Ultrafast laser microexplosions in bulk material create extreme conditions at mesoscopic scales and are essential to the synthesis of extraordinary matter structural phases and to light structuring beyond the diffraction limit. Observing the transformation cycle can elucidate their evolution. We discuss multiscale relaxation dynamics in the formation of nanoscale structures in laser-irradiated fused silica. Tightly focused ultrafast nondiffractive Bessel beams are used to generate microexplosions that lead to uniform voids. These trigger thermodynamic nonequilibrium conditions in one-dimensional geometries with record excitation confinement down to 100 nm and electronic pressures in the gigapascal range. Time-resolved phase-contrast microscopy on nanosecond to microsecond scales indicates that void formation is a slow process developing from low-viscosity phases after persistent plasma fluid stages signaled via nanosecond-long luminescence. The void evolution is not necessarily driven by rarefaction following initial pressure relaxation, but involves molecular kinetics and stress mechanisms that interfere with the evolution of the liquid phase and induce cavitation. Heat transport is also visualized. Higher energy leads to hydrodynamic instabilities and void fragmentation. The dynamic view helps us understand material transformation under confinement.
© 2017 Optical Society of America
Controlling the structural response of matter on the nanoscale is key in designing materials and functions, upscaling approaches for the synthesis of new forms and structures. When short spatial and temporal scales are involved, nonequilibrium conditions with strong confinement can determine nonstandard structural evolutions. To this end, microexplosions generated by ultrashort laser pulses in bulk transparent materials can lead to extreme conditions of temperature and pressure and thus to extraordinary material phases and topologies confined in the volume . In tight focusing geometries, energy concentrations of can be obtained from microjoule (μJ) pulse energies on a 100 fs timescale. This spatiotemporal confinement of energy was suggested to trigger extreme terapascal (TPa) pressure levels, high nonequilibrium, and ultra-high quenching rates () with a quasi-mesoscopic range of the explosion-like shock. Nanoscaled voids and topographies can thus be obtained. It is expectable that such stress values surpass the elastic limit and induce plastic rearrangements, lifting kinetic barriers for new polymorphs and polymorphs in the laser-affected regions. Prominent examples [2,3] refer to predicted, but not previously observed, body-centered cubic Al in laser-shocked and potentially superconductive tetragonal Si phases in silicon. In silica, data derived from densification experiments indicates densification ratios beyond the stishovite structure and the potential capability to tailor states with designed density and structural order [4,5]. It is thus foreseeable that this scenario can apply to laser microexplosions. The application of the microexplosion concept to silica glass  thus carries a strong interest. This is motivated on one hand by its technological potential in optics and microelectronics. Nanoscaled structures were also proposed as long-life data storage [7,8]. Fused silica is equally a marker for high pressure loads in geophysical high-energy interactions . With its structural flexibility giving a wide range of densities and polymorphs, it outlines a platform of study of significant fundamental interest. The possibility to build up structural arrangements in a flexible manner may define new mesoscopic properties and behaviors.
It is commonly considered that, in the conditions of ultrafast laser-induced microexplosions and formation of nanovoids, material relaxation proceeds via shock waves and follow-up rarefaction. This is related to high energy concentrations enabled by tight focusing geometries for the incoming laser pulses, bypassing diffraction and plasma defocusing. Low-density phases can then be generated in the explosion epicenter, with material rupture down to the formation of nanoscaled voids. In the surrounding areas, metastable high-density phases can be subjected to extreme compression rates. These rates critically depend on the mechanical and thermodynamical scenarios. The control of material evolution in microexplosion concepts involves, therefore, on one side, the achievement of particular interaction geometries and excitation levels and, on the other side, the understanding of the kinetics of matter movement. With respect to the former, if typically Gaussian beams of ultrashort durations were used to develop confined-volume microexplosions in transparent materials, a relatively novel range of beams is beginning to attract attention. To achieve record thermodynamic conditions and a high degree of confinement, nondiffractive beams  can offer extended possibilities. They have received attention for their potential to control energy deposition in one-dimensional geometries, where, associated to ultrashort duration, they lead to a qualitatively new level in energy concentration. The characteristic conical self-healed intersection of wavefronts is nonlinearly stable and less affected by preionization, allowing extreme confinement scales for the electronic excitation inside the transparent materials [11,12]. These features allow for extrapolating the microexplosion concept in one-dimensional geometries with the achievement of high-aspect-ratio voids and sections approaching the nanoscale [13,14].
The formation of nanoscaled voids and the possibility to generate structural states are intrinsically related. With respect to process evolution, two generation pathways are considered nowadays for the emergence of novel structural arrangements; a shock-related (displacive) synthesis, and growth in nonequilibrium plasma phases (nucleated). The shock/rarefaction model acts particularly fast. Here, for initial pressures in the TPa range given by electron heating , picosecond (ps) timescales are discussed for void opening in a quasi-solid material. If nucleation is activated, the achievable phases may not only be a direct consequence of pressure but equally of the entropic nonequilibrium relaxation towards close-packed structures. Intermediary stages are therefore key in establishing what and how fast material changes occur. Alternative scenarios are of interest as they may suggest different material evolution trajectories.
We discuss here the underlying dynamics in the formation of nanoscaled voids in fused silica following Bessel beam ultrashort laser irradiation. The dynamics is followed over a large time window, ranging from nanoseconds to microseconds (ns to μs). In this time domain, phase transitions, mechanical development, and cooling stages can be effectively observed. The relaxation dynamics and the characteristic timescales of the excited matter are considered paramount in defining the mechanisms for the transformation of the material. Particular one-dimensional excitation geometries are employed, with a preferential radial transport of energy and matter during relaxation. Ultrafast nondiffractive beams in tight focusing conditions are used to generate dense electron plasmas and subsequently uniform long voids with nanoscaled sections. Multiscale time-resolved microscopy techniques (operating on multiple time scales) are applied to visualize via optical phase mapping the dynamics of excited matter down to the permanent opening of the void. The technique allows us, via relative transient changes of the refractive index, to observe electronic excitation, the formation of high-density liquid phases, as well as cavitation phenomena from the liquid phase. We discuss drive forces of material relaxation in extreme thermodynamic conditions and strong deformation yields that may equally serve in developing strategies for material synthesis. We give particular evidence on a slow (ns to μs) void opening path passing through the development of plasmas, liquid phases, and cavitation, associated with a Rayleigh–Plesset growth model.
2. EXPERIMENTAL METHOD
A. Ultrafast Bessel Beams
Volume excitation of a fused silica sample is done using single 50 fs laser pulses at the central wavelength of 800 nm delivered from a regeneratively amplified ultrafast Ti:sapphire laser system. A zero-order Bessel beam  is generated using an axicon lens (apex angle of 179°) [15–17]. The beam encompasses a narrow intense central core sustained over a long distance (), a consequence of the conical interference of wavefronts induced by the conical phase at the axicon. The core is demagnified (demagnification factor 100) and imaged inside a sample (Corning 7980-5F) using a 4f afocal imaging system  with a microscope objective () as the final imaging element. Thus, the final core dimension is 0.75 μm full width at half-maximum (FWHM), and the nondiffractive length stays in the 100 μm range. Spatial filtering ensured no leakage from the axicon tip. The Bessel conical half-angle in glass reflects rather tight focusing conditions, sufficient to induce optical breakdown in the bulk. For observing the structures in situ, optical transmission (OTM) and positive phase-contrast (PCM) microscopy are employed, followed in specific cases by post-irradiation scanning electron microscopy (SEM) imaging.
B. Time-Resolved Microscopy
The multiscale relaxation dynamics following ultrashort pulse excitation of bulk fused silica in nondiffractive tight focusing conditions is observed using time-resolved imaging. The method allows the observation of the laser-induced object in phase and amplitude for a timescale ranging from ns to μs and with submicron spatial resolution. A two-color pump-probe microscopy technique sensitive to relative differences in the object optical phase, and having ns time resolution is used, collecting corresponding charts of material transformation at various illumination time moments. The pump pulse is represented by an 800 nm ultrashort laser pulse, shaped in nondiffractive mode. The imaging setup is based on an upright optical microscope (Olympus BX-51) operating in optical transmission and phase-contrast diascopy. In OTM, absorbing or scattering regions appear dark on a bright background. In PCM, phase shifts corresponding to negative and positive index changes appear bright or dark, respectively, on a grey background. The observation objective gives a spatial resolution of 650 nm at the observation wavelength. For illumination, a stroboscopic method is used, and the probe pulse cross-illuminates the interaction zone. To obtain high-resolution single-shot images with uniform background and low speckle noise, a low-spatial-coherence pulsed source is developed based on the random lasing effect, as described in Ref. . The laser gain medium consists of a colloidal solution of Rhodamine B (2.5 g/l) with immersed latex nanobeads (of 325 nm size) at a concentration of . The random lasing effect is obtained by irradiating the colloidal solution with 532 nm laser pulses in the absorption band of the solution. The exciting laser pulses in the solution are obtained from a frequency doubled Nd:YAG laser operating at 10 Hz and delivering 7 ns pulses in the millijoule (mJ) energy range. The random lasing effect is detected via the appearance of a strong spectral narrowing of the fluorescence band to around 13 nm bandwidth (FWHM) centered at 590 nm. Its pulse duration is similar to the excitation pulse in solution, i.e., 7 ns. The electronic synchronization between the ultrafast laser system and the ns laser system ensures the time synchronization between the exciting ultrashort laser pulse in fused silica and the random lasing illumination (probe) source, with delays measured by a fast photodiode. The excitation region in the glass sample is imaged in a perpendicular geometry using a Köhler illumination arrangement, providing at the same time information on the amplitude and the phase of the object. The probe enters the illumination path of the microscope via the microscope condenser, and the optical transmission and phase-contrast microscopy images are recorded in the image plane with a back-illuminated electron-multiplying charge-coupled device camera (Andor iXon Ultra 897 EMCCD). A 20 nm bandwidth-pass filter centered at the probing laser wavelength (590 nm) is used to cut parasitic incoherent light emitted by the specimen and the scattering from the pump.
In addition, time-resolved photoluminescence from the excited regions was collected in an imaging geometry with a gated intensified CCD (Pi-Max3 ICCD, Princeton Instruments). The acquisition is electronically synchronized with the excitation pulse and records images on the ns scale, with photoemission signal integrated over a gate duration of 5 ns.
3. RESULTS AND DISCUSSION
A. Energetic Balance in Void Formation
The results of single-shot Bessel irradiation are given in Fig. 1(a) in a range where uniform void structures are obtained (i.e., for an input energy of 3 μJ). The figure thus represents the side view of the permanent one-dimensional void of around 120 μm length in phase-contrast mode. The low-density region is depicted here by the bright color. At the margins, limited zones of index increase are obtained (dark zones). The inset shows a typical cross section of the void, measured by intersecting the surface with the Bessel beam and ensuring that the void penetrates deep into the sample. The typical dimensions lie in the range of 100–500 nm, below the diffraction limit. In these conditions, the rather tight focusing is able to overcome diffraction and, most importantly, plasma defocusing, and confine enough energy to trigger void formation. We recall that for moderate focusing conditions, ps pulse envelopes are required to form uniform one-dimensional void structures [12,13]. Similar structures are obtained for input energies ranging from 0.5 to 4 μJ/pulse suggesting that, even in these conditions [femtosecond (fs) pulses, tight focusing], plasma defocusing acts as a leveling factor for the incoming energies. The corresponding energetic conditions are retrieved numerically using a nonlinear optical propagation model based on the nonlinear Schrödinger equation [19,20] for a Bessel laser pulse defined as in interaction with bulk fused silica. The model detailed in [13,20,21] takes into account basic processes occurring in the presence of an intense laser field, such as photoionization, free-carrier absorption, and collisional multiplication, as well as self-induced effects such as self-phase modulation, self-focusing, and plasma defocusing. The resulting quantity of deposited energy (, volumic energy density) for a 3 μJ, 50 fs laser pulse is given in Fig. 1(b). A maximal value of is obtained, stored in the electronic system mainly via inverse Bremsstrahlung from the laser field. Assuming that all this energy density will be completely transformed into heat, this corresponds to a matrix temperature increase of , calculated using the cold value of specific heat capacity () and a glass density of . This value (with about 50% error bar due to model uncertainties) seems, at a first glance, sufficient to induce a rapid and significant gas-phase nucleation process. However, the typical boiling point (2500 K at ambient pressure) is expected to increase drastically with local pressure, decreasing the nucleation probability and thus rising a degree of incertitude on the process. The associated matrix thermoelastic pressure increase , with being the thermal expansion coefficient () and being the solid Young’s modulus, will then raise to 150 MPa, with the primary electronic pressure being estimated as at around 5–8 GPa. The latter is calculated for free electronic densities in the range of the critical density value in bulk silica as typically observed in void formation  and average electron energies of around 15–20 eV, estimated by dividing energy and number density charts . If the electronic pressure may act as an expansion drive, a value in the gigapascal (GPa) to tens of GPa range can overcome to a certain extent the mechanical resistance of the silica glass, and the mass initially located within the laser heated zone can be pushed outwards, creating a shell of compressed material. Mechanical shocking occurs above the value of the Young’s modulus. This reflects the laser microexplosion concept. Gamaly et al.  suggested a mechanism of ionic acceleration following electronic pressure gradients and a shock mechanism relaxing the initial pressure. The size of the void (the stopping distance of the shock, considering an absorption rate of 40% from the input laser energy) can be calculated  by assuming that internal energy in the volume inside the shock front is comparable to the absorbed pulse energy . A value around 200 nm is found, consistent with the size of the measured cavity. To make full use of an electronic pressure drive on matter movement, such a mechanism will require rapid energy transfer from the electrons to the ions. Recent reports  indicate slow ns-long electronic relaxation for excitation conditions leading to void formation whenever a low-viscosity phase is formed. This suggests the potential rapid generation of a liquid phase that should impede the self-trapping of the electrons, facilitating the appearance of a long-living excited fluid phase. Consequences in the mechanical response should be expected. We note at the same time that the ionic pressure component seems to have a much softer action range, with mechanical expansion rather than shocking. These specific thermodynamic conditions will have a defined influence on the relaxation of the excited material. We equally recall that the silica glass has significant structural flexibility, with potential anomalies occurring in the thermomechanical properties . Molecular kinetics and structural rearrangements in low-viscosity phases can play a role in the relaxation dynamics .
The initial thermodynamic conditions are critical for the mechanisms of void formation. An initial pressure drive surpassing the mechanical resistance of the material will create a shock evolution and a rarefaction wave with a sub-ns dynamics inducing a low-density phase. A void can thus be created in several tens of ps. Alternatively, a slower nucleation or cavitation mechanism in the liquid phase may take place. The following section will describe the observed dynamics in relation to the above-mentioned processes.
B. Multiscale Dynamics in Void Formation
We give below a dynamical perspective on the physical processes leading to the formation of laser-induced structural modifications in the regime of void formation. Time-resolved phase-contrast images are presented in Fig. 2(a) on a timescale ranging from 1 ns to 11 μs, when dynamics comes to rest. An input pulse energy of 1.8 μJ is used, close to the threshold for the formation of void-like depressed regions. Here, cross sections below 200 nm can be obtained. The zero delay is defined by the earliest appearance of a clear image of the excited region. Three different stages of void evolution are evident. The early stage: in the first few images, at less than 10 ns delay, a region of decreased refractive index (bright zone) is observed. Two physical phenomena can explain this appearance; a rapid void formation creating a decreased density zone or the presence of excited free electrons. Due to their light mass, the carrier oscillation in the light field corresponds to a decrease of the local refractive index proportional to their concentration (). We now make the hypothesis that the observed phase shift is related to an electronic presence, a fact that will be argued below. The intermediary stage: in the next temporal stage, for the time domain corresponding to several hundreds of ns, we observe the phase shift of an apparent index increase, and no further presence of free carriers. Neglecting the electronic influence, the typical refractive index variation depends explicitly on temperature (via the thermo-optic effect) and on density, and reads as . For fused silica the thermo-optic coefficient is positive ()  and the refractive index increases with temperature. The primary information therefore indicates a hot phase. The void development and stabilization stage: beyond 700 ns we notice the onset of a low-index phase with a dynamics that continues and stabilizes at above 1 μs. To render the dynamics more visible we create an orthogonal view of the stack of individual time-resolved images in Figs. 2(b) and 2(c). A section of about 6 μm large is selected in the various locations of the void denoted I, II, and III for each image corresponding to a given delay, and the resulting images are concatenated together. The zones I, II, III correspond to chosen locations where, at the end, we obtain either a void-like region, a refractive index increase, or an intermediary region with no visible modifications. Figure 2(b) summarizes the specific conditions of zone I, II, III with a partial blow-up of instantaneous images at 1 ns and 11 μs delay. In order to have an idea about the excitation strength and thus, from a relative standpoint, about the local temperature, the following approach is used. A section is made on the PCM image giving a curve (red line) that maps the axial spread of the free electron yield. This curve representing the intensity along the median line of the bright zone is proportional to the phase retardation of the electrons, and thus to the electron number. In a first approximation, the local temperature should be proportional to the number density of a thermalized electron plasma. Thus, an axial representation of local relative temperature is obtained, which will help in correlating the temperature elevation to different significant regions of the permanent refractive index change. Figure 2(c) then gives the orthogonal projections as a graph where the lateral dimension represents the lateral information in the specific images in Fig. 2(a) and the longitudinal size takes a temporal dimension. A logarithmic timescale is used. The zone corresponding to the void region is described in the left part of Fig. 2(c). The initial plasma decay is apparent in the first few ns (region 1) followed by the appearance of the hot high-index zone (region 2). A low-index part appears to be visible on the lateral sides, though a certain assignment is difficult due to the inherent appearance of halos in phase-contrast microscopy. A transient lateral enlargement of the high-index zone signalizes the diffusion of heat (region 3) with a time constant of 1 μs. This is consistent with a cooling time on the order of , considering a thermal diffusivity of and characteristic heat source transverse dimension below 1 μm. The onset of the final void structure (region 4) occurs on a few hundreds of ns and gradually increases laterally to achieve a stationary dimension when cooling has finished. The characteristic timescale is not consistent with the rapid shock and rarefaction scenario expected at high pressures. Equally, the nucleation scenario seems less probable, as the void growth occurs on the cooling phase. A probable mechanism is related to slow cavitation or fracture. Cavitation being facilitated in the liquid phase implies that the hot zone (2, 3) can be assimilated to a liquid region of increased density. The central compaction can equally create stress on the neighboring zones, which upon cooling may pull the opening of the void. We point out that on these timescales anomalous mechanical behaviors may be present. As the dynamics continue on timescales where temperature has significantly decreased, this late evolution resembles a stress-driven mechanical propagation. These dynamics suggest equally a mechanical drive related to a moderate thermoelastic pressure in the 100 MPa range rather than to the several GPa of electronic pressures. A stronger coupling between electronic and ionic systems and the achievement of a higher pressure level will, according to a Rayleigh–Plesset bubble growth mechanism (albeit the geometry), accelerate the void formation. To conclude, in the case of study here, the void formation is most probably driven by cavitation/spall from the liquid phase and propagation of the fracture.
An interesting observation is related to region II in Fig. 2. If the initial index behavior can be assimilated to a relaxing excited phase, the temperature in the region II is minimal. Upon relaxation [Fig. 2(c) right], besides heat diffusion (marked by an increasing lateral size of the high-index region) a permanent index increase is noticed (dark color). It has been argued before  that this is related to the appearance of defects, particularly nonbridging oxygen hole centers. A defect-assisted densification relies then on structural rearrangements associated to electron trapping and bond-breaking, and a more compact repacking of the matrix . The structural rearrangement is then the main cause of densification. In addition, the bound electron associated to the defect can equally create relative variations of the dielectric function. This positive index region survives the thermal cycle. The conclusion that can be drawn is that here the heat wave is unable to anneal the sample and relax the structural densification. However, in region III a higher temperature can be inferred. The effect here is that no permanent modification can be noticed at the end of the thermal cycle. We can suppose then that here heating was effective in annealing the region to the initial structure. According to , if the heat wave was able to cure the defect-induced structural distortions in 100 ns, the local temperature should be above the glass transition temperature and in the vicinity of 2400 K, making the central (region I) temperature come closer to a value of 4000 K. In this case, the fact that boiling has not yet started may be related to the timescale of nucleation or to an increase of boiling temperature at the given pressure. We note that at pressures ranging from 100 MPa to GPa, the combination of high temperatures and pressure may even facilitate the appearance of a plasma fluid without crossing the vapor boundary .
A particular situation is encountered in the case where the input energy exceeds the limit required for uniform void formation. A fragmentation process occurs, resulting in multiple small-scale voids arranged on the axis. The situation is depicted in Figs. 3(a) and 3(b), which give the time-resolved dynamics of the process. In this case, even though the initial excited region is uniform, the permanent result is a succession of small voids of spherical symmetry and 1–2 μm periodicity. The inspection of the sequence of time-resolved images shows that the initial electronic relaxation is longer, down to 500 ns, followed immediately by the onset of the voids in a discontinuous way [Figs. 3(a)]. The orthogonal view [Fig. 3(b)] details the process, with underlining the diffusive heat processes and the possible formation of regions of expansion around the central axis. These regions will start shrinking to the initial density values upon cooling, creating tensile stresses. The formation of fragmented voids seems to be partly the consequence of a Rayleigh–Plateau hydrodynamic instability in the liquid column that triggers the fragmentation of the liquid cylinder. Assuming this scenario, the characteristic time scale of Rayleigh–Plateau instability of a liquid cylinder can be described  as , where is the liquid density, the radius of the cylinder, and the surface tension. This moment, corresponding to the time for which surface-tension forces are balanced by inertia, can be estimated to 100–200 ns in our conditions. Here, with , , and . For , we obtain and for , we have . These values are in a good agreement with the observations reported in Fig. 3. In addition, lateral mechanical stresses generated during cooling are equally present. Most probably we are dealing with a combination of pulling forces originating from side tensile stress and minimization of surface energy.
Now we review the initial hypotheses concerning the assignment of the different zones. For the early negative index change assigned to electronic excitation and relaxation of free carriers, the different dynamics observed in Figs. 2 and 3 confirms this hypothesis. Equally, the fact that this index zone is either spread on regions where finally the void is not observed (Fig. 2 zone III), or that it has uniform appearance, whereas the final result is not uniform (Fig. 3), comes in support of the initial ansatz. We believe that it is less probable that this dynamics represents a first void opening due to rarefaction that will then disappear, being refilled with liquid. In this case, the negative pressure can force a flux of liquid material. The high-index zone showing subsequently hydrodynamic instability supports equally the assignment of this high-index zone as a fluid phase. However, it should be said that, if temperatures in the range of 3000 K can be sustained over tens to hundreds of ns, the viscosity of the fluid can drop to around 1 Pa  and it can become sufficiently low to allow ns-rapid void collapse . To further support the electronic presence, we have measured the time-resolved photoluminescence from the irradiation zone  in an imaging configuration using the gated ICCD (gate width of 5 ns). The results for the uniform void regime are given in Fig. 4 and show a luminescence decay within 100 ns. Its broadband spectrum is continuous and slowly decaying in the visible domain (not shown) in the transmission range of the imaging setup (400–800 nm). If at the first sight the luminescence cannot be unambiguously attributed to electrons due to the presence of a hot material phase, corroborated with the negative index change in the first 100 ns (Fig. 2), this leads to the highly probable conclusion that we are observing electron Bremsstrahlung emission. Thus, the first stage corresponds to an electronically excited liquid or fluid phase. During the electronic relaxation and energy feed-through to the lattice, no significant hydrodynamic evolution on observable scales can be noticed. The limited spectral range does not allow for an accurate determination of temperature, which is only estimated for a Planckian distribution to be above 5000 K at the peak of the emission (a couple of ns after fs excitation).
The final observation element corresponds to energetic regimes just below the threshold for the formation of depressed void-like regions. The optical phase evolution is given in Figs. 5(a) and 5(b), using a similar representation as in the previous figures. In this case solely the dynamic of the heat wave is observed with diffusion and cooling on a 1 μs timescale, corresponding to the standard diffusivity of fused silica. The temperature and the cycle duration are insufficient to determine heat-driven molecular kinetics and a rearrangement of the matrix towards a higher fictive temperature, suggesting thus local temperatures inferior to the softening point.
4. CONCLUSIONS AND OUTLOOK
In conclusion, time-resolved imaging techniques were used to observe the multiscale time evolution of laser-generated nanovoids in silica glass. We have shown that the dynamics that lead to the formation of nanoscaled voids rely on a slow cavitation mechanism in low-viscosity phases, which stabilizes on the cooling stage. This involves prior heating, compaction, and transitions to liquid phases, for initial thermoelastic ionic pressures in the 100 MPa range, below the Young’s modulus value. A ns-long plasma phase intermediates the process, before the void break-up in the liquid in hundreds of ns. The amount of energy deposition defines conditions for uniform one-dimensional voids or for the onset of hydrodynamic instabilities leading to fragmentation. The relation between permanent morphologies and dynamic evolution and the potential of observing matter in movement are of paramount assistance in proposing scenarios of material modification. Equally, understanding cavitation dynamics can lead to new developments in laser nanostructuring and material breakdown in confined conditions. New potential for quantifying transient processes is seen in the advent of quantitative phase techniques.
Agence Nationale de la Recherche (ANR) (ANR 2011 BS04010 NanoFlam, ANR 2011 BS09026 SmartLasir); Labex Manutech-SISE (ANR-10-LABX-0075), of the Université de Lyon, within the program Investissements d’Avenir (ANR-11-IDEX-0007) operated by ANR.
We thank A. Soleihac and R. Antoine for their support in the experimental set-up and random laser illumination.
1. S. Juodkazis, K. Nishimura, S. Tanaka, H. Misawa, E. G. Gamaly, B. Luther-Davies, L. Hallo, P. Nicolai, and V. T. Tikhonchuk, “Laser-induced microexplosion confined in the bulk of a sapphire crystal: evidence of multimegabar pressures,” Phys. Rev. Lett. 96, 166101 (2006). [CrossRef]
2. A. Vailionis, E. G. Gamaly, V. Mizeikis, W. Yang, A. V. Rode, and S. Juodkazis, “Evidence of superdense aluminium synthesized by ultrafast microexplosion,” Nat. Commun. 2, 445 (2011). [CrossRef]
3. L. Rapp, B. Haberl, C. J. Pickard, J. E. Bradby, E. G. Gamaly, J. S. Williams, and A. V. Rode, “Experimental evidence of new tetragonal polymorphs of silicon formed through ultrafast laser-induced confined microexplosion,” Nat. Commun. 6, 7555 (2015). [CrossRef]
4. Y. Shen, S. B. Jester, T. Qi, and E. J. Reed, “Nanosecond homogeneous nucleation and crystal growth in shock-compressed SiO2,” Nat. Mater. 15, 60–65 (2016). [CrossRef]
5. C. A. McCoy, M. C. Gregor, D. N. Polsin, D. E. Fratanduono, P. M. Celliers, T. R. Boehly, and D. D. Meyerhofer, “Shock-wave equation-of-state measurements in fused silica up to 1600 GPa,” J. Appl. Phys. 119, 215901 (2016). [CrossRef]
6. S. Juodkazis, S. Kohara, Y. Ohishi, N. Hirao, A. Vailionis, V. Mizeikis, A. Saito, and A. Rode, “Structural changes in femtosecond laser modified regions inside fused silica,” J. Opt. 12, 124007 (2010). [CrossRef]
7. E. N. Glezer and E. Mazur, “Ultrafast-laser driven micro-explosions in transparent materials,” Appl. Phys. Lett. 71, 882–884 (1997). [CrossRef]
8. J. Zhang, M. Gecevičius, M. Beresna, and P. G. Kazansky, “Seemingly unlimited lifetime data storage in nanostructured glass,” Phys. Rev. Lett. 112, 033901 (2014). [CrossRef]
9. G. A. Lyzenga and T. J. Ahrens, “Shock temperatures of SiO2 and their geophysical implications,” J. Geophys. Res. 88, 2431–2444 (1983). [CrossRef]
10. J. Durnin, J. J. Miceli, and J. H. Eberly, “Diffraction-free beams,” Phys. Rev. Lett. 58, 1499–1501 (1987). [CrossRef]
11. M. K. Bhuyan, F. Courvoisier, P. A. Lacourt, M. Jacquot, R. Salut, L. Furfaro, and J. M. Dudley, “High aspect ratio nanochannel machining using single shot femtosecond Bessel beams,” Appl. Phys. Lett. 97, 081102 (2010). [CrossRef]
12. M. K. Bhuyan, P. K. Velpula, J. P. Colombier, T. Olivier, N. Faure, and R. Stoian, “Single-shot high aspect ratio bulk nanostructuring of fused silica using chirp-controlled ultrafast laser Bessel beams,” Appl. Phys. Lett. 104, 021107 (2014). [CrossRef]
13. P. K. Velpula, M. K. Bhuyan, F. Courvoisier, H. Zhang, J. P. Colombier, and R. Stoian, “Spatio-temporal dynamics in nondiffractive Bessel ultrafast laser nanoscale volume structuring,” Laser Photon. Rev. 10, 230–244 (2016). [CrossRef]
14. L. Rapp, R. Meyer, R. Giust, L. Furfaro, M. Jacquot, P. A. Lacourt, J. M. Dudley, and F. Corvoisier, “High aspect ratio micro-explosions in the bulk of sapphire generated by femtosecond Bessel beams,” Sci. Rep. 6, 34286 (2016). [CrossRef]
15. J. Mcleod, “The axicon: a new type of optical element,” J. Opt. Soc. Am. 44, 592–597 (1954). [CrossRef]
16. G. Indebetouw, “Nondiffracting optical fields: some remarks on their analysis and synthesis,” J. Opt. Soc. Am. A 6, 150–152 (1989). [CrossRef]
17. R. M. Herman and T. A. Wiggins, “Production and uses of diffractionless beams,” J. Opt. Soc. Am. A 8, 932–942 (1991). [CrossRef]
18. A. Mermillod-Blondin, H. Mentzel, and A. Rosenfeld, “Time-resolved microscopy with random lasers,” Opt. Lett. 38, 4112–4115 (2013). [CrossRef]
19. L. Sudrie, A. Couairon, M. Franco, B. Lamouroux, B. Prade, S. Tzortzakis, and A. Mysyrowicz, “Femtosecond laser-induced damage and filamentary propagation in fused silica,” Phys. Rev. Lett. 89, 186601 (2002). [CrossRef]
20. I. M. Burakov, N. M. Bulgakova, R. Stoian, A. Mermillod-Blondin, E. Audouard, A. Rosenfeld, A. Husakou, and I. V. Hertel, “Spatial distribution of refractive index variations induced in bulk fused silica by single ultrashort and short laser pulses,” J. Appl. Phys. 101, 043506 (2007). [CrossRef]
21. P. Polesana, M. Franco, A. Couairon, D. Faccio, and P. Di Trapani, “Filamentation in Kerr media from pulsed Bessel beams,” Phys. Rev. A 77, 043814 (2008). [CrossRef]
22. E. G. Gamaly, S. Juodkazis, K. Nishimura, H. Misawa, B. Luther-Davies, L. Hallo, P. Nicolai, and V. T. Tikhonchuk, “Laser-matter interaction in the bulk of a transparent solid: confined microexplosion and void formation,” Phys. Rev. B 73, 214101 (2006). [CrossRef]
23. T. Deschamps, J. Margueritat, C. Martinet, A. Mermet, and B. Champagnon, “Elastic moduli of permanently densified silica glasses,” Sci. Rep. 4, 7193 (2014). [CrossRef]
24. G. Ghosh, “Model for the thermo-optic coefficients of some standard optical glasses,” J. Non-Cryst. Solids 189, 191–196 (1995). [CrossRef]
25. K. Mishchik, C. D’Amico, P. K. Velpula, C. Mauclair, A. Boukenter, Y. Ouerdane, and R. Stoian, “Ultrafast laser induced electronic and structural modifications in bulk fused silica,” J. Appl. Phys. 114, 133502 (2013). [CrossRef]
26. N. S. Shcheblanov and M. E. Povarnitsyn, “Bond-breaking mechanism of vitreous silica densification by IR femtosecond laser pulses,” Eur. Phys. Lett. 114, 26004 (2016). [CrossRef]
27. M. Lancry, B. Poumellec, A. Chahid-Erraji, M. Beresna, and P. G. Kazansky, “Dependence of the femtosecond laser refractive index change thresholds on the chemical composition of doped-silica glasses,” Opt. Mater. Express 1, 711–723 (2011). [CrossRef]
28. H. J. Melosh, “A hydrocode equation of state for SiO2,” Meteorit. Planet. Sci. 42, 2079–2098 (2007). [CrossRef]
29. J. Eggers, “Nonlinear dynamics and breakup of free-surface flows,” Rev. Mod. Phys. 69, 865–930 (1997). [CrossRef]
30. R. Taylor, C. Hnatovsky, and E. Simova, “Applications of femtosecond laser induced self-organized planar nanocracks inside fused silica glass,” Laser Photon. Rev. 2, 26–46 (2008). [CrossRef]
31. W. Lauterborn and T. Kurz, “Physics of bubble oscillations,” Rep. Prog. Phys. 73, 106501 (2010). [CrossRef]
32. C. W. Carr, H. B. Radousky, A. M. Rubenchik, M. D. Feit, and S. G. Demos, “Localized dynamics during laser-induced damage in optical materials,” Phys. Rev. Lett. 92, 087401 (2004). [CrossRef]