\u003ctitle\u003eRole of supercontinuum in the fragmentation of colloidal gold nanoparticles in solution\u003c/title\u003e

Role of supercontinuum in the fragmentation of colloidal Gold nanoparticles in solution

Fabian A Videla*a,d, Gustavo A Torchia a, Daniel S.. Schincaa,d , Lucía B Scaffardia,d

Pablo . Morenob, Cruz Méndezb Luis Rosob L. Giovanettic, Jose Ramallo Lopezc

aCentro de Investigaciones Ópticas (Conicet La Plata CIC) ,Casilla de Correos Nº3 Gonnet (1897) Buenos Aires Argentina.

b Servicio Láser Universidad de Salamanca Plaza de la Merced s/n 37008 Salamanca, Spain. c INIFTA Conicet La Plata Facultad de Ciencias Exactas, UNLP, Diagonal 113 y 64 La Plata

Buenos Aires Argentina (CP 1900) d Facultad de Ingeniería, Universidad Nacional de La Plata UNLP, Argentina Calle 1y 47 La Plata (1900) Buenos Aires

ABSTRACT

In this work we have studied the fragmentation of gold nanoparticles (NPs) after generation by femtosecond laser ablation of a solid target in deionized water. The fragmentation process was carried out using two different types of radiation: direct ultra-fast pulses and super-continuum radiation focused in the colloidal solution. In the former case, IR pulses were applied both in low and high fluences regime, while in the latter, super-continuum was generated by an external sapphire crystal. In this last case, to assess the effects of the different spectral bands present in the super-continuum for fragmentation, we have determined different efficiency regions. From the analysis of optical extinction spectra and Transmission Electron Microscopy (TEM) histograms we can conclude that the main mechanism is linear absorption in the visible region. Likewise, the super-continuum generated in water during fragmentation resulted more efficient than that obtained externally by the sapphire crystal. This fact can be attributed to the broadening of the water continuum band originated due to large intensity used for generation. TEM and Small Angle X-ray Scattering (SAXS) measurements support the results found from optical extinction spectroscopy.

Keywords: Fragmentation, supercontinuum, gold nanoparticles, Optical extinction, IR, TEM ,SAXS.

1. INTRODUCTION Fabrication of metallic Nps has been studied during last years due to their strong size and shape dependent optical properties. This dependence suggest the possibility of fabricating new Nps- based devices. Controlled size of Nps give place to applications in either industrial and biological areas. As it is well known Nps. can been obtained by means of chemical and laser ablation methods .The performance of both, refered to the control of size and distributions are different. In the first method, special laboratory conditions are required; in addition this kind of fabrication procedure includes stabilizer or encapsulation elements which are not adequate for biological or medical applications. In contrast, laser ablation has emerged as a clean method to produce Nps suitable for biological or medical applications [1,2,3].. Ultra-short laser ablation to produce metal Nps technique, avoids thermal effects mainly due to that energy can not be transferred to the lattice in times lower than 1ps. Once irradiated electrons in the skin of metal escape from the solid and create a strong electric field due to charge separation with the parent ions. This make that ions will been pulled out of the solid. [4] In this case once the particles have been fabricated, a second irradiation (fragmentation) was effectuated, so this last is a two step process. Other characteristic achieved using this treatment and reported in other works [5] is the stability or immunity to agglomeration by several months. In fragmentation, provided that the breakdown free condition

Instrumentation, Metrology, and Standards for Nanomanufacturing III, edited by Michael T. Postek, John A. Allgair,Proc. of SPIE Vol. 7405, 74050U · © 2009 SPIE · CCC code: 0277-786X/09/$18 · doi: 10.1117/12.831032

Proc. of SPIE Vol. 7405 74050U-1

was achieved, the method become independent of wavelength .To elucidate this degree of independence we perform a series of experiments exploring the fragmentation obtained for different radiation sources used separately. In one case IR ultrashort pulses radiation were used for gold Nps in solution. After optical extinction measurement, SAXS and TEM were also performed in order to fit size and distribution [6] of the gold Nps fragmented. In contrast, a similar sample was irradiated with supercontinuum generated by mean of a sapphire crystal Interaction between short pulses and the crystal produce nor linear effects. The obtained radiation was in turn applied with and without a edge cut optical filter in the range 760 820 nm. The sizes of the fragmented particles were obtained using the same techniques mentioned before. In summary we analyze 1) direct absorption of ultra short laser pulses 2) absorption from supecrontinuum filterd and unfiltered

2. EXPERIMENTAL 2.1 Samples treated with direct IR femtosecond laser at various fluences

Gold nanoparticles produced by femtosecond laser ablation were fabricated from 99.99% purity blanks of 12.5 mm diameter and 1 mm thickness inside of 3 ml of deionised water. To perform the laser ablation experiments we have used a 800 nm CPA system from Spectra Physics (Tsunami-Spitfire) with 120 fs pulse width, 1 KHz repetition rate and can deliver up to 1 mJ of energy per pulse. It is well known that using different laser fluences focused on the surface of the target, Nps generation take place, and then different colored solution can be obtained depending on the particle sizes. The experimental set-up used to fabricate and fragment gold Nps in solution is similar to that used in reference [8] and the update made for our experiments it is sketched in Figure 1. It must be noticed that the sapphire crystal only is used for fragmentation procedure.

Figure 1. Set-up used to fabricate gold Nps in solution (one step procedure) and post irradiation with IR femtoscond laser and continuum generated using a sapphire crystal

To produce the initial Nps solution, the sample was moved along lines at 50 μm/s writing different ablation tracks in all the surface of the sample avoiding rewriting along the lines. Details of these tracks describing meanders are shown in figure 2. During the process, water stains with a typical reddish color, which is attributed to the presence of an important number of Nps in the solution (Figure 2).

x z

Neutral Filter

Shuter

Power meter

B.S.

He Ne Laser Alignement

CCD Camera

Diaphragm

Cube (removed when Fs laser is applied)

Lens Half wave plate λ/2

Sapphire Crystal for continuum generation

y Cell with NPs in solution

Ti: Sa CPA System

Proc. of SPIE Vol. 7405 74050U-2

Figure. 2. Samples showing the reddish color obtained after fragmentation at different fluences.

On the other hand, the fluences used in our experiment for Nps generation were set to be lower than 300 J/cm2 in order to avoid the strong ablation regime which could give rise to undesirable droplets Nps [9][10]. To study the fragmentation process of the gold Nps, we exposed the initial solution to IR femtosecond laser radiation at different fluences, whose values were ranged from 1 to 700 J/cm2.

Figure 3. Microscopy of a typical meander described on the sample. The distance between groves was 50 μm. The total micro-machined surface was 1mm2. In the right side detail of a meander geometry .Time of micromachining was 20 minutes approximately.

Fluence can be calculated as

Where F: Fluence in J/cm2 E energy in J A: Spot Area in cm2 dw: beam wise at focus f: focal lens λ wavelength d spot diameter covering the lens. 2.2 Samples treated with direct super-continuum by using a sapphire crystal Super-continuum (SC) generation by using a Sapphire crystal was used to perform the fragmentation process. SC radiation was focused into Nps solution to analyze the effect over Nps size distribution after the light interaction. The set-up used for super-continuum (SC) generation is detailed in Figure 1 top. In this case the piece of sapphire crystal must be included. The radiation obtained shows noticeable components in the visible region of spectra as can be seen appreciated in the figure 4 a white spot surrounded by a rainbow colored ring ( fluences < 1 J/cm2). Optical extinction (OE) was registered by mean of a Shimadzu spectrophometer at room temperature. From the analysis of the plasmon resonance, and FWHM we can fit the size and size distribution of the Nps present in the solution.

Meander

)1(.44.2

42

22

⎟⎠⎞

⎜⎝⎛

=

⎟⎠⎞

⎜⎝⎛

==

df

EdE

AEF

w λππ

Proc. of SPIE Vol. 7405 74050U-3

In order to explore the size and morphology of the gold colloidal solution Small Angle X-ray Scattering (SAXS) and Transmission Electronic Microscopy (TEM) by means of Zeiss EM 900 system was performed. These experiments should be considered complementary with respect to OE measurements; they allow testing the size distribution of the gold Nps in water solution.

Figure 4. Aspect of the radiation (without focusing ) generated with a sapphire crystal when is illuminated by a femtosecond laser at 4μJ.

3. RESULTS 3.1 Generation and fragmentation by IR radiation

Figure 5 (a) shows the absorption spectrum corresponding to starting point gold Nps solution, fabricated with 300 J/cm2 inside deionized water. As it can be seen the plasmon resonance is peaked around to 531 nm with FWHM of 35 nm. The position of the plasmon resonance shows that there are large sizes (< 30 nm) of Nps present in the initial solution. As it is well known, the plasmon resonances characterize the shape of particles, in particular for isotropic particles this shape result spherical as can be seen analyzing the polarizability function [12, 13, 14]. Additionally, figure 5 (b) introduces SAXS measurements. From the SAXS distribution it is clear the presence of a broad range of particle sizes up to 40 nm. For this representation the vertical axis corresponds to the total volume of particles. This method corroborates the size of particles with a better statistic than TEM. Fig 5 (c) shows a histogram based on TEM pictures (which includes approximately 400 particles) corresponding to the same sample. Inspection of TEM, allows us, to recognize the two ablation regimes: gentle and strong even for low fluences used for generation. These regimes can be determined by fitting two Gaussian curves [5],[6]. In gentle regime, the ablation of material take place after the action of laser pulse, so the thermal regime became excluded. Small particles (4-10 nm) are associated to this regime. In the strong regime the particles are heated for the plasma itself [9]. From optical extinction spectrum (figure 5 (a)) mean size of the nanoparticles was ranged from 8 nm to 10nm [11], this value is in agreement with TEM histogram distribution from figure 5 (c). To compare SAXS against TEM and optical measurement, this axis should represent the number of particles instead of volume. For this, the SAXS vertical axis must be divided by the volume of each particle size as is shown in figure 5 (d). After this treatment, smaller values of Nps has been emphasized, however a little contributions of large sizes can be important to describe the real solution from the optical extinction spectroscopy [11].

Proc. of SPIE Vol. 7405 74050U-4

Figure 5. (a) Absorption spectrum corresponding to gold Nps fabricated with fluences of 300 J/cm2 deionised water. (b)

SAXS data .The resonance plasmon is peaked around to 531 nm. (c) TEM histogram of Nps diameters based on pictures corresponding to initial gold Nps. (d) SAXS modified representing number of particles instead volume.

To start the fragmentation procedure the Nps initial solutions was irradiated (the gold disk was retired) with the fundamental wavelength at several laser fluences. Figure 6 shows the normalized extinction spectra corresponding to gold Nps obtained after impinging the solution with fluences of 416 J/cm2 (b) 342 J/cm2 (c) and 0.8 J/cm2 (d). It is also included for comparison, the optical extinction spectrum corresponding to the initial gold Np solution (a).

Figure 6. Normalized extinction spectra corresponding to the fragmented gold Nps obtained with fluences of: 416 (b), 256

(c) and 0.8 (d) J/cm2.It is also included for comparison, the extinction spectrum for fabrication (a).

Proc. of SPIE Vol. 7405 74050U-5

The spectra corresponding to fragmented Nps show a blue shift of few nanometers, this can be appreciated observing the position of peaks from 531 nm to 521 nm (marked by vertical dashed lines). The largest blue shift (10 nm) was observed for the fragmented Nps by using the lowest fluence (d). However for the other fluences (b), (c), the blue shift is also important, the resonance plasmon shifts about 4-8 nm. For higher fluences the blue shift is smaller, getting a value of 2-3 nm. Under high regimes, above threshold breakdown and multi-photonic absorption in water, part (lower than 40%) of total energy provided by the IR source is lost by reflection or absorption in the plasma, so lower fluences are available for fragmentation process. From figure we can inferred that small Nps are obtained if fluence used to fragment decrease. As it was discussed in the paper written by [5] the main phenomena responsible for the fragmentation using lower fluences (<1J/cm2) have been attributed to the linear absorption of 800 nm laser radiation and the super-continuum which is produced by the IR radiation inside the water. Also, the authors in other work [16], suggest that the main responsible for fragmentation correspond to the visible band. In this sense, to elucidate the radiation responsible of the fragmentation we have performed experiments to fragment the initial gold Nps solution by using separately radiation coming from continuum light and IR laser. 3.2 Fragmentation by Super-Continuum radiation

For this purpose we have generated this radiation by means of ultra-short infrared radiation through a piece of sapphire crystal [7] (see Fig.1). Some of the characteristics parameters set for each experiment are shown, summarized in table I. The parameters used for the generation of the initial gold Nps solution were added in the first row in the table (shadow). The fragmentation cases using SC are shown in rows 2 ( SC Sapphire without filter), 3 (SC with filter) and 4 (SC water+ IR lowest fluences). For the two last cases the intensity values are below to the threshold value for optical breakdown (1.1e13 W/cm2)[5], while for the 2nd row this value is overcome. On the other hand, respect to multiphotonic absorption the threshold (2e12 W/cm2) [5] is reached in the cases for rows 2 and 4). Given that, the thresholds are overcame, breakdown in water could occur, but the reduction of the transmittance in the solution due to the plasma is 40% [15]. In this sense, is not possible discard it beforehand, the interaction between metallic Nps and IR due to total absorption of IR

Table I.: Experimental conditions for fragmentation experiments on a solution of gold Nps irradiated by using SC and IR

pulses. The shadow row corresponds to the initial experimental conditions for gold Nps generation.

In Figure 7 we compare the spectra generated in water by direct irradiation of femtosecond laser pulses and by the sapphire crystal .It could be appreciated the extra broadening in water[16]respect to continuum externally generated by using a sapphire crystal. The experimental conditions correspond to pulses of 4 μJ energy in 3cm3 of water .The spectrum was taken with an Advantes spectrograph.

F [J/cm2]

Power [W] Intensity [W/cm2]

Plasmon Peak [nm]

Experimental condition

256.6 2.5E9 2.13 E15 531 Starting point sample

3.4 3.3E7 2.86 E13 526 SC without IR filter

0.017 1.6E5 1.42 E11 524 SC with IR filter 0.855 8.3E6 7.12 E12 521 SC in water + IR

low fluence

Proc. of SPIE Vol. 7405 74050U-6

Figure 7. Spectra of white light generated by a femtosecond laser pulse in water (solid line) and the sapphire crystal filtered

at 750 nm (dashed line). It can be appreciated the extended blue shift of the water spectrum up to 400 nm. At rigth side, picture of the supercontinuum in water produced by the femtosecond laser.

By using white light, we carry out the fragmentation experiments. For this, we focused the super-continuum inside the Nps solution obtained previously with and without IR filter. Figure 8 shows the optical extinction spectra corresponding to the gold Nps fragmented by using super-continuum radiation obtained using the sapphire crystal (a) and (b) and water (c). For reference, in each picture the optical extinction spectrum corresponding to initial gold Nps marked as “before” (open circles) is included. Dashed line represents the white light fragmentation spectrum with filter, figure 8 (a), to suppress the fundamental laser radiation in the fragmentation process. Figure 8 (b) represents the laser radiation spectrum without IR filter. Additionally, we have added in this figures, the optical extinction spectrum corresponds to resultant gold Nps after fragmentation (marked as “after”).The optical extinction corresponding to the white light fragmented Nps showed a little difference when they were irradiated with (figure 8 (a)) or without IR filter (figure 8 (b)). As can be seen, they present peaks at 524 nm and 526 nm respectively. On the other hand, for IR fragmented particles at lower fluences (< 1 J/cm2), which includes the super-continuum generated in water; as it can be seen in figure 8 (c), the extinction spectrum has the maximum centered at 521 nm, this means the solution contains the lowest size gold Nps. In order to test the main Nps sizes distribution we have fitted this experimental curve by a theoretical data including a mixing of several Nps sizes (Figure 8 (d)). The inset shows a histogram corresponding to a size Nps distribution used to fit well the all extinction spectrum as it was described in reference [11]. This last fit was made considering the dielectric function (equation 2) corresponding to bulk composed by two terms .The first of the second member consider the free electrons component, the second the bounded electrons. Given that particles of size less than 10 nm appear during fragmentation process the dependence of refractive index with the particle size must be considered. For this, the dielectric function was corrected because the damping factor change taken into account the decreasing in the mean free path and in consequence the augment in collisions events .The formulae for calculation were

( ) ( ) ( ).ωεωεωε electronsboundedelectronsfreebulk −− += (2) In this expression the values of ε are complex.

The dielectric function for free electrons is ( ) .1 2

2

ωγωω

ωεfree

pelectronsfree i+

−=− (3)

Here )4(/ 2

0

em

VNp ε

ω = is the plasma frequency being N the amount of free electrons in the volume V, m is the

effective mass of an electron, e and 0ε are the electron charge and the permittivity in vacuum respectively. freeγ is the

damping factor in the Drude model. For particles less than 10 nm the value of freeγ must be corrected as

( ) )5(.)( RvCR Fbulkfree += γγ here, vF is the electron velocity at Fermi surface and C scattering constant. In

this case the corrected values are indicated by 'ε . For extinction determination (given by Qext) we apply

Proc. of SPIE Vol. 7405 74050U-7

( ) )6()()2.n (

.*n*RR,Q

2bulk

'2

220bulk

'1

bulk'23

03

ext εελ

ε

λ++

=

Where R is the radius of the particle, n0 is the refractive index of the medium (water) λ the wavelength at frequency ω

Figure 8. Optical extinction corresponding to the white light fragmented Nps when it was fragmented without IR filter (a)

peaked at 526 nm and with IR filter (b), peaked at 524 nm (c) extinction spectrum of IR fragmented particles at lower fluences, with the maximum centered at 521 nm. In (d) fitting of the curve shown in part (c) by means of a set of basis functions obtained for several sizes of gold Nps as it was performed in reference [6,11].

To complete the analysis we will show in the next section the histograms obtained from TEM pictures corresponding to the optical extinction spectra presented above. The distribution of relative abundance allows appreciate the effect of fragmentation .This is noticeable when this distribution is compared against the histogram in figure 8 (d). 3.3 Unfiltered SC fragmentation

The sample corresponding to gold NPs previously irradiated, whose spectrum shows a resonance peak in 531 nm (hollow circles), was irradiated again with super continuum from sapphire crystal. The resultant optical extinction has a peak in 526 nm. In this region (range 400-820 nm) the SC spectra (dashed line) exhibit a strong peak in 800nm surrounded by an IR band from 780 to 820 nm For fragmentation of Nps, there exists some mechanism that could originate it: linear absorption and multi-photonic absorption from visible and IR bands. In this sense, the interval 400-750 nm the corresponding Nps absorption is high and SC is low, while in the interval 750 820nm the situation is the opposite. We take in account this observation in the discussion. As a result the size of particles obtained shows no presence of large particles as it is observed in figure 5 (b),

Proc. of SPIE Vol. 7405 74050U-8

where a bimodal distribution appears. In this case the distribution tend to be monomodal. (possibly due to the absence of heating by the plasma itself [5]). In figure 9 is shown the relative abundance or fraction of particles referred to the total present in the TEM picture. The histogram shows a large quantity of particles with diameter between 8 and 10 nm.

Figure 9. Relative abundance or fraction of particles referred to the total present in a TEM picture. Can be observed a

monomodal distribution strongly concentrated around 8-10nm The fluences used in this case was 3.4 J /cm2

3.4 Filtered SC fragmentation

The experiment was developed similarly as previously described. In this case we have added a low pass filter (Thorlab FES0750) whose cut-off frequency is about 750 nm. In the wavelength band 450 and 700nm its transmittance is 80%, in consequence the energy is 20 % lower than the previous case (in the same band). The size distribution for the sample after irradiation was obtained by TEM. The relative abundance is similar to the first case. A monomodal distribution can be noticed again .The distribution is centered around 10 nm diameters.

Figure 10. . It is shown the relative abundance correspondingly .to fragmented Nps As the can be observed the

distribution is again monomodal In addition the spectrum corresponding to super-continuum generated in similar form as was previously indicated , using a filter (Thorlab FES0750 in the inset) whose cut off frequency is about 750 nm.

By comparison of figures 9 (a) and 10 (a) it is possible deduct that, the IR band (700 820 nm) present in the case without filter is less efficient than the visible band 400 -700 nm, so it can be discarded for fragmentation. This fact can be attributed to the decreasing of the ablation threshold fluence with the wavelength [4]

Proc. of SPIE Vol. 7405 74050U-9

3.5 Case with low fluence IR in water

After applying radiation whose fluence was about 1 J/cm2, as a result it was obtained an extinction spectrum that shows a shifting in its peak from 531 to 521nm (the peaks shifts more than in the previous case). From the histogram obtained by TEM microscopy can be observed a mean size around 8-10 nm. In comparison with the histograms before analyzed it can be appreciated that the relative abundance is smaller than in those cases. Also the distribution is almost monomodal, meaning that the rest of the sizes have been fragmented. In turn the area of the spectrum showed in Figure 11 (b), the range between 400- 760 nm (visible) represent 5 % of total radiation, however we obtained the lowest the gold Nps fragmented by this radiation. We can explain this fact taking into account the spectral response shown in figure 7, the large blue wing (extra broadening) in continuum generated in water, which allow us to obtain the small particles because of this wavelength range present the least fluence ablation thresholds. This result is in agreement with that obtained in the previous section related to the not effect of the IR band (760- 820 nm) for the fragmentation process.

Figure 11.: Histogram from the fragmented gold Nps by using the lowest fluences (1 μJ/cm2). The Optical Nps distribution

was obtained by averaging of several samples Also it is shown in detail, the spectrum measured from water.

4. CONCLUSIONS The main results for the fragmentation process by two step procedure were studied applying 3 complementary measurements: optical extinction, TEM and SAXS. We explored the fragmentation by IR ultra-short laser pulses focused inside the initial gold NPs solution observing the blue-shift in the plasmon of the optical extinction spectra is larger when the fluences are decreasing. The large shift was obtained for 1 J/cm2 of fluence where the peak was moved from 531 nm to 521 nm. As was suggested in other works [2, 9] we studied the effects in fragmentation by continuum generation alternatively by using an external sapphire crystal and that generated in water. To determine the relative importance of the different bands in fragmentation, we treat samples of gold Nps. solutions irradiating with continuum externally generated, with and without filter. After data analysis was possible to discriminate the response in both cases and reveals the efficiency of each one The result makes evident the strong effect of wavelengths belonging to the visible band in the range of low fluencies below (or closest) to the threshold of breakdown and multi-photonic absorption used in this work. On the other hand, for fluences closest to 1 J/cm2 for IR pulses in similar conditions respect of thresholds previously mentioned, continuum generation in water seems to have an important role over fragmentation of gold Nps samples; we think due to blue broadening effect super continuum generated in water at high intensities (1000GW/cm2) improve the efficiency. We can conclude that in spite of the IR pulses can be an excellent tool to generate gold Nps in water but also in vacuum[18], for fragmentation process its efficiency seem to be one not enough suitable for this purpose. In contrast super continuum radiation is necessary for fragmentation of gold Nps solution. SAXS measurements support the results obtained from optical extinction spectroscopy. Complementary by mean of TEM analysis we could estimate the sizes, abundance and shape of Nps.

Proc. of SPIE Vol. 7405 74050U-10

REFERENCES

[1]Mafuné, F., Kohno, J., Takeda Y., Kondow, T., Sawabe, H., “Formation of gold nanoparticles by laser ablation in aqueous solution of surfactant” J. Phys. Chem. B 105, 5114-5120(2001). [2]Kabashin, A.V, Meunier, M., Kingston, C., Luong J. T., “Fabricaton and characterization of gold nanoparticles by femtosecond laser ablation in aqueous solution of cyclodextrins” Phys. Chem. B 107, 4527-(2003). [3]Tsuji, T., Iryo, K., Watanabe, NTsuji, .M., “Preparation of silver nanoparticles by ablation in solution :influence of laser wavelength on particle size” Appl. Surf. Science 202, 80-85 (2002). [4]Gamaly, G., Rode, A.V., Luther Davies ,B. and Tikhonchuk, V.T., “Ablation of solids by femtosecond lasers: Ablation mechanism and ablation threshold for metals and dielectrics” Physics of Plasmas 9, 949-957 (2002). [5]Besner, S., Kabashin ,A.V.and Meunier, M.,“Fragmentation of coloidal nanoparticles by femtosecond induced super-continum generation” , Appl. Phys. Letters 89, 233122-1-3 (2006). [6]Torchia, G., Scaffardi, L., Mendez, C., Moreno, P., Tocho, J.and Roso,L., “Optical extinction for determining size distribution of gold nanoparticles fabricated by ultrashort pulsed laser ablation” Applied Physics A: Materials Science and Processing, 93, 967-971.(2008). [7]Schaffer, C., “Interaction of Femtosecond Laser Pulses with Transparent Materials” PhD Theses Harvard University, (2001). [8]Moreno, P., Méndez, C., García, A.,. Torchia, G. A, Delgado, D, Vázquez de Aldana, J. R., Arias, I. and Roso, L., “Synthesis of ceramic nanoparticles by ultrafast laser ablation of solid targets in water”, Journal of Nanoscience and Nanotechnology” 6, 1961–1967 (2006). [9]Kabashin, A.V., Meunier, M., “Synthesis of colloidal nanoparticles during femtosecond laser ablation in gold particles”, J. App. Phys. 94, 7941-7943 (2003). [10]Moreno, P., Méndez, C., García, A., Torchia, G. A,. Delgado, D., Vázquez de Aldana, J. R. Arias, I. and Roso, L., “Synthesis of ceramic nanoparticles by ultrafast laser ablation of solid targets in water”, Journal of Nanoscience and Nanotechnology” 6, 1961–1967 (2006). [11]Scaffardi, L. B., Pellegri, N., de Sanctis, O., and Tocho, J. O., “Size dependence of refractive index of gold nanoparticles Nanotechnology” 16, 158-163 (2005). [12]Bohren, C. F., Huffman, D. R., [Absorption and Scattering of Light by Small Particles] Wiley, New York 141-148 1998. [13]Van de Hulst, H.C., [Light Scattering by Small Particles] Dover, New York, 70-74 (1981).

[14]Fox, M., [Optical properties of solids] Oxford University Press, New York, 29-34 (2001).

[15]Noack, J., Vogel, A., “Laser Induced Plasma formation in water at nanosecond to femtosecond time scales: Calculation of thresholds absorption and energy density” IEEE Journal of Quantum Electronics 35, 1156-1167 (1999).

[16]Besner, S., Kabashin, A.V,. Meunier M., “Two step femtosecond laser ablation –based method for thr synthesis for stable and ultra-pure nanoparticles in water” App. Phys. A. 88, 269 272 (2007). [17]Brodeur A., Chin S.L.,“Ultrafast white light continuum generation and self focusing in transparent media” JOSA B 16, 637- 650(1999).

Proc. of SPIE Vol. 7405 74050U-11

[18]Noel, S., Hermann, J. and Itina, T., “Investigation of Nanoparticles generation during femtosecond laser ablation of metals” Applied Surface Science 253, 6310-6315(2007).

Proc. of SPIE Vol. 7405 74050U-12