b$ axis along the Fe–Fe chain direction decreases by $\sim $4.3%, whereas $a$ and $c$ axes decrease by $\sim $8.1% and 6.7%, respectively. Such an anisotropic compression behavior is consistent with the Q1D crystal structure, in which the Fe–Fe bonding along the chain direction ($b$-axis) is stronger. The pressure dependence of volume $V(P)$ in the monoclinic phase, as shown in Fig. 2(c), can be fitted to the third-order Birch–Murnaghan equation of state,[24] which yields a bulk modulus $B_{0}=38.3(3)$ GPa with $B_{1}=4$ fixed. A smaller $B_{0}=27(2)$ GPa was reported by Asgerov et al.[23] with fitting $V(P)$ in the pressure range 0–5.2 GPa. Such a discrepancy should arise from the different fitting pressure ranges.
cpl-37-4-047102-fig2.png
Fig. 2. (a) HP XRD patterns of TlFeSe$_{2}$ under various pressures up to 20.69 GPa. The bottom and middle vertical marks denote the Bragg positions of TlFeSe$_{2}$ in the low-pressure monoclinic structure and the HP orthorhombic structure, respectively. The blue dashed lines indicate the Bragg positions of ruby. The TlSe phase is marked by the asterisk. Pressure dependences of (b) lattice parameters and (c) unit-cell volume. The solid lines in (c) represent the Birch–Murnaghan fitting curves. The fitting results are listed inside the figure.
When the pressure is increased to 14.5 GPa and above, the XRD patterns are totally different from those below $P_{\rm c}$ and most diffraction peaks of the monoclinic phase disappear. This indicates that the crystal structure of TlFeSe$_{2}$ undergoes a significant reconstruction during the structural transformation. Unfortunately, the diffraction peaks in the HP phase become much weaker and broader so that the diffractions from ruby pressure marker dominate the pattern above $P_{\rm c}$ as indicated by the blue dashed lines in Fig. 2(a). Under such a circumstance, it is impossible to determine the crystal structure of HP phase by analyzing the available XRD data. We thus tentatively index the XRD patterns only. According to previous studies, several ternary $M$-III-VI$_{2}$ ($M$ is metal) compounds such as CuInSe$_{2}$ have been reported to undergo pressure-induced structural transformation to high symmetry structure upon compression.[25,26] After many trials, we are finally able to index the XRD patterns above 14.50 GPa with an orthorhombic unit cell with $a \sim 10.4$ Å, $b \sim 4.56$ Å and $c \sim 6.44$ Å, similar with the orthorhombic Tl$_{3}$Fe$_{2}$Se$_{4}$ (Pnma).[27] The pressure dependences of lattice parameters of the HP phase are given in Fig. 2(b). As seen in Fig. 2(c), the pressure-induced structural transformation at $P_{\rm c}$ is accompanied with a large volume collapse $\Delta V/V_{0}=8.3{\%}$. A least square fitting of $V(P)$ to the third-order Birch–Murnaghan equation of state gives a bulk modulus of $B_{0}=67.2$ GPa for the orthorhombic phase, which is about twice the value of the low-pressure monoclinic phase. The dramatic enhancement of bulk modulus indicates that the HP orthorhombic phase may have a three-dimensional crystal structural. The left panel of Fig. 3 shows the $\rho (T)$ of TlFeSe$_{2}$ under various pressures up to 12 GPa measured with a palm CAC apparatus. It is found that the $\rho (T)$ of TlFeSe$_{2}$ is very sensitive to pressure and decreases monotonically with increasing pressure. The resistivity at room temperature drops by four orders of magnitude from $\sim 10^{3}\,$m$\Omega$$\cdot$cm at ambient to $\sim 10^{-1}\,$m$\Omega$$\cdot$cm at 12 GPa, while the resistivity at the lowest temperature are reduced by six orders of magnitude. To see clearly the changes of $\rho (T)$ with increasing pressure, we have shown the data in three figures with different vertical scales in Figs. 3(a)–3(c). As can be seen in Fig. 3(a), the $\rho (T)$ at 2 GPa is totally different from that of 0 GPa; it first shows a metallic behavior in the temperature range from 300 to 175 K, and then experiences a broad minimum centered at $T_{\rm M} \sim 175$ K, below which the resistivity increases gradually followed by a plateau below $T_{\rm p} \sim 50$ K. The similar behavior is observed for the $\rho (T)$ at 4 GPa, except that the $T_{\rm M}$ and $T_{\rm p}$ are reduced to 160 and 30 K, respectively. When the pressure is increased to 6 GPa, as shown in Fig. 3(b), the metallic region is significantly extended down to $T_{\rm M} \sim 75$ K, below which the resistivity upturn is much reduced and a weak hump centered at $\sim $30 K appears. Upon further increasing pressure to 8 GPa and above, Fig. 3(c), all $\rho (T)$ curves exhibit metallic behavior in the whole temperature range. In addition, a broad concave curvature is observed in the high-temperature region, which is typical for correlated metals. The gradual evolution of the metallic behavior can be seen clearly from the normalized $\rho (T)$/$\rho$(290 K) in Fig. 3(d).
cpl-37-4-047102-fig3.png
Fig. 3. (a)–(c) Temperature-dependent resistivity $\rho (T)$ and (d) normalized resistivity $\rho (T)/\rho$(290 K) of TlFeSe$_{2}$ under various pressures from 2 to 12 GPa measured with a palm cubic anvil cell apparatus.
For the metallic phase at $P \ge 8$ GPa, we have tried to analyze the resistivity data at low temperatures with the power-law formula, i.e., $\rho (T)=\rho_{0}+ CT^{n}$, for which $n=2$ means a Fermi-liquid behavior. Interestingly, we observed a non-Fermi-liquid behavior with $n=1.72$ for $P=8$ GPa, where the low-temperature insulating behavior is completely suppressed. Since the Q1D monoclinic crystal structure is retained in this pressure range as confirmed by HP XRD, such a non-Fermi-liquid behavior may be caused by the low-dimensional fluctuations or associated with the suppression of long-range antiferromagnetic order. Further studies are needed to clarify if the antiferromagnetic order can be suppressed by pressure. To investigate the electrical transport properties of the HP orthorhombic phase above 13 GPa, which is beyond the pressure capability of CAC, we have employed a 300 µm-culet DAC apparatus to measure the resistance $R(T)$ of TlFeSe$_{2}$ from 15 to 45 GPa. Figure 4(a) shows the measured resistance $R(T)$ data normalized by that at 250 K. As can be seen, there is a broad hump centered around 100 K for $R(T)/R$(250 K) at 15.4 GPa. This hump feature is suppressed gradually and the typical metallic behavior is recovered upon further increasing pressure. It should be noted that the sample in DAC might experience a considerable shear stress because a solid PTM is used. Interestingly, we observed at 30.6 GPa a slight resistance drop starting at $\sim $1.7 K, which moves to higher temperatures and reaches about 2.8 K at 45.2 GPa. Meanwhile, the resistance drop becomes stronger and sharper with increasing pressure, as illustrated in the inset of Fig. 4(a). Since the observed resistance drop can be suppressed gradually by the application of an external magnetic field (as shown in the following), it signals the emergence of possible superconductivity in the HP phase.
cpl-37-4-047102-fig4.png
Fig. 4. (a) Temperature dependence of the normalized resistance $R(T)/R$(250 K) under various pressures from 15 to 45 GPa measured with a DAC. Inset shows an enlarge view of the resistance drop in the low-temperature region for the pressure range 30.6–45.2 GPa. (b) Temperature dependence of the upper critical field at 45.2 GPa. Bottom-left inset shows the $R(T)$ from 4 K down to 0.3 K under 45.2 GPa with zero magnetic field. Top-right inset shows the $R(H)$ curves at various temperatures under 45.2 GPa.
To verify the possible superconducting state, we have further conducted resistance measurements in a He-3 refrigerator at 45.2 GPa, which is the highest pressure in the present study. As shown in the bottom inset of Fig. 4(b), the resistance drop with $T_{\rm c}^{\rm onset} \sim 2.7$ K is reproduced, but zero resistance cannot be achieved down to 0.3 K, which is the lowest temperature in the present study. This should be attributed to the stress that is inevitably experienced by the sample in DAC with solid PTM. To study the upper critical field, we have measured a series of $R(H)$ curves at various temperatures from 0.3 to 2.7 K, the top inset of Fig. 4(b). As can be seen, the onset of superconducting transition is gradually suppressed by increasing either temperature to 2.7 K and/or magnetic field to $\mu_{0}H=2$ T. These results thus provide further evidences in support of superconductivity in TlFeSe$_{2}$ at high pressures. Here, we defined $T_{\rm c}$ ($H_{\rm c2}$) as the temperature (magnetic field) at the 90% of the normal state resistance, and plotted the temperature dependence of upper critical fields $\mu_{0}H_{\rm c2}$ in the main panel of Fig. 4(b). As can be seen, $\mu_{0}H_{\rm c2}(T)$ can be best fitted to a linear behavior with a slope of $dH_{\rm c2}/dT=-0.594$ T/K and $\mu_{0}H_{\rm c2}(0)=1.32$ T. The linear behavior of $H_{\rm c2}(T)$ over a wide temperature range has been observed in many iron-based superconductors[28] and Nb$_{2}$Pd$_{0.81}$S$_{5}$,[29] and has been attributed to the multiband effect. For type-II superconductors, the zero-temperature upper critical field $\mu_{0}H_{\rm c2}(0)$ can also be estimated from the Werthamer–Helfand–Hohenberg (WHH) equation[30] given by $\mu_{0}H_{\rm c2}(0)=-DT_{\rm c}(dH_{\rm c2}/dT)\big| _{T=T_{\rm c}}$, where $D$ is equal to $-0.693$ in the dirty limit. As indicated by the solid circle in Fig. 4(b), we obtained $\mu_{0}H_{\rm c2}(0)=0.918$ T based on the WHH model. It is lower than the fitted value shown above. A better agreement between the theoretical and experimental values may have been obtained if one takes an additional parameter characterizing spin-orbital coupling. Taking $\mu_{0}H_{\rm c2}(0)=1.32$ T as the upper limit, the superconducting coherence length can be estimated to be $\xi_{\rm GL}(0)=15.76$ nm according to the Ginzburg–Landau formula $\xi_{\rm GL}(0)=\{{\it\Phi}_{0}/[2\pi H_{\rm c2}(0)]\}^{1/2}$, where ${\it\Phi}_{0 }=h/2e$ is the magnetic flux quantum. Firstly, pressure-induced structural transition and superconductivity in the HP phase of TlFeSe$_{2}$ constitute the main findings of the present work. As mentioned above, the monoclinic TlFeSe$_{2}$ studied here has a distinct structure in comparison with the tetragonal $I\bar{4}m2$ phase used in Ref. [10]. Instead, the copper-based chalcopyrite Cu$XB_{2}$ ($X$ = Ga, In, $B$ = S, Se) adopts a similar tetragonal structure with space group $I\bar{4}2d$.[31,32,33] Previous studies have shown that Cu$XB_{2}$ can undergo a pressure-induced transformation to NaCl-type structure (space group $F\bar{4}3m$) with a random distribution of Cu$^{+}$ and $X^{3+}$ ions.[26,32,33] Although the HP phase above $P_{\rm c} \approx 13$ GPa can be indexed in an orthorhombic symmetry, the detailed structural information remains unknown at present. In addition, whether the orthorhombic structure of TlFeSe$_{2}$ will transform to higher symmetry phase such as the NaCl-type structure deserves further XRD studies at higher pressures. This structural information is essential for a better understanding of the observed superconductivity above 30 GPa, which does not coincident with the structural transition pressure of $P_{\rm c} \approx 13$ GPa. For this purpose, HP XRD data with better quality are needed. Then, it is better to utilize the gas PTM such as neon or helium rather than the silicon oil used in the present study. Alternatively, one can also resort to structural optimization and searching algorithm based on ab initio calculations under HP. Secondly, pressure-induced metallization of the monoclinic TlFeSe$_{2}$ in a relatively low pressure offers some insights into the peculiar electronic state of the Q1D Fe–Fe chains along the $b$ axis. At ambient pressure, the monoclinic structure of TlFeSe$_{2}$ is characterized by Q1D chain-like structure, as shown in inset of Fig. 1(a), in which the intrachain nearest-neighbor Fe–Fe distance of 2.7 Å is comparable to the Fe–Fe distance of 2.48 Å in the metallic iron, and is much shorter than the interchain Fe–Fe distance of about 6 Å. Hence, the physical properties of monoclinic TlFeSe$_{2}$ are expected to be dominated by chains of edge-shared FeSe$_{4}$ tetrahedra along the $b$ axis, which should possess a certain degree of itinerancy and 1D metallic behavior. However, the resistivity along the $b$-axis was found to be insulating, supported by the theoretical study by Luo et al.[19] However, based on a detailed analysis of the magnetic properties, Seidov et al.[15] argued that TlFeSe$_{2}$ should be 1D metal, which is hidden by the mechanical breaks and structural imperfections within the chains. Nishioka et al.[34] have pointed out similar situations in KFeS$_{2}$, in which the lowest resistivity was achieved on the annealed samples. Nevertheless, a metallic behavior has never been realized in these samples. In the present study, we confirmed that the metallic resistivity is readily realized in TlFeSe$_{2}$ under a moderate pressure of 2 GPa, when the chain-like Q1D crystal structure is still retained. Our results indicate that the pressure-induced reduction of the Fe–Fe intrachain/interchain distances gradually enhances the $3d$ electron delocalization and leads to the 1D metallic character in TlFeSe$_{2}$. We noted that Fe in TlFeSe$_{2}$ has a nominal valence of 3$+$, which is a hall-filled $d$ shell with no spin-orbit coupling. As is well known, an interaction-electron gas in 1D will be an insulator at half filling due to the Umklapp process,[35] which is expected to be irrelevant away from the half filling and the system is expected to be a conductor, such as (TMTSF)$_{2}X$ and (TMTTF)$_{2}X$ ($X$ = PF$_{6}$/Br/ClO$_{4}$).[36–38] There are many HP studies on these organic conductors, in which the degree of delocalization of charge carriers can be effectively tuned by hydrostatic pressure due to the enhancement of three-dimensional interchain interactions.[20,37,39] For example, the organic metal (TMTSF)$_{2}$PF$_{6}$ is a spin-density-wave insulator below 12 K at ambient pressure, but the SDW phase can be suppressed and a superconducting state is stabilized at low temperatures above a critical pressure $\sim $0.8 GPa.[39] Obviously, the interchain coupling in TlFeSe$_{2}$ will be strengthened under pressure as shown by the larger compressibility along the $a$ and $c$ axes, see Fig. 2(b). This will inevitably strengthen the three-dimensional interactions and lead to metallization of TlFeSe$_{2}$. In addition, the observed metallization of TlFeSe$_{2}$ indicates that an insulator-to-metal transition should be easily achieved in the family of monoclinic $A$FeSe$_{2}$ at ambient pressure through other routes such as the carrier doping. Finally, our results demonstrate that the structural and electronic properties of monoclinic TlFeSe$_{2}$ are highly tunable, making it an interesting platform to explore insulator-metal transition, dimensionality crossover, structural transformation, and superconductivity. Thus, the ternary $A$FeSe$_{2}$ ($A$ = Tl, K, Cs, and Rb) with the thermodynamically stable monoclinic symmetry, different from the tetragonal structure used in Ref. [10], also deserve further investigations. In summary, we have performed a comprehensive HP study on the monoclinic TlFeSe$_{2}$ crystals under various pressure up to 45 GPa. We find that TlFeSe$_{2}$ undergoes a pressure-induced structural transformation from the low-pressure monoclinic phase to a HP orthorhombic phase at $P_{\rm c} \approx$ 13 GPa, accompanied with a large volume collapse of $\Delta V/V_{0}=8.3{\%}$. The electrical transport properties of TlFeSe$_{2}$ below and above $P_{\rm c}$ were characterized via resistivity measurements with CAC and DAC, respectively. For $P < P_{\rm c}$, the insulating ground state is gradually suppressed by pressure and is converted to a metal completely at $P > 6$ GPa, where non-Fermi-liquid behavior was observed in resistivity. Whether the non-Fermi-liquid behavior is caused by the low-dimensional fluctuations or associated with the suppression of long-range antiferromagnetic order deserves further investigations. For $P > P_{\rm c}$, the sample is metallic and a possible superconducting state with $T_{\rm c}^{\rm onset} \sim 2$ K emerges gradually at $P > 30$ GPa. This work demonstrates that the structural and electronic properties of monoclinic TlFeSe$_{2}$ are greatly tunable so that the ternary $A$FeSe$_{2}$ should be regarded as an important material platform for explorations of interesting phenomena such as insulator-metal transition, dimensionality crossover, structural transformation, and superconductivity. References Possible highT c superconductivity in the Ba?La?Cu?O systemIron-Based Layered Superconductor La[O 1- x F x ]FeAs ( x = 0.05−0.12) with T c = 26 KSuperconductivity at 55 K in Iron-Based F-Doped Layered Quaternary Compound Sm[O 1- x F x ] FeAsSuperconductivity at 43 K in SmFeAsO1-xF xThe Challenge of Unconventional SuperconductivityPrediction of phonon-mediated high-temperature superconductivity in Li 3 B 4 C 2 Predicting Unconventional High-Temperature Superconductors in Trigonal Bipyramidal CoordinationsA possible family of Ni-based high temperature superconductorsTheoretical studies of superconductivity in doped BaCoSOAFeSe2 (A=Tl, K, Rb, or Cs): cuprate analogs of iron-based superconducting materialsThe antiferromagnetic structures of KFeS2, RbFeS2, KFeSe2, and RbFeSe2 and the correlation between magnetic moments and crystal field calculationsMagnetic structure of TlFeS2 and TlFeSe2 chalcogenidesSynthese und Kristallstrukturen von T1FeS2 und T1FeSe2Magnetic susceptibility and ESR study of the covalent-chain antiferromagnets TlFeS 2 and TlFeSe 2 On superparamagnetism in quasi-one-dimensional semiconductor antiferromagnets TlFeS2 and TlFeSe2Magnetic properties of chain antiferromagnets RbFeSe2, TlFeSe2, and TlFeS2Magnetic states of the five-orbital Hubbard model for one-dimensional iron-based superconductorsPressure-induced deconfinement of the charge transport in the quasi-one-dimensional Mott insulator ( TMTTF ) 2 As F 6 A new quasi-one-dimensional compound Ba3TiTe5 and superconductivity induced by pressureIntegrated-fin gasket for palm cubic-anvil high pressure apparatusHigh-pressure effect on the chain-like crystal structure of the semiconductors TlFeSe 2 and TlFeS 2Finite Elastic Strain of Cubic CrystalsHigh pressure transformations of ternary chalcogenides with chalcopyrite structure — I. Indium-containing compoundsStructural Studies of CuInS2 and CuInSe2 under High PressureMagnetism of iron-sulfur tetrahedral frameworks in compounds with thallium I. Chain structuresNearly isotropic superconductivity in (Ba,K)Fe2As2Superconductivity with extremely large upper critical fields in Nb2Pd0.81S5Temperature and Purity Dependence of the Superconducting Critical Field, H c 2 . III. Electron Spin and Spin-Orbit EffectsIdeal Weyl Semimetals in the Chalcopyrites CuTlSe 2 , AgTlTe 2 , AuTlTe 2 , and ZnPbAs 2 High-Pressure Behavior of Raman Modes in CuGaSe 2Pressure-induced phase transformations in the chalcopyrite-structure compounds: CuGa S 2 and AgGa S 2 Electric conduction in quasi-one-dimensional compound KFeS2Umklapp process and resistivity in one-dimensional fermion systemsTransverse transport in organic conductors: possible evidence for a Luttinger liquidElectron spin resonance studies on the organic linear-chain compounds ( TMT C F ) 2 X   ( C = S , Se ; X = PF 6 , AsF 6 , ClO 4 , Br ) Organic Superconductors: A Survey of Low Dimensional Phenomena
[1] Bednorz J G and Müller K A 1986 Z. Phys. B: Condens. Matter 64 189
[2] Kamihara Y, Watanabe T, Hirano M and Hosono H 2008 J. Am. Chem. Soc. 130 3296
[3] Ren Z A, Lu W, Yang J, Yi W, Shen X L, Li Z C, Che G C, Dong X L, Sun L L, Zhou F and Zhao Z X 2008 Chin. Phys. Lett. 25 2215
[4] Chen X H, Wu T, Wu G, Liu R H, Chen H and Fang D F 2008 Nature 453 761
[5] Norman M R 2011 Science 332 196
[6] Gao M, Lu Z Y and Xiang T 2015 Phys. Rev. B 91 045132
[7] Hu J P, Le C C and Wu X X 2015 Phys. Rev. X 5 041012
[8] Le C C, Zeng J F, Gu Y H, Cao G H and Hu J P 2018 Sci. Bull. 63 957
[9] Qin S S, Li Y X, Zhang Q, Le C C and Hu J P 2018 Front. Phys. 13 137502
[10] Zhao X, Ma F, Lu Z Y and Xiang T 2019 arXiv:1910.03545 [cond-mat.supr-con]
[11] Bronger W, Kyas A and Müller P 1987 J. Solid State Chem. 70 262
[12] Asgerov E B, Dang N T, Beskrovnyy A I, Madadzada A I, Ismayilov D I, Mehdiyeva R N, Jabarov S H and Karimova E M 2015 Semiconductors 49 879
[13] Kutoglu A 1974 Naturwissenschaften 61 125
[14]Makovetskii G I and Kasinskii E I 1984 Inorg. Mater. 20 1514
[15] Seidov Z, Krug von Nidda H A, Hemberger J, Loidl A, Sultanov G, Kerimova E and Panfilov A 2001 Phys. Rev. B 65 014433
[16] Veliyev R G 2012 Semiconductors 46 1263
[17] Seidov Z, Krug von Nidda H A, Tsurkan V, Filippova I, Günther A, Najafov A, Aliyev M N, Vagizov F G, Kiiamov A G, Tagirov L R, Gavrilova T and Loidl A 2017 Bull. Russ. Acad. Sci.: Phys. 81 885
[18]Kerimova E M, Seidov F M, Mustafaeva S N and Abdinbekov S S 1999 Inorg. Mater. 35 106
[19] Luo Q, Foyevtsova K, Samolyuk G D, Reboredo F and Dagotto E 2014 Phys. Rev. B 90 035128
[20] Pashkin A, Dressel M and Kuntscher C A 2006 Phys. Rev. B 74 165118
[21] Zhang J, Jia Y, Wang X, Li Z, Duan L, Li W, Zhao J, Cao L, Dai G, Deng Z, Zhang S, Feng S, Yu R, Liu Q, Hu J, Zhu J and Jin C 2019 NPG Asia Mater. 11 60
[22] Cheng J G, Matsubayashi K, Nagasaki S, Hisada A, Hirayama T, Hedo M, Kagi H and Uwatoko Y 2014 Rev. Sci. Instrum. 85 093907
[23] Asgerov E B, Dang N T, Ismayilov D I, Kichanov S E, Mehdiyeva R N, Madadzada A I, Jabarov S H, Kerimova E M and Lukin E V 2015 Mod. Phys. Lett. B 29 1550024
[24] Birch F 1947 Phys. Rev. 71 809
[25] Range K J, Engert G and Weiss A 1969 Solid State Commun. 7 1749
[26] Tinoco T, Polian A, Gomez D and Itie J P 1996 Phys. Status Solidi B 198 433
[27] Welz D, Deppe P, Schaefer W, Sabrowsky H and Rosenberg M 1989 J. Phys. Chem. Solids 50 297
[28] Yuan H Q, Singleton J, Balakirev F F, Baily S A, Chen G F, Luo J L and Wang N L 2009 Nature 457 565
[29] Zhang Q, Li G, Rhodes D, Kiswandhi A, Besara T, Zeng B, Sun J, Siegrist T, Johannes M D and Balicas L 2013 Sci. Rep. 3 1446
[30] Werthamer N R, Helfand E and Hohenberg P C 1966 Phys. Rev. 147 295
[31] Ruan J, Jian S K, Zhang D, Yao H, Zhang H, Zhang S C and Xing D 2016 Phys. Rev. Lett. 116 226801
[32] Gonzalez J and Chervin J C 1993 Jpn. J. Appl. Phys. 32 575
[33] Werner A, Hochheimer H D and Jayaraman A 1981 Phys. Rev. B 23 3836
[34] Nishioka S, Kuriyaki H and Hirakawa K 1995 Synth. Met. 71 1877
[35] Giamarchi T 1991 Phys. Rev. B 44 2905
[36]Balicas L, Behnia K, Kang W, Canadell E, Auban-Senzier P, Jérome D, Ribault M and Fabre J M 1994 J. Phys. I 4 1539
[37] Moser J, Gabay M, Auban-Senzier P, Jerome D, Bechgaard K and Fabre J M 1998 Eur. Phys. J. B 1 39
[38] Dumm M, Loidl A, Fravel B W, Starkey K P, Montgomery L K and Dressel M 2000 Phys. Rev. B 61 511
[39] Jerome D 1982 Mol. Cryst. Liq. Cryst. 79 511