betaFeedback

WorksPhysics and Astronomy

Composition-dependent ultrafast luminescence in Cu-Ni alloys: Combined experimental and ab initio study

Abstract

Properties of Cu-Ni solid solutions have long been studied in physical and materials sciences. Yet, their many-body properties have not been well understood. Here, we investigate ultrafast luminescence in near infrared region for Cu1−x_{1-x}Nix_x alloys. The luminescence intensity was the highest in Cu and decreased dramatically by adding Ni, approaching close to the value for pure Ni at x=0.45x=0.45. This composition dependence was well reproduced by calculations assuming two body scattering of the energetic electrons. The luminescent decay rate was not straightforward, i.e., it decreased first by adding Ni up to x=0.17x=0.17 and then started to increase approaching twice the initial value at x=0.45x=0.45. This behavior was in good agreement with ab initio calculations of electron-phonon (e-ph) coupling strength. This work provides a new perspective on the electron relaxation dynamics in solid solutions systems.

Introduction

Metal alloys are of significant importance both in application and solid state physics, because the crystal structure and phase stability [1, 2], electrical and thermal conductivity [3, 4, 5], magnetization [6, 7], and optical properties [8] can be greatly modified from the constituent metals by alloying. In addition to these macroscopic properties, it is of fundamental interest to understand the excited state dynamics and the mutual interactions between electrons and phonons. Many body properties in an ordered compound have been investigated in detail [9], but those in solid solutions are not well understood.

Cu-Ni alloy is a traditional solid solution system, utilized for coins with silver-like color for a long time. The nearly temperature independent resistivity and low thermoelectric power in Cu0.55Ni0.45\mathrm{Cu}_{0.55}\mathrm{Ni}_{0.45} alloy (so-called constantan) are utilized for thermocouples (known as copper-constantan) and standard register. Relatively low thermal conductivity in cupronickel (Cu0.7Ni0.3\mathrm{Cu}_{0.7}\mathrm{Ni}_{0.3}) is useful for cryogenic instrument. The advantage of this system is that Cu and Ni have the same crystal structure (FCC) with lattice constant close each other, ensuring homogeneous mixture at arbitrary composition. The lattice constant of Cu1−xNix\mathrm{Cu}_{1-x}\mathrm{Ni}_x alloys linearly decreases with the Ni concentration xx, satisfying the well-known Vegard’s law [1]. However, their magnetic and electronic properties exhibit more complex behaviors with increasing xx. The Cu-Ni alloys show no magnetization below x≃xc=0.44x \simeq x_c = 0.44, above which the ferromagnetic order starts to increase [10, 11]. The resistivity of Cu-Ni alloy is far larger than the end materials [11, 12, 13]. This nonmonotonic composition dependence of the residual resistivity is ascribed to impurity scattering [11] and the magnetic resistivity due to the electron scattering from spin antiparallel to parallel states [14]. Although the electron-electron (e-e) and electron-phonon (e-ph) interaction may contribute to the excited state dynamics, their composition-dependence of Cu-Ni alloys is not well understood.

The ultrafast relaxation dynamics following a pulsed laser excitation enables us to study the e-e and e-ph properties of solids. Recently, we have systematically studied ultrafast luminescence of various elemental metals [15, 16, 17, 18, 19]. The time evolution of the luminescence intensity can be decomposed into a quasi-instantaneous and an exponential decay component. We have shown that (i) the intensity of the quasi-instantaneous component is proportional to inverse of the Drude damping rate evaluated at excitation photon energy and (ii) the relaxation rate of the exponential component is well explained by the modified two-temperature model (TTM) [20, 21] combined with ab initio calculations for the e-ph coupling strength [19]. These indicate that the luminescence dynamics in the initial and relaxation stages are governed by the e-e and e-ph scattering, respectively. It should be noted that the luminescence properties were very different for Cu and Ni, i.e., the lifetimes were short (order of 100 fs) and the intensity was two orders of magnitude lower for Ni compared to Cu. It is unclear how the luminescence properties of Cu1−xNix\mathrm{Cu}_{1-x}\mathrm{Ni}_x behave for 0<x<10 < x < 1.

Based on the TTM [20] and/or modified TTM [21], the e-ph coupling constant has been estimated from the pump-probe experiments for elemental metals [22, 23] and high-TcT_c cuprates [24], whereas the effects of nonequilibrium electrons and phonons are also investigated by using more complex models [25, 26, 27, 28]. From a view point of material science, alloys are highly attractive, because we will have a chance to design the behavior of energetic electrons by controlling the composition and crystal structure. However, we can rarely find experimental or theoretical research on the ultrafast electron dynamics in alloys [29, 30] and it is rare to apply the TTM to the electron dynamics of alloys. Experimental methods—The pure metals (Cu and Ni), Cu0.7_{0.7}Ni0.3_{0.3} alloy (cupronickel) and Cu0.55_{0.55}Ni0.45_{0.45} alloy (constantan) were purchased from Nilaco Corporation as a form of plate. The samples with other compositions were prepared by co-melting cupronickel and Cu on a tungsten boat mounted in a vacuum chamber of a vapor deposition equipment. In order to suppress evaporation of Cu, the chamber was purged by Ar gas at a pressure of about 3 Torr. The obtained ingot was cut and polished with alumina polishing paper (#1000—#4000). Finally the surfaces were roughened by pressing with an alumina polishing film. The X-ray diffraction powder patters were taken for all samples including pure metals and purchased alloys. Trace of Mn was found in the purchased Cu0.55_{0.55}Ni0.45_{0.45} alloy at a level of 1 %, while no trace of metal impurity was found in cupronickel. It was verified that the lattice constant follows Vegard’s law, which guarantees formation of homogeneous solid solution in whole composition range. The element analysis was made with an electron-probe micro-analyzer (EPMA) for the surfaces of metals. Traces of C, O and Al were found in melt-grown samples. Carbon is a common impurity originating from hydrocarbons in vacuum system. Oxygen and aluminum are presumed to come from the polishing alumina powder, which has practically no effect on the ultrafast luminescence properties of metals [19].

We used an up-conversion technique, using amplified mode-locked pulses (1.19 eV, 600 mW, 100 MHz repetition rate) from an Yb-fiber laser. The pumping pulse (typically 150 fs duration, 200 mW) was focused on the sample with an approximate spot size of 20 μ\mum. The luminescence was collected and focused on a nonlinear optical crystal (lithium iodate, LIO) using two paraboloidal mirrors and mixed with a gating pulse. The anti-Stokes luminescence from the sample was removed by a Si-filter. The generated sum-frequency light was passed through a long-wave-pass edge filter, directed into a spectrometer consisting of tunable band-pass filters and detected by an avalanche photodiode (APD). The time resolution of this system was typically 240 fs (full width at half maximum). Detailed information of the experimental setup are presented in our previous reports [16, 19] and Supplementary Materials for Ref. [16]. All luminescence measurements were performed in air at room temperature.

Absorptivity of the samples was measured at 1.19 eV as an increase of temperature under irradiation by the Yb-fiber laser with reduced power (20 mW). The absorptivity (emissivity) at longer wavelengths, where luminescence signal appears, was measured by using tunable light from an optical parametric amplifier (OPA) operating at a repetition frequency of 1 kHz with an average power (10 mW) to avoid damages on the sample.

As done in our previous works [15, 16, 17, 18, 19], we first measured the absorptivity of the alloy surfaces by using calorimetry (see Fig. S1 [31]). These spectra were used to obtain internal spectra and intensity, by normalizing the observed luminescence spectra by the energy dependence of emissivity.

Computational details—To calculate the electron DOS, we performed electronic structure calculations for alloys. To model the crystal structure of Cu-Ni alloys, we considered a 3×3×33 \times3 \times3 supercell of the cubic unit cell, 108 atoms in total. For each Ni concentration, we created 50 structures, where Ni atoms randomly occupy the Cu sites. We first optimized the geometry by using M3GNet, a universal interatomic potential incorporating three-body interactions [32], and determined the Ni distribution that has the lowest energy among 50 samples.

We reoptimized the crystal structure based on density-functional theory (DFT) calculations, where the atomic positions are relaxed and the lattice constant aa follows the Vegard’s law. In the present work, we assumed a=(1−x)aCu+xaNia = (1-x)a_{\mathrm{Cu}} + xa_{\mathrm{Ni}}, where the experimental lattice constants of aCu=3.61a_{\mathrm{Cu}} = 3.61 Å for Cu and aNi=3.52a_{\mathrm{Ni}} = 3.52 Å for Ni were used. The DFT calculations were performed by using the Quantum ESPRESSO (QE) [33]. The pseudopotentials of Perdew, Burke, and Ernzerhof [34] in pslibrary1.0.0 [35] were used. The cutoff energies for the wavefunction and charge density were set to be 60 Ry and 600 Ry, respectively, and the smearing parameter was set to be 0.02 Ry [36]. The Γ\Gamma point in the Brillouin zone was sampled in self-consistent field (scf) calculations, and the kk grid was increased to 6×6×66 \times6 \times6 in non-scf calculations. The electron DOS was plotted with a Gaussian broadening parameter of 0.2 eV. The DOS and magnetization for various Ni content are provided in Fig. S2 [31]. The spin splitting occurs around Ni 40%, which is consistent with experiments (Ni 44%) [10, 11].

To study the e-ph coupling, we calculate the Eliashberg function given by [20]

α2F(ω)=1ℏNFN∑α,α′,k∑β,q∣gα,α′β(k,q)∣2\alpha^2 F(\omega) = \frac{1}{\hbar N_{\mathrm{F}}N} \sum_{\alpha,\alpha',k} \sum_{\beta,q} \left|g^{\beta}_{\alpha,\alpha'}(k,q)\right|^2
×δ(ϵF−ϵαk)δ(ϵF−ϵα′k+q)δ(ω−ωβq),(1)(1){}\times\delta(\epsilon_{\mathrm{F}}-\epsilon_{\alpha k})\delta(\epsilon_{\mathrm{F}}-\epsilon_{\alpha' k+q})\delta(\omega-\omega_{\beta q}), (1) \tag*{(1)}

where ℏ\hbar is the Planck constant, NFN_{\mathrm{F}} is the electron DOS at the Fermi level, and NN is the number of the unit cell. ϵαk\epsilon_{\alpha k} is the electron energy with the wavevector kk and the band index α\alpha, ωβq\omega_{\beta q} is the phonon frequency for the wavevector qq and the branch index β\beta, and gα,α′β(k,q)g^{\beta}_{\alpha,\alpha'}(k,q) is the matrix elements for the e-ph interaction Hamiltonian. ω\omega is the phonon frequency. Therefore, α2F(ω)\alpha^2F(\omega) can be regarded as the phonon DOS weighted by the e-ph matrix elements at the Fermi level. We calculate α2F(ω)\alpha^{2}F(\omega) based on the density-functional perturbation theory (DFPT) [37] implemented in QE [33]. We considered a 2×2×22 \times2 \times2 supercell including eight atoms in the unit cell and optimized the crystal structure. We used a 4×4×44 \times4 \times4 qq grid for the phonon wave vectors, an 8×8×88 \times8 \times8 kk grid for constructing the induced charge density and the dynamical matrix, and a 16×16×1616 \times16 \times16 kk grid for the double-δ\delta function with the electron energy restricted at the Fermi surface. We calculated α2F(ω)\alpha^{2}F(\omega) of Cu8_8, Cu7_7Ni1_1, Cu6_6Ni2_2, and Cu4_4Ni4_4. The crystal structures used are provided in Supplemental Material [31].

We calculated the e-ph coupling constant defined as

λ⟨ωn⟩=2∫α2F(ω)ωn−1dω(2)\lambda\langle\omega^{n}\rangle= 2 \int\alpha^{2}F(\omega)\omega^{n-1}d\omega \tag*{(2)}

with an integer nn and estimated the decay rate for the nonequilibrium electrons Γ=3ℏλ⟨ω2⟩/(2πkBTph)\Gamma= 3\hbar\lambda\langle\omega^{2}\rangle/(2\pi k_{\mathrm{B}}T_{\mathrm{ph}}) with the Boltzmann constant kBk_{\mathrm{B}} and the phonon temperature TphT_{\mathrm{ph}} [20, 21].

Luminescence intensity—Figure 1(a) shows the internal time-resolved luminescence spectra measured at time origin. The abscissa is proportional to photon numbers per second per unit energy interval and the spectra are normalized at maximum of each curve. The spectra for pure Cu and Ni are basically in agreement with the sandblasted samples in our previous publication [19]. The slight deviations are ascribed to the difference in the roughness and the excitation power density, which depends on the optics of up-conversion measurement equipment. We can clearly see that the weight of the spectrum moves to lower energy side by increasing Ni content from x=0x = 0 to x=0.45x = 0.45. The spectrum for Ni 45% alloy is almost equal to that of pure Ni. This result indicates that the electron temperature, or average excess kinetic energy of electrons is lower in Ni rich alloys.

Time-resolved luminescence spectra and luminescence decay profiles for Cu-Ni alloys

Figure 1. (a) Time-resolved luminescence spectra normalized by emissivity for Cu-Ni alloys. Each spectrum is normalized at their maxima and shifted vertically. (b) Luminescence decay profiles at 0.6 eV for alloys with x=0x = 0, 0.040.04, 0.100.10, 0.170.17, 0.300.30, 0.450.45, and 11. Each curve is normalized to 1000 at maxima.

The composition dependences of the internal luminescence intensity evaluated at t=0t = 0 ps are plotted for 0.90.9, 0.60.6 and 0.40.4 eV in Figure 2(a). For 0.90.9 eV, the intensity decreases to 1/160 when the Ni content is increased from 0 to 45%, while the decrease is milder for 0.60.6 and 0.40.4 eV.

Internal luminescence intensity, IDOS, and their dependence on Ni content

Figure 2. (a) Ni concentration dependence of the internal luminescence intensity at 0.9, 0.6 and 0.4 eV. The solid curves show phenomenological fittings to the experimental data. (b) The IDOS in Eq. (3) as a function of the Ni content. The spin splitting occurs around 40 Ni%. εmin⁡\varepsilon_{\min} and εmax⁡\varepsilon_{\max} were set to be −1.2-1.2 and 0.4 eV, respectively. The IDOS is normalized by its maximum. (c) Dependence of the internal luminescence intensity on IDOS. The dashed curve is an exponential fit with 17.4×IDOS−3.0317.4 \times\mathrm{IDOS}^{-3.03}.

In the previous work, we have demonstrated that the luminescence intensity is correlated with the inverse of the Drude damping rate [19]. This is based on a phenomenological model for the complex interplay between the surface plasmon polariton and hot electrons. However, the estimation of the Drude damping rate for Cu-Ni alloys is difficult due to the lack of the frequency-dependent optical constants. We believe that the underlying physics behind the instantaneous response is governed by the e-e scattering within the energy window around the Fermi level, where electrons with different energies are scattered to unoccupied states at elevated electron temperature conditions. Alternatively, we calculate the integrated DOS

IDOS⁡(εmin⁡,εmax⁡)=∫εmin⁡εmax⁡[D↑(ε)+D↓(ε)]dε,(3)\operatorname{IDOS}(\varepsilon_{\min},\varepsilon_{\max}) = \int_{\varepsilon_{\min}}^{\varepsilon_{\max}} [D_{\uparrow}(\varepsilon) + D_{\downarrow}(\varepsilon)]d\varepsilon, \tag*{(3)}

where Dσ(ε)D_{\sigma}(\varepsilon) is the electron DOS with energy ε\varepsilon and spin σ\sigma. This is a rough estimate of the number of the single-particle states involving the e-e scattering events in our excitation condition. We assume εmin⁡=−1.2\varepsilon_{\min} = -1.2 eV that is nearly equal to the photon energy of pump pulse (1.19 eV) with opposite sign and treat εmax⁡\varepsilon_{\max} as a parameter accounting for the number of excited electrons. We have considered εmax⁡=0.0,0.4,0.8,\varepsilon_{\max} = 0.0, 0.4, 0.8, and 1.21.2 eV and show our analysis for εmax⁡=0.4\varepsilon_{\max} = 0.4 eV below.

As shown in Figure 2(b), IDOS increases with the Ni content and the increase becomes moderate at higher Ni content. Figure 2(c) shows the observed luminescence intensity as a function of the calculated IDOS. We can see a good correlation between them, i.e., the intensity shows an IDOS−3.03^{-3.03} dependence. Note that the integrand of the e-e collision term in the Boltzmann transport equation is proportional to the cube of the electron DOS [38], and the e-e collision accelerates the electron thermalization [25, 28]. In addition, the luminescence intensity reflects the initial relaxation before the energy transfer to phonons [19]. Therefore, our model calculations demonstrate that the integrated value of the collision term of the Boltzmann equation is well approximated by the power function of IDOS, and allow to estimate the relevant energy window governing the relaxation process. When εmax⁡\varepsilon_{\max} is changed from 00 to 1.21.2 eV, the power law changes from −2.92-2.92 to −3.49-3.49 that are equal to −3-3 within an error of 17% (see Supplemental Material [31]). More quantitative discussion requires solving the Boltzmann equation, but such a study is beyond the scope of the present work.

Luminescence decay—Time development of luminescence intensity was investigated at 0.6 eV and shown in Figure 1(b), where the curves are normalized to 1000 at their maxima. In contrast to the monotonic concentration dependence of luminescence intensity at 0.9 eV, it shows a complicated behavior. The slope between 0 and 0.5 ps decreases from pure Cu by adding Ni and keeps similar value from Ni 4% to 17 % (i.e., the relaxation is slowed). Above 17 %, it starts to increase and reaches maximum value at Ni 100% (i.e., the relaxation becomes faster). To analyze this behavior quantitatively, we decomposed the decay profile into an instantaneous response and an exponentially decaying component [19]. The decay profiles are well fitted by a sum of a delta function and an exponential function with convolution by the instrumental response function.

The decay rates Γexp⁡\Gamma_{\exp} of the exponential component at 0.6 eV are shown in Figure 3(a). By adding Ni to pure Cu, Γexp⁡\Gamma_{\exp} decreases first till Ni 17% and turns to increasing over 17%. We also plotted the calculated Γ\Gamma. To correct the underestimation for the decay rate, the factor 1.7 was multiplied to Γ\Gamma [19]. It is clear that the calculated Γ\Gamma is nonmonotinic function for Ni content. This shows that

Luminescence decay rate and second moment of the phonon spectra

Figure 3. (a) Luminescence decay rate at 0.6 eV as a function of Ni content. Red and blue circles indicate the experimental and calculated data, respectively. The factor 1.7 is multiplied to the calculated Γ\Gamma. (b) The second moment of the phonon spectra as a function of Ni content.

the luminescence decay rate corresponding to the cooling rate is dominated solely by e-ph coupling and that the disorder originating from alloying has minor contribution to the decay process.

To understand the origin of the nonmonotonic behavior, we calculated the second moment of the phonon spectra, i.e., ⟨ω2⟩=2∫F(ω)ωdω\langle\omega^2 \rangle= 2 \int F(\omega)\omega\mathrm{d}\omega for Cu-Ni alloys, where F(ω)F(\omega) is the phonon DOS. The ⟨ω2⟩\langle\omega^2 \rangle monotonically increases with Ni content, as shown in Figure 3(b). One should notice that the ⟨ω2⟩\langle\omega^2 \rangle is obtained if α2F(ω)\alpha^2F(\omega) is replaced with F(ω)F(\omega) in Eq. (2). The nonmonotonic behavior is not due to the redistribution of the phonon spectra, but due to the weakened α2F(ω)\alpha^2F(\omega). We speculate that in dilute alloys the matrix elements ∣g∣2|g|^2 and/or the phase space for the e-ph scattering are decreasing function of Ni concentration, resulting in an anomalous decrease in ⟨ω2⟩\langle\omega^2 \rangle. The α2F(ω)\alpha^2F(\omega) and the phonon DOS for Cu-Ni alloys are provided in Supplemental Material [31].

We calculated ⟨ω2⟩\langle\omega^2 \rangle for Cu7Z1\mathrm{Cu}_7\mathrm{Z}_1 with period 4 metals ZZ from K to Zn. Among them, Cu7Ni1\mathrm{Cu}_7\mathrm{Ni}_1 has the smallest value of ⟨ω2⟩\langle\omega^2 \rangle. We also calculated ⟨ω2⟩\langle\omega^2 \rangle for X7Y1\mathrm{X}_7\mathrm{Y}_1 with X=CuX = \mathrm{Cu}, Ag, Au and Y=NiY = \mathrm{Ni}, Pd, Pt, Cu, Ag, Au. A small inclusion of group 10 atoms always gives rise to a decrease in ⟨ω2⟩\langle\omega^2 \rangle. The smallest ⟨ω2⟩\langle\omega^2 \rangle is 16.1 meV2^2 for Au7Pt1\mathrm{Au}_7\mathrm{Pt}_1. Our systematic investigation suggests that the relaxation dynamics can be manipulated by alloying. These results are provided in Supplemental Material [31].

*Conclusion—*Luminescence intensity and decay profiles were measured in Cu-Ni alloys and the composition dependence was compared with a theoretical prediction. Nonmonotonic composition dependence of the decay rate was found below x=0.2x = 0.2. Elongation of the luminescence lifetime by adding transition metal to noble metal is a rather unexpected result, because the increase of DOS near the Fermi level accelerates relaxation due to e-ph scattering. We emphasize that the quantitative prediction of the electron relaxation dynamics will promote alloy design for desired electronic and optical properties, e.g. to realize longer lifetime. Size effect in nanocrystals may help to modify the Eliashberg function to further reduce the e-ph interaction strength, leading to improved quantum efficiency of nanoparticle imager. The inverse effect, i.e., an increase in the e-ph coupling by foreign atom doping, may enhance the superconducting transition temperature of metals.

We also found that the smaller the IDOS around the Fermi level, the larger the luminescence intensity. This may reflect the phase space available for the e-e scattering. This concept will be useful for controlling the brightness of alloys.

Acknowledgments

This work was supported by JSPS KAKENHI (Grant Number JP20K03823). T.S. is indebted to the Institute for Solid State Physics at the University of Tokyo for its support with the Joint Research Program. Some of the numerical calculations were done using the facilities of the Supercomputer Center, the Institute for Solid State Physics, the University of Tokyo, and the supercomputer “MASAMUNE-IMR” at the Center for Computational Materials Science, Institute for Materials Research, Tohoku University.

References

  1. [1]L. Vegard, Die Konstitution der Mischkristalle und die Raumfüllung der Atome, Z. Phys. 5, 17 (1921).DOI
  2. [2]S. Curtarolo, D. Morgan, and G. Ceder, Accuracy of ab initio methods in predicting the crystal structures of metals: A review of 80 binary alloys, Calphad 29, 163 (2005).arxiv.org/abs/cond-mat/0502465
  3. [3]A. B. Bhatia and D. E. Thornton, Structural Aspects of the Electrical Resistivity of Binary Alloys, Phys. Rev. B 2, 3004 (1970).DOI
  4. [4]T. Ying, H. Chi, M. Zheng, Z. Li, and C. Uher, Low-temperature electrical resistivity and thermal conductivity of binary magnesium alloys, Acta Mater. 80, 288 (2014).DOI
  5. [5]C. Wei, N. Antolin, O. D. Restrepo, W. Windl, and J.-C. Zhao, A general model for thermal and electrical conductivity of binary metallic systems, Acta Mater. 126, 272 (2017).DOI
  6. [6]P. James, O. Eriksson, B. Johansson, and I. A. Abrikosov, Calculated magnetic properties of binary alloys between Fe, Co, Ni, and Cu, Phys. Rev. B 59, 419 (1999).DOI
  7. [7]Y. Iwasaki, R. Sawada, E. Saitoh, and M. Ishida, Machine learning autonomous identification of magnetic alloys beyond the Slater-Pauling limit, Commun. Mater. 2, 31 (2021).DOI
  8. [8]C. Gong and M. S. Leite, Noble Metal Alloys for Plasmonics, ACS Photonics 3, 507 (2016).DOI
  9. [9]A. Giri, M. V. Tokina, O. V. Prezhdo, and P. E. Hopkins, Electron-phonon coupling and related transport properties of metals and intermetallic alloys from first principles, Mater. Today Phys. 12, 100175 (2020).DOI
  10. [10]S. A. Ahern, M. J. C. Martin and W. Sucksmith, The spontaneous magnetization of nickel+copper alloys, Proc. R. Soc. Lond. A 248, 145 (1958).DOI
  11. [11]K. Ikeda and K. Tanosaki, Origin of Anomalous Resistivity in Ni-Cu Alloys, Trans. JIM 25, 447 (1984).DOI
  12. [12]H. M. Ahmad and D. Greig, Electrical resistivity and thermopower of nickel-copper alloys, J. Phys. Colloques 35, C4-223 (1974).
  13. [13]J. Paras and A. Allanore, Evidence of ordering in Cu-Ni alloys from experimental electronic entropy measurements, Phys. Rev. B 109, 104206 (2024).arxiv.org/abs/2311.06603
  14. [14]T. Kasuya, Electrical Resistance of Ferromagnetic Metals, Prog. Theor. Phys. 16, 58 (1956).DOI
  15. [15]T. Suemoto, K. Yamanaka, N. Sugimoto, Observation of femtosecond infrared luminescence in gold, Phys. Rev. B 100, 125405 (2019); Phys. Rev. B 104, 119905(E) (2021).DOI
  16. [16]T. Suemoto, K. Yamanaka, N. Sugimoto, T. Otsu, S. Tani, Y. Kobayashi and T. Koyama, Effect of emissivity on ultrafast luminescence spectra in silver, J. Appl. Phys. 128, 203103 (2020).DOI
  17. [17]T. Suemoto, K. Yamanaka, N. Sugimoto, Y. Kobayashi, T. Otsu, S. Tani, and T. Koyama, Relaxation dynamics of hot electrons in the transition metals Au, Ag, Cu, Pt, Pd, and Ni studied by ultrafast luminescence spectroscopy, J. Appl. Phys. 130, 025101 (2021).DOI
  18. [18]T. Suemoto, S. Ono, A. Asahara, T. Okuno, T. Suzuki, K. Okazaki, S. Tani, and Y. Kobayashi, Observation of infrared interband luminescence in magnesium by femtosecond spectroscopy, J. Appl. Phys. 134, 163105 (2023).DOI
  19. [19]T. Suemoto, S. Ono, A. Asahara, T. Okuno, T. Suzuki, K. Okazaki, S. Tani, and Y. Kobayashi, Comprehensive Study on the Luminescence Properties of Elemental Metals, Phys. Rev. B 111, 035150 (2025).DOI
  20. [20]P. B. Allen, Theory of thermal relaxation of electrons in metals, Phys. Rev. Lett. 59, 1460 (1987).pubmed.ncbi.nlm.nih.gov/10035240
  21. [21]V. V. Kabanov and A. S. Alexandrov, Electron relaxation in metals: Theory and exact analytical solutions, Phys. Rev. B 78, 174514 (2008).DOI
  22. [22]S. D. Brorson, A. Kazerounian, J. S. Moodera, D. W. Face, T. K. Cheng, E. P. Ippen, M. S. Dresselhaus, and G. Dresselhaus, Femtosecond room-temperature measurement of the electron-phonon coupling constant γ in metallic superconductors, Phys. Rev. Lett. 64, 2172 (1990).pubmed.ncbi.nlm.nih.gov/10041602
  23. [23]M. Obergefell and J. Demsar, Tracking the time evolution of the electron distribution function in copper by femtosecond broadband optical spectroscopy, Phys. Rev. Lett. 124, 037401 (2020).
  24. [24]C. Gadermaier, A. S. Alexandrov, V. V. Kabanov, P. Kusar, T. Mertelj, X. Yao, C. Manzoni, D. Brida, G. Cerullo, and D. Mihailovic, Electron-phonon coupling in high-temperature cuprate superconductors determined from electron relaxation rates, Phys. Rev. Lett. 105, 257001 (2010).
  25. [25]B. Y. Mueller and B. Rethefeld, Relaxation dynamics in laser-excited metals under nonequilibrium conditions, Phys. Rev. B 87, 035139 (2013).DOI
  26. [26]L. Waldecker, R. Bertoni, R. Ernstorfer, and J. Vorberger, Electron-phonon coupling and energy flow in a simple metal beyond the two-temperature approximation, Phys. Rev. X 6, 021003 (2016).
  27. [27]P. Maldonado, K. Carva, M. Flammer, and P. M. Openeer, Theory of out-of-equilibrium ultrafast relaxation dynamics in metals, Phys. Rev. B 96, 174439 (2017).DOI
  28. [28]S. Ono, Thermalization in simple metals: Role of electron-phonon and phonon-phonon scattering, Phys. Rev. B 97, 054310 (2018).DOI
  29. [29]G. A. Manoukian, O. Kizilkaya, S. Lendinez, L. D. B. Manuel, T. R. Leite, Karunya S. Shirali, W. A. Shelton, P. T. Sprunger, J. B. Baxter, and K. M. McPeak, Emergent properties from CuPd alloy films under near-infrared excitation, J. Chem. Phys. 157, 174702 (2022).
  30. [30]E. Thareja, K. M. McPeak, P. T. Sprunger, I. Vekhter, and W. A. Shelton, Trends in hot carrier distribution for disordered noble-transition metal alloys, J. Phys.: Condens. Matter 36, 335701 (2024).DOI
  31. [31]See Supplemental Material at (URL) for infrared absorptivity spectra, electron and phonon DOS, Eliashberg function, and lattice parameters of Cu-Ni alloys; the luminescence intensity versus IDOS for various energy window; and Eliashberg function of Cu₇Z₁ and X₇Y₁ alloys with period 4 metals Z, noble metals X, and transition and noble metals Y.
  32. [32]C. Chen and S. P. Ong, A universal graph deep learning interatomic potential for the periodic table, Nat. Comput. Sci. 2, 718 (2022).DOI
  33. [33]P. Giannozzi, O. Andreussi, T. Brumme, O. Bunau, M. B. Nardelli, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, and M. Cococcioni et al., Advanced capabilities for materials modelling with Quantum ESPRESSO, J. Phys.: Condens. Matter 29, 465901 (2017).
  34. [34]J. P. Perdew, K. Burke, and M. Ernzerhof, Generalized Gradient Approximation Made Simple, Phys. Rev. Lett. 77, 3865 (1996).pubmed.ncbi.nlm.nih.gov/10062328
  35. [35]A. Dal Corso, Pseudopotentials periodic table: From H to Pu, Computational Material Science 95, 337 (2014).DOI
  36. [36]N. Marzari, D. Vanderbilt, A. De Vita, and M. C. Payne, Thermal contraction and disordering of the Al(110) surface, Phys. Rev. Lett. 82, 3296 (1999).DOI
  37. [37]S. Baroni, S. Gironcoli, A. A. Dal Corso, and P. Giannozzi, Phonons and related crystal properties from density-functional perturbation theory, Rev. Mod. Phys. 73, 515 (2001).DOI
  38. [38]S. Ono and T. Suemoto, Ultrafast photoluminescence in metals: Theory and its application to silver, Phys. Rev. B 102, 024308 (2020).DOI

Paper details

vlnc:W_4kjpcm6j
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents