Open Access
Add to BibSonomyAbstract
Irradiation of non-luminescent silica with polarized IR femtosecond laser light produced a significant amount of luminescent defects. We have investigated the properties of luminescence produced by the defects using UV-VUV excitation experiment depending on the relative orientation of the laser polarization and its scanning direction. Silicon Oxygen Deficient Center (SiODC) is identified. SiODC related luminescence is much stronger when the excitation polarization is parallel to the sample scanning direction and moved at low velocity, regardless of the writing polarization direction. This indicates that the creation of this anisotropic defect is oriented by the movement of the femtosecond laser beam.
©2012 Optical Society of America
1. Introduction
Today advanced femtosecond laser systems offer a variety of material processing in silica-based glasses, from surface machining to 3D refractive index changes (isotropic or anisotropic) writing [1]. For laser processing, a key advantage of using femtosecond pulses, relative to longer pulses, is that one can rapidly and precisely deposit energy in solids. This allows 3D multi-component photonic devices to be fabricated in a single step within a variety of transparent materials [1–3]. These interactions potentially enable the development of a new generation of powerful, complex components for micro-optics, optical telecommunications, optical data storage, sensor technologies, material processing and much more. For many of these applications, silica is the preferred material, providing excellent physical and chemical properties such as optical transparency from IR to UV, a low thermal expansion coefficient, and a high resistance to laser induced damage.
Background on femtosecond laser matter interaction in silica: refractive index modifications in silica glass, induced by femtosecond laser irradiation have been reported in many papers. Depending on the exposure parameters, three different kinds of structural changes can be induced in fused silica [4]: an isotropic positive refractive index change (type-I); a form birefringence with negative index change [5] (type-II); and voids (type-III). The type-I and type-II modifications were analyzed as follows. Above a first threshold, T1, (i.e. 0.10μJ/pulse, 800nm, 160fs, 200kHz, 0.5 NA), the index change is permanent and isotropic (type-I). The maximum index change is 6∙10−3 in fused silica [6]. This is very large compared to the one induced by UV nanosecond lasers [7,8]. Above a second damage threshold, T2, (e.g. 0.31 μJ/pulse when laser polarization and movement are parallel, 800nm, 160fs, 200kHz 0.5 NA [4]), the characteristics are quite different (type-II). The anisotropic index change magnitude can be as large as 2∙10−2 [6] and resists decay during two hours at 1000°C [9]. This form birefringence originates from sub-wavelength nanogratings [10] and we have proved recently that index contrast is due to the fact that nanoplanes are nanoporous matter produced by a decomposition of SiO2 into SiO2(1-x) + x∙O2 [11–13].
From the fundamental point of view, the process is initiated by multiphoton ionization resulting in a highly nonlinear dependence on the light beam intensity. The laser light is absorbed by valence electrons and the optical excitation ends before the surrounding lattice is perturbed, which results in highly localized “damage” in the material [14]. Point defects such as NBOHC (Non Bridging Oxygen Hole Center), SiE’ (Si dangling bonds), peroxy linkage or radical, and interstitial oxygen (atoms or molecules) have been identified using luminescence, electron paramagnetic resonance (EPR) and other properties [15,16]. The formation mechanism of the photo-induced modifications is not understood, but one can classify them according to the required pulse energy. Most results are for pure silica glass and above T2 threshold. The first defect detected is an unknown one giving rise to a luminescence peaking at 540nm under 514nm excitation for pulse energy ranging between 0.1 to 1 μJ (800nm, 130fs, 1 kHz, NA = 0.5) [17]. Beside this one, NBOHC is clearly detected by many authors [17–21] from 0.4 μJ to 5 μJ (800nm, 130fs, 1-1000kHz, NA≈0.55-0.85) either from luminescence or light absorption at 4.5eV. From absorption spectrum or EPR, SiE’ and GeE’ are found above a few μJ (800nm, 120fs, 200kHz, NA = 0.15-0.5) [22]. The concentrations of most defects increase on dose. In [23], the authors reported permanent linear dichroism at 800nm and 1050nm in silica glass samples irradiated at 0.25 μJ/pulse (800nm, 160fs, 100 kHz, NA = 0.50) but they do not investigate the origin. In Ge-doped silica glass (8 mol% in Ge), strong blue luminescence (410nm) of GeODC defect states has been reported under fs multiphoton excitation around 0.05μJ (800nm, 120fs, 200kHz, NA = 0.5). In addition they reported the first evidence of anisotropic light scattering which peaks in the plane of light polarization in isotropic media [24].
2. Experimental details
The direct writing procedure using IR-fs (Infrared femtosecond laser) has been already described extensively [25]. In the following, silica glass (Suprasil 1 from Heraeus) plates of 0.5mm thick have been used. Considering that the propagation vector is along the direction, the laser beam (800nm, 1kHz or 100kHz, 120fs or 160fs) was focused 250μm below the entry surface using a 0.6NA objective. The sample was moved along a perpendicular direction (let us say) to the laser beam thereby tracing continuous lines. The linear polarization could be set along x (parallel to the sample displacement) or along y (perpendicular to the sample displacement). The laser pulse energy was fixed to 0.5 or 1µJ/pulse i.e. above the second damage threshold and thus within the nanogratings domain. The scanning speed was varied from 10μm/s up to 1000μm/s. The spacing between lines was 10 μm in order to have a surface density of photo-induced effects large enough for the synchrotron beam size. The parameters are summarized in Table 1 .
Table 1. Sample Laser Irradiation Conditions
The photo-luminescence (PL) measurements were performed on the wiggler line of the SOLEIL storage ring (DESIRS beamline). It provides light at almost 99% polarized in any state, with high intensity of excitation and low level of scattered light. We point out that the synchrotron beam power (below 1014photons cm−2 s−1) is much smaller than the laser intensity (typ. 1012W/cm2 i.e. 1032 photons cm−2 s−1), and therefore it is unable to induce a detectable transformation in glasses on our time scale. The linearly polarized synchrotron beam light is monochromatized (bandwidth = 0.1nm) before reaching the sample in the xy plan. Then, luminescence light is collected at 45° of the excitation beam direction by means of a silica lens into a pure silica fiber bundle and detected by a CCD detector after dispersion through a second monochromator. The collecting lens and the gain of the CCD detector are configured as to get a non-saturated spectrum, thus avoiding any distortion of the excitation spectra. In some cases a dichroic UV analyzer was inserted before the CCD detector in order to investigate the luminescence polarization degree P. The PL spectra were recorded in the UV-Vis range for various excitation wavelengths ranging between 4eV and 10eV.
A very important point is the polarization quality of the excitation monochromator and its spectral dependence. This is a crucial question for ensuring that the observed polarization values are indeed a property of the sample and not (in part) caused by the polarization response of instrumentation. Therefore, we determined the ellipse after processing of the beam through the various optics (mirrors, monochromator, windows). We obtained a pure linear polarization (vertical or horizontal) at the sample location. This was measured just upstream of the sample (after the last mirror) with a VUV polarimeter specially designed for this task and that can be inserted at any time in the beam in vacuum. Based on reflections on rotating prisms, this polarimeter can determine with precision (1% relative error) the four Stokes parameters of the incident beam. For a given photon energy, once the polarization measured in a test configuration of the wiggler, one determined by polarimetry the changes to induce to applied to the wiggler (magnetic fields and phase) to obtain a pure polarization at the sample. Thus following this procedure the purity of vertical and horizontal polarization at the sample is greater than 99% in the whole spectral range investigated. The same concerns the luminescence channel detection. The detection channel is based on an optical fiber bundle that is not sensitive to the input PL polarization orientation. Then the PL polarization state before the detecting monochromator is in general elliptical. After performing some tests using the well polarized synchrotron excitation beam as a source, we did not detect any significant impact (less than 1%) on the detection channel on our measurements (see Fig. 1 ).
The intensity of each spectrum was corrected in order to take into account the change of the incident light intensity with the wavelength or with other events like the synchrotron beam light decay and storage ring injections. For that purpose, we used the signal recorded through a gold grid and a pico-ampmeter. In addition, data have been also corrected to take into account the decrease of the excitation beam penetration depth (when smaller than sample thickness) in the VUV range using VUV absorption spectra shown in Fig. 2 .
Fig. 2 Unpolarized UV-VUV absorbance spectra before and after IR-fs irradiation of Suprasil type 1 plate (S1 sample). The full line is for initial spectrum. The empty and full circles are for spectra after irradiation with 0.5μJ/pulse and 1μJ/pulse respectively. The interaction length was estimated to be around 100μm based on optical and electronic microscopy observations performed on the sample cross-section.
UV-VUV absorption spectroscopy was achieved by means of a Jobin Yvon LHT 30 spectrometer equipped with its two PM detectors. A schematic diagram of the dual-beam optical setup developed in this work is shown in [26]. The first beam was reflected by a MgF2 beam-splitter and then focused on the reference photomultiplier. The second beam was transmitted through the beam-splitter, the sample and then focused on the signal PM. The light source was a 30W D2 lamp. The vacuum chamber is evacuated by turbo molecular pumps down to a pressure on the order of 10−6 Torr. To reduce radiation-induced damage to the samples resulting from exposure to the lamp light for a long time, the signal beam was blocked most of the time by means of a shutter except for the time when the optical density spectra were recorded. Spectra were recorded in the 4eV to 10eV energy range with an energy resolution of 10−2 eV.
3. Results
Figure 2 shows absorption spectra of Suprasil Type I plate irradiated with IR-fs laser. More specifically, the full line corresponds to initial attenuation of the pristine sample, the empty and full circles are for the spectra after an exposure at 0.5μJ/pulse and 1μJ/pulse respectively. The initial absorption spectrum contains absorption bands at 5eV with a very small intensity of about 0.5cm−1 and at 7eV with an intensity of about 6cm−1 as well as strong absorption above 7.3eV. The laser exposure induced absorption bands peaking at 5eV, 5.8eV, 6.8eV associated with the growth of the tail of a strong absorption feature peaking at higher photon energy.
Notice that thickness of the irradiated zone is smaller than the sample thickness. We have performed some optical and electronic microscopy observations on the sample cross-section. Based on the observed permanent damages (e.g. permanent refractive index changes), we have used the length of the observed laser tracks (typ. 100 microns) as the path length over which the sample has been effectively modified to calculate the absorption coefficient of irradiated samples.
Figure 3 shows a bird’s-eye view of the PL emission spectra recorded at 300K for excitation energies in the UV-VUV range in the same irradiation conditions used for the absorption study. One can observe two emission bands in this figure. Luminescence bands at 465nm or 2.7eV (labelled as blue) and 290nm or 4.3eV (labelled as UV) can be excited either at 5.0eV or 6.8eV. For energies higher than 8eV the yield of the luminescence is almost zero. Figure 4 shows two typical luminescence spectra excited at 5eV and 6.8eV respectively at 300K. Notice that the two PL spectra exhibit similar shape. In agreement with what is usually observed in silica [27–30], the emission intensity of the 2.7eV emission band is weak (typ. 10 times smaller in our experiments) compared to the 4.3eV emission band [27–32].
Fig. 3 A bird’s-eye view of the PL spectrum mapping of the IR-fs irradiated Suprasil type 1 S1 sample. The excitation energy spanned from 4eV up to 8eV while the PL was recorded between 2eV and 6.5eV. The linear traces on the right hand side of the picture are ghosts of the excitation light.
Fig. 4 Samples of PL spectra of the IR femtosecond irradiated S1 sample. The PL energy was recorded between 2eV and 6.5eV. The dashed line for excitation at 6.8eV while the full line is for excitation at 5eV. Detection is not analyzed.
It should be noted that the maximum band position is subjected to a small shift (less than 0.1eV) together with a small variation of the FWHM in our excitation energy range. In the following, we have thus chosen to present luminescence excitation spectra at one fixed output energy of 4.3eV ± 0.1eV (corresponding to the UV band) since the blue band intensity is weaker. Indeed, monitoring this band as a function of the excitation wavelength is a good way for tracking changes in defect population excitable in the VUV-UV range.
An example of the luminescence excitation spectra recorded in SiO2 is shown in Fig. 5 . The two curves correspond to the spectrum before (full circles) and after (empty circles) exposure to IR femtosecond laser light at 1μJ/pulse. As it can be seen, the pristine sample exhibits no significant PL. The initial absorption spectrum (see Fig. 2) possesses a trace of the band peaking around 7eV with intensity at the level of 5cm−1. The initial low intensity of this absorption band correlates with the fact that the initial photoluminescence intensity is close to zero. In contrast, after IR-fs laser irradiation we detect luminescence with a high yield from the silica sample. Evidently, luminescence centers are created by the IR-fs laser irradiation, giving rise to two excitation bands peaking around 5eV and 6.8eV. This is in agreement with results reported in the 4eV - 6eV excitation range [33].
Fig. 5 UV band UV-VUV excitation spectra before and after IR femtosecond laser irradiation at 1μJ/pulse (S1 sample). The laser polarization x was parallel to the sample displacement. ● is for the pristine sample while ❍ is after exposure. Detection is not analyzed.
Figure 6 shows the two excitation spectra for the two excitation polarization taken at 300K after the exposure of the SiO2 plate to x-polarized femtosecond laser light at 800nm (the sample scanning direction is x). The parameter of the experiment is the polarization orientation of the synchrotron beam probe with regards to the one of the IR-fs irradiation and the scanning direction. We can observe some differences in the shape of the spectra according to the synchrotron beam polarization direction. More precisely, when the synchrotron beam polarization is parallel (along x, full circles) to the linear polarization of the IR-fs laser, the intensity Ix of the 6.8eV excitation band is much higher than the intensity Iy for a synchrotron beam polarization along y (empty circles).
Fig. 6 UV-VUV excitation spectra of the UV emission band after IR-fs laser irradiation of S1 sample with the laser writing polarization parallel to the laser scanning. The parameter of the experiment is the polarization direction of the synchrotron beam probe relatively to the writing direction: ● is for synchrotron beam polarization parallel to the writing direction x while ❍ is for perpendicular orientation. Detection is not analyzed.
The excitation polarization degree (Ix-Iy)/(Ix + Iy) (EPD) of the 6.8eV band is quite high i.e. 21%. In contrast, the EPD for the 5eV excitation band is significantly smaller (typ. 10%). This linear dichroism in excitation indicates an anisotropic defect partially oriented. In addition as reported in Table 2 , we obtained the same EPD when the laser writing polarization is perpendicular (along y) or parallel (along x) to the sample scanning direction x. Therefore, the defect anisotropy is related to the direction of the laser scanning direction but is independent of the IR laser polarization orientation. Here, we used the luminescence band at 4.3eV but we have obtained similar results for the blue luminescence band. A quite interesting result is also that the anisotropy is writing-speed dependent. The slower is the writing speed, the higher is the anisotropy degree. Indeed as shown in Table 3 , the EPD of the UV band increases from 11% up to 27% when the writing speed changes from 1000μm/s down to 100μm/s. The blue luminescence band follows similar trend.
Table 2. EPD’s (Ix-Iy)/(Ix + Iy) of the Excitation of the Luminescence Bands in Irradiated Silica Samples*
Table 3. EPD (Ix-Iy)/(Ix + Iy) of the 6.8 eV Excitation Band in Irradiated Silica Samples*
4. Discussion
Band assignment: It has been demonstrated that after the generation of electron-hole pairs and then excitons, the defect formation starts from the relaxation of self-trapped excitons [34]. Self Trapped Excitons (STE) are formed in a few ps [35–37]. In SiO2, besides radiative recombination, they may relax into SiE’ and NBOHC [34]. SiE’ (Silicon dangling bond, Si•) absorb at 5.8eV [38] as we observed in Fig. 1 but do not give rise to luminescence. NBOHC (Non-Bridging Oxygen Hole center, Si-O•) absorb at 4.8eV and 2.0eV and give rise to luminescence at 1.9eV [39,40]. Their 4.8eV band can contribute to the 5.0eV band observed in Fig. 2. This defect has not been tracked in this paper but its PL at 1.9eV has been recently observed in similar experimental conditions (fs laser irradiated silica glass excited at 5.0eV) by Watanabe et al. [33]. Then SiE’ and NBOHC can recombine into SiODC(II) and POR (Peroxy linkage Si-O-O-Si or peroxy radical Si-O-O•) that absorbs in the VUV [40–42]. There is also another relaxation channel that can produce SiODC(I) and interstitial oxygen [43].
SiODC(II) (twofold coordinated silicium defect, Si: [27]) absorb at 5.0eV (S0→S1) [40,44] and 7eV (S0→S2) [31,45]. As it is the case in Fig. 3, both excitation channels (at 5.0 or 6.8eV) give rise to two luminescence bands at 2.7eV (labelled as blue) and 4.3eV (labelled as UV) as it is already seen in [29,31,32]. Emission band centred at ~4.3eV under ~5.0eV excitation (usually referred to as the α band having short lifetime (typ. 5 ns)) is due, in SiODC(II), to the radiative transition between the first singlet state S1 and the ground state S0. The emission band centred at ~2.7eV under the ~5.0eV excitation, named β band (emission lifetime ≈10 ms [27,46], ), is due to the spin-orbit forbidden triplet-singlet transition T1→S0. This last is weaker than the UV emission band. The triplet state T1 is populated from S1 via a phonon assisted Inter-System Crossing (ISC) with a high activation energy barrier. This is well described in [46].
SiODC(I) (oxygen monovacancy, Si-Si) absorbs around 7.6eV [40,47] and gives rise to a similar luminescence as SiODC(II) except that UV emission band is here weaker than blue band. However, the weight of excitation channel (at 5.0eV for SiODC(II) or 7.6eV for SiODC(I)) depends on the chemical composition of the glass [48–54]. As we did not observe any luminescence when the samples were excited above around 7.6eV no contribution can be ascribed to SiODC(I). A similar observation has been reported in [29]. The strong absorption band peaking above 7eV can thus be likely ascribed to O2 [55,56].
We find in our experiment that both emission bands (α and β) can be excited at ~5.0eV as it is the case for SiODC(II) (S0→S1 band). Similarly, excitation around 6.8eV is from singlet S0 to singlet S2 of the same defect. These considerations lead us to assign the photoluminescence bands to only SiODC(II) defects. The formation of these defects and of O2 gaz suggests silica oxide dissociation that we have previously revealed in similar irradiation conditions [11]. This dissociation leads to under-stoichiometric material (and thus related to ODC defects) and oxygen nanobubbles [13].
Beside the luminescence reported in this paper, Watanabe [33] mentioned also a 2.2eV (564nm) PL band. We have not searched for this band, but it is useful to note that it is observed under various conditions like electron irradiation when the sample is crushed, i.e. each time that a surface is involved. In the writing conditions used in Watanabe et al. [57], it produces nanogratings and a lot of cracks. Therefore, this band likely arises from surface defects rather than from bulk defects [58].
Polarization data: Now, we shall discuss the luminescence polarization properties. Due to random orientations of individual luminescence centers in glass, luminescence is usually not polarized if the excitation is not polarized. However when excitation is done by polarized light, the luminescence yield can be polarized even in isotropic sample. This indicates that the respective excitation and emission transitions are not totally independent and some symmetry relation exists between their transition dipole moments. More precisely, polarization studies allow obtaining information on the anisotropy degree of the oscillators, their orientation, the point group (n fold symmetry) of luminescent centers, and the involved spectral transition. It can also provide information on the interactions with the surrounding medium.
Let us compute the luminescence polarization degree (LPD) in the case for which the directions of the absorbing and emitting oscillators make an angle α between them. It is defined by (I//- I⊥)/(I// + I⊥) with I// and I⊥, the luminescence intensities measured with the analyzer parallel or perpendicular to the electric vector of the excitation light. Note that the LPD concerns the defect properties and the direction distribution statistics (it is relative to the polarization of the luminescence emission, the excitation being fixed) whereas what we call the EPD is relative to dependence of the luminescence intensity on the polarization of the excitation, the luminescence emission being not analyzed. When exciting with linearly polarized light, it can be shown [59] that the limiting value of the luminescence polarization degree is (3cos2 α −1) / (cos2 α + 3) for randomly oriented centers, α is the angle between the direction of the absorbing and the emitting oscillator. The theoretical upper limit is 0.5 when α = 0 i.e. when absorbing centers and emitting oscillator are parallel. This is almost never approached in practice [27,30,46]. However, if the medium contain partially oriented anisotropic emitting centers as it is suggested in Fig. 6, the LPD may be considerably higher and exceed 0.5 depending on the specific structure of the center and its neighborhood [59]. In the specific case of complete orientation of the C2v centers that has usually attributed to SiODC(II) [27], the LPD upper limit would be 1 [59].
Figure 7 shows the LPD of the 4.4eV luminescence band as function of the excitation energy. P is positive (typ. + 0.2) around both 5eV and 7eV. The non-zero LPD indicates that both photoluminescence excitation bands are optically anisotropic. For 5eV excitation, the sign and the magnitude of P are in good agreement with those previously published [30,46,60] in isotropic SiO2 (i.e. non irradiated). In contrast, for 6.8eV excitation, the LPD is quite high compared to previous publications [30]. This can be explained by a preferential orientation of anisotropic defects along the laser scanning direction as shown in Fig. 6. In addition as reported in Table 2, the EPD is significantly higher for 6.8eV excitation (typ. 21%) when compared to 5.0eV excitation pathway (typ. 7%). So how to explain this discrepancy between these two excitation bands whereas they are related to the same defect SiODC(II). The explanation can be found in the orientation distribution of the defects combined with the symmetry of the excited states. Assuming that SiODC(II) exhibit a C2v local symmetry [46].
Fig. 7 Polarization degree LPD of the α band versus the excitation wavelength. Detection at 4.4eV is analyzed. The writing polarization was parallel to the laser scanning direction x (sample S1). The probe polarization of the excitation was parallel to the laser scanning direction (x).
The ground state is 1A1 and the first excited singlet is 1B1. Transition from one to the other is due to an electron transiting from an orbital with symmetry a1 to an orbital with symmetry b1 built mainly on the “non-bonding” Si 3px atomic orbital (x being perpendicular to the molecular plane). The S0-S1 transition is thus polarized along x direction. In contrast, the second excited singlet term S2 (corresponding to the excitation band at 6.8eV) is related to transition from 1A1 to 1B2 spectroscopic level related to an electron transiting from a a1 orbital to a b2 orbital built mainly on the Si3p-O2p anti-bonding molecular orbital in the molecular plan. This results in transition between S0 and S2 that is polarized in the O-Si-O plane [27]. Then, if excitation polarization reveals an anisotropy (i.e. excitation dichroism) larger for 1B2 (6.8eV excitation) than for 1B1 (5.0eV excitation), this means that the “molecules” are more aligned following b2 symmetry than following b1 symmetry, and the transition operator with b2 symmetry corresponds to y axis of the molecule that goes through the oxygen atoms. As the luminescence intensity excited at 6.8eV is larger when probe polarization (or the dipole operator) is parallel to the direction of writing, this means that the axis passing through the oxygen atoms is aligned with the laser scanning direction. This is summarized in Fig. 8 .
Fig. 8 Symmetry of the excited states of SiODC(II). If S0→S2 more activated when s pol // written line ⇒ O-O axis is partially aligned along the scanning direction. The anisotropic defect is partially oriented, not by writing laser polarization but rather by the scanning.
However, the fact that the EPDs of the blue luminescence band is not the same as the EPDs of the UV luminescence band (see Table 2) is problematic taking into account the knowledge cumulated for more than 50 years. The spectroscopic data that we have presented above are consistent with the property of only one defect SiODC(II) but the observed EPDs show that they should be several excitation channels giving rise to two luminescence bands, one in the blue range (at 2.7eV) and one in the UV range (at 4.4eV). This is commonly encountered in doped silica [44] but not in pure silica. This is a point to clear with specific experiment that we have currently proposed to synchrotron facilities.
Mechanisms of defects orientation: The orientation of the defects is shown to be determined only by the laser scanning direction and not by the writing polarization direction; it is thus dependent on writing speed. More specifically, the efficiency of orientation is larger as the speed is smaller. The overlap between two consecutive pulses should thus be large. However, relation between two consecutive pulses cannot occur otherwise than via the matter as the pulses are largely separated from each other. On the other hand, it has been shown [61] that photo-induced defects ensure a memory role. We can imagine thus that the ionization from a matter already irradiated is larger than from a virgin matter. This could induce a gradient in the free electron plasma density that gives rise to DC electric field and defect orientation. However, taking into account the dipole orientation of ODC(II), this does not correlate with orientation of O-O axis in O-Si-O with the scanning direction since the static dipole of the defect is oriented perpendicularly.
Another contributing force could be a photo-induced shear stress [25,62]. Indeed, it is well known that shear stress favours defect centre formation and it can also induce a large degree of alignment for the randomly oriented anisotropic defects [63]. As a matter of fact, in similar irradiation conditions (100TW/cm2), from the polarization and scanning direction dependence of the surface topography after cleaving of the laser tracks (related stress relaxation) we have previously revealed a shear stress contribution dependent on only scanning direction and not on direction of the excitation polarization. It is something like a solid drag effect [64,65]. We suggest thus that this solid drag effect induces a large degree of alignment (typ. 20%) for the randomly oriented anisotropic defects.
Beside, it is useful to precise the relation with the formation of porous nanoplanes within the laser tracks that are obtained in the same conditions [11–13]. Defect creation could be the result of the non-spherical shape of the nanopores which forms the nanogratings. Indeed the nanopores exhibit oblate shape with a small axis (typ. 20nm) oriented in the writing laser polarization direction and a long axis (typ. 30-50nm) oriented in the perpendicular direction [11–13]. Such mesoporous nanostructures should be accompanied with the formation of SiODC defects at the pore-background material interface since the background material is under-stoichiometric due to the silica oxide dissociation and the formation of nanopores filled with O2 [11–13]. Then, vertical (excitation light aligned with the laser polarization) linearly polarized light selectively excites the defects localized around the long axis resulting in a stronger PL as we observed. However when we turn the writing laser polarization (i.e. we turn also the nanogratings orientation), this should affect the anisotropic luminescence as well which is not the case here. Indeed the luminescence remains more efficient in the same direction i.e. the direction of the sample displacement whatever the writing polarization may be.
5. Conclusion
Through analysing the polarization properties of the luminescence induced by IR-fs irradiation in pure silica, we show the creation of oriented SiODC(II) defects. The irradiation conditions used falls in the domain of nanogratings formation. We have previously shown that the nanoplanes at the base of this nanostructure are made by the decomposition of SiO2 into understoichiometric defects and O2 nanobubbles. Therefore, it was expected that SiODC(II) would be the product of this decomposition. However, although these nanostructures depend on the IR-fs laser polarization, SiODC(II) orientation does not depend on the polarization direction but on the writing direction. They are appearing as uncorrelated with nanogratings.
Nevertheless, we have also shown that besides nanogratings, there are other nanostructures that are not dependent on polarization but on the direction of writing. They are also associated to shearing of the matter or to some drag effect according to laser beam structure. In addition, we show that they are dependent on the writing speed indicating a memory effect from pulse to pulse.The next study will be to map the defect distribution along the laser track in order to make a correlation with one component of shear stress.
Acknowledgments
This work has been achieved in the frame of FLAG (Femtosecond Laser Application in Glasses, www.flag.u-psud.fr) consortium project with the support of several organisations: the Agence Nationale pour la Recherche (ANR-09-BLAN-0172-01), the PRES UniverSud Paris (Pôle de Recherche et d'Enseignement Supérieur, 2008-39), the RTRA Triangle de la Physique (Réseau Thématique de Recherche Avancée, 2008-056T), and the Essonne administrative Department (ASTRE2007). Experiments were performed at DESIRS beamline at SOLEIL synchrotron, France. We are grateful to Nelson De Oliveira for assistance and to the SOLEIL staff for smoothly running facility.
References and links
1. K. Itoh, W. Watanabe, S. Nolte, and C. Schaffer, “Ultrafast processes for bulk modification of transparent materials,” MRS Bull. 31(08), 620–625 (2006). [CrossRef]
2. M. Ams, G. Marshall, P. Dekker, M. Dubov, V. Mezentsev, I. Bennion, and M. Withford, “Investigation of ultrafast laser–photonic material interactions: challenges for directly written glass photonics,” IEEE J. Sel. Top. Quantum Electron. 14(5), 1370–1381 (2008). [CrossRef]
3. J. Qiu, K. Miura, and K. Hirao, “Femtosecond laser-induced microfeatures in glasses and their applications,” J. Non-Cryst. Solids 354(12-13), 1100–1111 (2008). [CrossRef]
4. B. Poumellec, M. Lancry, A. Chahid-Erraji, and P. Kazansky, “Modification thresholds in femtosecond laser processing of pure silica: review of dependencies on laser parameters [Invited],” Opt. Mater. Express 1(4), 766–782 (2011). [CrossRef]
5. E. Bricchi, B. G. Klappauf, and P. G. Kazansky, “Form birefringence and negative index change created by femtosecond direct writing in transparent materials,” Opt. Lett. 29(1), 119–121 (2004). [CrossRef] [PubMed]
6. M. Lancry, B. Poumellec, A. Chahid-Erraji, M. Beresna, and P. Kazansky, “Dependence of the femtosecond laser refractive index change thresholds on the chemical composition of doped-silica glasses,” Opt. Mater. Express 1(4), 711–723 (2011). [CrossRef]
7. M. Lancry, P. Niay, and M. Douay, “Comparing the properties of various sensitization methods in H2-loaded, UV hypersensitized or OH-flooded standard germanosilicate fibers,” Opt. Express 13(11), 4037–4043 (2005). [CrossRef] [PubMed]
8. M. Lancry and B. Poumellec, “Multiphoton absorption processes & UV laser processing of silica-based materials,” Phys. Rep. (to be published).
9. E. Bricchi and P. Kazansky, “Extraordinary stability of anisotropic femtosecond direct-written structures embedded in silica glass,” Appl. Phys. Lett. 88(11), 111119 (2006). [CrossRef]
10. Y. Shimotsuma, P. G. Kazansky, J. Qiu, and K. Hirao, “Self-organized nanogratings in glass irradiated by ultrashort light pulses,” Phys. Rev. Lett. 91(24), 247405 (2003). [CrossRef] [PubMed]
11. M. Lancry, B. Dufaure, and B. Poumellec, “Self-aligned porous nanoplanes photo-induced by IR femtosecond irradiation in silica glass,” XII Conference on the Physics of Non-Crystalline Solids, Foz do Iguaçu, Brazil (2009).
12. J. Canning, M. Lancry, K. Cook, A. Weickman, F. Brisset, and B. Poumellec, “Anatomy of a femtosecond laser processed silica waveguide [Invited],” Opt. Mater. Express 1(5), 998–1008 (2011). [CrossRef]
13. M. Lancry, K. Cook, J. Canning, and B. Poumellec, “Nanogratings and molecular oxygen formation during femtosecond laser irradiation in silica,” in The International Quantum Electronics Conference (IQEC)/The Conference on Lasers and Electro-Optics (CLEO) Pacific Rim (2011).
14. D. Du, X. Liu, G. Korn, J. Squier, and G. Mourou, “Laser-induced breakdown by impact ionization in SiO2 with pulse-widths from 7 ns to 150 fs,” Appl. Phys. Lett. 64(23), 3071–3073 (1994). [CrossRef]
15. J. W. Chan, T. Huser, S. Risbud, and D. M. Krol, “Structural changes in fused silica after exposure to focused femtosecond laser pulses,” Opt. Lett. 26(21), 1726–1728 (2001). [CrossRef] [PubMed]
16. H. Sun, S. Juodkazis, M. Watanabe, S. Matsuo, H. Misawa, and J. Nishii, “Generation and recombination of defects in vitreous silica induced by irradiation with a near-infrared femtosecond laser,” J. Phys. Chem. B 104(15), 3450–3455 (2000). [CrossRef]
17. J. Chan, T. Huser, S. Risbud, and D. Krol, “Modification of the fused silica glass network associated with waveguide fabrication using femtosecond laser pulses,” Appl. Phys., A Mater. Sci. Process. 76(3), 367–372 (2003). [CrossRef]
18. W. Reichman, D. Krol, L. Shah, F. Yoshino, A. Arai, S. Eaton, and P. Herman, “A spectroscopic comparison of femtosecond-laser-modified fused silica using kilohertz and megahertz laser systems,” J. Appl. Phys. 99(12), 123112 (2006). [CrossRef]
19. W. Reichman, D. Krol, L. Shah, F. Yoshino, A. Arai, S. Eaton, and P. Herman, “Fluorescence and Raman microscopy of waveguides fabricated using kHz and MHz repetition rate femtosecond lasers,” in Lasers and Electro-Optics / Quantum Electronics and Laser Science Conference (CLEO/QELS) (2006).
20. A. Zoubir, M. Richardson, L. Canioni, A. Brocas, and L. Sarger, “Optical properties of infrared femtosecond laser-modified fused silica and application to waveguide fabrication,” J. Opt. Soc. Am. B 22(10), 2138–2143 (2005). [CrossRef]
21. A. Zoubir, M. Richardson, T. Cardinal, L. Canioni, A. Brocas, and L. Sarger, “Nonlinear optics of femtosecond laser-modified fused silica: applications to waveguide fabrication,” in Conference on Lasers and Electro-Optics/Quantum Electronics and Laser Science and Photonic Applications Systems Technologies, Technical Digest (CD) (Optical Society of America, 2005), paper CFG2.
22. H. Hosono, K. Kawamura, S. Matsuishi, and M. Hirano, “Holographic writing of micro-gratings and nanostructures on amorphous SiO2 by near infrared femtosecond pulses,” Nucl. Instrum. Methods Phys. Res. 191(1-4), 89–97 (2002). [CrossRef]
23. A. Couairon, L. Sudrie, M. Franco, B. Prade, and A. Mysyrowicz, “Filamentation and damage in fused silica induced by tightly focused femtosecond laser pulses,” Phys. Rev. B 71(12), 125435 (2005). [CrossRef]
24. L. Sudrie, M. Franco, B. Prade, and A. Mysyrowicz, “Study of damage in fused silica induced by ultra-short IR laser pulses,” Opt. Commun. 191(3-6), 333–339 (2001). [CrossRef]
25. B. Poumellec, L. Sudrie, M. Franco, B. Prade, and A. Mysyrowicz, “Femtosecond laser irradiation stress induced in pure silica,” Opt. Express 11(9), 1070–1079 (2003). [CrossRef] [PubMed]
26. M. Lancry, B. Poumellec, P. Niay, M. Douay, P. Cordier, and C. Depecker, “VUV and IR absorption spectra induced in H2-loaded and UV hyper-sensitized standard germanosilicate preform plates through exposure to ArF laser light,” J. Non-Cryst. Solids 351(52-54), 3773–3783 (2005). [CrossRef]
27. L. Skuja, A. N. Streletsky, and A. Pakovich, “A new intrinsic defect in amorphous SiO2: twofold coordinated silicon,” Solid State Commun. 50(12), 1069–1072 (1984). [CrossRef]
28. R. Tohmon, Y. Yamasaka, K. Nagasawa, Y. Ohki, and Y. Hama, “Cause of the 5.0 eV absorption band in pure silica glass+,” J. Non-Cryst. Solids 95-96, 671–678 (1987). [CrossRef]
29. A. Trukhin and H. Fitting, “Investigation of optical and radiation properties of oxygen deficient silica glasses,” J. Non-Cryst. Solids 248(1), 49–64 (1999). [CrossRef]
30. A. Trukhin, B. Poumellec, and J. Garapon, “Study of the germanium luminescence in silica: from non-controlled impurity to germano-silicate core of telecommunication fiber preforms,” J. Non-Cryst. Solids 332(1-3), 153–165 (2003). [CrossRef]
31. H. Nishikawa, E. Watanabe, D. Ito, and Y. Ohki, “Decay kinetics of the 4.4-eV photoluminescence associated with the two states of oxygen-deficient-type defect in amorphous SiO2.,” Phys. Rev. Lett. 72(13), 2101–2104 (1994). [CrossRef] [PubMed]
32. N. Nishikawa, Y. Miyake, E. Watanabe, D. Ito, K. S. Seol, Y. Ohki, K. Ishii, Y. Sakurai, and K. Nagasawa, “Photoluminescence of oxygen-deficient-type defects in α-SiO2,” J. Non-Cryst. Solids 222, 221–227 (1997).
33. M. Watanabe, S. Juodkazis, H. Sun, S. Matsuo, and H. Misawa, “Luminescence and defect formation by visible and near-infrared irradiation of vitreous silica,” Phys. Rev. B 60(14), 9959–9964 (1999). [CrossRef]
34. S. Mao, F. Quere, S. Guizard, X. Mao, R. Russo, G. Petite, and P. Martin, “Dynamics of femtosecond laser interactions with dielectrics,” Appl. Phys., A Mater. Sci. Process. 79(7), 1695–1709 (2004). [CrossRef]
35. P. Martin, S. Guizard, P. Daguzan, G. Petite, P. D'Oliveira, P. Meynadier, and M. Perdrix, “Subpicosecond study of carrier trapping dynamics in wide-band-gap crystals,” Phys. Rev. B 55(9), 5799–5810 (1997). [CrossRef]
36. M. Lancry, N. Groothoff, S. Guizard, W. Yang, B. Poumellec, P. Kazansky, and J. Canning, “Femtosecond laser direct processing in wet and dry silica glass,” J. Non-Cryst. Solids 355(18-21), 1057–1061 (2009). [CrossRef]
37. M. Lancry, N. Groothoff, B. Poumellec, S. Guizard, N. Fedorov, and J. Canning, “Time-resolved plasma measurements in Ge-doped silica exposed to infrared femtosecond laser,” Phys. Rev. B 84(24), 245103 (2011). [CrossRef]
38. R. Weeks, “Paramagnetic resonance of lattice defects in irradiated quartz,” J. Appl. Phys. 27(11), 1376–1381 (1956). [CrossRef]
39. H. Hosono, K. Kajihara, T. Suzuki, Y. Ikuta, L. Skuja, and M. Hirano, “Vacuum ultraviolet optical absorption band of non-bridging oxygen hole centers in SiO2 glass,” Solid State Commun. 122(3-4), 117–120 (2002). [CrossRef]
40. L. Skuja, M. Hirano, H. Hosono, and K. Kajihara, “Defects in oxide glasses,” Phys. Status Solidi C 2(1), 15–24 (2005). [CrossRef]
41. G. Pacchioni and G. Ierańo, “Ab initio theory of optical transitions of point defects in SiO2,” Phys. Rev. B 57(2), 818–832 (1998). [CrossRef]
42. B. Stefanov and K. Raghavachari, “Photoabsorption of the peroxide linkage defect in silicate glasses,” J. Chem. Phys. 111(17), 8039–8042 (1999). [CrossRef]
43. L. Skuja, K. Kajihara, T. Kinoshita, M. Hirano, and H. Hosono, “The behavior of interstitial oxygen atoms induced by F2 laser irradiation of oxygen-rich glassy SiO2,” Nucl. Instrum. Methods Phys. Res. B 191(1-4), 127–130 (2002). [CrossRef]
44. L. Skuja, “Optically active oxygen-deficiency-related centers in amorphous silicon dioxide,” J. Non-Cryst. Solids 239(1-3), 16–48 (1998). [CrossRef]
45. E. O'Reilly and J. Robertson, “Theory of defects in vitreous silicon dioxide,” Phys. Rev. B 27(6), 3780–3795 (1983). [CrossRef]
46. L. Skuja, “Isoelectronic series of twofold coordinated Si, Ge, and Sn atoms in glassy SiO2: a luminescence study,” J. Non-Cryst. Solids 149(1-2), 77–95 (1992). [CrossRef]
47. H. Hosono, Y. Abe, H. Imagawa, H. Imai, and K. Arai, “Experimental evidence for the Si-Si bond model of the 7.6-eV band in SiO2 glass,” Phys. Rev. B Condens. Matter 44(21), 12043–12045 (1991). [CrossRef] [PubMed]
48. S. Agnello, R. Boscaino, M. Cannas, A. Cannizzo, F. M. Gelardi, S. Grandi, and M. Leone, “Spectral heterogeneity of oxygen-deficient centers in Ge-doped silica,” Radiat. Meas. 38(4-6), 645–648 (2004). [CrossRef]
49. A. Anedda, C. Carbonaro, F. Clemente, R. Corpino, and A. Serpi, “Excitation pattern of the blue emission in Ge-doped silica,” J. Non-Cryst. Solids 315(1-2), 161–165 (2003). [CrossRef]
50. J. Garapon, “Etude théorique des défauts déficients en oxygène dans la silice pure ou dopée,” PhD thesis dissertation (Université de Paris 11, Orsay, France, 2001).
51. M. Lancry, B. Poumellec, and M. Douay, “UV excited luminescence behavior in OH-flooded Ge-doped silica preform plates exposed to 193nm laser light,” in Bragg Gratings, Photosensitivity, and Poling in Glass Waveguides, OSA Technical Digest (CD) (Optical Society of America, 2007), paper JWA58.
52. M. Leone, R. Boscaino, M. Cannas, and F. Gelardi, “The landscape of the excitation profiles of the αE and β emission bands in silica,” J. Non-Cryst. Solids 245(1-3), 196–202 (1999). [CrossRef]
53. B. Poumellec, V. Mashinsky, A. Trukhin, and P. Guenot, “270 nm absorption and 432 nm luminescence bands in doped silica glasses,” J. Non-Cryst. Solids 239(1-3), 84–90 (1998). [CrossRef]
54. B. Poumellec, T. Taunay, P. Bernage, R. Cortes, and J. Krupa, “Defect population in silica glasses studied by luminescence VUV excitation spectroscopy,” J. Lumin. 72-74, 442–445 (1997). [CrossRef]
55. K. Kajihara, M. Hirano, L. Skuja, and H. Hosono, “Vacuum-ultraviolet absorption of interstitial O2 and H2O molecules in SiO2 glass,” J. Non-Cryst. Solids 352(23-25), 2303–2306 (2006). [CrossRef]
56. K. Kajihara, T. Miura, H. Kamioka, A. Aiba, M. Uramoto, Y. Morimoto, M. Hirano, L. Skuja, and H. Hosono, “Diffusion and reactions of interstitial oxygen species in amorphous SiO2: a review,” J. Non-Cryst. Solids 354(2-9), 224–232 (2008). [CrossRef]
57. C. Hnatovsky, R. Taylor, E. Simova, P. Rajeev, D. Rayner, V. Bhardwaj, and P. Corkum, “Fabrication of microchannels in glass using focused femtosecond laser radiation and selective chemical etching,” Appl. Phys., A Mater. Sci. Process. 84(1-2), 47–61 (2006). [CrossRef]
58. V. A. Radzig, V. N. Bagratashvili, S. I. Tsypina, P. V. Chernov, and A. O. Rybaltovskii, “Photoinduced reactions of oxygen deficient centers with molecular hydrogen in silica glasses,” J. Phys. Chem. 99(17), 6640–6647 (1995). [CrossRef]
59. P. Feofilov, The Physical Basis of Polarized Emission (Consultants Bureau, 1961).
60. M. Lancry, B. Poumellec, and M. Douay, “Anisotropic luminescence photo-excitation in H2-loaded Ge-doped silica exposed to polarized 193nm laser light,” J. Non-Cryst. Solids 355(18-21), 1062–1065 (2009). [CrossRef]
61. S. Richter, F. Jia, M. Heinrich, S. Döring, U. Peschel, A. Tünnermann, and S. Nolte, “The role of self-trapped excitons and defects in the formation of nanogratings in fused silica,” Opt. Lett. 37(4), 482–484 (2012). [CrossRef] [PubMed]
62. B. Poumellec, M. Lancry, J. C. Poulin, and S. Ani-Joseph, “Non reciprocal writing and chirality in femtosecond laser irradiated silica,” Opt. Express 16(22), 18354–18361 (2008). [CrossRef] [PubMed]
63. C. Wu, P. Tsay, H. Cheng, and S. Bai, “Polarized luminescence and absorption of highly oriented, fully conjugated, heterocyclic aromatic rigid-rod polymer poly-p-phenylenebenzobisoxazole,” J. Appl. Phys. 95(2), 417–423 (2004). [CrossRef]
64. P. Kazansky, W. Yang, E. Bricchi, J. Bovatsek, A. Arai, Y. Shimotsuma, K. Miura, and K. Hirao, “'Quill' writing with ultrashort light pulses in transparent materials,” Appl. Phys. Lett. 90(15), 151120 (2007). [CrossRef]
65. M. Lancry, W. Yang, B. Poumellec, and B. Bourguignon, “Scan speed dependence of quill writing with ultrashort laser pulses in fused silica,” OSA topical meeting, Femtosecond Laser Microfabrication (2009).
