Magnetic Alloy Nanoparticles from Laser Ablation in Cyclopentanone

Apr 15, 2010 - The generation of nonoxidized magnetic alloy nanoparticles is still a challenge using conventional chemical reduction methods. However ...
0 downloads 0 Views 4MB Size
Download PDF
pubs.acs.org/Langmuir © 2010 American Chemical Society

Magnetic Alloy Nanoparticles from Laser Ablation in Cyclopentanone and Their Embedding into a Photoresist Jurij Jakobi, Svea Petersen, Ana Menendez-Manjon, Philipp Wagener, and Stephan Barcikowski* Laser Zentrum Hannover e. V., Hollerithallee 8, Hannover D-30419, Germany Received August 6, 2009. Revised Manuscript Received April 9, 2010 The generation of nonoxidized magnetic alloy nanoparticles is still a challenge using conventional chemical reduction methods. However, because these nanoparticles are currently attracting much attention, alternative methods are required. In this context, the applicability of femtosecond laser ablation, which has evolved as a powerful tool for the generation of colloidal metal nanoparticles, has been investigated using the example of Ni48Fe52 and Sm2Co17 ablation in cyclopentanone. Besides stability and size measurements, the focus has been placed on the analysis of the elemental composition of nanoparticles, which proved the preservation of the stoichiometry of the target in Ni-Fe nanoparticles but not in Sm-Co. It is assumed that this is due to a greater difference in the heat of evaporation of the bulk alloy components in Sm-Co than in Ni-Fe. Hence, the successful generation of magnetic alloy nanoparticles is possible for alloys composed of elements with similar heats of evaporation. This one-step approach allows the fabrication of nanomagnetic polymer composites (e.g., with application prospects in microtechnology such as microactuators).

Introduction In the last few decades, intensive research has been done on the development of novel methods for the generation of nanoparticles.1-8 Magnetic nanoparticles in particular have attracted a great deal of attention because of their widespread application prospects in biomedicine and information technology, such as nanomarkers for magnetic resonance imaging or polymer composites with adjustable magnetic properties.9-16 Nowadays, the generation of nonoxidized magnetic nanoparticles consisting of metals such as encapsulated Co and goldcoated Fe or alloys such as FePt and SmCo has been investigated *Corresponding author. Phone: þ49 511 2788-377. Fax: þ49 511 2788-100. E-mail [email protected]. (1) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Soc., Chem. Commun. 1994, 801. (2) Sun, Y.; Wiley, B.; Li, Z.-Y.; Xia, Y. J. Am. Chem. Soc. 2004, 126, 9399. (3) Grass, R. N.; Stark, W. J. J. Mater. Chem. 2006, 16, 1825. (4) Binns, C.; Trohidou, K. N.; Bansmann, J.; Baker, S. H.; Blackman, J. A.; Bucher, J.-P.; Kechrakos, D.; Kleibert, A.; Louch, S.; Meiwes-Broer, K.-H.; Pastor, G. M.; Perez, A.; Xie, Y. J. Phys. D: Appl. Phys. 2005, 38, R357. (5) Tartaj, P.; del Puerto Morales, M.; Vtemillas-Verdaguer, S.; GonzalezCarreno, T.; Serna, C. J. J. Phys. D: Appl. Phys. 2003, 36, R182. (6) Kim, D. K.; Zhang, Y.; Voit, W.; Rao, K. V.; Muchammed, M. J. Magn. Magn. Mater. 2001, 225, 30. (7) Lai, J.; Shafi, K. V. P. M.; Ulman, A.; Loos, K.; Popovitz-Biro, R.; Lee, Y.; Vogt, T.; Estournes, C. J. Am. Chem. Soc. 2005, 127, 5730. (8) Tsuji, T.; Hamagami, T.; Kawamura, T.; Yamaki, J.; Tsuji, M. Appl. Surf. Sci. 2005, 243, 214. (9) Jiguet, S.; Judelewicz, M.; Mischler, S.; Bertch, A.; Renaud, P. Microelectron. Eng. 2006, 83, 1273. (10) Feldmann, M.; Demming, S.; B€uttgenbach, S. Proc. Nanotech. 2007, 3, ISBN 1-4200-6184-4. (11) Waldschik, A.; Feldmann, M.; B€uttgenbach, S. Proc. of Actuator 2008, 669–672. (12) Hu, W.; Wilson, R. J.; Koh, A.; Fu, A.; Faranesh, A. Z.; Earhart, C. M.; Osterfeld, S. J.; Han, S.-J.; Xu, L.; Guccione, S.; Sinclair, R.; Wang, S. X. Adv. Mater. 2008, 20, 1479. (13) Petersen, S.; Jakobi, J.; Barcikowski, S. Appl. Surf. Sci. 2009, 255, 5435. (14) Petersen, S.; Barcikowski, S. Adv. Funct. Mater. 2009, 19, 1. (15) Hahn, A.; Barcikowski, S. J. Laser Micro/Nanoeng. 2009, 4, 51. (16) Nakata, K.; Hu, Y.; Uzun, O.; Bakr, O.; Stellacci, F. Adv. Mater. 2008, 9999, 1. (17) Eggeman, A. S.; Petford-Long, A. K.; Dobson, P. J.; Wiggins, J.; Bromwich, T.; Dunin-Borkowski, R.; Kasama, T. J. Magn. Magn. Mater. 2006, 301, 336. (18) Cho, S. J.; Kauzlarich, S. M.; Olamit, J.; Liu, K.; Grandjean, F.; Rebbouh, L.; Long, G. J. J. Appl. Phys. 2004, 5, 11. (19) Aslam, M.; Fu, L.; Li, S.; Dravid, V. P. J. Colloid Interface Science 2005, 290, 444.

6892 DOI: 10.1021/la101014g

using wet chemical synthesis and the milling of powders.17-19 However, besides the reproducibility concerns, the use of these methods may lead to the contamination of the product as a result of chemical precursors or the abrasion of grinding elements.20 Laser ablation in an aqueous solution offers an alternative approach to the generation of pure nanoparticle colloids of various materials without the use of precursers.21-27 Besides using water as a liquid matrix,28,29 the laser fabrication of nanoparticles in nonpolar organic solvents with and without the addition of surfactants or ligands has been reported.30-32 Until now, mainly metal and ceramic colloids have been produced using this method, including several studies on laser-generated magnetic nanoparticles.8,21,33 The laser-based generation of colloidal magnetic alloys has not been reported so far. It has been shown that, in contrast to nanosecond pulsed laser ablation in air, femtosecond laser ablation allows the production of alloy nanoparticles with smaller deviations in stoichiometry compared to that of the bulk material.34 This work indicates that the duration of interaction of the ultrashort laser pulse (20) Dahl, J. A.; Maddux, B. L. S.; Hutchison, J. E. Chem. Rev. 2007, 107, 2228. (21) Kabashin, A. V.; Meunier, M.; Kingston, C.; Luong, J. H. T. J. Phys. Chem. B 2003, 107, 4527. (22) Dolgaev, S. I.; Simakin, A. V.; Voronov, V. V.; Shafeev, G. A.; Bozon-Verduraz, F. Appl. Surf. Sci. 2002, 186, 546. (23) Menendez-Manjon, A.; Jakobi, J.; Schwabe, K.; Krauss, J.; Barcikowski, S. J. Laser Micro/Nanoeng. 2009, 4, 95. (24) Liu, Z.; Yuan, Y.; Khan, S.; Abdolvand, A.; Whitehead, D.; Schmidt, M.; Li, L. J. Micromech. Microeng. 2009, 19, 1. (25) Amendola, V.; Meneghetti, M. Phys. Chem. Chem. Phys. 2009, 11, 3805. (26) Barcikowski, S.; Hahn, A.; Kabashin, A.; Chichkov, B. Appl. Phys. A 2007, 87, 47. (27) Liang, C.; Shimizu, Y.; Sasaki, T.; Koshizaki, N. J. Mater. Res. 2004, 19, 1551. (28) Besner, S.; Kabashin, A.; Winnik, F.; Meanier, M. J. Phys. Chem. C 2009, 113, 9526. (29) Tsuji, T.; Thang, D.-H.; Okazaki, Y.; Nakanishi, M.; Tsuboi, Y.; Tsuji, M. Appl. Surf. Sci. 2008, 254, 5224. (30) Compagnini, G.; Scalisi, A. A.; Puglisi, O. J. Appl. Phys. 2003, 94, 7874. (31) Compagnini, G.; Scalisi, A. A.; Puglisi, O. J. Mater. Res. 2004, 19, 2795. (32) Amendola, V.; Rizzi, G. A.; Polizzi, S.; Meneghetti, M. J. Phys. Chem. B 2005, 109, 23125. (33) Kazakevich, P. V.; Voronov, V. V.; Simakin, A. V.; Shafeev, G. A. Quantum Electron. 2004, 34, 951. (34) Koch, J.; von Bohlen, A.; Hergenr€oder, R.; Niemax, K. J. Anal. At. Spectrom. 2004, 19, 267.

Published on Web 04/15/2010

Langmuir 2010, 26(10), 6892–6897

Jakobi et al.

Letter

with an alloy could be more advantageous for alloy particle generation than using a laser with a long laser pulse. In this context, we have investigated the applicability of femtosecond laser ablation of magnetic alloys for the generation of magnetic alloy nanoparticles with the same stoichiometry and their embedding into thermosetting polymers. This could be interesting for microtechnology applications such as the development of micromotors or magnetic micropumps.10,11 We have focused our research on two magnetic alloys, permanent magnetic Sm2Co17 and soft magnetic Ni48Fe52. Both materials have many applications in the nanoparticulate state, for example, in magnetic resonance imaging, magnetic sealing systems, and data storage media as well as for polymer reinforcement in microstructuring.35-39 Sm2Co17 magnets are well known because of their strong, permanent magnetic properties, high coercitivity, and high magnetic remanence. In contrast, the Ni-Fe alloy has high remanence and low coercitivity. Because of its tendency toward oxidation, the preparation of nanoparticles consisting of pure Sm-Co or Ni-Fe is often disabled by synthesis in an aqueous solution. Laser ablation in inert atmospheres or organic liquids may reduce this risk of oxidation; hence we accomplished the laser ablation of Sm2Co17 and Ni48Fe52 in cyclopentanone. Because it has been reported that laser-generated metal nanoparticles are electrostatically stabilized in aqueous solutions,40,41 we assume that laser ablation in polar organic liquids such as cyclopentanone is also an appropriate physical method for the production of stable magnetic nanoparticles without the need to add stabilizing agents such as oleic acid or sodium dodecyl sulfate, which is generally afforded.42 The generated cyclopentanonenanoparticle colloids are subsequently characterized by their zeta potential, size, elemental composition, and magnetic behavior. Besides the expected stability, the generation of nanopartitcles in organic solution enables the transfer of generated nanoparticles in monomers simply by mixing the components. In the next step, nanocomposites are produced by the subsequent polymerization of the monomer.

Experimental Section Generation Process. Laser ablation of Ni48Fe52 and Sm2Co17 (the alloy also contains small amounts of Fe, Cu, and Zr. For the exact composition, see the Supporting Information) targets is performed using a pulsed femtosecond laser system (Spitfire Pro, SpectraPhysics). The laser parameters used were 120 fs laser pulses at 800 nm (central wavelength) with a repetition rate of 5 kHz and a pulse energy of 300 μJ/pulse. The laser beam was focused on the target, which was positioned at the bottom of a crystallization dish filled with 13 mL of cyclopentanone. The diameter of the crystallization dish was 60 mm, and the liquid layer covering the target was 2 mm. The dish was placed on an axis system that moved at a constant speed of 1 mm s-1 in a spiral with an outer radius of 2 mm and an inner radius of 0.4 mm. Characterization. The zeta potential and dynamic light scattering measurements were performed using a Zetasizer ZS (Malvern). Each sample was subjected to three measurements, and the mean values are presented. (35) Jung, C. W.; Jacobs, P. Magn. Reson. Imaging 1995, 13, 661. (36) Raj, K.; Moskowitz, R. J. Magn. Magn. Mater. 1990, 85, 233. (37) Raj, K.; Moskowitz, B.; Casciari, R. J. Magn. Magn. Mater. 1995, 149, 174. (38) Hasegawa, M.; Uchida, K.; Nozawa, Y.; Endoh, M.; Tanigawa, S.; Sankar, S. G.M; Tokunaga, M. J. Magn. Magn. Mater. 1993, 124, 325. (39) Popplewell, J.; Rosensweig, R. E.; Siller, J. K J. Magn. Magn. Mater. 1995, 149, 53. (40) Sylvestre, J. P.; Poulin, S.; Kabashin, A. V.; Sacher, E.; Meunier, M.; Luong, J. H. T. J. Phys. Chem. B 2004, 108, 16864. (41) Sylvestre, J. P.; Kabashin, A. V.; Sacher, E.; Meunier, M.; Luong, J. H. T. J. Am. Chem. Soc. 2004, 126, 7176. (42) Wang, Y.; Li, Y.; Rong, C.; Ping Liu, J. Nanotechnology 2007, 18, 465701.

Langmuir 2010, 26(10), 6892–6897

A transmission electron microscope (TEM Jeol 2010F) was used to obtain micrographs and EDX measurements of the colloids. Samples were prepared by placing a drop of the corresponding solution onto a carbon-coated, Formvar-covered copper grid (400 Mesh) and then dried at room temperature. A superconducting quantum interference device (MPMS-XL SQUID) was used to measure the magnetic properties of the nanoparticles generated. Samples were prepared by placing a drop of the solution on a silicon target (5  5 mm2) and drying at room temperature under a nitrogen atmosphere. Subsequently, the particles are covered with a water glass and measured after solidification. For the preparation of the polymer composites, the nanoparticle colloid (ca. 0.5 mL containing 2 mg of nanoparticles) generated in cyclopentanone was mixed with 2 mL of the SU8 resist. The mixture was incubated at room temperature under UV light (254 nm) for 24 h. For polymerization in an external magnetic field, the sample was positioned between two NdFeB magnets (flux density ca. 1 T) under the same experimental conditions.

Results and Discussion The ablation process is limited by the process duration because an increasing concentration of nanoparticles in the liquid leads to the scattering of the incident laser beam (by the nanoparticles) and subsequently less photon energy reaches the target. As a consequence, the ablated mass of Ni48Fe52, determined by weighing the target before and after ablation, increases with ablation time, and the ablation rate continuously decreases (Figure 1a). Compared to Ni48Fe52, the determination of the ablated mass of Sm2Co17 is more difficult because of its strong magnetic properties and the resulting interference with a microscale. The approach to weighing the targets using a spacer between the scale and target also leads to the falsification of the measurements. Therefore, we compared the concentration of the colloids using UV-vis spectrometry. The generated nanoparticles do not show surface plasmon resonance in the UV-vis region, so the scattering of light determines the excitation spectra. The total scattering intensity is proportional to the particle concentration. The light scattering by small particles can be described by Mie’s solution of Maxwell’s equations or in terms of Rayleigh scattering, which is inversely proportional to the wavelength raised to the fourth power. For this reason, the UV-vis spectra can be expressed by the following function A ¼ y0 þ kλ - 4 with y0 being the background absorption, λ being the wavelength, and k being a factor that is proportional to the particle concentration. The calculated fit curve for the UV-vis spectra of Sm-Co nanoparticles ablated at 80 min can be seen in Figure 1b. Hence, we compared the UV-vis spectra of both colloids. (For the UV-vis spectra, see Supporting Information Figure S1.) Figure 1c shows the change in concentration-dependent factor k as a function of ablation time. In contrast to the laser ablation process for Ni48Fe52, the concentration of the nanoparticles, obtained after the laser ablation of Sm2Co17, starts to decrease after 50 min. This observation might be explained by the redeposition of ablated material on the target surface as a result of strong magnetic attraction. The fragmentation of generated nanoparticles via interaction with the laser beam, leading to higher transmittance of the colloid, could be a second explanation.8 To clarify this, the influence of the process duration on the nanoparticle size was investigated using dynamic light scattering measurements (DLS). As displayed in Figure 1d, there is no significant change DOI: 10.1021/la101014g

6893

Letter

Jakobi et al.

Figure 1. (a) Ablated mass of Ni-Fe nanoparticles. (b) Fit curve of the UV-vis spectra of Sm-Co colloids calculated by the Rayleigh equation. (c) Concentration-proportional factor of Ni-Fe and Sm-Co as an estimation of the ablated mass. (d) Hydrodynamic particle diameter and (f) zeta potential of laser-generated magnetic nanoparticles from NiFe and Sm2Co17 targets in cyclopentanone as a function of the ablation time.

in the hydrodynamic size distribution for varying process durations. The mean size of the Ni-Fe nanoparticles is 66 ( 33 nm whereas for nanoparticles generated by the ablation of Sm2Co17 it is 23 ( 7 nm. Because no significant difference in hydrodynamic diameter is detected with varying ablation time, the decrease in absorption in the case of Sm-Co is probably due to magnetically induced nanoparticle redeposition on the permanent magnetic Sm2Co17 target. In contrast, the Ni-Fe colloid has the same absorption tendency as does the ablated mass. However, besides nanoparticle size, the further applicability of nanoparticles (e.g., the transfer and embedding of nanoparticles into a polymer matrix) might also depend on the stability of the colloid. The electrostatic stabilization of nanoparticles generated using laser ablation in polar organic liquids omitting the addition of surfactants can be estimated via zeta potential measurements. Results indicate stable colloids for both materials by high zeta potential values of -68 to -90 mV, independent of the laser ablation time (Figure 1f). It has already been shown that the charging of the laser-generated nanoparticles occurs because of the partial oxidation of the nanoparticle surfaces.40,41 Because complete isolation of oxygen from the solvent was impossible, we assume that the surface charge is affected by the M-O- species (M = Ni, Fe, Cm, and Co). Furthermore, strong laser irradiation (43) Yang, S.; Cai, W.; Zeng, H.; Xu, X. J. Mater. Chem. 2009, 19, 7119.

6894 DOI: 10.1021/la101014g

might lead to the decomposition of the solvent. Although ultrashort pulses are reported to minimize the thermal affection,43 Besner et al. showed that even femtosecond pulses could cause thermal decomposition of solvated surfactants and additives such as biomolecules.28 The decomposition process can cause the formation of charged species that can adsorb on the nanoparticle surface and lead to the negative charge. Representative particle size distributions (DLS and transmission electron microscopy) resulting from the laser ablation of Sm2Co17 and NiFe are displayed in Figure 2. The average DLS diameter (log-normal fit) of nanoparticles produced from the Sm2Co17 target is 30 ( 6 nm, and it is 60 ( 6 nm for Ni-Fe nanoparticles. Generally, DLS data indicate the hydrodynamic size of nanoparticles, including the solvation shell. To get a better understanding of the magnetic core size (Feret diameter), we prepared TEM samples. The TEM images of the samples indicate a Feret diameter (log-normal fit, peak maximum) of 6 nm for the nanoparticles generated from both materials (Figure 2a,b). Besides the dispersed primary nanoparticles, we also detected chains of nanoparticles with a length of over 1300 nm, consisting of 10 nm nanoparticles (Figure 2c, d, peak maximum). Because DLS measurements, which are based on the diffusion coefficient of the nanoparticles, did not show any large particles in the colloid (over 100 nm), we assume that the formation of nanoparticle chains occurs during the drying process on the TEM grids. Langmuir 2010, 26(10), 6892–6897

Jakobi et al.

Letter

Figure 2. Comparison of the hydrodynamic and Feret particle size distribution (log-normal fit) of primary particles (a) Sm-Co and (b) Ni-Fe and aligned particles (c) Sm-Co and (d) Ni-Fe.

The formation of the chains in this case can be caused by the magnetic forces of prepared nanoparticles. Because the size distributions of nanoparticles in aligned chains are slightly broader compared to the size distributions of primary nanoparticles, the presence of aligned and dispersed nanoparticles let us assume that two nanoparticle fractions are generated during pulsed laser ablation in liquid, one of which is permanently magnetic. However, because the size of the generated nanoparticles determine the magnetic behavior, size control is important for possible applications. The size reduction can be reached by the fragmentation of the generated nanoparticles through focusing of a laser beam in the nanocolloid.44,45 More difficult is the size increase of the laser-generated nanoparticles, but there are also ways, as shown by Amendola et al., via the addition of surface-active substances and subsequent irradiation with the laser beam.25,46 But in our case, the possibility of size control of the laser-generated alloy nanoparticles has to be developed. Besides these options, as shown by Yang et al., the centrifugation of the generated colloids can also be used for the classification of the particles.47 Regarding the hydrodynamic sizes of the two materials, the nanoparticles produced from the Sm2Co17 target are smaller than the particles from NiFe (Figure 1c) whereas their Feret diameters are similar (Figure 2a,b). If we consider that the particles have comparable zeta potentials, then the difference in hydrodynamic diameter might be explained by different interactions between the solvent and the nanoparticle material. Besides size and stability, the elemental composition of nanomagnets is important for their potential applications because it determines their magnetic behavior. It has been reported that nanoparticles produced during pulsed laser ablation of an alloy (44) Mafune, F.; Kohno, J.; Takeda, Y.; Kondow, T. J. Phys. Chem. B 2002, 106, 7575. (45) Besner, S.; Kabashin, A. V.; Meunier, M. Appl. Phys. Lett. 2006, 89, 233122. (46) Amendola, V.; Meneghetti, M. J. Mater. Chem. 2007, 17, 4705. (47) Yang, S.; Cai, W.; Zhang, H.; Xu, X.; Zeng, H. J. Phys. Chem. C 2009, 113, 19091. (48) Tsuji, T.; Tatsuyama, Y.; Tsuji, M.; Ishida, K.; Okada, S.; Yamaki, J. Mater. Lett. 2007, 61, 2062.

Langmuir 2010, 26(10), 6892–6897

Figure 3. Composition of Ni-Fe nanoparticles generated using the laser ablation of a Ni48Fe52 target in cyclopentanone. Transmission electron micrograph (TEM) and energy-dispersive X-ray analysis (EDX) of a Ni-Fe nanoparticle in two separated regions (A and B).

may have the same elemental composition as the bulk material48 whereas the phase separation of alloys depends on the laser pulse duration and vapor pressure of each element.34 The composition of generated nanoparticles is analyzed using energy-dispersive X-ray spectroscopy (EDX). Figure 3 displays a TEM image of a nanoparticle generated using femtosecond laser ablation of an Ni48Fe52 target in cyclopentanone and the corresponding EDX analysis in two separate regions. The compositions of both regions are similar, with no significant difference compared to the target (Ni 46 ( 2.01 atom % and Fe 54 ( 1.82 atom % in area A, Ni 47 ( 1.25 atom % and Fe 53 ( 1.48 atom % in area B; for EDX analysis of the target, see the Supporting Information). This is in agreement with results published by Liu et al., who reported the successful generation of Ni-Fe nanoparticles with defined stoichiometry using femtosecond laser ablation in a vacuum chamber.49 In contrast to Ni-Fe, EDX analysis of the particles generated from Sm2Co17 indicates the enrichment of samarium in large particles and cobalt in small particles (Figure 4, Sm 97 ( 2.64 atom % and Co 3.4 ( 0 83 atom % in area A, Sm 96 ( 2.12 atom % and Co 4.0 ( 0.73 atom % in area B, Co 79 ( 4.04 atom % and Fe 21 ( 2.25 atom % in area 1). (49) Liu, B.; Hu, Z.; Chen, Y.; Sun, K.; Pan, X.; Che, Y. Proc. SPIE 2007, 6460, 646014-1–.

DOI: 10.1021/la101014g

6895

Letter

Jakobi et al.

Figure 4. Composition of nanoparticles generated using the laser ablation of a Sm2Co17 target in cyclopentanone. Transmission electron micrograph (TEM) and energy dispersive X-ray analysis (EDX) of nanoparticles in three separate regions (A, B, and 1).

Figure 5. Magnetization determined by SQUID at 5 and 300 K for laser-generated nanoparticles from NiFe and from Sm2Co17 targets.

The small amount of iron detected in the particles was already present in the target, where EDX analysis revealed the exact composition to be Sm2(Co, Cu, Fe, Zr)17 with 19.7% iron (for EDX analysis of the target, see the Supporting Information). This composition is generally summarized by Sm2Co17. The amount of Cu and Zr in the target is very low; and these elements could not be detected by EDX measurements of nanoparticles. As can be seen in Figure 4 (analysis of areas A and B), the small amount of cobalt was still detectable in the large nanoparticles. Because EDX analysis represents only single particles (Figures S2 and S3 in Supporting Information show more TEM pictures were the EDX analysis was done), we performed a quantitative elemental analysis of both nanoparticle colloids using inductively coupled plasma mass spectrometry (ICP-MS). In the case of Ni-Fe nanoparticles, the concentrations of iron and nickel were 16 and 13 mg/L, respectively, corresponding to 55 and 45%, which is in accordance with the target composition. In contrary, ICP-MS analysis of the colloid produced during the laser ablation of Sm2Co17 revealed concentrations of 97 mg/L samarium, 33 mg/L cobalt, and 16 mg/ L iron (66% Sm, 23% Co, and 11% Fe). Compared to the target, the composition of the colloid changed. As a result of ICP-MS analysis, the difference in element composition was 54.5% more samarium, 26.2% less cobalt, and 8.7% less iron. This confirms the enrichment of samarium in the liquid phase, which can be explained by magnetic precipitation of the cobalt and iron-rich particles from the solution because the complete target was covered by redeposited nanoparticles. In accordance to EDX data, we observed a different composition of the colloid compared to the target in the case of Sm-Co but similar composition in the case of Ni-Fe. Summarizing the ICP-MS and EDX analyses, we assume that both elements in Ni-Fe are ablated in amounts correlated to their percentage in the target and that the stoichiometry of nanoparticles is similar to that of the target. In contrast, Sm-Co seems to disproportionate during femtosecond laser ablation, resulting in nanoparticles with different stoichiometries. It is known that the laser ablation process causes fast vaporization of the metal, where different elements in the plume may (50) Tsuji, T.; Tsuboi, Y.; Kitamura, N.; Tsuji, M. Appl. Surf. Sci. 2004, 229, 365.

6896 DOI: 10.1021/la101014g

segregate into slow and fast components.49-52 The change in the composition of nanoparticles generated from Sm2Co17 compared to the target might be due to the large difference (132%) in the heat of evaporation of Sm and Co (Sm 164.8 kJ/mol and Co 382.4 kJ/mol). In comparison, the difference in the heat of evaporation in the case of Ni-Fe nanoparticles is only 10% (Ni 374.8 kJ/mol and Fe 340.2 kJ/mol).53 As a consequence, Sm requires less energy to evaporate than does Co, which leads to element disproportionation until it solidifies during cooling and quenching caused by liquid confinement effects. In conclusion, it seems that femtosecond laser ablation in cyclopentanone is a one-step method for the generation of stable colloidal magnetic alloy nanoparticles for elements with similar heats of evaporation. However, in the case of elements with different heats of evaporation, the stoichoimetry changes completely. To analyze the magnetic properties of generated nanoparticles, we prepared superconducting quantum interference device measurements (SQUID). Figure 5 shows the resulting magnetization of the nanomaterials produced. The coercivity of Ni-Fe nanoparticles is 140 Oe at 5 K and changes to 59 Oe at 300 K. The measured remanence ratio (Mr/Ms) of this material is 0.2 at 5 K and 0.16 at 300 K. The nanoparticles generated by the ablation of an Sm2Co17 target have a coercivity of about 170 Oe at 5 K and 30 Oe at 300 K with a remanence ratio changing from 0.2 at 5 K to 0.11 at 300 K. The data does not reveal the characteristic high coercivity that is expected for permanent magnetic Sm-Co alloys.54,55 This coercivity is similar to the coercivity that was obtained by Grass at al. for cobalt nanoparticles.3 This result emphasizes the results of the EDX and ICP-MS measurements showing that femtosecond laser ablation of Sm2Co17 leads to disproportionation of the material. (51) Yang, G. W. Prog. Mater. Sci. 2007, 52, 648. (52) Hermann, J.; Noel, S.; Itina, T. E.; Axente, E.; Povarnitsyn, M. E. Laser Phys. 2008, 18, 374. (53) Holleman, A.; Wiberg, E. Lehrbuch der Anorganischen Chemie; Walter de Gruyter: Berlin, 1995; Vol. 101, Tafel IV-V (54) Hou, Y.; Xu, Z.; Peng, S.; Rong, C. J. P.; Liu, S. S. Adv. Mater. 2007, 19, 3349. (55) Sreenivasulu, G.; Gopalan, R.; Chandrasekaran, V.; Markandeyulu, G.; Suresh, K. G.; Murty, B. S. Nanotechnology 2008, 19, 335701.

Langmuir 2010, 26(10), 6892–6897

Jakobi et al.

Letter

Figure 6. Transmission electron micrographs of an Sm-Co nanoparticle/SU-8 composite polymerized (a) without an external magnetic field and (b) in an external magnetic field.

To demonstrate prospective applications, we investigated embedding laser-generated magnetic nanoparticles into an epoxy resin (SU-8) that is widely used in photolithography and microtechnology. To transfer the nanoparticles to the resist, the nanoparticle-cyclopentanon colloid was mixed with SU-8. Subsequently, we performed polymerization with and without applying an external magnetic field; TEM images of Sm-Co nanoparticle polymer composites are displayed as an example in Figure 6. The difference in both nanocomposites is visible. Although particles are arranged only in short chains when no magnetic field is applied during polymerization (Figure 6a), long chains following the field lines form in the composite, polymerized in an external magnetic field (Figure 6b).56 The remaining magnetic behavior of the colloid prepared from the Sm2Co17 target is sufficient to enable line formation in the polymer. Because the magnetic properties of nanoparticles prepared from Sm2Co17 targets are similar to those of the Ni-Fe nanoparticles, we assume the same behavior for Ni-Fe nanoparticles during embedding into polymer. (For a similar light microscope picture of the Ni-Fe nanoparticle/SU-8 composite texture, see Figure S6 in the Supporting Information.)

Conclusions The successful generation of magnetic Ni-Fe alloy nanoparticles using femtosecond laser ablation in cyclopentanone was demonstrated. The resulting stoichoimetry of the nanoparticles depends on (56) Pazos-Perez, N.; Rodrı´ guez-Gonzalez, B.; Hilgendorff, M.; Giersig, M.; Liz-Marzan, L. M. J. Mater. Chem. 2010, 20, 61.

Langmuir 2010, 26(10), 6892–6897

the elemental composition of the target. One determining parameter seems to be the heat of evaporation of the elements contained. If the difference in the heat of evaporation is relatively small, as for Ni-Fe, then the stoichiometry of the target can be preserved during the ablation process. For higher differences, as in Sm-Co, an enrichment of Sm with a lower heat of evaporation was observed. The SQUID measurements showed coercivities of 59 Oe for Ni-Fe nanoparticles and 30 Oe for nanoparticles obtained from the Sm2Co17 target. The change in stoichiometry for Sm-Co was confirmed by measurements of the magnetic properties, which also deviated from the expected values. Subsequent transfer and embedding into a polymer resin without using chemical additives could be shown for both colloids, independent of the nanoparticle composition. Acknowledgment. This work was supported by funding from the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) for the Cluster of Excellence REBIRTH and ZFM Center for Solid State Chemistry and New Materials, Leibniz University Hannover, Hannover, Germany. We thank S. Maas from Material Analytischer Service Freiburg for the TEM and EDX analysis. We also acknowledge D. Goll (Max-PlanckInstitut f€ur Metallforschung, Stuttgart, Germany) for the SQUID measurements. Supporting Information Available: Additional UV-vis spectra, TEM images of generated nanoparticles, and XRD analyses of targets. This material is available free of charge via the Internet at http://pubs.acs.org.

DOI: 10.1021/la101014g

6897