Planckian scattering and parallel conduction channels in the iron chalcogenide superconductors FeTe 1-x Se x
Ralph Romero III, Hee Taek Yi, Seongshik Oh, N. P. Armitage · 2025
Read Planckian scattering and parallel conduction channels in the iron chalcogenide superconductors FeTe 1-x Se x by Ralph Romero III, Hee Taek Yi, Seongshik Oh, N. P. Armitage (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-00623.txt.
Introduction
Ralph Romero, Hee Taek Yi, Seongshik Oh, and Nicholas Armitage probe how electrons scatter in an iron-based superconductor, in a 2025 paper. Copper oxide superconductors show a famously unexplained resistance that grows linearly with temperature. Whether that odd behavior ties to the pairing state itself remains a central question.
Using terahertz spectroscopy on two samples of iron telluride selenide, the team separated the current into two parallel channels. One channel is broad and barely changes with temperature. The other is sharp, and its scattering rate tracks the so-called Planckian limit set by temperature alone. When the samples turned superconducting, the condensate drew mainly from that Planckian channel. The finding links the strange scattering directly to the pairing that produces superconductivity, at least in this material family.
Catalog record
| CREATOR | Ralph Romero III, Hee Taek Yi, Seongshik Oh, N. P. Armitage |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 23 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-00623
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Planckian scattering and parallel conduction channels in the iron chalcogenide superconductors FeTe 1-x Se x Ralph Romero III, Hee Taek Yi, Seongshik Oh, N. P. Armitage Abstract The remarkable linear in temperature resistivity of the cuprate superconductors, which extends in some samples from T_{c} to the melting temperature, remains unexplained. Although seemingly simple, this temperature dependence is incompatible with the conventional theory of metals that dictates that the scattering rate, 1/\tau , should be quadratic in temperature if electron-electron scattering dominates. Understanding the origin of this temperature dependence and its connection to superconductivity may provide the key to pick the lock of high-temperature superconductivity. Using time-domain terahertz spectroscopy (TDTS) we elucidate the low temperature conducting behavior of two FeTe 1-x Se x (FTS) samples, one with almost equal amounts of Se and Te that is believed to be a topological superconductor, and one that is more overdoped. Constrained with DC resistivity, we find two conduction channels that add in parallel, a broad one in frequency with weak temperature dependence and a sharper one whose scattering rate goes as the Planckian limited rate, \sim kT/h . Through analysis of its spectral weight we show the superconducting condensate is mainly drawn from the channel that undergoes this Planckian scattering. The chalcogenide superconductor series FeTe 1-x Se x (FTS) provides a powerful platform to explore strongly correlated superconductivity in proximity to other ordered states [ 1 , 2 ] . The end member FeTe does not superconduct and possesses a tetragonal to monoclinic structural transition accompanied by an antiferromagnetic (AFM) transition around 70K [ 3 , 4 ] . The other end member FeSe undergoes a tetragonal to orthorhombic structural transition into a nematic phase at 90K and superconducts at 9 K [ 5 ] . At an “optimal” doping, i.e. close to equal parts Se/Te ( x=0.45 ), T_{c} is elevated to 14.5 K with upper critical fields reaching 50T [ 2 ] . Along with this maximized T_{c} , samples with optimal doping have amassed a compelling body of evidence [ 6 , 7 , 8 , 9 , 10 , 11 , 12 ] suggesting FTS belongs to a class of topological superconductors able to host Majorana zero modes (MZMs), that are proposed to be a platform for topological quantum computation [ 13 ] . Figure 1: THz conductivity of the Se = 0.45 and 0.35 samples. Real and imaginary parts of the THz conductivity from room temperature down to 1.4K measured from 0.2 to 3 THz for the x=0.45 sample a), b) and the x=0.35 sample c), d). The iron chalcogenide superconductors have a phenomenology reminiscent of the cuprate superconductors, namely a superconducting dome whose peak straddles a quantum critical point (QCP) [ 14 ] . While this is interesting, it should come as little surprise as out of the iron based superconductors and other pnictides, FeTe and FeSe have the highest relative mass enhancements [ 15 ] , suggesting correlations play a large role in their behavior. In retrospect it is also not surprising that a combination of these correlated parent compounds gives birth to unconventional superconductivity, but what is so special about the optimal ( x=0.45 ) doping that gives rise to an increased transition temperature and ultimately such a unique superconducting state? In the cuprates it is often proposed that T_{c} is maximized at the doping that coincides with a zero temperature phase transition. It has been conjectured that proximity to this QCP may give rise to strong quantum fluctuations, which manifest themselves as deviations from the Fermi liquid picture, namely resistivity or equivalently the charge scattering rate that goes as as Planckian rate, ~\sim kT/h [ 16 ] (and linear in frequency single particle self-energies). Such a scenario presents challenges however. In FTS, it was initially conjectured that critical fluctuations stemmed from an AFM QCP associated with the transition in FeTe [ 17 ] . However, recently it was shown that optimal doping for FTS occurs near a QCP associated with a nematic order [ 14 ] and theoretically a clean nematic QCP is expected to give a imaginary self-energy that goes like \omega^{2/3} [ 18 ] . A time-resolved technique such as TDTS allows the experimenter to measure the magnitude and phase of the transmitted electric field simultaneously. This allows one to measure both parts of the complex conductivity in the THz range, \sigma_{1}(\nu) and \sigma_{2}(\nu) , without having to use the Kramers-Kronig relations. TDTS was performed on two 100 nm FeTe 1-x Se x films with x=0.45 and 0.35 grown by molecular beam epitaxy on 0.5 mm thick CaF 2 substrates, capped with 100 nm of Se. The two samples have similar transition temperatures of 13.2 and 11.8 K, respectively, measured via DC resistivity (Extended Data Fig. 1 ). Fig. 1 shows \sigma_{1}(\nu) (top) and \sigma_{2}(\nu) (bottom) for the two samples. The left side shows the complex conductivity for the optimally doped sample, while the right side corresponds to the x=0.35 sample. Both samples show the same qualitative behavior. At high temperatures, T\gg T_{c} , \sigma_{1}(\nu) is broad and featureless while \sigma_{2}(\nu) is small and slowly increasing with frequency. This is canonical behavior of a metal whose scattering rate is larger than the measurement window (i.e. \Gamma\gg 3 THz). As temperature decreases \sigma_{1}(\nu) begins to increase and acquire negative slope reminiscent of a Drude peak. Once the superconducting transition temperature is crossed, \sigma_{1}(\nu) decreases while \sigma_{2}(\nu) begins to show 1/\nu dependence, corresponding to the onset of superconductivity as spectral weight moves below our measurement range into the delta function at \nu=0 . In a conventional scenario one would expect most of the spectral weight below the gap energy, 2\Delta=3.5kT_{c}\approx 1 THz, to condense to \nu=0 , much like what has been seen in disordered NbN thin films [ 19 ] . However, in the present case even at the lowest temperature T = 1.4 K, there is still considerable spectral weight within the gap. Figure 2: Fits to the THz conductivity of the Se = 0.45 sample at various temperatures. a), b) \sigma_{1} , \sigma_{2} at a temperature (38 K) where a single Drude term suffices. c), d) A temperature (26 K) where a constant offset, \sigma_{0} , must be included alongside the Drude term. e), f) A temperature, above T_{c} , (16 K) where the two fluid model must be invoked. g), h) The lowest temperatures where the conductivity becomes mainly sensitive to the condensate at \nu=0 . Figure 3: Fit parameters along with DC resistivity. The left column correspond to the x=0.45 sample, while the right to x=0.35 sample. a) and c) show the scattering rate, \Gamma , with the vertical dashed line indicating T_{c} and the horizontal dashed line denoting the edge of our spectral range. Linear in temperature scaling can be observed with slopes on the order of 3kT/h . b) and d) Show the spectral weight of the Drude term, S , (left axis) and the constant offset, \sigma_{0} , (right axis). Figure 4: Superfluid density, three ways. The superfluid density measured using three methods for both samples. The left column corresponds to the optimally doped sample, while the right to x=0.35 sample. a) and c) show fits