Ultrafast optical excitation induced transient modification on the energy-band structures in tungsten, which resulted in the expansion and shift toward the Fermi-level of d-band. This process led to enhanced interband transitions at reduced photon energies. Meanwhile, enhanced interband excitation led to increased electron density above the Fermi level, resulting in enhanced optical scattering by localized surface plasmon resonance (LSPR). These mechanisms are responsible for balancing the direct heating of bulk electrons by optical pulses. The corresponding studies not only revealed the physics for the electronic dynamics in tungsten carbide, but also proposed that the modified electronic and electron-phononic interactions are one of the important responsible mechanisms for the enhanced laser-damage threshold of the hard-metal coating. Furthermore, the nanostructured hard-metal coating integrates functions of enhancement of the damage-threshold and anti-reflection coating, which is important for exploring new tools or materials in laser engineering.
© 2016 Optical Society of America
Cemented carbide is a composite material made of refractory metallic carbides (WC, TiC, TaC, etc.) and cemented by transition metals (Fe, Co, Ni), where the WC-Co system is the most widely employed configuration [1–3]. Owing to their excellent performances in hardness, modulus, strength, and wear resistance, the sintered cemented carbides have been long used as tool materials for metal machining, mining, tunneling, rock drilling, and for exploring high-performance die materials and wear-resistant parts.
Tungsten carbide has been deposited onto the surface of metallic components with a coating thickness of tens to hundreds of microns, providing much improved wear and corrosion resistance for substrate work piece [4,5]. The WC-Co coating is generally fabricated by thermal spraying, which is significantly effective in the surface protection and repairing of large engineering components in the fields of aviation, steel metallurgy, oil drilling, printing, and electric power generation. Recently, cemented carbide coatings have been produced using nanostructured WC-Co powders as the starting material . The resultant WC-Co coatings possess further improved hardness, toughness, and wear/corrosion resistance due to the nanoscale effects.
Furthermore, the nanostructured WC-Co coatings also exhibit high resistance to high-power laser damage, shielding the base metal from being modified by the laser radiation, as observed in our experiments. These performances may not only trigger extended photophysical and electronic investigations on such compounds, but also introduce new applications of the hard-metal coatings. The electronic dynamics under optical excitation were found to have played important roles in achieving such performances. In this work, we investigate the electronic dynamics in the WC-Co coating under strong laser pulse excitation using ultrafast spectroscopy based on reflective femtosecond transient absorption (TA). Multiple plasmonic electron-photon interactions led to balanced transient optical processes and consequently reduced electronic heating effects, which resulted from band-structure modulation in the composite material, optical scattering by enhanced localized surface plasmon resonance, and mitigated laser heating by the nanostructured metal surface. Such mechanisms made contributions both to the enhanced laser-damage threshold and to the reduced optical reflection by the metal surface. This WC-based coating can be applied effectively in laser-machining work piece, optical engineering with high damage-resistance, or in protection of equipment from strong laser irradiation.
2. Experimental methods
2.1 Nanostructured cemented carbide thermal-sprayed onto low-carbon steel
The WC-Co composite powder synthesized by in situ reactions of tungsten oxide, cobalt oxide, and carbon black powders was used as the starting material , which was then mixed with 1.2wt.% polyvinyl alcohol (PVA), 1.5wt.% polyethylene glycol (PEG), and 30% distilled water to form a slurry. The slurry was spray-dried and then heat-treated at 1200 °C for 5 hours. After the crushing and screening processes, the agglomerated WC-Co nanoparticles in the size ranging from 15 to 45 μm were obtained, producing the thermal-spray feedstock powder. In the thermal spray process, the WC-Co feedstock particles were heated and accelerated in the spraying flame (~3000 °C) and then impacted, spread, and solidified onto the substrate, forming the coating layer on the low-carbon steel.
Figures 1(a) and 1(b) show the scanning electron microscopic (SEM) and transmission electron microscopic (TEM) images of the prepared WC coating, respectively, where the dominant grain size ranges from sub-100 nm to larger than 300 nm. The selected area diffraction patterns (SADPs) are shown on the right panel of Fig. 1(b) for sites A and B, as indicated in the left panel, which correspond to crystalline WC nanoparticle and amorphous Co–binder phase, respectively.
2.2 Transient spectroscopic characterization
Transient absorption spectroscopic measurements were performed using femtosecond pump-probe, where a Ti:sapphire amplifier at 800 nm supplies 150-fs pump pulses with a repetition rate of 1 kHz. A portion of the 800-nm pulses were sent to a quartz cell containing heavy water with a thickness of 1 mm to produce supercontinuum in a spectral range from shorter than 300 nm to longer than 1200 nm, which was employed as the probe. The pump beam has a diameter of about 5 mm on the sample surface, whereas, the probe was focused onto the sample into a spot smaller than 400 µm in diameter at the center of the pump spot. The probe pulses reflected by the metal surface were sent to a spectrometer to measure the modulation dynamics at different delays from the pump pulses. In the practical pump-probe measurement, the pump pulse was delay-adjusted to approach the probe. Thus, for a positively chirped probe pulse, the pump pulse overlapped shorter-wavelength probe pulse earlier than the longer. Transient absorption spectrum (ΔA) was measured as a function of time delay (Δτ) between the pump and probe pulses.
3. Ultrafast electronic excitations in WC by femtosecond laser pulses
3.1 Excitation of bulk electrons and modulation on the light reflection
The free electrons excited by femtosecond light pulses in metals generally evolve in three stages . The first stage involves pure electronic processes or electron-electron (e-e) scattering processes taking place within the first 500 fs. This process is followed by the electron-phonon (e-p) interaction within a few picoseconds in the second stage and the phonon-phonon (p-p) interaction lasting longer than hundreds of picoseconds in the third stage. The hot electron gas produced by femtosecond optical excitation and its interaction with phonons enhances the reflection or reduces the transient absorption of light. Following the two-temperature model [9,10], above mechanisms can be explained well by: ΔR∝α⋅ΔTe + β⋅ΔTl, where ΔR is the change in the reflectance due to the optical excitation induced electronic heating, ΔTe and ΔTl are the increase in the temperatures of the electrons and the lattices, α and β are the linear coefficients. Due to electronic heating, enhanced reflection (ΔR>0) and reduced reflective optical extinction (ΔA<0) will be observed in transient absorption.
3.2 Modification on the electronic band-structures of tungsten
Although the electronic band structures of WC have been modified relative to pure tungsten (W) by the addition of C into the crystalline lattices, the performance of the valence electrons of the d-band still resembles to a large extent that of W. An important modulation lies in the shift of the electron density from W to C and in the shift of the Fermi level (EF) to lower energy levels in WC with respect to W. Thus, a wider unfilled portion of the d-band is produced in WC as compared to W and the total band structure of W is broadened by the addition of C due to the mixing of C-2p with W-5d and W-6s bands .
Furthermore, electronic heating by femtosecond pulses induces an expansion and a shift toward higher energies of the d-band of W, which leads to an increase of the screening effect with the augmentation of the electronic temperature, as has been investigated theoretically in . This process is illustrated schematically in Fig. 2(a), where the broadening and the shift of d-band results from the laser-pulse-excitation induced electronic heating in W with an electronic temperature increase of ΔTe. As a result, the interband transition requires a reduced photon energy (hν2<hν1), which may correspond to a higher density of states (N) within the d-band. This scheme implies enhanced interaction between optical electric field and free electrons.
Based on above physics, WC has a higher-located d-band than W and femtosecond laser excitation induces expansion and further shift toward higher levels of the d-band. These are mechanisms responsible for the strong modulation on the transient absorption spectroscopic response of the optical reflection by the WC coating. It is thus understandable that under strong electronic excitation by femtosecond laser pulses, the interband transitions will shift to the red and become stronger with increasing the excitation intensity. In the TA spectrum and dynamics, positive values of ΔA will be measured as function of time delay. Apparently, this mechanism is a pure electronic process, which should start earlier and evolve faster than the electron-phonon and phonon-phonon scattering processes, as shown in Fig. 2(b).
Figure 2(b) shows the TA dynamics measurement on a polished surface of bulk W with a pump intensity of about 25 μJ/cm2. We observe a “sharp” positive TA peak followed by a “fast” transition to an equally large negative signal. The negative TA results from the bulk electronic excitation, which evolves into the e-e and e-p processes. Clearly there are two stages of the evolution of the negative TA signal, as marked by the dashed green lines, which are located before and after a delay of Δτ≈2 ps. The earlier stage that evolves much more rapidly than the later. It is understandable that the electronic band-structure modulation (bs-m) is observed within the strong excitation process by the pump pulse. This process corresponds to a time scale of the cross-correlation between the pump and probe pulses, which is about τbs-m = 290 fs for a pump pulse duration of about τP≈150 fs. This dynamics is overlapped with the bulk e-e process within the first 500 fs and followed by the subsequent e-p and p-p processes in the bulk tungsten, as shown in Fig. 2(b). Obviously, the bulk e-e and e-p processes induced negative values of ΔA. Furthermore, the positive-to-negative transition actually involves the competition between the bs-m and e-e processes, as marked roughly in the inset of Fig. 2(b).
3.3 Modified inter-/intra-band transitions and enhanced localized surface plasmons
As shown in Fig. 1, the hard-metal coating is composed of Co-cemented WC nanoparticles in the size ranging from sub-100 nm to about 1 μm. Localized surface plasmon resonance (LSPR) can be thus excited when light is incident onto the nanostructured surface. As a result, the incident light beam will be strongly scattered or absorbed by the randomly distributed nanostructures and an amount of light energy will be spread onto a large amount of nanostructured sites on the top surface of the coating layer. In particular, the broadening and red-shift of the d-band due to strong optical excitation by the pump pulse enhances both the interband and intraband transitions, resulting in significantly enhanced optical scattering and absorption by LSPR. Furthermore, excitation by femtosecond pump pulses results in red shift of the spectrum of LSPR, the transient optical absorption/scattering by LSPR will be thus modulated in a style of the differential of the optical extinction spectrum of LSPR. Depending on the spectral shape of LSPR, negative side lobes (ΔA<0) centered around a positive peak (ΔA>0) will be generally observed in transient absorption . Moreover, the evolution dynamics of the TA signal also starts with a fast e-e (femtoseconds to sub-picosecond) interaction process followed by slower e-p (picoseconds) and p-p processes (>100 ps) in localized surface plasmons.
4. Results and discussions
4.1 Transient absorption spectroscopic response of WC coating and comparison with that of low-carbon steel
Figures 3(a) and 3(b) show the transient absorption spectra in the first 2.6 ps measured on the low-carbon steel without and with hard-metal coating, respectively. The blue-filled spectra correspond to negative TA data and the magenta-filled correspond to positive values of TA. This definition applies to all transient spectra in figures in the following sections. The TA spectra for the substrate in Fig. 3(a) were collected with a delay step of 50 fs within the first 1.0 ps and with steps of 100 fs in the delay range of 1.0~2.6 ps. A pump fluence of about 600 μJ/cm2 was employed in the TA measurements in Fig. 3. Negative values of ΔA were observed over the whole spectrum from 400 to 1000 nm in Fig. 3(a), although higher spectral intensity was observed close to the pump spectrum. The excited bulk electrons in the low-carbon steel by the pump laser pulse are responsible for the reduced absorption or enhanced reflection of the steel surface, which is a general phenomenon in laser-metal interaction . Largely increased amount of hot electrons induced enhanced optical reflection of the metal surface. The red curve in Fig. 3(c) shows the TA dynamics within the first 300 ps for the uncoated sample at a wavelength of 750 nm with a closer look within the first 5 ps shown in the inset of Fig. 3(c). The corresponding tail fitting to the exponential decay indicates two lifetimes with τ1≈4 ps and τ2≈90 ps, which imply dominant e-p and p-p coupling processes in the decay dynamics, respectively . In short, the laser-pulse induced heating of the bulk electrons is responsible for the broad-band negative TA signal measured on the steel substrate. We did not resolve the e-e scattering processes with a sub-ps lifetime due to the strong electronic heating process.
In contrast, the TA spectrum is dominated by a positive feature peaked at about 750 nm and at a delay smaller than 2.1 ps for the WC coating, as shown in Fig. 3(b), which is followed by a negative signal in the relatively long-term delay. The chirp in the probe pulse was not compensated, because it provides us with opportunities to identify plasmon resonance more clearly by the selective spectral features. This can be verified by a comparison between Figs. 3(a) and 3(b). For transient spectra without selective features, i.e. the data for low-carbon steel in Fig. 3(a), the chirp did not introduce much modification on the broadband spectrum, slight difference was observed only within the first 300 fs, negative spectra extending over the whole studied band (400~1000 nm) can be observed consistently. However, for the transient spectra in Fig. 3(b), positive TA spectra with an average bandwidth smaller than 50 nm exhibit strong selectivity at different time delays within the first 1 ps. The peak wavelength of the positive TA spectrum was tuned from 400 to nearly 900 nm as the time delay was increased from 0 to 2.1 ps, as shown in Fig. 3(b). Different bandwidths of the positive TA signal can be observed at different delays, implying different lifetimes at different wavelengths. Furthermore, these spectra have very small overlap in time domain at a fixed wavelength, implying very short lifetime of the TA signal at the corresponding wavelength. Such a tuning performance of the narrow-band TA spectrum enables much better observation of LSPR. Therefore, the TA signals within different spectral bands can be attributed to different laser-matter interaction processes. The blue curve in Fig. 3(c) shows the TA dynamics measured on the sample with WC coating at 750 nm. The inset of Fig. 3(c) magnifies the dynamic curves in the first 5 ps, making a clear comparison between the ultrafast optical responses of the two samples. The positive TA signal lasts for less than 1 ps, before it drops down below zero. Then, the negative signal by the blue curve evolves slowly to get overlapped with that by the red, implying that the phonon-phonon processes become dominating the dynamics beyond 100 ps. Furthermore, the minimum of the negative TA signal delays about 850 fs from the positive peak, as shown in the inset of Fig. 3(c), indicating that the bs-m and LSPR e-e processes evolved much earlier and faster than the bulk e-e and e-p process. Such an evolution dynamics provides opportunities for bs-m and LSPR to balance the direct bulk electronic heating processes.
The amplitude of the TA signal is much reduced in Fig. 3(b) as compared with that in Fig. 3(a), which is also verified by the comparison between the blue and red curved in Fig. 3(c). The TA signal for the substrate has an initial amplitude of −9 mOD, however, that for the WC-coated sample has a positive amplitude of about 3 mOD and a negative amplitude smaller than −2 mOD. Balance between the positive and negative values of ΔA has reduced the final amplitude of TA, which was based on the interactions between multiple mechanisms described in section 3.
Furthermore, we plot the TA spectra at delays (τ) of 350 fs, 650 fs, and 1.15 ps in Fig. 3(d). The spectra at τ = 350 and 650 fs have positive peaks at about 430 and 490 nm, respectively, with negative side lobes. Such features are very typical for the transient absorption spectrum of LSPR due to laser-excitation-induced red-shift , as has also been discussed in section 3.2. However, the transient spectrum changed at 1.15 ps, where a broad positive spectrum follows a negative one, which is different from those at τ = 350 and 650 fs. Therefore, we attribute the positive TA signals in the shorter-wavelength spectrum (e.g. <600 nm) to the LSPR of the WC nanostructures and the positive ones extending in the near infrared to the band-structure modulation by the pump laser pulse. Meanwhile, e-e and e-p processes induced by pump laser pulse heated free electron are always accompanying and competing with the LSPR and band-structure modulation processes. Figure 3(e) shows TA dynamics at 580 and 755 nm within the first 2 ps by the yellow and red curves, respectively, which have been normalized to the positive peak. The yellow curve shows a shorter lifetime of the positive signal and a larger negative amplitude than the red one, implying different photophysical processes. As has been discussed above, LSPR of small WC nanoparticles are the main mechanism responsible for the TA spectra at wavelengths shorter than 600 nm. However, smaller metallic nanostructures generally induce weaker plasmon resonance  with shorter lifetimes . Therefore, the experimental results in Fig. 3(e) support the proposed processes combining LSPR and bulk e-e/e-p processes. Above mechanisms can be further verified by the experimental results in section 4.2. It should be noted that the bs-m should have influenced the whole spectral band in above studies. Therefore, even the LSPR in the shorter-wavelength range involving intraband transitions has been enhanced by the bs-m process, which can be further verified by the experimental results in the following section.
4.2 Grain-size dependence of the TA spectra measured on WC coating
To resolve and verify further the proposed mechanisms in section 3, we perform TA measurements on the differently sized WC micro- and nano-structures. Figure 4(a) shows the SEM image of WC microstructures in size larger than 10 μm, implying that we do not expect LSPR in the visible spectral range. Figure 4(b) shows the SEM image of another sample consisting of WC particles in the size ranging from 100 nm to sub-1 μm. Figures 4(c) and 4(d) show the TA spectra measured on the samples in Figs. 4(a) and 4(b), respectively, where a pump fluence of about 600 μJ/cm2 has been employed.
For the measurement results in Fig. 4(c), the TA signals with positive values of ΔA can be observed only for wavelengths longer than 600 nm, which extends to longer than 950 nm. However, the positive TA spectra can be observed from shorter than 500 nm to longer than 900 nm in Fig. 4(d). The main difference between samples in Figs. 4(a) and 4(b) is the size of the WC particles, where LSPR in the visible can only be excited for nanostructured WC in Fig. 4(b). Therefore, LSPR-related processes are responsible for the TA spectra at wavelengths shorter than 600 nm. Since the mean grain size in Fig. 4(a) is larger than 10 μm, we do not need to consider LSPR effects in the visible. However, the e-e and e-p processes should have induced enhanced reflection or negative values of ΔA, we have to think about the bs-m process, where the strong pump pulse has modulated the band-structure of WC, making the d-band expanded and shifted to higher-energy levels and enabling enhanced absorption of photons with lower energies. These positive TA spectra are also overlapped with bulk e-e and e-p processes.
4.3 Enhanced damage threshold by femtosecond laser pulses in WC coating
Mitigation of direction bulk electronic heating by distributed LSPR will undoubtedly increase the damage threshold of the coated work piece. Furthermore, optical extinction by LSPR is dominated by scattering of light for nanostructures larger than 100 nm, which has been employed to increase light-matter interaction distance and enhance light absorption by solar cells . In this work, such effects favor scattering of incident light energy into free space and result in reduction in the direct light-matter interaction, leading to increased laser-damage threshold. Two control experiments were performed by focusing femtosecond laser pulses onto the work piece of hard-metal coated low-carbon steel and pure low-carbon steel. The laser beam of femtosecond pulses with an average power of 100 mW, a repetition rate of 1 kHz, a pulse length of about 150 fs were focused into a spot with a diameter of about 300 μm at FWHM, corresponding to a peak intensity of about 103 GW/cm2. Figure 5 shows optical microscopic images of the laser-irradiated samples of low-carbon steel with (a) and without (b) nanostructured hard-metal coating. The black regions in Fig. 5 are burns by the laser irradiation. These samples have been polished by a same process. Larger burn areas under the same laser irradiation intensity imply low damage thresholds. The nanostructured hard-metal coating on low-carbon steel base shows very tiny burns, possessing much higher damage threshold. This not only verified LSPR-based mechanisms, but also confirmed that the hard-metal coating supplied much enhanced protection of the work pieces from laser damage.
4.4 Understanding the mechanisms for the damage-threshold enhancement
The enhancement of the laser damage threshold due to the coating of nanostructured WC can be understood by evaluating the absorbed light energy and the net light intensity penetrating into the metal. For a polished metal surface without WC nanostructured coating, the femtosecond pulse experiences reflection and absorption processes. Thus, the absorbed light energy can be simply calculated by , where E0 is the incident pulse energy and ER is the reflected pulse energy. Assuming that the laser pulse with a pulse length of τP is focused into an area of A, the net light intensity that is interacting with the metal can be evaluated by . However, for a metal surface coated with a layer of nanostructured WC, the incident light pulse experiences enhanced reflection and enhanced optical scattering by plasmonic nanostructures, so that the absorbed light energy can be written as:, where ES is the scattered light energy. The factor σ is included to take into account the increased reflection due to the enlarged surface area through nanostructuring. The spot area within the laser focus remains unchanged, however, the nanostructures enlarged the interacting area due to the significantly enlarged total surface area, which can be simply evaluated by ρ⋅A with ρ much larger than 1. However, we also need to consider the different reflection properties using a factor γ between the metal substrate and the hard metal coating, therefore, we have σ = ρ⋅γ. However, in the studied spectrum, the value of γ does not have much difference for different metals in this work. Therefore, we also have σ>1. Consequently, the laser intensity penetrating into the nanostructured surface can be expressed as Apparently, is much smaller than and the laser intensity that is used to heat the metal surface is reduced significantly. This rough evaluation explains the much enhanced laser-damage threshold.
To demonstrate above mechanisms more clearly, we simulated the optical electric field below and above the metal/air interface when a similar femtosecond laser pulse at 800 nm was sent to a steel work piece with a smooth and a nanostructured surface along the normal of the sample surface, as shown in Figs. 6(a) and 6(b), respectively.
We need to note that the reflection by the nanostructures is not symmetric with the incidence about the normal of the substrate due to the complicated surface morphology, as can also be understood by looking at Fig. 6(b). Thus, the enhanced reflectance by the nanostructured coating cannot be measured directly using the far-field reflected light beam. In fact, it is difficult to evaluate the reflection and scattering by the nanostructured metal separately and they work together to send the light into the free space close to the surface of the metal, as shown in Fig. 6(b). We propose that the enhancement in the reflection and in the plasmonic scattering by the nanostructured WC coating have played important roles in the enhanced laser-damage threshold.
Additionally, when a laser beam is sent to the surface of a metal, the bulk electron gas is heated directly in a very thin layer on the top surface. The penetration depth of a laser beam to a bulk metal may be characterized by , where α is the absorption coefficient, ω is the light frequency, c is the velocity of light, and is the complex refractive index of metal. For low-carbon steel, we may employ the value of α for ferrum, which is 5.725 × 105/cm. This implies that the intensity of the laser beam at 800 nm will be reduced to 1/e at a depth of about 17.5 nm. This implies very fast and intensive heating of the bulk electrons concentrated in a very small volume by an ultrafast laser pulse. Such a penetration depth into plasmonic nanostructures on the metal surface will be definitely changed. The light intensity distribution along the penetration depth will also modify the ultrafast heating process and consequently make contributions to the improvement in laser-damage threshold. However, due to the lack of available parameters of WC required for such evaluations, we do not include investigations on this mechanism in this work.
We investigated ultrafast optical response of the cemented carbide coating under excitation of intensive femtosecond laser pulses. In addition to the generally observed strongly excited electronic and electron-phononic processes in bulk metals, we revealed the following processes that dominated the interaction between the nanostructured tungsten carbide and the femtosecond laser pulses in the early stage within picoseconds: (1) Modulation on the electronic band structure, which consists of the expansion and shift to higher energy of the d-band due to electronic heating by the laser pulses and lowering of the fermi level due to the insertion of C into W lattice, induced red-shifted and enhanced transient optical absorption. (2) Strong optical excitation by the pump pulses induced redshift of localized surface plasmon resonance of the WC nanostructure, leading to a differential-like modulation on the TA spectrum, which is based on the excitation of the localized electron-electron and electron-phonon interactions. (3) Enhanced scattering of light due to enhanced LSPR made contributions to the reduced reflective TA. Above processes have different lifetimes and dominate TA dynamics at different stages. These multifold processes balanced the laser-metal interactions, reducing optical reflection and screening direct electronic heating simultaneously, thus, contributing to the enhancement of the damage threshold of the WC-coated steel substrate.
Program 973 (2013CB922404); National Natural Science Foundation of China (NSFC) (11434016, 11574015, 51425101).
The authors acknowledge the Beijing Key Lab of Microstructure and Property of Advanced Materials for the support.
References and links
1. G. Upadhyaya, “Materials science of cemented carbides - an overview,” Mater. Des. 22(6), 483–489 (2001). [CrossRef]
2. J. M. Tarragó, J. J. Roa, V. Valle, J. M. Marshall, and L. Llanes, “Fracture and fatigue behavior of WC-Co and WC-CoNi cemented carbides,” Int. J. Refract. Met. Hard Mater. 49, 184–191 (2015). [CrossRef]
3. E. Soares, L. F. Malheiros, J. Sacramento, M. A. Valente, and F. J. Oliveira, “Microstructures and properties of submicrometer carbides obtained by conventional sintering,” J. Am. Ceram. Soc. 95, 951–961 (2012).
4. Z. Geng, S. Li, D. L. Duan, and Y. Liu, “Wear behaviour of WC-Co HVOF coatings at different temperatures in air and argon,” Wear 330–331, 348–353 (2015). [CrossRef]
5. P. Chivavibul, M. Watanabe, S. Kuroda, and K. Shinoda, “Effects of carbide size and Co content on the microstructure and mechanical properties of HVOF WC-Co coatings,” Surf. Coat. Tech. 202(3), 509–521 (2007). [CrossRef]
6. G. J. Yang, P. H. Gao, C. X. Li, and C. J. Li, “Simultaneous strengthening and toughening effects in WC-(nanoWC-Co),” Scr. Mater. 66(10), 777–780 (2012). [CrossRef]
7. X. Wang, X. Song, X. Liu, X. Liu, H. Wang, and C. Zhou, “Orientation relationship in WC-Co composite nanoparticles synthesized by in situ reactions,” Nanotechnology 26(14), 145705 (2015). [CrossRef] [PubMed]
8. J.-Y. Bigot, V. Halté, J.-C. Merle, and A. Daunois, “Electron dynamics in metallic nanoparticles,” Chem. Phys. 251(1-3), 181–203 (2000). [CrossRef]
9. M. I. Kaganov, I. M. Lifshitz, and L. V. Tanatarov, “Relaxation between electrons and the Crystalline Lattice,” Sov. Phys. JETP 4, 173–177 (1957).
10. H. D. Wang, W. G. Ma, Z. Y. Guo, X. Zhang, and W. Wang, “Experimental study of ultra-fast conduction process in metals using femtosecond laser thermal reflection method,” J. Eng. Thermophys. 32, 465–468 (2011).
11. R. J. Colton, J.-T. J. Huang, and J. W. Rabalais, “Electronic structure of tungsten carbide and its catalytic behavior,” Chem. Phys. Lett. 34(2), 337–339 (1975). [CrossRef]
12. E. Bévillon, J. P. Colombier, V. Recoules, and R. Stoian, “Free-electron properties of metals under ultrafast laser-induced electron-phonon nonequilibrium: A first-principles study,” Phys. Rev. B 89(11), 115117 (2014). [CrossRef]
13. X. P. Zhang, B. Q. Sun, J. M. Hodgkiss, and R. H. Friend, “Tunable ultrafast optical switching via waveguided gold nanowires,” Adv. Mater. 20(23), 4455–4459 (2008). [CrossRef]
14. C. A. Antonelli, B. Perrin, B. C. Daly, and D. G. Cahill, “Characterization of mechanical and thermal properties using ultrafast optical metrology,” MRS Bull. 31(08), 607–613 (2006). [CrossRef]
16. J.-H. Klein-Wiele, P. Simon, and H.-G. Rubahn, “Size-dependent plasmon lifetimes and electron-phonon coupling time constants for surface bound Na clusters,” Phys. Rev. Lett. 80, 0319007 (1998).
18. R. Feynman, The Feynman Lectures on Physics, 2nd ed. (Addison-Wesley, 2005).