Frequency mixing in boron carbide laser ablation plasmas

Frequency mixing in boron carbide laser ablation plasmas

Applied Surface Science 336 (2015) 53–58 Contents lists available at ScienceDirect Applied Surface Science journal homepage: www.elsevier.com/locate...

1MB Sizes 1 Downloads 62 Views

Applied Surface Science 336 (2015) 53–58

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Frequency mixing in boron carbide laser ablation plasmas ˜ M. Oujja, A. Benítez-Canete, M. Sanz, I. Lopez-Quintas, M. Martín, R. de Nalda ∗ , M. Castillejo Instituto de Química Física Rocasolano, CSIC, C/Serrano 119, 28006 Madrid, Spain

a r t i c l e

i n f o

Article history: Received 1 July 2014 Received in revised form 19 September 2014 Accepted 20 September 2014 Available online 6 October 2014 Keywords: Nonlinear frequency mixing Harmonic generation Laser ablation plasma Optical emission spectroscopy Boron carbide

a b s t r a c t Nonlinear frequency mixing induced by a bichromatic field (1064 nm + 532 nm obtained from a Qswitched Nd:YAG laser) in a boron carbide (B4 C) plasma generated through laser ablation under vacuum is explored. A UV beam at the frequency of the fourth harmonic of the fundamental frequency (266 nm) was generated. The dependence of the efficiency of the process as function of the intensities of the driving lasers differs from the expected behavior for four-wave mixing, and point toward a six-wave mixing process. The frequency mixing process was strongly favored for parallel polarizations of the two driving beams. Through spatiotemporal mapping, the conditions for maximum efficiency were found for a significant delay from the ablation event (200 ns), when the medium is expected to be a lowionized plasma. No late components of the harmonic signal were detected, indicating a largely atomized medium. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Boron carbide (B4 C) is a material with outstanding properties, and applications in high-technology industries: one of the hardest materials known, it displays very high resistance to chemical agents, high melting point, low density and a high neutron absorption cross section [1]. Deposition of boron carbide thin films with controlled properties is intensely sought for, and several techniques like chemical vapor deposition (CVD) [2], laser-assisted CVD [3] or magnetron sputtering [4] have been attempted, but both the stoichiometry and crystallinity of the deposits are very sensitive to variations in the deposition conditions. Pulsed laser deposition (PLD) of boron carbide has been explored only to a limited extent [5–8], and may be of high value since laser ablation easily overcomes the difficulties associated with the high melting point of B4 C and allows for a broad range of deposition conditions through the control of the laser properties, together with the atmosphere, geometry and temperature. In this context, the in-situ study of laser ablation plasmas of B4 C is of interest as a guide for controlled thin film synthesis through PLD. Nonlinear optical processes in laser-produced plasmas were first observed in the 1970s [9], and the idea that nonlinear optical processes can be used as a diagnostic of complex media like plas-

∗ Corresponding author. Tel.: +34 917459553; fax: +34 915642431. E-mail address: [email protected] (R. de Nalda). http://dx.doi.org/10.1016/j.apsusc.2014.09.119 0169-4332/© 2014 Elsevier B.V. All rights reserved.

mas or gas flows was proposed in the pioneering works of Zheltikov et al. [10–12]. These studies are based on the use of tight focusing geometries where the driving fields are employed as probes of local density in these media that act as 3D microscopes. In some of these works [11], resonances have been used for enhanced sensitivity to particular species. In less stringent conditions, recent work by our group has shown that low-order harmonic generation is sensitive to the presence of atoms, molecules, clusters and nanoparticles in a laser ablation plasma [13–15], and can, in some cases, be used as a probe of their density. A significant body of work has been constructed over the last years on high-order harmonic generation in laser ablation plasmas ([16,17] and references therein), most of which have concentrated on high-order harmonic generation of intense ultrashort laser pulses. Harmonic generation in plasmas using bichromatic driving fields has also been studied in references [18–20]. Generation of odd low-order harmonics of an IR driving laser beam in B4 C plasmas was explored by our group in previous work [15]. Symmetry prevents the emission of even harmonics in centersymmetric systems if a single-color driving beam is employed. This restriction is lifted if bichromatic driving fields are employed, which have the additional advantage that they permit the exploitation of a broader range of resonances in the nonlinear species. This work explores the generation of a frequency-mixed beam at the frequency of the fourth harmonic of the fundamental, resulting from a parametric process in a NIR ns laser-produced plasma of the B4 C material. For that purpose, a bichromatic field composed of a NIR fundamental at 1064 nm and a visible second harmonic at 532 nm

54

M. Oujja et al. / Applied Surface Science 336 (2015) 53–58

Fig. 1. Scheme of the experimental setup employed for frequency mixing experiments in a B4 C plasma (a), and typical result (b). The ablation laser interacts with the B4 C target at normal incidence and generates an ablation plasma that expands away from the surface. The driving laser is first frequency-doubled and later split in two arms in order to generate a bichromatic (1064 nm + 532 nm) driving beam. This beam interacts with the plasma after an electronically controlled delay, and generates a range of harmonics and frequency-mixed beams that copropagate with the driving beam. A set of dichroic mirrors discards the fundamental beams and drives the beam of interest toward the spectrometer, at the exit of which sits an ICCD device. The typical result is the spectrum shown in part (b), where the FH 266 nm beam can be detected, together with dominant spontaneous emission in this range, corresponding to the excited B atom.

was employed as the driving beam, and in these conditions a UV beam at 266 nm, which will be called FH for “fourth harmonic”, was generated collinearly. The spatiotemporal characterization of this FH beam provided a map of the presence of nonlinear optical species in the plasma, and the behavior as a function of the intensities of both driving beams provided clues as to the nature of the nonlinear frequency mixing process responsible for FH generation [21]. 2. Material and methods The setup employed was a variation of a system that had been reported previously [13]. It is schematically depicted in Fig. 1(a). The B4 C target (from Tech Supplies LTD, purity 99.9%) was mounted on a rotating holder to minimize cratering, inside a vacuum chamber (<10−5 mbar). A plasma was generated near the surface of the target by laser ablation with the fundamental radiation of a Q-switched Nd:YAG laser (Spectra Physics, Quanta Ray Indi-HG, 1064 nm, 7 ns full width at half maximum (FWHM), 10 Hz) that typically delivered on-target fluences in the region of 3 J cm−2 at normal incidence. After an electronically controllable delay, a second Q-switched Nd:YAG laser system (Lotis TII LS-2147, 15 ns FWHM, 10 Hz) was fired as the driving source for harmonic generation and frequency mixing in the B4 C plasma. Its fundamental output at 1064 nm was first frequency doubled in a type I KDP crystal, yielding a 532 nm beam with 4% efficiency, and later split in two arms with dichroic mirrors. Separate energy and polarization control in each arm was possible with variable attenuators and wave retarders for each wavelength. The NIR (1064 nm) and visible (532 nm) beams were later recombined with another dichroic element so that they copropagated parallel to the surface of the target and toward the plasma. No additional elements for phase control were introduced, because the optical path difference of the NIR and visible driving beams was significantly longer than the coherence length of the Nd:YAG laser, so no fixed phase relation can be expected. The beams were focused

inside the plasma with a 20 cm focal length lens, attaining intensities in the region of 0.2 TW cm−2 and 3 GW cm−2 for the NIR and the visible beam, respectively. Some degree of chromatic focal mismatch (fvis–NIR ∼ 5 mm) was present but it was measured to be within the Rayleigh range (bvis ∼ 9 mm, bNIR ∼ 4 mm). The distance from the focal spot to the B4 C surface could be varied along the x-axis, but for most experiments it was fixed at 0.6 mm. Also, the position of the focus with respect to the plasma along the z axis (the driving laser propagation coordinate) could be scanned. In these conditions, a beam at 266 nm (FH), resulting from frequency mixing, was generated in copropagation with the driving beams, and was separated from them with two dichroic mirrors that drove it to the entrance of a spectrograph (Bentham, TMc300, 300 lines per millimeter grating), at the exit of which sat a time-gated ICCD camera (Andor Technologies, 2151). The gate was typically set at 100 ns width, synchronous with the driving laser. Spectra resulting from the accumulation over 125 laser shots were acquired in the spectral region of the vicinity of the FH wavelength, with the result that is shown in Fig. 1(b). It is important to note that the FH was detected only in the presence of both driving beams at 1064 nm and 532 nm. Together with the FH signal, the strong spontaneous emission of excited boron atoms in the plasma in the 2 S1/2 → 2 P3/2 transition [22] was visible at 249.8 nm. Additional experiments were carried out to obtain a separate diagnostic of the B4 C-ablated medium, which consisted of the detection of spontaneous emissions from electronically excited species formed in the ablation plasma. For these experiments, the driving laser was blocked, and an imaging lens was placed in the detection arm so that the image of the plasma was formed on the entrance slit plane of the spectrometer. This allowed us to obtain either 2D projections of the emissions, by using the diffraction grating of the spectrometer as a mirror (i.e. at zero order), or 1D spatially-resolved spectra. Temporal resolution was achieved by adequately gating the ICCD device.

M. Oujja et al. / Applied Surface Science 336 (2015) 53–58

Optical emission gate delay (ns) 1

Optical emission (arb. units)

FH intensity (arb. units)

(a)

55

0.1

100

Ablation-driving time delay (ns)

(b)

Fig. 2. Spontaneous emissions from the boron carbide plasma. (a) False-color images of the total visible emission from electronically excited species in a B4 C plasma generated by 1064-nm ablation, as a function of the delay with respect to the ablation event. The temporal width of the acquisition gate is 50 ns. The color code is not comparable among images, each has been normalized to its maximum value. The dashed vertical lines across the images indicates the region of interaction with the driving laser. Note that the vertical and horizontal scales are not comparable. The total vertical range is 7.5 mm in all cases. (b) Typical spontaneous emission spectrum of the B4 C plasma in the 220–460 spectral range, with assignment of the emitting species. This spectrum was acquired for a delay of 50 ns with respect to the ablation event and at a distance of 2 mm from the surface. (* ) stands for detection of the emission at 206.7 nm at its second diffraction order.

3. Results and discussion 3.1. Observation of spontaneous emissions from the laser ablation plasma We will first describe the spatially and temporally resolved measurement of spontaneous emissions from the B4 C plasma created by the ablation laser and in the absence of the driving laser. This provides information on the dynamics of electronically excited species that emit fluorescence in the full active spectral region of the spectrometer, which is set at zero order so that no spectral selection is applied. The images obtained in these measurements are shown in Fig. 2, where the delay value indicated in each frame sets the opening of the gate, which is always kept open for 50 ns. The total spatial window of observation along the plume expansion direction is 6 mm. The figure shows that the excited species constitute a fast-expanding medium, with the region of maximum luminosity moving away from the surface at a speed of the order of 105 ms−1 . Spectral analysis, of which the spectrum in the right of the figure is an example, revealed that spontaneous emission is dominated by transitions of the boron cation B+ , especially the intense 2s4f (3 F) → 2s3d(3 D) transition at 412.2 nm. In the visible range, a weak contribution from the C2 Swan bands could also be observed. At early times, and distances close to the surface, an intense emission from B++ (1s2 2p (2 P) → 1s2 2s (2 S), at 206.7 nm, detected at the second diffraction order) is also important, but dies off very fast with either delay or distance. There is also a contribution from either excited neutral B atoms (2s2 3s (2 S) → 2s2 2p (2 P)), emitting at 249.8 nm, or carbon atoms in their 2s2 2p3s (1 P) → 2s2 2p2 (1 S) transition at 247.8 nm. All these transitions possess oscillator strengths of the order of 108 s−1 , and therefore the radiative lifetimes of the excited states are below 10 ns. This means that

1000

FWHM 4.8 mm

-10

-5

0

5

z distance (mm)

10

0

1

2

x distance (mm)

3

Fig. 3. Spatiotemporal analysis of the FH emitted beam from the B4 C laser-generated plasma. (a) Temporal study. The black circles show the FH beam generated as a function of the delay between the ablation event and the arrival of the driving beam to the plasma, for an on-axis configuration (z = 0) at a distance of x = 0.6 mm from the surface of the target. In the same temporal scale, red triangles show the integrated spontaneous emission originating from the region of interaction. The grey band indicates the range of background fluctuation. (b) Spatial study. (Left panel) z-scan: intensity of the FH beam as a function of the position of the driving beam focus along the propagation axis z, for x = 0.6 mm. (Right panel) x-scan: intensity of the FH beam as a function of the distance of the driving beam to the surface of the target, for an on-axis configuration (z = 0). The delay was optimized for each position.

the speed estimated above (105 ms−1 ) cannot be considered as the speed of individual excited particles, which would be de-excited rapidly and close to the surface (<1 mm), but to an effective global speed of the excited medium, constituted by species that may gain electronic excitation far from the surface, mainly due to electron impact. The dashed lines across images indicate the approximate region of interaction with the driving laser in the frequency-mixing experiments described below. The result of integrating the emission signal obtained in this region is plotted in Fig. 3 as a function of time with empty circles. It seems clear that all light electronically excited species visible in the spontaneous emission spectra will be practically absent from the region of interaction with the driving laser (see next section) in less than 100 ns after the ablation event.

3.2. Frequency mixing in a boron carbide laser ablation plasma. We now turn to the behavior of the plasma as nonlinear medium for harmonic generation. The moderate intensities reached with the ns Nd:YAG laser (0.2 TW/cm2 ) restrict the study to low-order processes. The third and fifth harmonics of the NIR beam alone in a laser-produced B4 C plasma have been described in a previous publication [15]. The addition of the second driving beam, in the visible (532 nm), produces changes in the harmonic spectra, the most notable of which is the generation of light beams with frequencies that are even multiples of the frequency of the fundamental

56

M. Oujja et al. / Applied Surface Science 336 (2015) 53–58

(1064 nm). It has been checked that the visible driving beam alone does not produce any measurable harmonic generation. The relative phase between the fundamental (1064 nm) and the second harmonic (532 nm) driving beams could in principle play a role on the FH generation efficiency. In fact, control of this parameter has been reported in [19,23], where significant modulation in the intensity of the generated beam was found as a function of the relative phase. In our case, given that the difference in path length is notably longer than the coherence length of the Nd:YAG laser, it can be assumed that the results shown correspond to averaging over all possible phase differences, even on a shot-to-shot basis. The data presented below correspond to the description of the generation of a 266 nm, frequency mixed beam in the B4 C plasma medium, which we will call FH for convenience, even though the unambiguous determination of the process responsible for the emission of this beam has been elusive and will be discussed below. The experimental configuration described above, where the delay between the ablation event and the driving laser pulse can be electronically controlled in a broad temporal range, allows for a detailed exploration of the spatiotemporal evolution of the species responsible for harmonic emission or frequency mixing. In this case, for a fixed distance of the driving laser, 0.6 mm above the B4 C surface, the temporal evolution of the FH signal showed a simple behavior, which is shown in Fig. 3(a). The study was performed in conditions where the polarization directions of the two driving beams were at an intermediate angle, but it was checked that spatiotemporal distributions were independent of relative polarizations. As is clear from the figure, the distribution has a single, relatively narrow component, peaking around 200 ns. This delay would correspond to average velocities in the region of 3 × 103 ms−1 . Even though it is significantly lower than the average displacement of the spontaneous emitters per unit time (105 ms−1 ), it is expected to correspond to the arrival of light species (atoms, small molecules) with high kinetic energy content. Atomic C+ and B+ cations should be practically absent at the time of maximum FH generation, given typical velocities for ions, as well as electronically excited species, obtained through optical emission measurements, represented in Fig. 3(a) with red triangles. This allows us to identify the neutral B and C atoms, together with small molecules like C2 , as the most likely candidates to constitute the nonlinear medium in this case. The fact that only one temporal component appears under B4 C ablation at 1064 nm is consistent with an analogous finding, presented in [15], for third harmonic generation of B4 C plasmas. There, it was found that a second, delayed component in the temporal evolution of the third harmonic only appeared in the presence of a buffer gas (Kr or Xe). This second component was attributed to heavier species (clusters or nanoaggregates), whose formation is strongly favored if the ablation plasma expands in the presence of a buffer gas. The 2D spatially resolved FH generation was also studied by changing the focus position of the driving beam with respect to the axis of the plasma expansion (z direction) and the distance from the driving laser propagation axis to the surface of the target (x direction). These measurements provide indications of the regime in which the process takes place, the phase matching conditions and the nature of the plasma expansion. Whether the process takes place under strong or weak focusing conditions depends on the value of b/L, b being the Rayleigh range, and L, the width of the density distribution medium. The spatial analysis of the FH signal is presented in Fig. 3(b), where the left panel shows the result of the z-scan for an x distance of 0.6 mm, and the right panel shows an x-scan for the on-axis z configuration. The z-scan shows a single maximum with a FWHM of ∼5 mm. This value is much larger than the expected effective length of the plasma medium at a distance of 0.6 mm from the

surface (L ∼ 1 mm), which seems to indicate that it is a reflection of the intensity profile of the driving laser (with bNIR ∼ 4 mm) rather than the density profile of the nonlinear medium. This measurement confirms the idea that the experiment was conducted in soft focusing conditions, close to the plane wave limit. The right panel of Fig. 3(b) shows the efficiency of the FH generation as a function of the distance from the target surface (the delay was optimized for each case); the fall of the signal intensity can be fitted with a function of the form A x−2 , as is expected for near-isotropic expansion of the nonlinear plasma medium and in the plane wave limit [24]. Once the optimized temporal delay was found, the intensity of the FH beam was measured as a function of the intensity of each of the driving beams. This measurement is crucial to unravel the order of the parametric process responsible for FH generation, since, in the perturbative limit, the nonlinear polarization scales as the product of the electric fields involved. In this case, the relative polarizations of the two driving beams were set to be parallel. The lowest-order frequency mixing process that can generate 266 nm with a bichromatic (1064 nm + 532 nm) driving beam would be a sum-frequency process involving four waves: ωvis + 2ωNIR → ωFH (see Fig. 4(c), left panel); for this process, the expected behavior would be linear as a function of the green (532 nm) beam intensity, and quadratic as a function of the NIR (1064 nm) beam intensity. The experimental power laws are shown in Fig. 4. The left panel shows the FH signal as a function of the 1064 nm beam, for a fixed value of the 532 nm beam at 5 mJ. It is displayed on a log–log plot, with a fitted slope of 1.87 ± 0.08, reasonably compatible with the expected value of 2, corresponding to a quadratic dependence. The right panel shows the FH signal as the green beam energy is changed, while maintaining a fixed IR energy of 300 mJ. Clearly, the behavior is not characterized by a single power law; a fit to the lower intensity region yields a value of the fitted slope of 2.8 ± 0.3; this is surprising, and incompatible with a four-wave sum frequency mixing ωvis + 2ωNIR → ωFH process. As is the case in harmonic generation, parity conservation requires that the total number of driving photons involved by the two laser fields is odd [25]. Therefore, beyond the four-wave mixing mechanism, the lowest order process that would be allowed by symmetry would be a six-wave difference frequency mixing process of the form 3ωvis − 2ωNIR → ωFH [26] (see Fig. 4(c), right panel). This is a higher order process that should in principle be much less efficient than the four-wave mixing mechanism, especially for a relatively weak visible beam, but unlike the latter, it is consistent with the measured power law with respect to the green driving laser energy. It remains unclear why the fifth-order process seems to dominate over the third-order process in this case. Resonances that could play either a detrimental role for the third order process, or an enhancing role for the fifth order processes, have not been found for the atomic cations or neutrals of B and C, although they could exist for larger species. In particular, C2 is common in flames and plasmas of carbon-containing materials, and must be considered as an important candidate for FH generation. Degenerate four-wave mixing (DFWM, a difference frequency mixing process of third order) has been reported in C2 present in flames [27], and was shown to be very significantly enhanced by the resonances of the Swan bands in the 515 nm region ( = 0 bands). The 532 nm visible beam employed here would be in resonance with excited rovibrational transitions of the  = −1 bands of the C2 Swan system. Since C2 is one of the likely candidates for FH generation in the B4 C plasma, we cannot discard that this may be playing a role, although the reason why it would favor the fifth-order process vs. the third order process remains obscure. The collective, as opposed to the singlespecies response needs to be considered too. It is known that phase matching is much more favorable for difference frequency mixing schemes than for sum frequency mixing schemes [28,29], but

M. Oujja et al. / Applied Surface Science 336 (2015) 53–58

(a)

57

90 60

120

30

150

180

0

330

210

240

(b)

300 270

Fig. 5. FH beam intensity as a function of the relative polarization angle between the NIR and the visible driving beams. Angles are defined so that “0” and “180” correspond to parallel polarizations, and “90” and “270”, to perpendicular polarizations. The solid line is a cos2 function and is intended as a guide to the eye.

(c)

was much favored (7:1) when the polarizations of the two driving beams were parallel. The difference in sensitivity of the acquisition system as a function of the polarization of the FH beam, which may vary as the NIR driving field is rotated, was checked independently with a linearly polarized 266 nm laser beam, and was found to be within 15%; therefore, it does not introduce a significant bias in the measurement. Early works reported that the efficiency of frequency mixing processes is higher for parallel polarizations of the driving beams, relative to perpendicular polarizations [30,31], but a more complex picture emerged later. Kim et al. [32] demonstrated that, for some harmonics, perpendicular polarizations were favored, and this was a function of both the single-atom response (via the relative phase between the driving lasers) and collective effects (better or worse phase matching due to different degree of ionization). In the case studied here, the measurements show an unambiguous enhancement of the frequency mixing process for parallel polarizations of the driving beams. 4. Conclusions

Fig. 4. Experimental power laws for the FH generation process, and some possible mixing schemes responsible for FH generation. (a) Intensity of the FH signal as a function of the NIR driving beam energy, for a fixed value of the visible driving beam energy of 5 mJ, plotted in log–log scale. The line is a fit to the data, and shows nearquadratic character. (b) Intensity of the FH signal as a function of the visible driving beam energy, for a fixed value of the IR driving beam energy of 300 mJ, plotted in log–log scale. The lines are fits to the data in the low and high energy regions explored. A gradual change from near-cubic to near-linear dependence is found. (c) Sum-frequency mixing (left) and difference frequency mixing (right) schemes for FH generation.

in our view, this fact alone should not be expected to explain a reversal of the expected probabilities of the two processes. Further investigations would be required on this point. The intensity of the FH beam was measured as a function of the relative polarizations of the fundamental and second harmonic driving beams. For this purpose, the polarization of the NIR beam was rotated with a half-wave plate before recombination with the 532 nm beam. The result is shown in Fig. 5, where the “zero” angle means parallel polarization, and “90” is perpendicular polarization. As the Figure shows, the FH generation process in the B4 C plasma

A nonlinear optical frequency mixing process resulting in the emission of a UV beam (266 nm) from the mixture of a NIR (1064 nm) and a visible (532 nm) driving beams has been studied in a NIR laser-generated boron carbide plasma, in conditions close to the plane wave limit. Power laws point toward a six-wave mixing process as responsible for the emission, which is found to be notably favored by the use of parallel polarizations of the driving beams. A space- and time-resolved study mapped the abundance of nonlinear optical species in the plasma. Optimum generation occurs for a significant delay (200 ns), which indicates that a high degree of ionization in the plasma is detrimental to the efficiency of the generation process. Neutral atomic (B and C) and small molecular species (C2 ) have been identified as the main constituents of the nonlinear medium. No late temporal components were found, indicating a low proportion of aggregated species in the boron carbide plasma. Acknowledgements Funding has been provided by Ministry of Economy and Competitiveness, Spain (MINECO) under Projects CTQ2010-15680 and CTQ2013-43086. I.L.Q. thanks support from the FPI 2011 programme of MICINN. M.O. thanks CSIC for a contract. A. B. C. thanks

58

M. Oujja et al. / Applied Surface Science 336 (2015) 53–58

JAE-TEC 2010 program (CSIC) for a contract. Dr. Esther Rebollar and Prof. R.A. Ganeev are thanked for fruitful discussions. References [1] F. Thévenot, Boron carbide—a comprehensive review, J. Eur. Ceram. Soc. 6 (1990) 205. [2] Ali O. Sezer, J.I. Brand, Chemical vapor deposition of boron carbide, Mater. Sci. Eng., B 79 (2001) 191. [3] O. Conde, A.J. Silvestre, J.C. Oliveira, Influence of carbon content on the crystallographic structure of boron carbide films, Surf. Coat. Technol. 125 (2000) 141. [4] L.G. Jacobsohn, R.D. Averitt, C.J. Wetteland, R.K. Schulze, M. Nastasi, L.L. Daemen, Z. Jenei, P. Asoka-Kumar, Role of intericosahedral chains on the hardness of sputtered boron carbide films, Appl. Phys. Lett. 84 (2004) 4173. [5] O. Derkach, S. Pentzien, W. Kautek, Pulsed laser deposition of boron carbide: Spectroscopic study of the laser ablation plasma, SPIE Proc. 2991 (1997) 48. [6] W.J. Pan, J. Sun, H. Ling, N. Xu, Z.F. Ying, J.D. Wu, Preparation of thin films of carbon-based compounds, Appl. Surf. Sci. 218 (2003) 297. [7] D. Zemsky, R. Shneck, P.J. Dagdigian, I. Bar, Structure and morphology of pulsed laser deposited boron carbide films: influence of deposition geometry, J. Appl. Phys. 102 (2007) 104309. [8] H.A. Castillo, E. Restrepo-Parra, J.M. Vélez, W. de la Cruz, Substrate temperature influence on boron carbide coatings grown by the PLD technique, Surf. Coat. Technol. 205 (2011) 3607. [9] P. Lee, D.V. Giovanielli, R.P. Godwin, G.H. McCall, Harmonic generation and frequency mixing in laser-produced plasmas, Appl. Phys. Lett. 24 (1974) 406. [10] A.B. Fedotov, N.I. Koroteev, A.N. Naumov, D.A. Sidorov-Biryukov, A.M. Zheltikov, Coherent four-wave mixing in a laser-preproduced plasma: optical frequency conversion and two-dimensional mapping of atoms and ions, J. Nonlinear Opt. Phys. Mater. 6 (1997) 387. [11] D.A. Akimov, A.B. Fedotov, N.I. Koroteev, A.N. Naumov, D.A. Sidorov-Biryukov, A.M. Zheltikov, R.B. Miles, One-dimensional coherent four-wave mixing as a way to image the spatial distribution of atoms in a laser-produced plasma, Opt. Lett. 24 (1999) 478. [12] A.N. Naumov, D.A. Sidorov-Biryukov, A.B. Fedotov, A.M. Zheltikov, Thirdharmonic generation in focused beams as a method of 3D microscopy of a laser-produced plasma, Opt. Spectrosc. 90 (2001) 778. [13] M. Oujja, R. de Nalda, M. Lopez-Arias, R. Torres, J.P. Marangos, M. Castillejo, CaF2 ablation plumes as a source of CaF molecules for harmonic generation, Phys. Rev. A: At. Mol. Opt. Phys. 81 (2010) 043841. [14] R. de Nalda, M. Lopez-Arias, M. Sanz, M. Oujja, M. Castillejo, Harmonic generation in ablation plasmas of wide bandgap semiconductors, Phys. Chem. Chem. Phys. 13 (2011) 10755.

[15] I. Lopez-Quintas, M. Oujja, M. Sanz, M. Martin, R.A. Ganeev, M. Castillejo, Loworder harmonic generation in nanosecond laser ablation plasmas of carbon containing materials, Appl. Surf. Sci. 278 (2013) 33. [16] R.A. Ganeev, Generation of harmonics of laser radiation in plasmas, Laser Phys. Lett. 9 (2012) 175. [17] R.A. Ganeev, Harmonic generation in laser-produced plasmas containing atoms, ions and clusters: a review, J. Mod. Opt. 59 (2012) 409. [18] R.A. Ganeev, H. Singhal, P.A. Naik, I.A. Kulagin, P.V. Redkin, J.A. Chakera, M. Tayyab, R.A. Khan, P.D. Gupta, Enhancement of high-order harmonic generation using a two-color pump in plasma plumes, Phys. Rev. A: At. Mol. Opt. Phys. 80 (2009) 033845. [19] R.A. Ganeev, H. Singhal, P.A. Naik, J.A. Chakera, H.S. Vora, R.A. Khan, P.D. Gupta, Systematic studies of two-color pump-induced high-order harmonic generation in plasma plumes, Phys. Rev. A: At. Mol. Opt. Phys. 82 (2010) 053831. [20] R.A. Ganeev, G.S. Boltaev, T. Usmanov, Third and fourth harmonics generation in laser-induced periodic plasmas, Opt. Commun. 324 (2014) 114. [21] W. Theobald, C. Wülker, F.P. Schäfer, B.N. Chichkov, High-order harmonic generation in carbon vapor and low charged plasma, Opt. Commun. 120 (1995) 177. [22] G. Tachiev, C. Froese Fischer, Breit-Pauli energy levels, lifetimes and transition data: Boron-like spectra, J. Phys. B 33 (2000) 2419. [23] W.-J. Chen, R.-Y. Lin, W.-F. Chen, C.-K. Lee, C.-L. Pan, Coherent control of thirdharmonic generation by a waveform-controlled two-colour laser field, Phys. Rev. A: At. Mol. Opt. Phys. 10 (2013) 065401. [24] A. Lago, G. Hilber, R. Wallenstein, Optical-frequency conversion in gaseous media, Phys. Rev. A: At. Mol. Opt. Phys. 36 (1987) 3827. [25] M.B. Gaarde, P. Antoine, A. Perrson, B. Carré, A. L’Huillier, C.-G. Wahlström, High order tunable sum and difference frequency mixing in the XUV region, J. Phys. B: At. Mol. Opt. Phys. 29 (1996) L163. [26] L. Misoguti, I.P. Christov, S. Backus, M.M. Murnane, H.C. Kapteyn, Nonlinear wave-mixing processes in the extreme ultraviolet, Phys. Rev. A: At. Mol. Opt. Phys. 72 (2005) 063803. [27] C.F. Kaminski, I.G. Hughes, P. Ewart, Degenerate four-wave mixing spectroscopy and spectral simulation of C2 in an atmospheric pressure oxy-acetylene flame, J. Chem. Phys. 106 (1997) 5324. [28] P.L. Shkolnikov, A.E. Kaplan, A. Lago, Phase matching for large-scale frequency upconversion in plasma, Opt. Lett. 18 (1993) 1700. [29] S. Meyer, H. Eichmann, T. Menzel, S. Nolte, B. Wellegehausen, B.N. Chichkov, C. Momma, Phase-matched high-order difference-frequency mixing in plasmas, Phys. Rev. Lett. 76 (1996) 3336. [30] M.D. Perry, J.K. Crane, High-order harmonic emission from mixed fields, Phys. Rev. A: At. Mol. Opt. Phys. 48 (1993) R4051. [31] H. Eichmann, A. Egbert, S. Nolte, C. Momma, B. Wellegehausen, W. Becker, S. Long, J.K. McIver, Polarization-dependent high-order two-color mixing, Phys. Rev. A: At. Mol. Opt. Phys. 51 (1995) R3414. [32] I.J. Kim, C.M. Kim, H.T. Kim, G.H. Lee, Y.S. Lee, J.Y. Park, D.J. Cho, C.H. Nam, Highly efficient high-harmonic generation in an orthogonally polarized twocolor laser field, Phys. Rev. Lett. 94 (2005) 243901.