Using time-domain terahertz spectroscopy we performed direct studies of the photoinduced suppression and recovery of the superconducting gap in a conventional BCS superconductor NbN. Both processes are found to be strongly temperature and excitation density dependent. The analysis of the data with the established phenomenological Rothwarf-Taylor model enabled us to determine the bare quasiparticle recombination rate, the Cooper pair-breaking rate and the electron-phonon coupling constant, λ=1.1±0.1, which is in excellent agreement with theoretical estimates.

Observations of radiation-enhanced superconductivity have thus far been limited to a few type-I superconductors (Al, Sn) excited at frequencies between the inelastic scattering rate and the superconducting gap frequency 2Δ/h. Utilizing intense, narrow-band, picosecond, terahertz pulses, tuned to just below and above 2Δ/h of a BCS superconductor NbN, we demonstrate that the superconducting gap can be transiently increased also in a type-II dirty-limit superconductor. The effect is particularly pronounced at higher temperatures and is attributed to radiation induced nonthermal electron distribution persisting on a 100 ps time scale.

We studied the superconducting (SC) state depletion process in an electron doped cuprate Pr1.85Ce0.15CuO 4−δ by pumping with near-infrared (NIR) and narrow-band THz pulses. When pumping with THz pulses tuned just above the SC gap, we find the absorbed energy density required to deplete superconductivity, A dep , matches the thermodynamic condensation energy. Contrary, by NIR pumping A dep is an order of magnitude higher, despite the fact that the SC gap is much smaller than the energy of relevant bosonic excitations. The result implies that only a small subset of bosons contribute to pairing.PACS numbers: 74.40. Gh, The quest for a pairing boson in cuprate hightemperature superconductors has been one of the key topics of solid state physics ever since the discovery of superconductivity in the cuprates. Recently, numerous femtosecond (fs) real-time studies of carrier dynamics in high-T c superconductors have been performed aiming to find the coupling strengths between the electrons and other degrees of freedom (high and low frequency phonons, spin fluctuations, electronic continuum) [1][2][3][4][5][6][7][8]. In this approach, fs optical pulses are used to excite the electronic system, while the resulting dynamics are probed by measuring the changes in optical constants [2][3][4][5][6] or the electronic distribution near the Fermi energy [1,7,8]. To connect the measured relaxation timescales to the electron-boson coupling strengths, the multitemperature models are commonly used [9,10]. These are based on the premise that the electron-electron (ee) thermalization is much faster than the electron-boson relaxation. While these models are commonly used to extract e.g. the electron-phonon (e-ph) coupling strengths, numerous inconsistencies have been noted (even for the case of simple metals) [11][12][13][14]. An alternative timedomain approach, based on the dynamics in the superconducting state, has been put forward [15,16]. Under the assumption that the absorbed optical energy is distributed between quasiparticles and high frequency ( ω > 2∆) bosons on the sub-picosecond timescale, and taking into account the nonlinearity of relaxation processes (pairwise recombination of quasiparticles), the electron-boson coupling strength is determined by studying the excitation density dependence of the Cooper pair-breaking process [15,16]. While this approach has been successfully applied to conventional superconductors [16,17], the results on cuprates show that the energy density required to suppress superconductivity exceeds the thermodynamic condensation energy, E c , by an order of magnitude [2,[18][19][20]. Therefore, the assumption that the absorbed energy is distributed between QPs and the coupled high frequency bosons fails. Considering the possible energy relaxation pathways, this discrepancy in the hole-doped high-T c cuprates has been attributed to the fact that the superconducting gap, 2∆, lies well in the range of optical phonons [19]. It has been argued that ≈ 90 % of the absorbed energy is directly released to ω < ...

No abstract