Ultrafast laser filament induced breakdown spectroscopy is a very promising method for remote material detection. We present characteristics of plasmas generated in a metal target by laser filaments in air. Our measurements show that the temperature of the ablation plasma is clamped along the filament channel due to intensity clamping in a filament. Nevertheless, significant changes in radiation intensity are noticeable, and this is essentially due to variation in the number density of emitting atoms. The present results also explain the near absence of ion emission but strong atomic neutral emission from plumes produced during fs LIBS in air.
© 2015 Optical Society of America
Filamentation of ultrafast lasers, resulting from a dynamic interplay between optical-Kerr effect and free electrons generated by multiphoton and tunneling ionization processes, has been a subject of interest, especially in the context of remote sensing . Self-guided filaments generated during ultrafast laser propagation are narrow beams of laser light with a nearly constant beam-waist over distances significantly exceeding the Rayleigh length. The robustness of the ultrafast laser filaments leads to numerous applications include remote sensing of atmospheric gases and aerosols, lightning control, laser-induced spectroscopy, coherent anti-stokes Raman scattering, and the generation of THz radiation . The supercontinuum radiation generated during filamentation process in air is also an ideal light source for absorption LIDAR [2,3]. Durand et al.  recently demonstrated filamentation up to a distance 1 km.
Laser induced breakdown spectroscopy (LIBS) has been shown to be a versatile technique for stand-off detection in hostile environments such as the identification of nuclear radioactive waste [5,6]. However, in conventional LIBS, which relies on nanosecond laser pulses, the diffraction and atmospheric turbulence effects ultimately limit the maximum distance achievable for remote sensing and nonlinear effects do not impact beam propagation or focusing. Filamentation of femtosecond laser pulses, capable of achieving intensities in the range of 1013 W/cm2, is sufficient to ablate and excite solid samples at large stand-off distances. Compared to focused nanosecond laser pulses, the filaments can propagate over very large distances without suffering from diffraction effects, demonstrating its potential for remote sensing.
LIBS using fs lasers has been studied either using sharply focused beams or using filamentation [7–11]. Though it has been reported that filamentation LIBS (fLIBS) is a very promising method for long-distance material detection, the properties of fLIBS plumes are not well understood. It is well-known that the properties of any laser generated plasmas strongly depend on laser irradiation conditions . A better understanding of the plume properties and their control will ultimately be important for improving the figures of merit of any analytical technique including fLIBS. The present study aims to characterize the plasma generated on a metal target by filaments generated using loosely focused fs pulses at various locations along the filament channel. Our results show that the temperature of the plasma generated by the fLIBS is clamped along the filament channel and the changes in signal intensity can be attributed mainly due to changes in number density of emitting species.
2. Experimental details
In our experiments, the filaments were formed by a loosely focused linear-polarized fs laser beam from a chirped pulse amplification Ti:sapphire laser system producing ~30 fs pulses at 10 Hz repetition rate, and having a central wavelength of ~800 nm. A laser pulse energy of 17 mJ was used for the present work, and all experiments were performed in ambient air at room temperature. Filaments were approximately 2.5 m in length and generated using a 4 m focal length plano-convex lens positioned on a 2 m optical rail. The use of a loose focusing geometry was required by laboratory space constraints since much larger distances would have been required for filamentation resulting from laser self-focusing alone. A brass target was used and its surface was exposed at various locations along the filament channel to generate a plasma plume. A plano-convex lens with a focal length of 5 cm and aperture 2.2 cm was positioned at an angle ~25 degrees with respect to target normal for gathering the plasma emission and a 400 µm multimode fiber optic cable was used to direct the emission light to an echelle Spectrograph (Catalina Scientific, Model EMU 120/65). An intensified CCD (ICCD) was used for detection and the combination of spectrograph and ICCD provided a spectral resolution better than 12,000. All spectral measurements were made in atmospheric air and in a space integrated manner. To keep the collection efficiency constant, the target and collection optics were held fixed and the focusing lens was moved along an optical rail that was well aligned to keep the filament-target intersection fixed as well.
3. Results and Discussion
Self-focusing occurs when the input laser power Pin exceeds the critical power (Pcr) which is given by Equations (1-4) embody key elements of the formation and stabilization of laser filaments, but are more qualitative than quantitative. Equation (4) shows that higher electron density can be generated within the filamentation plasma column by increasing the intensity of the filamenting laser pulse, which is consistent previous reports . The filaments, typically carry 10-15% of total pulse energy (~a few mJ), are clad by the rest of the laser energy which acts as a photon bath or energy reservoir . According to Rodriguez et al. , when the laser pulse power reaches 10-100 Pcr, multiple filaments can co-exist in a propagating filament. In that scenario, the filament beam will propagate nonlinearly with a turbulent exchange between nucleating and dying filaments and the photon bath itself . Hence, the implicit proportionality of the electron density on clamped intensity given in Eq. (4) is not exact considering two other key features are not represented: the photon bath outside the filament and formation of multiple filaments.
A distinct aspect of these filaments is the availability of light intensity far above the ablation threshold of most materials that exists along several meters of the laser propagating channel, which far exceeds the Rayleigh range. In our experiments, the beam diameter (at 1/e2) of the incident Ti:sapphire beam was 6 mm, giving a Rayleigh range of 0.415 m for a 4 m focal length lens and we have measured emission signals from a filament for over 1.65 m. Typical plasma emission spectra obtained in the visible region at various locations along the filament channel are given in Fig. 1. The distances given in the figure correspond to the measurement locations with respect to the 4 m focusing lens. The identified prominent transitions are marked in Fig. 1 and are mainly contributed by Cu and Zn neutral lines. It has been reported previously that fs LIBS predominantly provides neutral atomic plumes in contrast to the strong ion emission seen in nanosecond LIBS [18, 19].
We evaluated the changes in plasma emission intensity along the filament channel and results are given in Fig. 2 for Cu I 510 nm (3d94s2-3d104p) transition. The emission intensity from the plasma generated using filaments showed significant variation along the channel and the maximum is found at ~3.40 m from the lens with the signal levels falling off either direction along the channel from the optimal emission region. The signal levels recorded at the geometrical focus (f = 4 m) is found to be nearly 2 orders less compared to the emission signal level corresponding to the optimal position (~3.4 m), although the measurable plasma emission is evident throughout the studied channel length (~1.7 m). This indicates that the properties of filamentation assisted plasma will be different from traditional LIBS plasmas generated using short focal length lenses where highest emission intensity can be expected at or near the geometrical focus.
The significant variation in line emission intensity along the filament channel is linked to changes in plasma properties. Assuming local thermodynamic equilibrium (LTE) exists within the filament LIBS plume, the intensity of line radiation Imn for a transition from mth to nth level is given by 6]. Hence the temperature and density measurements were made in ablation plasmas generated at various locations along filament channel using optical emission spectroscopy (OES) which is a reasonably accurate and non-intrusive method for measuring various plasma parameters. Temperature calculations were accomplished by using the Boltzmann plot method assuming the plasma is in LTE. The emission lines used for generating Boltzmann plot include Cu I transition at 427.51, 465.11, 510.55, 515.32, 521.82 nm. The estimated temperature of the LIBS plume at various locations along the filament channel length is given in Fig. 3. The measurements were taken with 100 ns delay after the onset of plasma generation and with 2 μs integration time. Figure 3 clearly shows the measured temperature of the plasmas at various locations of the filament channel is constant within the error bars; or in other words, the temperature of the filament assisted ablation plasma is clamped.
Even though the temperatures of the ablation plasmas generated using filaments are clamped, the line emission intensity is found to vary significantly along the channel. Based on Eq. (5), apart from the temperature, the density of the emitting species can also affect the intensity of a line radiation. The population level of a transition is influenced by electron density of the plasma which is strongly dependent on both the excitation (electron impact) and relaxation (recombination) processes. The changing electron density is an indicator of the change in atomic number density. So, electron density measurements are obtained for plasmas generated along the channel length. Stark broadened line profiles of the Zn I spectral line centered at 481 nm (4s5s3S1-4s4p3P2) is used to infer the electron density and results are given in Fig. 4. The electron impact parameter values are obtained from ref . The measured density of the plasma generated along the filament channel approximately follows the changes in emission intensity profile. These results indicate that the variation in LIBS emission intensity generated by the filament assisted plasma is correlated with the electron density. However, the recorded emission signal intensity change is significantly higher (more than two orders) compared to measured electron density variation (factor of 3) along the filament channel. Both the temperature and density measurements given in Figs. 3 and 4 are time-integrated, and it is well known that the characteristics of all laser-produced plasmas are highly dynamic. Hence the values given in Figs. 3 and 4 should be considered as the average temperature and density during plasma lifetime. Considering the dynamic nature of the plasma parameters and its relation to excitation and de-excitation mechanisms, the direct relation of emission intensity of a line radiation to electron density of the plasma requires further investigation.
Measurements have also been made on the time evolution of temperature and density at certain locations along the filament channel. These results are given in Fig. 5. The lens-target distances selected include 3.45 m, which exhibits the optimal LIBS signal intensity, and two positions (3.05 and 3.85 m) located either side of the maximum intensity. The temperature estimate shows a maximum ~8500 ± 500 K at 100 ns, and drops to 6500 ± 500 K at 2 µs. The temperature also shows nearly constant values at various locations in the filament channel. The electron density exhibits similar time dependence at all three probed distances, whereas its initial value is highest at a lens-target distance of 3.45 m lens-target distance. This corresponds to the position in the filament channel where plasma emission signal intensity is optimal.
The constant temperature of ablation plumes created using filaments along the filament channel can be attributed to laser intensity clamping. However, this intensity clamping can also happen even for sharply focused (SF) laser beam using shorter focal length lens in air although the length of the filament column is limited . So, spectral measurements were taken by focusing 17 mJ pulses using a 0.2 m focal length lens. The spectral measurements (not given) showed that the line emission intensities as well as the plasma persistence increase considerably with the use of sharply focused beam instead of using loosely focused beam filaments. The temporal evolutions of Cu I emission line intensity at 510 nm are given in Fig. 6 for fIBS and SFLIBS which show the persistence of sharply focused beam produced plumes are significantly higher. For example, the recorded spectral features persist ~2 μs for filament assisted plasmas, while in the case of SF beams spectral features can be seen even after 5 μs. The measured time evolution of temperature and density of the plasma generated using SF system are given in Fig. 7, which shows a rise in the fundamental parameters compared to the properties of filament assisted plasma. For example, the measured temperature and density for SF beam generated plasmas at early time (~100 ns) are ~9600 ± 400 K and 6.4 ± 0.4 × 1017 cm−3, while the respective values for filament assisted LIBS are ~8700 ± 500 K and 5 ± 0.3 × 1017 cm−3. Hence, the improved persistence of various species in SF laser plume can be directly related to hotter and denser plasma generation. The higher temperature and density recorded for SF beam plasmas may be due to increased laser-target coupling. It has been reported that for tightly focused geometry, the peak intensities can exceed the clamping intensity value by at least one order of magnitude . However, it has to be mentioned that in SF laser beam the intensity clipping may be resulting from a one-off stripping of trailing pulse energy due to excessive plasma defocussing by the “overdense” plasma created by the leading part of the pulse and it could be different from stable filaments formed in loosely focused beams. Recently Valenzuela et al.  showed that material removal rate by filament assisted ablation is smaller compared to SF beam generated plasma, where higher clamped intensity was available for ablation.
The emission from both SFLIBS and fLIBS plasmas in air environment is predominantly from excited neutral species (atomic plume) regardless of focusing conditions. The observation of atomic plume during fs LIBS is strikingly different from ns LIBS where laser-plasma coupling leads to further heating and ionization, which in turn increase the continuum background emission (free-free and free-bound) along with emission from higher excited levels including ions [19, 24]. For fs-LA, Coulomb explosion, phase explosion, fragmentation and thermal vaporization are the main mechanisms responsible for the ablation . One of the reasons proposed for the occurrence of an atomic plume during fs LA is thermal vaporization, as temperatures of the emitted species are close to the vaporization point of the bulk and such species are far enough below the surface that ionization by the laser pulse is minimized [26, 27]. The observation of atomic plumes in SFLIBS or fLIBS in air could also be related to temperature clamping caused by intensity clamping in a filament. This is also consistent with our recent results which showed significant variation in ionization fraction (which strongly depends on plasma temperature) for sharply focused fs laser plasmas during a lens-scan experiment performed in vacuum conditions where intensity clamping is absent .
The changes in signal intensity at various locations along the filament channel can be correlated to atomic number density, which is directly connected to ablation efficiency. Weidman et al.  noticed the craters generated during filament assisted ablation are larger in diameter than the filament core (typically ~100 µm) and explained that the laser energy contained in the energy reservoir may also be used for ablation. Hence the ablation efficiency during filament assisted plasma generation depends both on clamped filamentation intensity and the rest of the energy distributed in the filament reservoir. They also noticed the ablation crater volume is more or less constant in a large propagation distances (~30 m). However, the present results show significant changes in plasma emission signal intensity as well as atomic density although the temperature is clamped. This indicates that the differences in focusing conditions and the laser energy levels may also affect both the properties of the filaments and subsequent generated plasmas. It is worth pointing out that by increasing the intensity of the filamenting laser pulse, a higher electron density for the filamentation plasma can be achieved  (ref. Equation (4)), and hence the electron plasma will have a stronger effect to counter the self-focusing effect of a more intense laser beam. Because of this dynamic balance between self-focusing and plasma defocusing, initial conditions (such as strong focusing, loose focusing, or non-focusing geometry to initiate filaments in air) are important as they ultimately determine the filament diameter, intensity, electron plasma density as well as the properties of fLIBS.
In this letter we have shown that the temperature of filament assisted ablation plasmas in the filament propagation channel is clamped due to intensity clamping. However changes in spectral intensity are seen along the filament channel and this is correlated to the changes in the plasma density. Filament-assisted plasma also shows lower temperature and density compared to sharply focused fs beam under similar laser energy conditions. The present results also explain the occurrence of atomic plume using fs generated plasmas in air, where laser intensity clamping controls the plasma temperature.
This work was supported by the DOE/NNSA Office of Nonproliferation and Verification Research and Development (NA-22). Pacific Northwest National Laboratory, a multi-program national laboratory operated by Battelle for the U.S. Department of Energy under Contract DE-AC05-76RL01830. The authors thank to Dr. Burt Beardsley for technical help as well as lending Echelle spectrograph and Dr. Bret D. Cannon for valuable discussions while preparing the manuscript.
References and links
2. L. Bergé, S. Skupin, R. Nuter, J. Kasparian, and J. P. Wolf, “Ultrashort filaments of light in weakly ionized, optically transparent media,” Rep. Prog. Phys. 70(10), 1633–1713 (2007). [CrossRef]
4. M. Durand, A. Houard, B. Prade, A. Mysyrowicz, A. Durécu, B. Moreau, D. Fleury, O. Vasseur, H. Borchert, K. Diener, R. Schmitt, F. Théberge, M. Chateauneuf, J. F. Daigle, and J. Dubois, “Kilometer range filamentation,” Opt. Express 21(22), 26836–26845 (2013). [CrossRef] [PubMed]
5. K. K. Ayyalasomayajula, V. Dikshit, F. Y. Yueh, J. P. Singh, and L. T. Smith, “Quantitative analysis of slurry sample by laser-induced breakdown spectroscopy,” Anal. Bioanal. Chem. 400(10), 3315–3322 (2011). [CrossRef] [PubMed]
6. S. S. Harilal, P. K. Diwakar, N. L. LaHaye, and M. C. Phillips, “Spatio-temporal evolution of uranium emission in laser-produced plasmas,” Spectrochem. Acta B 111, 1–7 (2015). [CrossRef]
7. A. Valenzuela, C. Munson, A. Porwitzky, M. Weidman, and M. Richardson, “Comparison between geometrically focused pulses versus filaments in femtosecond laser ablation of steel and titanium alloys,” Appl. Phys. B 116(2), 485–491 (2014). [CrossRef]
8. K. Stelmaszczyk, P. Rohwetter, G. Mejean, J. Yu, E. Salmon, J. Kasparian, R. Ackermann, J. P. Wolf, and L. Woste, “Long-distance remote laser-induced breakdown spectroscopy using filamentation in air,” Appl. Phys. Lett. 85(18), 3977–3979 (2004). [CrossRef]
9. S. Tzortzakis, D. Anglos, and D. Gray, “Ultraviolet laser filaments for remote laser-induced breakdown spectroscopy (LIBS) analysis: applications in cultural heritage monitoring,” Opt. Lett. 31(8), 1139–1141 (2006). [CrossRef] [PubMed]
10. E. J. Judge, G. Heck, E. B. Cerkez, and R. J. Levis, “Discrimination of Composite Graphite Samples Using Remote Filament-Induced Breakdown Spectroscopy,” Anal. Chem. 81(7), 2658–2663 (2009). [CrossRef] [PubMed]
11. P. Rohwetter, J. Yu, G. Mejean, K. Stelmaszczyk, E. Salmon, J. Kasparian, J. P. Wolf, and L. Woste, “Remote LIBS with ultrashort pulses: characteristics in picosecond and femtosecond regimes,” J. Anal. At. Spectrom. 19, 437–444 (2004). [CrossRef]
12. S. S. Harilal, P. K. Diwakar, M. P. Polek, and M. C. Phillips, “Morphological changes in ultrafast laser ablation plumes with varying spot size,” Opt. Express 23(12), 15608–15615 (2015). [CrossRef] [PubMed]
13. A. Couairon and A. Mysyrowicz, “Femtosecond filamentation in transparent media,” Phys. Rep. 441(2-4), 47–189 (2007). [CrossRef]
14. J. Bernhardt, W. Liu, F. Theberge, H. L. Xu, J. F. Daigle, M. Chateauneuf, J. Dubois, and S. L. Chin, “Spectroscopic analysis of femtosecond laser plasma filament in air,” Opt. Commun. 281(5), 1268–1274 (2008). [CrossRef]
15. J. V. Moloney, “Intense femtosecond pulse propagation with applications,” Proc. SPIE 6261, 626102 (2006). [CrossRef]
16. M. Rodriguez, R. Bourayou, G. Méjean, J. Kasparian, J. Yu, E. Salmon, A. Scholz, B. Stecklum, J. Eislöffel, U. Laux, A. P. Hatzes, R. Sauerbrey, L. Wöste, and J. P. Wolf, “Kilometer-range nonlinear propagation of femtosecond laser pulses,” Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 69(3), 036607 (2004). [CrossRef] [PubMed]
17. M. Mlejnek, M. Kolesik, J. V. Moloney, and E. M. Wright, “Optically turbulent femtosecond light guide in air,” Phys. Rev. Lett. 83(15), 2938–2941 (1999). [CrossRef]
19. S. S. Harilal, N. Farid, J. F. Freeman, P. K. Diwakar, N. L. LaHaye, and A. Hassanein, “Background gas collisional effects on expanding fs and ns laser ablation plumes,” Appl. Phys., A Mater. Sci. Process. 117(1), 319–326 (2014). [CrossRef]
20. H. R. Griem, Principles of Plasma Spectroscopy (Cambridge University Press, Cambridge, 1997).
21. M. S. Dimitrijevic and S. Sahal-Brechot, “Stark broadening of neutral zinc spectral lines,” Astron. Astrophys. Sup. 140, 193–196 (1999). [CrossRef]
22. A. V. Afonasenko, D. V. Apeksimov, Y. E. Geints, S. S. Golik, A. M. Kabanov, and A. A. Zemlyanov, “Study of filamentation dynamics of ultrashort laser radiation in air: beam diameter effect,” J. Opt. 16(10), 105204 (2014). [CrossRef]
24. S. S. Harilal, B. E. Brumfield, and M. C. Phillips, “Lifecycle of laser-produced air sparks,” Phys. Plasmas 22(6), 063301 (2015). [CrossRef]
25. E. G. Gamaly, Femtosecond Laser-Matter Interaction: Theory, Experiments and Applications (Pan Stanford, Singapore, 2011).
26. R. F. W. Herrmann, J. Gerlach, and E. E. B. Campbell, “Ultrashort pulse laser ablation of silicon: an MD simulation study,” Appl. Phys., A Mater. Sci. Process. 66(1), 35–42 (1998). [CrossRef]
27. R. Stoian, D. Ashkenasi, A. Rosenfeld, and E. E. B. Campbell, “Coulomb explosion in ultrashort pulsed laser ablation of Al2O3,” Phys. Rev. B 62(19), 13167–13173 (2000). [CrossRef]
28. M. Weidman, K. Lim, M. Ramme, M. Durand, M. Baudelet, and M. Richardson, “Stand-off filament-induced ablation of gallium arsenide,” Appl. Phys. Lett. 101(3), 034101 (2012). [CrossRef]