femtosecond laser induced density changes in geo2 and sio2 ... · femtosecond laser induced density...

Femtosecond laser induced densitychanges in GeO2 and SiO2 glasses:fictive temperature effect [Invited]

Lena Bressel,1 Dominique de Ligny,1 Camille Sonneville,1

Valerie Martinez,1 Vygantas Mizeikis,2 Ricardas Buividas,3 andSaulius Juodkazis3,∗

1 Universite Lyon 1, Universite de Lyon, Laboratoire de Physico-Chimie des MateriauxLuminescents, UMR 5620, 12 rue Ada Byron, 69622 Villeurbanne, France

2Division of Global Research Leaders, Research Institute for Electronics, Shizuoka University,3-5-1 Johoku, Naka-ku, Hamamatsu 432-8561, Japan

3 Centre for Micro-Photonics, Faculty of Engineering and Industrial Sciences,Swinburne University of Technology, Hawthorn, VIC, 3122, Australia

∗[email protected]

Abstract: Density changes of GeO2 and SiO2 glasses subjected toirradiation by tightly focused femtosecond pulses are observed by Ramanscattering. It is shown that densification caused by the void formation inGeO2 glass is very similar to the changes under hydrostatic pressure. Incontrast, the experimental observations in SiO2 glass could be explainedby pressure effect or by the fictive temperature anomaly, i. e., a resultantsmaller specific volume of the glass (a denser phase) at a high thermalquenching rate. Density changes of GeO2 and SiO2 glasses are oppositeupon close-to-equilibrium heating; this gives new insights into the mech-anisms of densification under highly non-equilibrium conditions: fs-laserinduced micro-explosions, heating and void formation. The pressure andtemperature effects of glass modification by ultra-short laser pulses arediscussed considering applications in optical memory, waveguiding, andformation of micro-optical elements.

© 2011 Optical Society of America

OCIS codes: (160.2750) Glass and other amorphous materials; (350.3850) Materials process-ing; (140.3390) Laser materials processing; (140.3440) Laser-induced breakdown.

References and links1. G. Cheng, K. Mishchik, C. Mauclair, E. Audouard, and R. Stoian, “Ultrafast laser photoinscription of polarization

sensitive devices in bulk silica glass,” Opt. Express 17, 9515–9525 (2009).2. G. Cerullo, R. Osellame, S. Taccheo, M. Marangoni, D. Polli, R. Ramponi, P. Laporta, and S. D. Silvestri,

“Femtosecond micromachining of symmetric waveguides at 1.5μm by astigmatic beam focusing,” Opt. Lett. 27,1938–1940 (2002).

3. A. Benayas, D. Jaque, B. McMillen, and K. P. Chen, “High repetition rate UV ultrafast laser inscription ofburied channel waveguides in sapphire: Fabrication and fluorescence imaging via ruby R lines,” Opt. Express 17,10076–10081 (2009).

4. S. M. Eaton, H. Zhang, M. L. Ng, J. Z. Li, W. J.Chen, S. Ho, and P. R. Herman, “Transition from thermaldiffusion to heat accumulation in high repetition rate femtosecond laser writing of buried optical waveguides,”Opt. Express 16, 9443–9458 (2008).

5. W. Gawelda, D. Puerto, J. Siegel, A. Ferrer, A. R. de la Cruz, H. Fernandez, and J. Solis, “Ultrafast imagingof transient electronic plasmas produced in conditions of femtosecond waveguide writing in dielectrics,” Appl.Phys. Lett. 93, 121109 (2008).

#147200 - $15.00 USD Received 10 May 2011; revised 6 Jul 2011; accepted 6 Jul 2011; published 13 Jul 2011(C) 2011 OSA 1 August 2011 / Vol. 1, No. 4 / OPTICAL MATERIALS EXPRESS 605

6. S. Hirono, M. Kasuya, K. Matsuda, Y. Ozeki, K. Itoh, H. Mochizuki, and W. Watanabe, “Increasing diffractionefficiency by heating phase gratings formed by femtosecond laser irradiation in poly(methyl methacrylate),”Appl. Phys. Lett. 94, 241122 (2009).

7. D. M. Krol, “Femtosecond laser modification of glass,” J. Non-Cryst. Solids 354, 416–424 (2009).8. S. Nolte, M. Will, J. Burghoff, and A. Tunnermann, “Femtosecond waveguide writing: a new avenue to three-

dimensional integrated optics,” Appl. Phys. A 77, 109–111 (2003).9. J. Siebenmorgen, K. Petermann, G. Huber, K. Rademaker, and S. N. A.Tunnermann, “Femtosecond laser written

stress-induced Nd:Y3Al5O12(Nd:YAG) channel waveguide laser,” Appl. Phys. B 97, 151–255 (2009).10. Y. Bellouard, M. Dugan, A. A. Said, and P. Bado, “Thermal conductivity contrast measurement of fused silica

exposed to low-energy femtosecond laser pulses,” Appl. Phys. Lett. 89, 161911 (2006).11. T. Kudrius, G. Slekys, and S. Juodkazis, “Surface-texturing of sapphire by femtosecond laser pulses for photonic

applications,” J. Phys. D: Appl. Phys. 43, 145501 (2010).12. G. T. Skublov, Y. B. Marin, V. M. Semikolennykh, S. G. Skublov, and Y. N. Tarasenko, “Volkhovite: A new type

of tektite-like glass,” Geol. Ore Deposits 49, 681–696 (2007).13. V. Bouska, Z. Borovec, A. Cimbalnıkova, I. Kraus, A. Lajcakova, and M. Pacesova, Natural Glasses (Academia,

Prague and Ellis Horwood, London, 1993).14. A. Koike and M. Tomozawa, “IR investigations of density changes of silica glass and soda-lime silicate glass

caused by micro-hardness indentation,” J. Non-Cryst. Solids 353, 2318–2327 (2007).15. M. Malinauskas, A. Zukauskas, G. Bickauskaite, R. Gadonas, and S. Juodkazis, “Mechanisms of three-

dimensional structuring of photo-polymers by tightly focussed femtosecond laser pulses,” Opt. Express 18,10209–10221 (2010).

16. M. Malinauskas, P. Danilevicius, and S. Juodkazis, “Three-dimensional micro-/nano-structuring via direct writepolymerization with picosecond laser pulses,” Opt. Express 19, 5602–5610 (2011).

17. M. Sakakura, M. Terazima, Y. Shimotsuma, K. Miura, and K. Hirao, “Thermal and shock induced modificationinside a silica glass by focused femtosecond laser pulse,” J. Appl. Phys. 109, 023503 (2011).

18. C. W. Ponader, J. F. Schroeder, and A. M. Streltsov, “Origin of the refractive-index increase in laser-writtenwaveguides in glasses,” J. Appl. Phys. 103, 063516 (2008).

19. S. Juodkazis, S. Kohara, Y. Ohishi, N. Hirao, A. Vailionis, V. Mizeikis, A. Saito, and A. Rode, “Structural changesin femtosecond laser modified regions inside fused silica,” J. Opt. 12, 124007 (2010).

20. J. Morikawa, A. Orie, T. Hashimoto, and S. Juodkazis, “Thermal and optical properties of the femtosecond-laser-structured and stress-induced birefringent regions of sapphire,” Opt. Express 18, 8300–8310 (2010).

21. S. Juodkazis, K. Yamasaki, V. Mizeikis, S. Matsuo, and H. Misawa, “Formation of embedded patterns in glassesusing femtosecond irradiation,” Appl. Phys. A 79, 1549–1553 (2004).

22. E. Vanagas, I. Kudryashov, D. Tuzhilin, S. Juodkazis, S. Matsuo, and H. Misawa, “Surface nanostructuring ofborosilicate glass by femtosecond nJ energy pulses,” Appl. Phys. Lett. 82, 2901–2903 (2003).

23. A. Marcinkevicius, V. Mizeikis, S. Juodkazis, S. Matsuo, and H. Misawa, “Effect of refractive index-mismatchon laser microfabrication in silica glass,” Appl. Phys. A. 76, 257–260 (2003).

24. T. Hashimoto, S. Juodkazis, and H. Misawa, “Void recording in silica,” Appl. Phys. A 83, 337–340 (2006).25. S. Juodkazis, H. Misawa, T. Hashimoto, E. Gamaly, and B. Luther-Davies, “Laser-induced micro-explosion

confined in a bulk of silica: formation of nano-void,” Appl. Phys. Lett. 88, 201909 (2006).26. L. Bressel, D. de Ligny, C. Sonneville, V. Martinez-Andrieux, and S. Juodkazis, “Laser-induced structural

changes in pure GeO2 glasses,” J. Non-Cryst. Solids 357, 2637–2640 (2011).27. A. Perriot, Nanoindentation de couches minces deposes sur substrat de verre de silice (English title: Nanoinden-

tation of thin films deposited on vitreous silica). PhD thesis, Universite Paris 6, 21 Dec. 2005.28. R. L. Parc, B. Champagnon, P. Guenot, and S. Dubois, “Thermal annealing and density fluctuation in silica glass,”

J. Non-Cryst. Solids 293-295, 366–369 (2001).29. T. M. Gross and M. Tomozawa, “Fictive temperature of GeO2 glass: its determination by IR method and its

effects on density and refractive index,” J. Non-Cryst. Solids 353, 4762–4766 (2007).30. A. Agarwal and M. Tomozawa, “Surface and bulk structural relaxation kinetics of silics glass,” J. Non-Cryst.

Solids 209, 264–272 (1997).31. R. Bruckner, “Properrties and structure of vitreous silica I,” J. Non-Cryst. Solids 5, 123–175 (1970).32. J. E. Shelby, “Properties and structure of vitreous silica,” J. Non-Cryst. Solids 349, 331–336 (2004).33. H. Kakiuchida, N. Shimodaira, E. H. Sekiya, K. Saito, and A. J. Ikushima, “Refractive index and density in F-

and Cl-doped silica glasses,” Appl. Phys. Lett. 86, 161907 (2005).34. M. Born and E. Wolf, Principles of Optics, 7th ed. (Cambridge University Press, 2002).35. S. Juodkazis, H. Misawa, E. G. Gamaly, B. Luther-Davis, L. Hallo, P. Nicolai, and V. Tikhonchuk, “Is the nano-

explosion really microscopic?,” J. Non-Cryst. Solids 355, 1160–1162 (2009).36. D. Durben and G. Wolf, “Raman spectroscopic study of the pressure-induced coordination change in GeO, glass,”

Phys. Rev. B 43, 2355–2363 (1991).37. M. Micoulaut, L. Cormier, and G. Henderson, “The structure of amorphous, crystalline and liquid GeO2,” J.

Phys.: Condens. Matter 18, R753–R784 (2006).38. C. Martinet, V. Martinez, C. Coussa, B. Champagnon, and M. Tomozawa, “Radial distribution of the fictive


temperature in pure silica optical fibers by micro-Raman spectroscopy,” J. Appl. Phys. 103, 083506 (2008).39. B. Champagnon, C. Martinet, C. Coussa, and T. Deschamps, “Polyamorphism: Path to new high density glasses

at ambient conditions,” J. Non-Cryst. Solids 353, 4208–4211 (2007).40. V. Martinez, R. L. Parc, C. Martinet, and B. Champagnon, “Structural studies of germanium doped silica glasses:

the role of the fictive temperature,” Opt. Mater. 24, 59–62 (2003).41. K. M. Davis, K. Miura, N. Sugimoto, and K. Hirao, “Writing waveguides in glass with a femtosecond laser,”

Opt. Lett. 21, 1729–1731 (1996).42. Y. Hayasaki, M. Isaka, A. Takita, and S. Juodkazis, “Time-resolved interferometry of femtosecond-laserinduced

processes under tight focusing and close-to optical breakdown inside borosilicate glass,” Opt. Express 19, 5725–5734 (2011).

43. S. Juodkazis, V. Mizeikis, S. Matsuo, K. Ueno, and H. Misawa, “Three-dimensional micro- and nano-structuringof materials by tightly focused laser radiation,” Bull. Chem. Soc. Jpn. 81, 411–448 (2008).

44. Y. Bellouard and M.-O. Hongler, “Femtosecond-laser generation of self-organized bubble patterns in fused sil-ica,” Opt. Express 19, 6807–6821 (2011).

45. S. Maruo and K. Ikuta, “Three-dimensional microfabrication by use of single-photon-absorbed polymerization,”Appl. Phys. Lett. 76, 2656–2658 (2000).

46. M. Thiel, J. Fischer, G. von Freymann, and M. Wegener, “Direct laser writing of three-dimensional submicronstructures using a continuous-wave laser at 532 nm,” Appl. Phys. Lett. 97, 221102 (2010).

1. Introduction

Structural modifications of glasses and crystals by direct laser writing with ultra-short laserpulses has a wide range of applications in waveguiding, lasing, and micro-/opto-fluidics [1–10].Densification of glasses and, consequently, an increase of the refractive index is useful forwaveguiding and fabrication of micro-optical elements. However, high intensity laser pulses arecausing similar structural damage of dielectric matrices as γ-rays or particle beams (electronsand neutrons) [11] since light-field driven electrons can reach high energies of up to ∼ 0.5 eVat breakdown conditions. Structural point defects like atomic voids, interstitials and the densityredistribution of constituent atoms/ions due to pressure and temperature effects need a betterunderstanding for micro-optical applications and fabrication of photonic devices. This researchcould also provide answers to unknown mechanisms of the formation of natural glasses, calledtectites, which are believed to be created by meteorite impacts [12, 13].

A pure silica glass has an anomalous fictive temperature, Tf, behavior [14], since the largestmass density is observed at elevated formation temperatures (high cooling rates of glass forma-tion). Hence, ultra-fast thermal quenching can help to create densified shells around the voidsformed by in-bulk micro-explosions in pure silica glasses and in quartz. The 3D localized ther-mal annealing realized via a multi-photon seeding of an avalanche ionization, as observed inphoto-polymers [15,16], creates an effective 3D localized hot-spot which can be used to directlyheat silica glass and cause an increase of its density after fast quenching. In view of recentlyobserved densification of silica below the threshold of shock wave generation at tight focusingand the simultaneous heating [17], formation of a dense phase due to the anomalous fictivetemperature effect is highly probable. How an ultra-fast rate of thermal quenching after themicro-explosion affects the glass structure and at which fictive temperature, Tf , [14] the glassis retrieved has to be better understood [18,19]. The Tf of shock amorphised crystals around thevoid-structures and their density could provide a method to estimate the pressure-temperature(p,T)-conditions of glass formation.

SiO2 and GeO2 are two technologically important materials for the direct laser writing ofwaveguides using fs-laser pulses. Structural analysis and understanding of the fundamental den-sification mechanisms of glass [18] in fs-laser structured regions will provide insights for thebest compositional formulation of the glass and exposure conditions for the waveguide record-ing. Pressure induced birefringence around fs-laser structured volumes [20] or the densificationof glass can be used to create a set of micro-optical elements and tools for microfluidics.

Here, we demonstrate that the compression of GeO2 glass network around the micro-


explosion sites is induced by fs-pulses. This is the first direct observation of densificationin GeO2 glass confirmed by corresponding changes of several characteristic Raman lines. Inparticular, the D2-band of three-ring tetrahedral-(GeO)3 increases in intensity and its positionmoves to the larger Raman shift wavenumbers. Comparison with hydrostatic pressure inducedchanges of Raman bands corroborates the presence of augmented pressure around fs-laser irra-diated sites. The present analysis explains the densification of silica reported recently [17]. Wediscuss how formation of waveguides by laser writing could be controlled via a glass compo-sition, which defines the fictive temperature behavior of the material, and/or by thermal condi-tions: heating and quenching rates. This can open new ways of material processing.

2. Methods

2.1. Sample Preparation and Characterization

Femtosecond laser pulses, 800 nm/150 fs (Hurricane, Spectra Physics), were tightly fo-cused [21, 22] with an objective lens of numerical aperture NA = 1.4 at 5–10 μm depth in-side glass samples in order to minimize spherical aberration and axial elongation of the focalregion [23]. Only the regions without strong crack formation were investigated. The regionswere laser modified at pulse energies larger than those required for the void formation [24,25];a single-pulse-per-void irradiation mode was used if not indicated otherwise. Separation of∼ 5 μm between adjacent irradiation spots was chosen in order to avoid thermal annealing ofpreviously exposed locations (Fig. 1(a)). The effects of close proximity of irradiation spots,thermal annealing, and crystallization we have reported elsewhere [26].

Raman scattering was measured using Aramis (Nano-Optic Devices) and Renishaw dedi-cated microscope-based setups at a NA = 0.5−0.9 focusing. The wavelength of excitation forRaman scattering measurements was 633 nm or 532 nm whichever caused less of a backgroundluminescence.

Germanium dioxide from Alfa Aesar labeled 99.98% was used as a starting product. Theglass sample was prepared by melting the starting powder at 1400◦C in a Pt crucible during36 h and quenched by dripping the bottom of the crucible in water. A perfectly transparentbubble free glass was obtained. Permanently densified samples were obtained at 9.6 GPa and15.4 GPa in a diamond anvil cell as described in ref. [26]. The laser-structured vitreous silicasamples were compared to a permanently densified and quenched samples. The silica glassof ρ = 2.25 g/cm3 density was obtained in a belt press at 3 GPa and 400◦C as described inref. [27]. The quenched sample was heat treated at 1500◦C for ∼ 1 h and then quenched inwater [28].

2.2. Theory: Fictive Temperature

The fictive temperature, Tf , is an imprint of the thermal history of glass preparation. It is thelast temperature at which glass reached equilibrium state with the supercooled liquid before arapid quench to room conditions. Usually, at slow thermal cooling rate a more dense glass canbe obtained, i. e., the specific volume is smaller in Volume ∝ Temperature dependence uponcooling from liquid phase as observed in most of silicate glasses [29].

There is a known exception from this behavior in the case of pure SiO2 glass [30–32] withthe highest density observed at Tf = 1500◦C [31]. The following dependence

ρ [g/cm3] = 2.1898+9.3×10−6Tf [◦C]. (1)

has been confirmed for the Tf range between 1000 and 1500◦ [14] (in the normal behaviorEq. (1) has a “-” sign.). This anomalous behavior is reversed in silica with 3 mol% of F or


2

m

(a)

(b)

0

-5

-10

-15

y (

m)

-15

0

5

0

5

s

-105x

single puls

-50( m)

ses

5 0

0.560

0.600

0.640

0.

0.

0.

0.

1.

Inte

nsi

ty (

arb

.un

its)

0

0

0

(c)

200

.00

.25

.50

.75

.00

Ge

sym

st

Ge

0 4

e-O-Ge

mmetric

retching

R

D1-motion

400

aman

600

Ge-O

-Ge

b

D2breathin

(GeO)3

n shift

nonlase9.6 15.4

800

Ge

bending

ng of

3-rings

(cm-

n-compreer irradiatGPa (53

4 GPa (5

100

Ge-O-G

asymme

stretchin

1)

essed (63ted regio

32 nm)532 nm)

00

e

etric

ng

33 nm)on (633 nnm)

Fig. 1. (a) An optical image of a GeO2 region modified by single pulses of 300 nJ/pulse en-ergy (at the entrance of microscope), 800 nm wavelength and 150 fs pulse duration focusedat 10 μm depth. (b) Map of the region boxed in (a) at the 520 cm−1 D2-band. (c) Ramanspectra of laser irradiated regions at different pulse energies 200, 300, and 400 nJ and atdifferent hydrostatic pressures; measured using 532 and 633 nm wavelength illumination.Arrows in (c) shows the observed tendencies with increasing pulse energy and/or pressure.Wavelengths of laser irradiation for Raman measurements are denoted.

Cl [33]. It has been established that GeO2 glass [29] as well as silicates show larger density atsmaller Tf , the most common behavior.

In silica, as the Tf increases the following changes in Raman scattering can be recognized: (i)the shift of broad band at 440 cm−1, the (Si-O-Si) stretching mode, to a higher Raman shifts, (ii)the area under the specific D1 and D2 bands increases (the D1 and D2 corresponds to the 4 and 3tetrahedral ring structures, respectively), (iii) the TO component of the fundamental stretchingof the Si-O-Si bridges shifts to lower Raman shifts, (iv) the Si-O-Si bond angle decreases, (v)the Rayleigh scattering is increasing [28].

The mass density, ρ , is related to the refraction index, n, via Lorentz-Lorenz equation, which

for the molar refractivity, A, reads [34]: A = 4π3 Nα = W

ρn2−1n2+1

, where N is the molecular numberdensity, α is the polarizability, W is the molecular weight. In terms of a refractive index change,

Δn, induced by the density change, Δρ , it reads [17]: Δn =n4

0−14n0

Δρρ0

, where 0-index indicatesthe initial properties of material. As mass density increases, the refractive index becomes larger.Hence, by controlling Tf it is possible to tune the refractive index. In the case of fs-laser writing,high flexibility in choice of writing geometry or energy deposition helps to form waveguidesin very different materials: crystals and glasses. The unique feature of fs-laser structuring isthat thermal quenching is very fast and a non-equilibrium state of matter can be quenched andretrieved to room conditions. Materials can be shock-liquidized by fs-pulses and afterwardsthermally quenched at a record high rate [35]. We put forward a conjecture that some of theamorphous phases in the fs-laser structured regions could behave as high-Tf glasses. In silicathis can explain formation of waveguiding regions at the center of inscribed line (at the centerof the focus) and is consistent with recent results [17].


(a

5

a)

5 m

(b)

Inte

nsi

ty a

t 5

20

cm

-1 (

arb

.un

its)

00.575

0.600

0.625

0.650

0.675

2

Cross

4

s section, y

800 nJ 700 nJ 600 nJ 500 nJ 400 nJ 300 nJ2 x 300 n

6

y ( m)

nJ

8

Fig. 2. (a) An optical image of a GeO2 region modified by single pulses of 300 nJ/pulse en-ergy (at the entrance of microscope), 800 nm wavelength and 150 fs pulse duration focusedat 10 μm depth. (b) Cross section of the two void structures (marked in (a)) at the 520 cm−1

D2-band vs pulse energy. Arrows in (b) denote main tendency of Raman intensity vs pulseenergy.

3. Results and Discussion

3.1. Femtosecond Laser Structured GeO2

Figure 1 shows observed changes in Raman spectrum caused by fs-laser pulses and permanentdensification retrieved upon compression up to 15 GPa in GeO2 glass. The pressure inducedchanges [36, 37] have the same trend as those observed from irradiated regions. Namely, thetetrahedral distortion, a decrease of symmetry is signified by a broad angle distribution de-crease of the 420 cm−1 band (symmetric stretching), an increase of the 860/970 cm−1 bands(asymmetric stretching), broadening of the 860 cm−1 band, a Raman shift increase of 420 cm−1

band, and a Raman shift decrease of 860/970 cm−1 bands. In addition, D2 band at 520 cm−1,assigned to 3-membered GeO3-rings, increases in intensity and Raman shift proportionally tothe stress applied (c). Mapping of the laser-structured location (see (b)) at this D2 band qualita-tively confirms presence of the denser regions at the irradiation sites.

The laser power dependence on densification is depicted in Fig. 2. The intensity of D2-bandbecome stronger for the highest pulse energies above 600 nJ. At the highest pulse energy of700 nJ/pulse (before an onset of strong crack formation) crystallization of GeO2 is observedjudging from an increase of the D2 band in between of the two void structures [37]. This mightbe caused by a heat effect due to comparatively close proximity of irradiation spots. High laserenergies or small spot separations can induce crystallization, quartz-like GeO2, and lattice rup-ture as we reported elsewhere [26]. This modification between the void-structures is of thermalnature and also might be due to the fictive temperature effect. Apart of this long-range phe-nomenon at the energies higher than 600 nJ, a recognizable increase of the D2 band intensity isobserved as well as a shift of the 860 cm−1 band towards the lower wavenumbers for the ener-gies higher than 500 nJ (not shown here). This implies that the structural changes of glass canbe controlled by changing the pulse energy. Interestingly, there were no measurable differenceof D2 band intensity for the single and double pulse per void-structures at the smallest pulseenergy (Fig. 2).

3.2. Femtosecond laser structured SiO2

In silica modified by tightly focused fs-laser pulses, densification can be caused by micro-explosion triggered shock-compaction or due to ultra-fast thermal quenching and resultant high


400 450

0.55

0.60

400 600 8000.0

0.1

0.2

0.3

0.4

0.5

0.6

Inte

nsity

(ab

.uni

ts)

Raman shift (cm

D2

mai

n-ba

nd

mai

n

D1

500

1000 1200

m-1)

initial laser treated quenched densified

550 600 650

0.1

0.2

D2D1

Fig. 3. Raman spectra from different SiO2 glasses (Eqs. (1,2)): fs-laser treated (den-sity ρ = 2.201 g/cm3, fictive temperature Tf = 1175◦C), hydrostatic pressure-densifiedρ = 2.25 g/cm3, thermally quenched (ρ = 2.202 g/cm3, Tf = 1350◦C) and initial silica(ρ = 2.198 g/cm3, Tf = 925◦C). Pulse energy was 200 nJ, wavelength 800 nm, pulse dura-tion 150 fs, lateral separation between irradiation sites was 2 μm and intra layer separation2.5 μm (ten layer structure). The inset shows an optical image of a 80 × 80 μm2 areapacked with void-structures.

Tf . Figure 3 shows changes of Raman spectra from fs-laser structured regions. The used pulseenergy was larger than that required for the void formation. The observed modifications, the Ra-man shift increase of the band at 440 cm−1 and the increase of D2 band intensity, are consistentwith both the high temperature [38] and the high pressure effects [39]. No changes in the D1

band seems to be more consistent with a moderate densification of the sample, ρ = 2.25 g/cm3,however, it is difficult to discriminate densification from a pure thermal quenching effect. Evo-lution of volume with thermal quenching rate is schematically represented in Fig. 4 for thenormal (a) and anomalous (b) glass transitions. In agreement with the anomalous behavior ofsilica, the denser silica glass ρ2 > ρ1 is formed at fast quenching rate and larger fictive tem-perature Tf 2 > Tf 1 (see, the ABC path in Fig. 4). The formation of more dense silica regionsaround voids will therefore benefit from both effects: high Tf as well as high pressure induceddensification.

A direct correlation between the maximum of the band at 440 cm−1 assigned to the (Si-O-Si)stretching mode, and the Tf was proposed [38]:

ν [cm−1] = 0.02Tf [◦C]+419.5. (2)

By assuming in first approximation that thermal effect is predominant in the laser irradiatedregion, it is therefore possible to estimate the Tf of the sample (Fig. 3) as follow: the Tf of theinitial silica was 925± 50◦C, the laser treated regions 1175± 50◦C, and thermally quenchedsilica 1350±50◦C using Eq. (2) and relate it to the density changes according to Eq. (1).

The change of fictive temperature can also be related to the induced change of refractiveindex since dn/dTf was estimated to be 1.6×10−6K−1 [31]. Applied here, the detected indexvariation associated with the laser treatment is 4× 10−4 (Δn � +0.03 %) in silica glass forsingle-pulse irradiation. It is important to recall that the volume occupied by void-structures issmaller than that measured by Raman scattering and that all the obtained results are averagedbetween the initial glass and the void-structure. Therefore the laser treatment has increased the


(a) (b)

density 1

density 2 > 1

Temperature Tm

slow cooling

fast cooling Volu

me

Silica:

liquid silica

super-cooledliquid

Crystal

density 2

density 1 > 2

Temperature Tm

slow cooling

fast cooling V

olu

me Glass:

liquid glass

super-cooledliquid

Tf1Crystal

Tf

Tf2 Tf2Tf1

A

B

C

Fig. 4. Typical normal (a) and anomalous (b) glass transition Volume ∝ Temperaturedependence observed in most of glasses (a) and silica (b), respectively (adopted fromref. [13, 31]). For silica: higher Tf corresponds to the larger density glass (anomalous be-havior); The crystal melting is a first order phase transition and marked by the dotted-line;Tm is the melting temperature. Note, volume is increasing upwards (a) and downwards (b);a darker shade background corresponds to the higher density, ρ . The cycle ABC schemati-cally shows how laser-induced melting and fast quenching results in waveguiding structure(higher mass density glass) in material with anomalous fictive temperature behavior.

refractive index in silica glass by at least 0.03 %. Locally some regions with higher refractiveindex and the mass density are present. The density change can explain compaction of the shellmaterial around the void. Possibility to create glasses at high thermal quenching rate is heredemonstrated.

It would be informative to measure Raman scattering from the inner part of the void-structures using a higher spatial resolution and sensitivity of detection. New insights into themechanisms of glass transition could be obtained. A possibility to tune the refractive indexchange should exist using Ge-doped SiO2 glass. Indeed Ge addition, an heavier atom, signifi-cantly increases the refractive index. As demonstrated earlier [40], the fictive temperature hasthe same effect in these glasses as in silica. This can be utilized to create waveguides in Ge-doped SiO2 glasses via the fictive temperature anomaly when the Ge-to-Si ratio is suitable. Bycoupling the pressure and fictive temperature effects induced by laser treatment new methodsto control the final density and refractive index can be obtained. Presented here analysis of Ra-man scattering and determination of fictive temperature, Tf , in pure silica and germania glassesis consistent with the first observation of refractive index increase in fs-laser inscribed waveg-uides [41] and more recent thermal effects of fs- and ps-laser writing in glasses [17, 42–44].

The two effects useful for waveguide formation: the anomalous fictive temperature and com-pression are very difficult to distinguish and can probably coexist on a scale of ∼ 100 nm belowthe used Raman spatial resolution. The main difference between the two process is a change ofthe width of the main-band which decreases significantly under densification and do not changewith the fictive temperature (see, Fig. 3). As the pressure caused change is a normal behaviorof glass, the anomalous behavior could be related to the anomalous fictive temperature effect.Further studies are required to advance an understanding of this potentially useful behavior inglasses.


4. Conclusions

Observation of GeO2 compression by fs-laser triggered micro-explosions is confirmed whenseparation between laser irradiated regions is large (> 5 μm). Structural changes are propor-tional to the pulse energy above 500 nJ as observed by intensity increase of the D2 and a shiftof the 860 cm−1 band towards lower wavenumbers. Modifications of SiO2 glass induced byfs-laser irradiation at tight focusing are consistent with formation of the high Tf -regions. Thoseregions are prospective for waveguiding since silica is an anomalous-Tf medium and a highertemperature causes formation of the higher density glass.

Further studies of Ge-doped silica are required to quantify the contribution of pressure den-sification and transition from anomalous to normal Tf behavior. This would allow to find anoptimal Ge-doping concentration for fs-recording of waveguides. The outlined mechanism ofcreating required density of a glass structure (a waveguide or an optical element) via controlledthermal treatment using direct laser writing could open new avenues in material engineering forwaveguides, sensors, and formation of 3D optical elements. The proposed mechanism is con-sistent with already reported results [17, 41]. The very same principles of 3D direct writing byhot-spot polymerization [16, 45, 46] can be extended to structuring of glasses, glass-ceramics,and crystals; e.g., a sol-gel transition leading to densification and glass formation can be lasercontrolled by a laser hot-spot scanning. The controlled fast thermal quenching can be usedto recover functional optical structures and devices from glasses whose composition and fic-tive temperature behavior are pre-engineered and optimized for fs-laser structuring - a novelapproach in engineering of optical materials.

Acknowledgments

This work was supported by CNRS, CECOMO and LLP - Leonardo da Vinci. SJ is thankful forthe visiting professorship at Lyon-I University and for a laser access at Photon Process lab ofHokkaido University for preparation of some of the samples. Authors are thankful to ProfessorsMicheline Boudeulle for discussions of structural properties of silicas, to Bernard Champagnonand Sylvie Le Floch for the densified silica glass sample.


femtosecond laser induced density changes in geo2 and sio2 ... · femtosecond laser induced density...

Documents