Using automated laser pulse temporal shaping we report on enhancing spectral emission characteristics of ablation plasmas produced by laser irradiation of brass on ultrafast time scales. For different input irradiance levels, control of both atomic and ionic species becomes possible concerning the yield and the excitation state. The improved energy coupling determined by tailored pulses induces material ejection with lower mechanical load that translates into hot gas-phase regions with higher excitation degrees and reduced particulates.
© 2010 Optical Society of America
The spectral characteristics of laser-induced plasmas in matter ablation conditions on ultrafast time scales play a paramount role in applications related to remote sensing, breakdown spectroscopy (LIBS), and pulsed laser deposition (PLD) techniques. Accurate elemental analysis can be done with minimal invasive effects by spectrally evaluating the ejected material [1–5]. At the same time the kinetic, excitation, or reactivity properties of the ablation plume determine the morphology, composition, and the structure of deposited films in femtosecond laser thin films deposition and material transfer techniques [6–9]. As they depend on the efficiency of various ablation mechanisms, these properties can be improved by judiciously distributing the excitation energy in the time domain at specific material relaxation rates, thus allowing better irradiation conditions.
For typical ablation ranges upon ultrashort pulse irradiation, fast laser heating leads to a swift succession of phase transitions and to a strong hydrodynamic motion accompanied by the reverse transport of a rarefaction wave [10, 11]. The complex thermal and mechanical dynamics induce a transient variation of the optical properties on ultrafast scales due to the rapid formation of thermal and density gradients. We can indicate for example the role of temperature in electronic collisions and related optical damping or density alterations between aggregational states in different time domains corresponding to the formation of dense, solid or low density phase expanding material. As the laser pulse acquires additional time-domain features, the reflectivity changes allow a modulated energetic coupling and determine the amount of energy deposited initially in the material . The relaxation of different quantities of energy on various times scales establishes the thermodynamic trajectories for the different ablated layers and can have interesting consequences for the material ejection . Of main interest for the ablation regime are either the disintegration of the supercritical fluid or the gas-phase nucleation in the metastable two phase coexistence region, as they are characterized by different thermo-mechanical behaviors. A key point becomes therefore the resulting temperature-pressure states and subsequent temperature and density gradients under laser exposure. This has consequences for the aggregational state of ejected matter, its density evolution, and its kinetic and energetic behavior, with far reaching implications for the behavior of ultrafast laser-generated plasmas.
The evolution of the ablated material, including the emissivity of the ejected species, depends on the amount and distribution of the energy stored in the material and it is drastically influenced by the laser fluence, wavelength, pulse duration, and the size of the irradiation spot. Improvements considering various criteria of plasma evolution and material transfer were seen by using multipulse sequences or post-irradiation plasma treatment [14–20]. As long as some information exists on the dynamics of phase transitions, specific pulse forms can be designed that exploit the transient material changes and the underlying thermal transport. However, considering the complexity of processes occurring between the initial excitation and the final structural transformation and their lack of coherence, the task of defining apriori the possible control factors becomes difficult. This indicates a strong requirement for more efficient improvement procedures, capable of optimizing ultrafast laser-induced processes even when the physical information at hand is rather scarce. Here beam engineering methods in the time domain can deliver a significant advantage.
Building on the above-described scenario we intend to show that by controlling the energy feedthrough on ultrafast scales [12, 21, 22] particular thermodynamic states can be achieved in the expanding material, carrying also an improved spectral signature. Using automated temporal pulse shaping integrated in adaptive loops excitation rates can be optimized to control the excitation degree and, to a certain extent, the emission characteristics of the ablation plume. Largely employed in coherent control studies for molecular dissociation and bond breaking , the technique is particularly powerful in regulating incoherent material transformations on mesoscopic to macroscopic scales characteristics to laser ablation. The effect was previously analyzed for silicon and aluminium samples [12, 21, 22], indicating kinetic and excitation control upon regulated laser radiation. It will be extended here to multicomponent metallic alloys (brass) [24–27] in view of the technologic and scientific interest posed by multicomponent samples in laser ablation and analytics experiments. Two main issues will be followed here: relative excitation of various species and absolute excitation and ionization degrees in the plume. We note that excitation and ionization control may significantly impact on current PLD and LIBS techniques.
The paper is organized as follows. The experimental section describes the methodological approach and the utilized instruments. The following sections describe the use of designer pulses and adaptive optimization procedures in different fluence regimes, intended to improve specific features of various emission products from ablated brass. These include, as indicated above, absolute and relative spectral emission yields, and nanoparticle ejection. The results of various temporal excitation pulse forms in thin film deposition techniques, particularly their morphologies, are accompanying the spectral emissivity study. Hydrodynamic simulations are used to illuminate specific thermodynamic influences leading to spectral control of the ablation products. Effects are identified in the probability of liquid nanoparticle ejection and the temperature of the plasma front. A conclusion section summarizes the results.
Brass (composition 62%Cu and 38%Zn) samples were irradiated with ultrashort near infrared Ti:Sapphire laser pulses (Fourier transform-limited duration 150 fs) generated by an 1 kHz, 800 nm oscillator-amplifier laser system. The choice was motivated by the different thermal and atomic properties of the components (e.g. higher phase transition points for Cu and higher ionization potential for Zn). The samples were placed in a vacuum chamber at a pressure of 10-5 Pa and irradiated at oblique incidence (45°) with p-polarized laser radiation down to an elliptical spot of approximately 7740μm2. The irradiation sequence contains 45 pulses in a row at a repetition rate of 1 kHz. A programmable temporal pulse shaping unit can generate arbitrary wave forms in a time window of 15 ps, inducing corresponding plasma plumes. The technique relies on spectral phase modulation via complex filtering of pulse frequency components in dispersive systems equipped with pixelated liquid-crystal spatial light modulators. Using a phase-only approach, the energy in the laser sequence is always conserved. The resulting pulses were characterized by a second order background free intensity cross-correlation technique using a short reference pulse retrieved from the laser beam before the shaping procedure . The spectral emissivity of the laser ablation plume was detected and quantified by a 300 to 1200 lines/mm grating spectrometer equipped with a time-gated intensified charge coupled device (ICCD) camera. An optical system, similar to that described in Ref. , collects the emitted light and focuses it on the entrance of an optical fiber coupled with the spectrometer in such a way that light is gathered from the most intense region of the plasma, close to the sample surface, as represented in Fig. 1(a). The optical signal corresponds thus to a time and space averaged signal of neutral or ionic emission. The spectral information is analyzed and, based on a maximization criteria of various spectral lines, a figure of merit (fitness) is assigned. The recorded spectrum corresponds to a recording time of 300 ns at a delay of 100 ns with respect to the pulse arrival and serves then as feedback for the closed loop created between the detection system and the pulse tailoring unit. The feedback loop is driven by an evolutionary strategy that manipulates discrete spectral phase masks on the modulator (and therefore the corresponding pulse forms) using genetic propagators to maximize the experimental output; an approach similar to the one used in Ref. . The optimization result is a laser pulse that provides the best fitness spectrum with respect to the imposed constraints discussed later in the text.
The irradiation procedure produces typical spectra of the plasma core as shown in Fig. 1(a,b) in the range of 300 – 700 nm. The spectral assignment indicates transitions schematically represented in Fig. 1(c) with complete spectral evaluation given in . The lines employed in this study were chosen to have a variety of upper levels, some of them more susceptible to be collisionally pumped and thus sensitive to the electronic temperature. In addition to integrated signals, we focus therefore respectively on the neutral transitions around 330 nm and on the ionic transitions located around 490 nm.
3.1. Material removal with designed laser pulses
A first spectral diagnosis is facilitated by irradiation with various input energies while employing short pulse (SP) laser radiation (corresponding to the Fourier transform-limited value of 150 fs), as the initial energy deposition defines to a great extent the follow-up behavior. As a function of the incident laser fluence the integrated neutral Cu-I and Zn-I signals or, respectively, ionic Cu-II, Zn-II emissions detected in the range of 300–700 nm show a non-uniform behavior depicted in Fig. 2(a,b), indicating several regions of interest. All these regions are well situated in the ablation regime above the modification threshold of F = 0.16 J/cm2 (measured by the spot regression method) and, as well, above the plasma detection threshold of F = 0.5 J/cm2. While improving the signal-to-noise ratio, the chosen fluences with rather elevated values with respect to the ablation threshold help in keeping the eventual elemental fractionation effects at high number of pulses per site at a low level . The low fluence region located between the observed plasma generation threshold of 0.5 J/cm2 and a value of 1.5 J/cm2 average fluence corresponds to a plasma with a dominant component consisting of neutral species, increasing in yield in the moderate fluence range and beginning to slow down its increase upon the onset of strong ionization (2.5 J/cm2). The spectroscopic temperature for the neutrals determined using Boltzmann regression on several Cu-I lines vary little with the fluence and stabilizes at a value of approximately 4000 K. This corresponds to a spatially and temporally averaged value and can only be used as a qualitative and relative property of different irradiation regimes. The increasing fluence is then inducing a higher yield of excited species by developing stronger thermal gradients and a higher volume of ablated material. In this case the fluence serves as an indicative parameter for establishing particular evolution regimes that will be discussed below.
In the respective regimes we have made an attempt to analyze the effect of a pulse shape on the ablation characteristics by employing simple designer pulses. This involves usually stretched pulses and double pulse sequences as simpler assumptions may be made regarding their influence on the material evolution. Attempting to modify the spectral signatures of atomic and ionic species, the underlying idea is to exploit the electronic relaxation and its effect on heat diffusion, the lattice heating dynamics on the scale of electron-phonon coupling, together with the potential generation of a liquid phase and a density gradient before the end of the pulse envelope. The subsequent hydrodynamic movement delivers reflectivity transients that can be exploited by the proposed double pulse series with adjustable separation or chirped pulses. The use of simplified pulses permits a first intuitive insight into the topological optimization space, as the individual pulse effects can be better followed as the material evolves.
Exploiting this simplicity, we have tested pulse lengths and separations of several picoseconds; more specifically double pulses (DP) with 14 ps separation and long pulses (LP) with 10 ps duration (FWHM) while, at the same time, monitoring neutral and ionic Cu and Zn emission. The respective delay ranges are the maximal domains given by the spectral resolution of the light modulator; note that particular phenomena may take place up to the nanosecond scale [30, 31]. The results are given as well in Fig. 2(a) and, respectively, Fig. 2(b) for defined fluence regimes and they allow the following observations. In the low and intermediate fluence ranges a swift increase of neutral emission was seen which levels out at approximately 2 J/cm2. The observed relative increase in the low fluence regime is rather similar for all the lines investigated in this work within the measurement uncertainties, suggesting that the average temperature in the neutral region does not vary significantly. The neutral emission (DP and LP) is superior to the SP case, with a higher multiplication factor corresponding to a similarly increasing difference between different observed lines, irrespective of their spectral position. The augmentation of the neutral emission becomes smaller for fluences superior to 2.5 J/cm2, saturating below the SP level. At this point a notable observation is related to the significant boost in the ion yield in the case of the tailored pulses as the neutrals stop increasing [Fig. 2(b)]. That shows that ionic augmentation is made on the costs of the neutral population and a reasonable factor could be related to the temperature upshifting with respect to the ionization potential in the plasma front. This also leads to ionic signals significantly higher than for SP, however a slight decrease in efficiency is observed at the highest fluences.
A comprehensive overview of DP and LP effects relative to SP is given in Fig. 3 for various pulse separations and lengths in the above-mentioned fluence regimes, observing this time restricted spectral domains around main lines instead of the whole range of 300 – 700 nm. The recorded spectral ranges for neutrals and ions are indicated on the figure and correspond to (a,e,v,w,x,y,z,aa) lines in Fig. 1(b,c). Both the LP and DP neutral signals are increasing in the low fluence regime [Fig. 3(a,b) for 1.3 J/cm2]. When the fluence is shifted in the moderate region (approximately 2.6 J/cm2), the neutral increase slows down and a significant manifold ion increase becomes visible [Fig. 3(c,d)], augmenting at different paces with the pulse duration or separation. As actually no SP ion yield is being seen by the spectroscopic method, this can be assigned as a strong ionic enhancement. At the same time the ionic emission increase is steeper for the double pulse sequences, showing already a notable difference for a delay of 1 ps but it saturates at levels less than those of the long pulse sequence. In the high fluence regime [Fig. 3(e,f) for 4.8 J/cm2] we notice a slight decrease of neutral emission below the corresponding SP level together with an observable increase of ionic emission for the LP and DP sequences. The ionic rise may slow down with further increasing the fluence, with a dynamics that is fluence dependent. These observations allow us to make a few first evaluation statements. They indicate that the pulse design allows to obtain neutral and, especially, ionic emission at fluences well below the SP level. At the same time they determine in the high fluence range ionic signals not attainable by SP radiation. The beneficial action of chosen pulse forms was already discussed for promoting strong gas-phase excitation [14,20,21,32,33] even in the presence of reduced ablation rates. Besides this, the tailored pulses (e.g. DP) can equally influence the angular distribution of the ablation products . The next question is if more performant pulses can be designed using adaptive optimization procedures.
3.2. Emissivity enhancement via adaptive optimization
The closed loop mentioned earlier in the experimental section was used to improve the plume fluorescence using a spectrally defined fitness . A first attempt was made to change the relative ratio of the different chemical species (Cu or Zn) present in the detected region at 4.2 J/cm2 using the lines of Cu-I(d) at 328.27 nm and Zn-I(e) at 330.3 nm. The relatively low sensitivity obtained within the experimental uncertainties allows us to believe that many spectral lines behave in a similar manner, indicating that a common temperature regulates the evolution of the different components. Particular relaxation dynamics that may affect the relative species sensitivity are averaged by the 300 ns detection gating window. This limits the possibility to selectively enhance the individual components by control of relative quantities of ejected material and requires special care in selecting the lines. It also serves as a further indication of the occurrence of homogeneous and congruent ablation in spite of different thermodynamic points and ionization potentials for Cu and Zn, with a selectivity that depends on the recombination dynamics of particular lines. No selective ablation, as usually expected in the case of slow, equilibrium thermal processes, was noticed. As it will be seen in the discussion section, the species may follow different spatial spreading, but behave according to similar temperature gradients.
The second type of optimization concerns the amount of excited species. The fact that a Fourier transform limited pulse does not optimize the ablation fluorescence was already indicated  and reconfirmed by the use of designed pulse forms in the previous sections. The follow-up idea is whether the feedback loop can provide the irradiation features that can further alter this behavior. The optimization runs were performed at different fluence regimes and had the purpose of maximizing the neutral yield in the low fluence regime or the ionic components in the high fluence domain, using a certain range of criteria defined below.
Specifically, this second experimental procedure involves the maximization of different fitness parameters involving neutral lines in low fluence regimes, absolute yield associated to ionic lines at higher fluences, or ionic increase relative to the neutral signal. Figure 4 shows the results of the adaptive approach on neutral and ionic lines in various fluence regimes. An initial regime at F = 1.2 J/cm2 was used to enhance the overall, mainly neutral optical signal, as consequences are relevant for spectroscopy and elemental analysis. A feedback signal based on Cu-I and Zn-I lines [rays (a to f) in Fig. 1(b,c)] was used to measure the global intensity within a spectral window of 66 nm centered at 330 nm [see Fig. 4(a)]. The fitness was chosen in the form of f 1 = (ΣI neutrals)2. The optimized form OP1 is a noisy intensity envelope extending over 14 ps [Fig. 4(b)]. The aspect is dominated by multipeaks separated by a relatively regular ps spacing. The multipeak structure can originate from relatively high amount of spectral phase modulation delivered to the pulse in a discrete manner. The emission increase factor was measured to be around 2.5, similar to that previously observed with LP and DP sequences.
The second feedback type concerns both absolute and relative ion yields [Fig. 4(c–f)]. A higher fluence regime at F = 4.2 J/cm2 was used, where plasma emission is divided between neutral and ionic species. The fitness involves the Cu-II and Zn-II emission rays in a 6 nm domain centered at 493 nm [(v-z), (aa-ac) in Fig. 1(b,c)]. Two types of fitness parameters were defined, the first one being related to the absolute ion emission intensity f 2–1 = (ΣI ions)2, while the second involves the ratio between the ionic emission and the neutral emission [Zn-I(s) line at 481.05 nm] f 2–2 = (ΣIions)2 /(ΣIneutrals)2. Similar optimal forms (OP2-1, OP2-2) were obtained in the two cases; picosecond sequences extended on a 15 ps scale [Fig. 4(d) and (f) respectively]. Both optimized forms deliver similar species improvement results, indicating an intrinsic coupling between the ionic and neutral populations. The yield augmentation result closely resembles the designer pulses results presented before. This shows that a large class of solutions can be defined and the parameter space is not very selective, as expected from a temperature-driven behavior. This is particularly true for quasi-linear materials where the solid to liquid transitions and the collisional electronic effects on the coupling efficiency do not alter significantly the optical properties as compared to density variations upon exposure. The optimization results in addition to the observations related to the designer pulses show the following fact. The solution space appears to be defined by the condition that the pulse sequence energy is sufficiently high and distributed in such a way that a phase transition can be prepared with the front part and that it can also enable significant absorption of the trailing part in the emerging hydrodynamic movement. It is also interesting to note that in this regime, while the ionic population augments, the neutral one decreases as would be expected from a temperature shift closer to the ionization potential.
A final observation relates to the specific threshold for species emission. If the neutral emission threshold seems quasi-similar for the short and optimized sequences (Fig. 2), once in ablation regime, the SP sequence requires more energy to obtain, if possible at all, signal levels comparable with the OP sequences. A factor of two in the energy requirements for ions onset is noticed between SP and OP, suggesting a more efficient energy coupling in the expanding phase for the optimized distribution. This is a relevant aspect for applications requiring a highest signal level with a minimal energy consumption and less invasive to the material such as remote spectroscopy applications.
As relevance for PLD was equally suggested, a short note can be added in that concerns deposited CuZn films in the conditions discussed above. We have deposited films with various pulse shapes including the optimal pulse sequences. The films were formed by 52 min exposure at 1 kHz pulse repetition rates and show thicknesses well in the sub-micron range. The films also show an excess in Cu (above 70%Cu) with respect to the original target, with variations related to the measurement position, fluence and pulse shape. The optimal pulses lead to larger Cu distributions (as well as the higher fluences) and the resulting films are in general less thick than the SP films. This is consistent with measured ablation rates, maximal for short pulses, and also indicative of a higher gas content in the ablation products as compared to SP. At F = 1.2 J/cm2, the OP1 ablation rate with approximately 30 nm/pulse is about 1.5 smaller than the SP value, while the film thickness is reduced by a factor of three. The morphological differences are indicated in Fig. 5(a,b) for the low fluence range and confirm the reduction in the particulates sizes, with a tendency to aggregation of clusters for SP. We expect that this derives from the high density of nanoparticles ejected during SP exposure. These trends, as well as the differences in the ablation rates, attenuate with increasing laser fluences. Note that the increase in the ablation rate with the energy for the SP translates into a superlinear increase of the atomic species yield and therefore of the gas-phase component, reducing the efficiency of increasing the film effective thickness.
To attempt an explanation of the observed behavior regarding the spectral emission under short and tailored irradiation, we have employed a non-equilibrium two temperature hydrodynamic code (Esther) . The Lagrangian code treats the material as a continuum two temperature system, solving the fluid equations for the conservation of mass, momentum and energy for electronic and ionic species. The radiation impact is calculated by evaluating the Helmholtz wave equation in the inhomogeneous media. Electronic temperature effects dominate the initial optical transients via the carrier scattering frequency, where typical behaviors are given in . The thermodynamic properties of the material (energy, temperature, pressure, density, heat of fusion and vaporization) are described by the Bushman-Lomonosov-Fortov multiphase equation of state (EOS) spanning a large range of densities and temperatures from the cold condensed state to a hot plasma . In the absence of reliable EOS and non-equilibrium properties for brass, the simulation were run for a copper (Cu) sample. Though the approach may seem disputable, we expect that some similar relative tendencies will occur, delivering partial insights into the transformation factors. This expectation is supported by the relative ion increase using optimal pulses from copper targets detected using time-of-flight mass spectrometry  (Fig. 6) employing this time a higher resolution shaper. Note that, typical to time/velocity resolved spectra [12, 20, 21, 37], the signal indicates the behavior of ions with a certain velocity, in this case a high velocity (2.6×104 m/s), located approximately to the front of the plume. The comparative prospect is also sustained by the note that, dealing with incoherent processes and thermomechanical dynamics, the class of optimized pulses is rather large (note similar DP effects in  in conditions of moderate plasma shielding) and potentially extendable to other materials as well. However, the simulation results should be taken as an indicative parameter as differences between copper and brass exist. We indicate notably a higher thermal conductivity for Cu (approximately four times) and a Cu electron-phonon coupling that strongly augments with the electronic temperature  while brass may preserve an alloy character. This has consequences on the efficiency of the energy confinement, threshold levels, and, via the electron-phonon interaction increase, on the dynamic optical absorption properties. In the present case the irradiation regimes were calculated relative to the theoretical ablation threshold for Cu of approximately Fth = 0.5 J/cm2.
The simulation results are shown in Fig. 7 for SP and for the experimentally determined optimal shapes (OP1, OP2–1, denoted simply as OP) presented in Fig. 4. The figure indicates time-resolved spatio-temporal density and temperature profiles of the expanding material under shaped and optimal irradiation for a lower fluence regime (6× the calculated threshold), moderate intensities (8× the threshold), and high irradiance (16× the threshold). The three situations are depicted in Fig. 7(a), Fig. 7(b), and Fig. 7(c) respectively, where the scales of simulations were chosen to match as closely as possible the experimental observations (the large dimensional scales imped the clear observations at early times, notably for the temperature maps). At lower fluences we observe the long evolution of ejected dense fluid material with a thickness of few nanometers (seen as evolving nanolayers at liquid density in the left column of Fig. 7), driven by the high initial pressure. This is accompanied by a moderate plume temperature (Fig. 7 right column). The evolving liquid layers can further evaporate or fragment into droplets as a function of the neighboring pressure conditions , a situation which is not developed here. A slight decrease of the relative liquid quantity in the plume is observed for higher fluences.
The overall scenario changes when temporally shaped pulses are being used. As observed in the figure, two factors can be noted in the presence of optimal pulses that may be related to the observed excited gas-phase. Firstly, at low fluences we note a reduction of the nanoliquid content/size and a weak relative augmentation in temperature as the fluence starts to increase. This is connected to an increasing energy absorption and deposition in the surface vicinity where thermal conductivity drops with the gradual density changes. Usually a time of a few picoseconds is required to ignite the plasma expansion, being related to the strong temperature dependence of the electron-phonon coupling. With the subsequent energy confinement the remaining solid phase starts to be screened. Interaction with the liquid nanolayer takes place as well on these time scales. The lower particulate content, directly expelled from the liquid phase can be related in a simplified manner to a lower efficiency of the laser-induced rarefaction wave that cannot ensure the confinement of liquid layers of nanoscale dimensions between expanding gas layers . This can explain the high amount of gas-phase species noticed for optimal pulses and the subsequent behavior of spectral emissivity for the observed lines. Secondly, as the fluence is significantly increased [Fig. 7(c)], the gas pressure rise in front of the surface slows down the nanoparticle evolution. For the optimal pulse a higher concentration of heat load in the plume front along with a relatively constant value in the vicinity of the surface is indicated. This is particularly observable on the right hand side of the figure corresponding to spatially-resolved axial temperature profiles where the calculation time reflects the mean experimental time window observations. Eventually these temperatures may become similar at very high fluence levels, beyond typical ablation regimes. The observed heat confinement behavior can be determined by a restricted heat transport in the lower density plasma phase that forms during the pulse exposure. This accentuates with the input energy (as already noted in Fig. 2 by the gradual augmentation of excited and then ionized species). If in the low energy case the energy is given in the nascent hydrodynamic movement, reducing the mechanical compression of the liquid and the rarefaction wave, in the high energy range the laser energy is already coupled to the gas phase, leading to a visible high temperature expansion front. The consequences are related to the possibility of achieving hotter states during the hydrodynamic evolution, states that are closer to the cohesion limit and susceptible to be more effectively transformed into the gas-phase . Supercritical regimes and atomization of the expanding fluid  are easier to be reached with a judicious tailoring of the laser pulse by channeling the laser radiation in the expanding material. As confirmed by the thermodynamic states given by the present simulation, this leads eventually to a preferential gas-phase transformation as compared to gas-droplet mixtures typical to short pulse irradiation. Nevertheless, as the energy deposition is not homogeneous, droplets may appear as well in the OP case for specific conditions. A comprehensive discussion of thermomechanical scenarios on an aluminum sample and preferential transformation of mechanical into thermal energy involving rarefaction was given in Refs. [12, 35]. This scenario implies in the present case excess heating and supercritical expansion for picosecond sequences as compared to advance into the metastable two-phase region for short pulses. Above a certain fluence level, as the kinetic expansion will be as well enhanced, the preferential accumulation of energy in the expanding front, with consequence in disintegration of the hot fluid, will limit the occurrence of other transformation mechanisms by confining the absorption mainly in the superficial layers. The previously observed differences between DP and LP sequences can be explained by taking into account the steepness of the density gradient. In general, as a function of the temperature level, this optimized evolution may either increase the quantity of neutral gas in low fluence regimes or determine a more effective ionization at higher energy input levels.
We note that restrictions may apply to these observations due to the real three-dimensional evolution of the plume against the one-dimensional character of the simulation approach, however we believe that the relative behavior remains valid.
An observation that may seem important in the context, also the temperature axial profiles T(z) play an important role (Fig. 7 right column). The temperature spatial gradient along the plume expansion axis that accentuates towards the front determines the excitation efficiencies or the resulting ionization degree as a function of the chemical nature of the species. Provided that the temperature may become high enough and the density does not vary, this leads to distinct regions of different abundances of neutral and ionic species even in the conditions of congruent ablation, creating an impression of spatial segregation. Resuming, the increasing temperature profile together with the corresponding ionization degrees determines the formation of neutral or ionized regions according to the ionization potentials of the individual species, leading as well to a spatially-selective agglomeration of the excited or ionized chemical species. Cu-I, Zn-I species are mainly located in the vicinity of the surface, decoupled from the high temperature regions. Closer to the plasma front the neutral zone is leaded by Cu-II and the Zn-II domains. An experimental demonstration of this particular situation was already indicated  but was mainly explained based on ambipolar diffusion. This segregation appears to originate in the temperature gradient and follows the axial temperature profiles. Additional influences such as charge effects and specific flight times will further accentuate this tendency.
These studies indicate that more synergetic interaction aspects can be achieved when the energy coupling efficiency can be managed by the laser pulse temporal form. Pulse tailoring and optimization loops based on spectral detection were implemented to control the spectral signature of excited and ionized states in plasmas generated from brass. Different spectral features were obtained in various fluence regimes. Simple designer pulses were used to enhance the spectral signature, where energy spreading on several ps was found beneficial. Additionally, optimal irradiation sequences achieved in an adaptive manner determine the change of gas-phase fraction in the plume with controllable excitation and ionization gradients. As indicated in the hydro-dynamic simulations, this was related to reducing the nanoparticle content or to coupling more energy to the emerging low density phases. The latter shows hot thermodynamic trajectories while keeping the thermal level in the remnant solid at a lower value, less efficient to support nucleation of the gas-liquid mixtures as in the case of short pulses. The emission enhancement as well as the corresponding excitation degree can thus be realized by pulses at lower energies as compared to standard short pulses, indicating a more efficient use of energy.
This work has been supported in part by the Ultraviolet Laser Facility operating at IESL-FORTH in the frame of the EC project ”Laserlab-Europe” (FP6 Contract RII3-CT-2003-506350 and FP7 Grant Agreement 212025). The support received from the Agence Nationale de la Recherche is equally acknowledged.
References and links
1. A. Giakoumaki, K. Melessanaki, and D. Anglos,“Laser-induced breakdown spectroscopy (LIBS) in archaeological science-applications and prospects,” Anal. Bioanal. Chem. 387, 749–760 (2007). [CrossRef]
2. A. Assion, M. Wollenhaupt, L. Haag, F. Mayorov, C. Sarpe-Tudoran, M. Winter, U. Kutschera, and T. Baumert, “Femtosecond laser-induced-breakdown spectrometry for Ca2+ analysis of biological samples with high spatial resolution,” Appl. Phys. B: Las. Opt. 77, 391–397 (2003). [CrossRef]
3. E. L. Gurevich and R. Hergenröder, “Femtosecond Laser-Induced Breakdown Spectroscopy: Physics, Applications, and Perspectives,” Appl. Spect. 61233A–242A (2007). [CrossRef]
4. F. Courvoisier, V. Boutou, V. Wood, A. Bartelt, M. Roth, H. Rabitz, and J.-P. Wolf, “Femtosecond laser pulses distinguish bacteria from background urban aerosols,” Appl. Phys. Lett. 87, 063901/1–3 (2005). [CrossRef]
5. H. L. Xu, G. Méjean, W. Liu, Y. Kamali, J.-F. Daigle, A. Azarm, P. T. Simard, P. Mathieu, G. Roy, J.-R. Simard, and S. L. Chin, “Remote detection of similar biological materials using femtosecond filament-induced breakdown spectroscopy,” Appl. Phys. B: Las. Opt. 87, 151–156 (2007). [CrossRef]
6. J. Perriere, E. Millon, W. Seiler, C. Boulmer-Leborgne, V. Craciun, O. Albert, J. C. Loulergue, and J. Etchepare, “Comparison between ZnO films grown by femtosecond and nanosecond laser ablation,” J. Appl. Phys. 91, 690–696 (2002). [CrossRef]
7. C. Ristoscu, G. Socol, C. Ghica, I. N. Mihailescu, D. Gray, A. Klini, A. Manousaki, D. Anglos, and C. Fotakis, “Femtosecond pulse shaping for phase and morphology control in PLD: Synthesis of cubic SiC,” Appl. Surf. Sci. 252, 4857–4862 (2006). [CrossRef]
8. F. Garrelie, N. Benchikh, C. Donnet, R. Y. Fillit, J. N. Rouzaud, J. Y. Laval, and V. Pailleret, “One-step deposition of diamond-like carbon films containing self-assembled metallic nanoparticles, by femtosecond pulsed laser ablation,” Appl. Phys. A: Mater. Sci. Process. 90, 211–217 (2008).
9. R. Teghil, A. Santagata, A. De Bonis, A. Galasso, and P. Villani, “Chromium carbide thin films deposited by ultra-short pulse laser deposition,” Appl. Surf. Sci. 255, 7729–7733 (2009). [CrossRef]
10. D. von der Linde and K. Sokolowski-Tinten, “The physical mechanisms of short-pulse laser ablation,” Appl. Surf. Sci. 154–155, 1–10 (2000). [CrossRef]
11. N. M. Bulgakova, I. M. Bourakov, and N. A. Bulgakova, “Rarefaction shock wave: Formation under short pulse laser ablation of solids,” Phys. Rev. E , 63, 046311/1–5 (2001). [CrossRef]
12. J. P. Colombier, P. Combis, A. Rosenfeld, I. V. Hertel, E. Audouard, and R. Stoian, “Optimized energy coupling at ultrafast laser-irradiated metal surfaces by tailoring intensity envelopes: Consequences for material removal from Al samples,” Phys. Rev. B 74, 224106/1–16 (2006). [CrossRef]
13. P. Lorazo, L. J. Lewis, and M. Meunier, “Thermodynamic pathways to melting, ablation, and solidification in absorbing solids under pulsed laser irradiation,” Phys. Rev. B 73, 134108/1–22 (2006). [CrossRef]
14. A. Semerok and C. Dutouquet, “Ultrashort double pulse laser ablation of metals,” Thin. Sol. Films 453–454, 501–505 (2004). [CrossRef]
15. T. Gunaratne, M. Kangas, S. Singh, A. Gross, and M. Dantus, “Influence of bandwidth and phase shaping on laser induced breakdown spectroscopy with ultrashort laser pulses,” Chem. Phys. Lett. 423, 197–201 (2006). [CrossRef]
16. S. Singha, Z. Hu, and R. J. Gordon, “Ablation and plasma emission produced by dual femtosecond laser pulses,” J. Appl. Phys. 104, 113520/1–10 (2008). [CrossRef]
17. T. Donnelly, J. G. Lunney, S. Amoruso, R. Bruzzese, X. Wang, and X. Ni, “Double pulse ultrafast laser ablation of nickel in vacuum,” J. Appl. Phys. 106, 013304/1–5 (2009). [CrossRef]
18. A. Klini, P. A. Loukakos, D. Gray, A. Manousaki, and C. Fotakis, “Laser Induced Forward Transfer of metals by temporally shaped femtosecond laser pulses,” Opt. Express 16, 11300–11309 (2008). [CrossRef] [PubMed]
19. A. Santagata, R. Teghil, G. Albano, D. Spera, P. Villani, A. De Bonis, G. P. Parisi, and A. Galasso, “Fs/ns dual-pulse LIBS analytic survey for copper-based alloys,” Appl. Surf. Sci. 254, 863–867 (2007). [CrossRef]
20. M. Spyridaki, E. Koudoumas, P. Tzanetakis, C. Fotakis, R. Stoian, A. Rosenfeld, and I. V. Hertel, “Temporal pulse manipulation and ion generation in ultrafast laser ablation of silicon,” Appl. Phys. Lett. 83, 1474–1476 (2003). [CrossRef]
21. R. Stoian, A. Mermillod-Blondin, N. M. Bulgakova, A. Rosenfeld, I. V. Hertel, M. Spyridaki, E. Koudoumas, P. Tzanetakis, and C. Fotakis, “Optimization of ultrafast laser generated low-energy ion beams from silicon targets,” Appl. Phys. Lett. 87, 124105/1–3 (2005). [CrossRef]
22. M. Guillermin, C. Liebig, F. Garrelie, R. Stoian, A.-S. Loir, and E. Audouard, “Adaptive control of femtosecond laser ablation plasma emission,” Appl. Surf. Sci. 255, 5163–5166 (2009). [CrossRef]
23. A. Assion, T. Baumert, M. Bergt, T. Brixner, B. Kiefer, V. Seyfried, M. Strehle, and G. Gerber, “Control of Chemical Reactions by Feedback-Optimized Phase-Shaped Femtosecond Laser Pulses,” Science 282, 919–922 (1998). [CrossRef] [PubMed]
24. A. De Giacomo, M. Dell’Aglio, O. De Pascale, R. Gaudiuso, R. Teghil, A. Santagata, and G. P. Parisi, “ns- and fs-LIBS of copper-based-alloys: A different approach,” Appl. Surf. Sci. 253, 7677–7681 (2007). [CrossRef]
25. O. V. Borisov, X. L. Mao, A. Fernandez, M. Caetano, and R. E. Russo, “Inductively coupled plasma mass spectrometric study of non-linear calibration behavior during laser ablation of binary Cu-Zn Alloys,” Spectrochim. Acta Part B 54, 1351–1365 (1999). [CrossRef]
27. V. Margetic, A. Pakulev, A. Stockhaus, M. Bolshov, K. Niemax, and R. Hergenröder, “A comparison of nanosecond and femtosecond laser-induced plasma spectroscopy of brass samples,” Spectrochim. Acta Part B 55, 1771–1785 (2000). [CrossRef]
28. M. Guillermin, “Study of the femtosecond laser ablation plume, control and optimization of processes,” PhD Thesis, Université Jean Monnet, Saint Etienne (2009) (http://tel.archives-ouvertes.fr/tel-00395196/en/).
29. C. C. Garcia, H. Lindner, A. von Bohlen, C. Vadlab, and K. Niemax, “Elemental fractionation and stoichiometric sampling in femtosecond laser ablation,” J. Anal. At. Spectrom. 23, 470–478 (2008). [CrossRef]
30. D. Scuderi, O. Albert, D. Moreau, P. P. Pronko, and J. Etchepare, “Interaction of a laser-produced plume with a second time delayed femtosecond pulse,” Appl. Phys. Lett. 86, 071502/1–3 (2005). [CrossRef]
31. V. Piñon, C. Fotakis, G. Nicolas, and D. Anglos, “Double pulse laser-induced breakdown spectroscopy with femtosecond laser pulses,” Spectrochim. Acta Part B 63, 1006–1010 (2008). [CrossRef]
32. S. Noël, E. Axente, and J. Hermann, “Investigation of plumes produced by material ablation with two time-delayed femtosecond laser pulses,” Appl. Surf. Sci. 255, 9738–9741 (2009). [CrossRef]
33. X. Wang, S. Amoruso, and J. Xia, “Temporally and spectrally resolved analysis of a copper plasma plume produced by ultrafast laser ablation,” Appl. Surf. Sci. 255, 5211–5214 (2009). [CrossRef]
34. Ph. Rohwetter, J. Yu, G. Méjean, K. Stelmaszczyk, E. Salmon, J. Kasparian, J.-P. Wolf, and L. Wöste, “Remote LIBS with ultrashort pulses: characteristics in picosecond and femtosecond regimes,” J. Anal. At. Spectrom. 19, 437–444 (2004). [CrossRef]
35. J. P. Colombier, E. Audouard, P. Combis, A. Rosenfeld, I. V. Hertel, and R. Stoian, “Controlling energy coupling and particle ejection from aluminum surfaces irradiated with ultrashort laser pulses,” Appl. Surf. Sci. 255, 9597–9600 (2009). [CrossRef]
36. A. V. Bushman, I. V. Lomonosov, and V. E. Fortov, “Models of wide-range equations of state for matter under conditions of high energy density,” Sov. Tech. Rev. B: Therm. Phys. Rev. 5, 1 (1993).
37. R. Stoian, H. Varel, A. Rosenfeld, D. Ashkenasi, R. Kelly, and E. E. B. Campbell, “Ion time-of-flight analysis of ultrashort pulsed laser-induced processing of Al2O3,” Appl. Surf. Sci. 165, 44–55 (2000). [CrossRef]
38. Z. Lin, L. V. Zhigilei, and V. Celli, “Electron-phonon coupling and electron heat capacity of metals under conditions of strong electron-phonon nonequilibrium,” Phys. Rev. B 77, 075133/1–17 (2008). [CrossRef]
39. B. Chimier and V. T. Tikhonchuk, “Liquid-vapor phase transition and droplet formation by subpicosecond laser heating,” Phys. Rev. B 79, 184107/1–10 (2009). [CrossRef]