Low-dimensional systems often display rich physical phenomena, such as charge density wave (CDW), spin density wave (SDW), and superconductivity. Many superconductors were found at low dimensions where the superconductivity coexists with or adjoins another ordered state and has an unconventional nature. The high transition temperature (<italic>T</italic><sub>C</sub>) superconductivity has been realized in two-dimensional copper oxides and iron pnictides, where the superconducting phase is located in the vicinity of magnetic and nematic orders. This raises an interest in finding unconventional superconductivity in low-dimensional materials. Ta<sub>4</sub>Pd<sub>3</sub>Te<sub>16</sub> with a quasi-one-dimensional crystal structure was discovered to be superconducting with <italic>T</italic><sub>C</sub>∼4.6 K. It consists of PdTe<sub>2</sub> chains, TaTe<sub>3</sub> chains, and Ta<sub>2</sub>Te<sub>4</sub> double chains along the crystallographic <italic>b</italic> axis. Band structure calculations indicate that Ta<sub>4</sub>Pd<sub>3</sub>Te<sub>16</sub> is an s-wave superconductor with pairing from phonons associated with the Te-Te p bonding and a Fermi-surface associated with Te p bands. Although scanning tunneling microscopy (STM) found that the superconducting gap structure in this system is more likely anisotropic without nodes, nodal gap behaviors were claimed by thermal conductivity and specific heat measurements. STM study suggested that the system is in the vicinity of an ordered state that shows a periodic modulation. This suggestion is consistent with the observation that the magnetoresistance shows an <italic>H</italic>-linear behavior without saturation up to 50 T. In this paper, we report nuclear magnetic resonance and nuclear quadrupole resonance investigations on Ta<sub>4</sub>Pd<sub>3</sub>Te<sub>16</sub>. The spin-lattice relaxation rate (1/<italic>T</italic><sub>1</sub>) divided by the temperature, 1/<italic>T</italic><sub>1</sub><italic>T</italic>, of <sup>125</sup>Te is almost a constant and obeys the Korringa relationship as in a normal metal, but 1/<italic>T</italic><sub>1</sub><italic>T</italic> of <sup>181</sup>Ta increases dramatically below 80 K. These results indicate strong electric-field-gradient (EFG) fluctuations, since <sup>181</sup>Ta has a nuclear spin <italic>I</italic>=7/2 with a large nuclear quadrupole moment that couples to EFG, but <sup>125</sup>Te with spin <italic>I</italic>=1/2 can only relax by magnetic interactions. Upon cooling, both the full width at half maximum of the spectra and 1/<italic>T</italic><sub>1</sub><italic>T</italic> show a sudden change at <italic>T</italic>=20 K, indicating a CDW transition takes place at <italic>T</italic><sub>CDW</sub>=20 K. This is consistent with the electronic structure calculation which implies that Ta<sub>4</sub>Pd<sub>3</sub>Te<sub>16</sub> prefers the CDW rather than the SDW. In the superconducting state, a Hebel-Slichter coherence peak appears in the temperature dependence of 1/<italic>T</italic><sub>1</sub> of <sup>125</sup>Te just below <italic>T</italic><sub>C</sub>, which indicates that Ta<sub>4</sub>Pd<sub>3</sub>Te<sub>16</sub> is a fully gapped superconductor. 1/<italic>T</italic><sub>1</sub> can be fitted by using a BCS gap with <inline-formula><mml:math xmlns:mml=http://www.w3.org/1998/Math/MathML display=inline><mml:mrow><mml:mi>Δ</mml:mi><mml:mo stretchy=false>(</mml:mo><mml:mn>0</mml:mn><mml:mo stretchy=false>)</mml:mo><mml:mo>=</mml:mo><mml:mn>1.76</mml:mn><mml:msub><mml:mi other=0>k</mml:mi><mml:mtext other=1>B</mml:mtext></mml:msub> <mml:msub><mml:mi other=0>T</mml:mi><mml:mtext other=1>C</mml:mtext> </mml:msub></mml:mrow></mml:math></inline-formula>. Although Ta<sub>4</sub>Pd<sub>3</sub>Te<sub>16</sub> is a multiband system, the main contribution to the DOS is from Te p orbitals. Therefore, the full gap derived from the <sup>125</sup>Te NMR is consistent with the suggestion from the band calculation that Ta<sub>4</sub>Pd<sub>3</sub>Te<sub>16</sub> should be an s-wave superconductor mediated by phonons associated with the Te-Te bonding if there are no spin fluctuations. Due to coexistence of CDW and superconducting, strong EFG fluctuations eliminate the coherence peak in the temperature dependence of 1/<italic>T</italic><sub>1</sub> of <sup>181</sup>Ta.
Read full abstract