Patterning of Magnetic Bimetallic Coordination Nanoparticles of

Article pubs.acs.org/Langmuir

Patterning of Magnetic Bimetallic Coordination Nanoparticles of Prussian Blue Derivatives by the Langmuir−Blodgett Technique Miguel Clemente-León,*,† Eugenio Coronado,*,† Á ngel López-Muñoz,† Diego Repetto,† Laure Catala,‡ and Talal Mallah‡ †

Instituto de Ciencia Molecular, Universidad de Valencia, Calle Catedrático José Beltrán 2, 46980 Paterna, Spain Institut de Chimie Moléculaire et des Matériaux d’Orsay, Université Paris-Sud 11, CNRS, 91405 Orsay, France

‡

S Supporting Information *

ABSTRACT: We report a novel method to prepare patterns of nanoparticles over large areas of the substrate. This method is based on the adsorption of the negatively charged nanoparticles dispersed in an aqueous subphase onto a monolayer of the phospholipid dipalmitoyl-L-α-phosphatidylcholine (DPPC) at the air−water interface. It has been used to prepare patterns of nanoparticles of Prussian blue analogues (PBA) of different size (K0.25Ni[Fe(CN)6]0.75 (NiFe), K0.25Ni[Cr(CN)6]0.75 (NiCr), K0.25Ni[Co(CN)6]0.75 (NiCo), Cs0.4Co[Cr(CN)6]0.8 (CsCoCr), and Cs0.4Co[Fe(CN)6]0.9 (CsCoFe)). The behavior of DPPC monolayer at the air−water interface in the presence of the subphase of PBA nanoparticles has been studied by the compression isotherms and Brewster angle microscopy (BAM) images. Atomic force microscopy (AFM) of the transferred films on mica substrates shows that patterns of the nanoparticles are observed for a 10−4 M concentration of the subphase, based on the nanoparticle precursors, at surface pressures between 1 and 6 mN/m and transfer velocities from 10 to 80 mm/min. Vertical, horizontal, or tilted fringes of the nanoparticles with respect to the transfer direction can be obtained depending on the transfer velocity and surface pressure.

■

INTRODUCTION In the past few years there has been an important advancement in the synthesis of nanoparticles of a wide range of materials with excellent control over particle size and shape. The next step, which is a fundamental requirement for the exploitation of their functionalities at the nanoscale, is the capability to deposit these objects on surfaces and, moreover, to do it in a controlled manner. This positioning is usually achieved by direct assembly in templates fabricated by top-down approaches.1 Still, a selfassembly method is highly desirable because of its simplicity and compatibility with heterogeneous integration processes.2 An elegant approach to arrange molecules or nanoparticles into well-organized monolayers or multilayered films is the Langmuir−Blodgett (LB) technique.3 Although most research efforts in nanoparticle LB films have been focused on the preparation of high-density or even close-packed nanoparticle arrays,4,5 few groups have studied low-density nanoparticle monolayers and their transfer onto substrates as this can lead to ordered alignment of the nanoparticles at the micrometer scale by controlled dewetting of the monolayer.6 These methods rely on the preparation of a floating monolayer of the nanoparticles, which requires a hydrophobic capping of the nanoparticles. In this paper we propose an alternative possibility that can be applied to hydrophilic negatively charged water-soluble nanoparticles. It is based on the adsorption of the nanoparticles dispersed in a water subphase on a floating phospholipid monolayer. Transfer of this monolayer onto a substrate by the LB technique gives rise to patterned structures of the © 2012 American Chemical Society

nanoparticles. This method is based on the work of Chi et al. that described the preparation of periodic stripe patterns of the phospholipid dipalmitoyl-L-α-phosphatidylcholine (DPPC) with feature sizes down to 100 nm over large surface areas through self-organization using the LB technique.7−9 These authors have used these patterned surfaces as templates to direct the self-assembly of nanoparticles.7,10 In the present case, the adsorption of negatively charged nanoparticles onto the DPPC Langmuir monolayer allows obtaining a pattern of the nanoparticles in one single step using similar conditions. In recent works, a similar strategy has been employed to grow gold11 and iron oxide12 nanoparticles on transferred DPPC monolayers, but a nanoparticle pattern has not been achieved. Bimetallic nanoparticles of Prussian blue analogues (PBA) have been chosen to test this method. The interest of these nanoparticles stems from the diversity of the properties observed in the bulk materials (ferro- and ferrimagnetism, photomagnetism, electrochromism, ion selective membranes, catalysis, charge transfer, etc.).13 Several groups have studied these nanoparticles to explore the size reduction effects on the magnetic properties14 and to create new properties at the nanometer scale (photomagnetism,15 multifunctionality16). Although the growth of thin films of PBA nanoparticles on different substrates has been achieved by different methods Received: December 15, 2011 Revised: February 7, 2012 Published: February 7, 2012 4525

dx.doi.org/10.1021/la2049508 | Langmuir 2012, 28, 4525−4533

Langmuir

Article

including the LB technique,17 there are very few reports on the selective attachment of PBA nanoparticles to specific positions on the surfaces and they are in all cases top-down approaches.18 Recently, liquid phase grafting was combined with direct FIB (focused ion beam) patterning to deposit Cs0.4Ni[Cr(CN)6]0.9 (CsNiCr) nanoparticles on selected portions of Si surfaces.19 Here, we present a different bottom-up approach based on the LB technique that leads, using a simple and fast process, to regular fringes of PBA nanoparticles with submicrometer-scale lateral dimensions over large areas of the substrate.

■

nanoparticles K 0.25 Ni[Fe(CN) 6 ] 0.75 (NiFe), K 0.25 Ni[Cr(CN)6]0.75 (NiCr), K0.25Ni[Co(CN)6]0.75 (NiCo), Cs0.4Co[Cr(CN)6]0.8 (CsCoCr), and Cs0.4Co[Fe(CN)6]0.9 (CsCoFe) have been prepared. The first step to fabricate a pattern of these nanoparticles on a substrate is the preparation of a monolayer of DPPC at the air−water interface. The negatively charged nanoparticles dispersed in the aqueous subphase are adsorbed onto the cationic part of this zwitterionic monolayer by electrostatic interactions as observed for discrete polyoxometalate anions.22 The second step is the transfer of this monolayer onto a mica substrate by using the LB technique. We will discuss first the behavior of DPPC monolayer at the air−water interface in the presence of the subphase of PBA nanoparticles that has been studied with the help of the compression isotherms and Brewster angle microscopy (BAM) images. Several concentrations of nanoparticles have been tried by diluting the original aqueous solution of nanoparticles to find the best conditions for the patterning. The concentration that gives a pattern of the transferred monolayer is 10−4 M with respect to the [B(CN)6]3− (B = Fe, Co, and Cr) and M2+ (Ni and Co) precursors (10−3 M in the original solution of the nanoparticles). Therefore, the original nanoparticle solution was diluted 10 times to prepare the water subphase used for the formation of the DPPC monolayer. In the case of Cs0.4Ni[Cr(CN)6]0.9, patterning could not be achieved at any concentration. DLS experiments confirm that the nanoparticles are preserved at this concentration. π−A Isotherms. Figure 1 shows the compression isotherms at 22 °C corresponding to the DPPC monolayer on a subphase

EXPERIMENTAL SECTION

Cs0.7Ni[Cr(CN)6]0.9 (CsNiCr) nanoparticles were prepared as reported in the literature.14a,c The following types of PBA nanoparticles K0.25Ni[Fe(CN)6]0.75 (NiFe), K0.25Ni[Cr(CN)6]0.75 (NiCr), K0.25Ni[Co(CN)6]0.75 (NiCo), Cs0.4Co[Cr(CN)6]0.8 (CsCoCr), and Cs0.4Co[Fe(CN)6]0.9 (CsCoFe) were prepared as follows. K0.25Ni[B(CN)6]0.75 (B = Cr, Co, and Fe). An aqueous 2 × 10−3 M solution of K3B(CN)6 (B = Cr, Co, and Fe) was added to an aqueous solution containing 2 × 10−3 M NiCl2·6H2O. The suspension was stirred for 1 h at room temperature. Cs0.4Co[Fe(CN)6]0.9. An aqueous 2 × 10−3 M solution of K3Fe(CN)6 was added to an aqueous solution containing 2 × 10−3 M CoCl2·6H2O and 4 × 10−3 M CsCl. The suspension was stirred for 1 h at room temperature. Cs0.4Co[Cr(CN)6]0.8. An aqueous 2 × 10−3 M solution of K3Cr(CN)6 was added to an aqueous solution containing 2 × 10−3 M CoCl2·6H2O and 4 × 10−3 M CsCl. The suspension was stirred for 1 h at 2 °C. Subphases were prepared by diluting the original solution of the nanoparticles 10 times to reach a final concentration of M2+ and B(CN)63− of 10−4 M. No problem of colloidal stability was observed for these solutions over a period of several hours, much longer than the time used for Langmuir experiments. DPPC dissolved in CHCl3 with a 1 mM concentration was used as spreading solution. An appropriate amount of this solution was carefully spread onto the aqueous subphase of the nanoparticles, and the spreading solvent was allowed to evaporate for 10 min prior to compression. Monolayers at the mica surface were prepared by immersing a freshly cleaved mica plate in the subphase, after which the DPPC solution was spread on the subphase. After compressing to the target surface pressure and waiting less than 2 min, the monolayer was transferred on the mica plate by the vertical lifting method with an upward motion. Isotherms were obtained with a NIMA trough (type 702BAM) equipped with a Wilhelmy plate and maintained at a constant temperature (22 or 19 °C). A KSV3000 trough has been used to prepare the LB films. Millipore water with a resistivity higher than 18 MΩ cm was used in all the experiments. A EP3-BAM from NFT has been used for the Brewster angle microscopy experiments. A commercial atomic force microscope (AFM) (Multimode SPM by Veeco) has been employed for surface sample characterization. AFM images were processed by using WSxM 5.0.20 Dynamic light scattering (DLS) experiments have been performed on a Malvern Nanozetasizer apparatus (equipped with a backscattering mode) on 1.5 mL of an aqueous solution. The particle size distribution from dynamic light scattering (DLS) is derived from a deconvolution of the measured intensity autocorrelation function of the sample. This is accomplished using a non-negatively constrained least-squares (NNLS) fitting algorithm. The number profile is used to estimate the size corresponding to the main peaks.

Figure 1. Compression isotherm of a monolayer of DPPC in the presence of a subphase of NiFe (black), NiCr (dark blue), NiCo (dark green), CsCoFe (red), CsCoCr (blue), CsCoCr (clear blue), and pure water (clear green) subphase. The concentration of nanoparticles in these subphases is 10−4 M.

in absence and presence of the five types of nanoparticles at a 10−4 M concentration. The isotherm of DPPC in pure water shows three distinguishable phases: the liquid-expanded (LE) region at higher areas per molecule, a plateau corresponding to the transition from LE to liquid-condensed (LC) where both phases coexist, and the LC phase at lower areas per molecule in good agreement with previously published results.9,11,12,22a The compression isotherms in presence of the nanoparticles show a very similar behavior with small differences in areas per molecule (less than 8 Å2/molecule) in the three phases of the compression isotherm and small increases in the collapse pressure for some of the nanoparticles (less than 5 mN/m). Although the changes of area per molecule are small, there is

■

RESULTS AND DISCUSSION We have used a simple procedure to obtain the stabilization of the Cs0.7Ni[Cr(CN)6]0.9 (CsNiCr) nanoparticles in an aqueous subphase.14a,c This method leads to the stabilization of negatively charged particles with different sizes depending on the metals and alkaline ions.16,19,21 The following types of PBA 4526


Langmuir

Article

Figure 2. BAM images of a DPPC monolayer, at the air−water interface on pure water (top) and on a 10−4 M concentrations of NiFe nanoparticles in the subphase (down) under different surface pressures. Real size: 457 × 350 μm2. The intensity scale is the same for all these pictures except for that with NiFe nanoparticles at 10.3 mN/m which has been adjusted due to the much higher brightness.

the number of adsorbed nanoparticles is too small to induce a pattern. This could indicate that the mechanism of the pattern of PBA nanoparticles is closely associated with that of DPPC alone reported by Chi et al.9 Brewster Angle Microscopy (BAM). BAM uses polarized light passing through media with dissimilar refractive indexes. It allows the direct observation of phase changes of amphiphilic molecules in a large length scale (μm). BAM images of DPPC on pure water have been reported in the literature.9,23 At low surface pressures, they show a homogeneous LE phase (see image at 5 mN/m, Figure 2), followed by the appearance of small and irregular LC domains at 7.0 mN/m (see image at 7.5 mN/m, Figure 2). Beyond the LE−LC transition region (8.1 mN/m), the LC domains occupy the entire image (see image at 8.1 mN/m, Figure 2). The LC domains then merge together with further compression (see image at 21.3 mN/m). For DPPC/NiFe, a homogeneous LE phase is observed at surface pressures below the LE−LC transition (see image at 5 mN/m, Figure 2) as observed for DPPC on water, but a different behavior is observed at higher surface pressures. Thus, at the LE−LC transition region an inhomogeneous phase with brighter points appears instead of the well-defined LC domains

some interaction between DPPC and the nanoparticles as shown by BAM Images (see below). This behavior contrasts with that observed for polyoxometalates at a 10−4 M concentration,22a which shift the compression isotherm of DPPC toward larger areas per molecule in the LE and LC zones of the isotherm. In this case, the shift of the isotherm in the presence of anions is explained by an increase of the repulsions between DPPC molecules, perhaps by an induced change in the orientation of the lipid polar head along the interface. To see a similar effect with PBA nanoparticles, it is necessary to increase the concentration of the nanoparticles subphase to the maximum (10−3 M in NiFe). Thus, at this concentration the isotherm is shifted toward larger areas in the LC and LE zones as observed for polyoxometalates, which could be interpreted in the same way (Figure SI1). However, at this concentration, the patterning is not achieved, and neither when the concentration of the nanoparticles is decreased to 10−5 M. Therefore, we can conclude that the concentration of the nanoparticles plays an important role in the formation of patterns on the transferred monolayer. If this concentration is too high, it causes great changes on the DPPC monolayer that prevents patterning of the nanoparticles, while if it is too low, 4527


Langmuir

Article

Figure 3. DPPC/NiFe patterns on mica vs surface pressure and withdrawn speed at 22 °C.

Preparation of Patterns of PBA Nanoparticles. The process of patterning formation is analogous to that described by Chi et al. for DPPC on pure water.9 After compressing the Langmuir monolayer to the liquid-expanded (LE) phase region, the monolayer is deposited onto a mica substrate by the LB technique. The mica substrate, which is immersed in the nanoparticle subphase before spreading, is withdrawn at a constant speed. Different patterns of the nanoparticles are obtained depending on the speed of the substrate and the surface pressure. As the clearest patterns have been obtained with NiFe nanoparticles, a more exhaustive study has been performed with these nanoparticles. Atomic Force Microscopy (AFM). The morphology of these LB films has been studied by atomic force microscopy (AFM). Patterns of NiFe nanoparticles are observed for a 10−4 M concentration of the subphase at surface pressures between 1 and 6 mN/m and transfer velocities from 10 to 80 mm/min at 22 °C as shown in Figure 3. In general, vertical fringes of the nanoparticles that run parallel to the transfer direction (perpendicular to the three-phase contact line) are obtained at low transfer velocity (10 mm/min) at almost all transfer pressures (1, 2, 3, 4, and 5 mN/m) and at high transfer velocity (80 mm/min) and low surface pressure (1 mN/m). Almost horizontal fringes running perpendicular to the transfer direction are obtained at high transfer velocity (80 mm/min) and intermediate surface pressures (3 and 4 mN/m) while tilted fringes are obtained at lower or higher surface pressures (2, 5, and 6 mN/m) at high transfer velocity (80 mm/min) and at 6 mN/m for low transfer velocity (10 mm/min). In some cases, different types of patterns have been obtained in the same substrate. This occurs at 5 mN/m and 10 mm/min, which is a condition intermediate between those ones giving rise to vertical fringes or tilted fringes. This could indicate a transition between two types of domains. These fringes of nanoparticles are separated by quasi-empty zones. The maximum height differences between the fringes and the lowest zones are around 25 nm and a mean height of 17 nm in good coincidence with the mean size of the nanoparticles calculated from DLS (19 mm, see Figure SI2). The nanoparticles have an apparent mean diameter of 25 nm. The features in an AFM image come from the convolution between tip and sample and the lateral resolution depends on the radius of the tip employed for the measurements. Since we use super sharp tips (radius less than 5 nm), the agreement between apparent and real diameter is rather good. A close look at these fringes (see images at 3 mN/m and 10 mm/min, 2

on a homogeneous LE background observed on pure water (see images at 7.3 and 7.5 mN/m, Figure 2). At increasing surface pressure, the reflectivity of these brighter points increases, and finally a more homogeneous and brighter phase is observed beyond the plateau of the isotherm (see image at 10.3 mN/m, Figure 2). Further compression leads to an increase of the reflectivity of this continuous phase. The reflectivity of the images in the LC region is much higher than that of DPPC on pure water. The behavior of DPPC on the other nanoparticle subphases is similar. The increase of reflectivity of the LC zones could indicate the adsorption of the negatively charged nanoparticles dispersed in the subphase in the cationic part of the zwitterionic DPPC. A similar effect has been observed for the cationic surfactant, DODA (dioctadecyldimethylammonium bromide) in the presence of CsNiCr nanoparticles. In this case, a continuous very bright phase appears for all surface pressure at similar concentration of the nanoparticles (10−4 M) attributed to the adsorption of a continuous layer of particles under the DODA monolayer, which was observed on transferred LB films.17 In the case of DPPC, it is possible to increase the amount of adsorbed nanoparticles by increasing the concentration of NiFe nanoparticles to 10−3 M. Thus, BAM images of DPPC monolayer at this concentration show a higher reflectivity from the beginning of the compression, which is much higher in the LC zone, but as mentioned below, a patterning is not obtained. On the contrary, a lower concentration of nanoparticles (10−5 M) gives rise to BAM images of similar morphology but lower reflectivity in the LE−LC transition and LC zones. Finally, the behavior of DPPC on a subphase containing the negatively charged precursors of the nanoparticles, K3Fe(CN)6 and K3Cr(CN)6, at the same concentration as in the nanoparticle solution (10−4 M), has been studied as a reference. The compression isotherm, BAM images, and reflectivity values are very similar to those of DPPC on pure water. This indicates that the changes in the morphology and the reflectivity of BAM images of the monolayer are caused by the large anionic species of the nanoparticles and not by the anionic precursor that could result from partial decomposition of the nanoparticles. BAM images are consistent with the adsorption of the nanoparticles on the DPPC monolayer. Furthermore, they confirm that a high or too low adsorption of the nanoparticles in the LE and LC zones prevents the pattern formation. This indicates that the mechanism of pattern formation of the PBA nanoparticles is related to that of DPPC on pure water. 4528


Langmuir

Article

that found for patterns of DPPC.9 According to their interpretation, the higher zones would be composed by DPPC molecules in the LC phase while the lower zones (the channels) could be formed by DPPC molecules in the LE phase in which the alkyl chains have a larger tilt angle compared to condensed DPPC. In contrast to this, the tilted fringes obtained at 2 mN and 80 mm/min are inside channels with a depth of 2 nm in the lower zones. At these conditions, the stripes separating these channels present a smoother surface and are much more regular than those obtained at 3 mN/m and 10 mm/min. It seems that in this case, the higher DPPC domains are more condensed than those obtained at lower surface pressure. The height difference of 2 nm corresponds to the length of DPPC. This could indicate that the lower parts of the channels correspond to the naked substrate while the higher ones with a height of around 1 nm could correspond to DPPC molecules in the LE state. These results confirm again that the pattern formation of the nanoparticles is related to that of DPPC alone. However, these channels are not observed for all the conditions. This could be due to the difficulty to get AFM images with good resolution of both the channels and the nanoparticles, which may depend on the tip quality (usually even an ultra sharp tip get blunt quickly) or on the possible relaxation of the LC DPPC domains to the LE state on the transferred films as observed for other patterns of phospholipids obtained with the LB technique.24 The width of the fringes is between 0.6 and 1.0 μm in the major part of the fringe. However, it is not constant and branches are often observed. The distance between the vertical, horizontal or tilted fringes is around 10 μm with irregularities due to the branches as observed in an AFM image taken in a larger zone of substrate (103 × 103 μm) for a LB film with horizontal fringes transferred at 3 mN/m and 80 mm/min (see Figure SI3). These distances are similar to those found for patterns of pure DPPC running parallel to the transfer direction at low transfer velocities (from around 20 to 10 μm), but they contrast with those found for the horizontal stripes of pure DPPC that present shorter widths of less than 3 μm.25 Transfer at other concentrations of the subphase does not give rise to patterns of the nanoparticles. Thus, images of LB films prepared on a 10−5 M subphase show mainly an empty surface with very few isolated NiFe nanoparticles (see Figure SI4). It is interesting to notice that the presence of NiFe nanoparticles at this low concentration prevents the formation

mN/M and 80 mm/min, or 5 mN/m and 80 mm/min, Figures 4, 5, and 6) reveals that they are formed by aggregates of close-

Figure 4. Phase AFM image (40 μm × 40 μm) of DPPC/NiFe pattern on mica at 3 mN/m and 10 mm/min at 22 °C.

packed nanoparticles with some empty zones. The measured heights indicate that the fringes are formed by a single layer of nanoparticles, whereas larger heights are due to larger nanoparticles. It should be mentioned that for this particular combination of sample and substrate (DPPC and nanoparticles on mica) it was not easy at all to get a good quality AFM image and, depending on the tip (new or old) and other parameters (applied force, feedback and scanning speed, oscillation amplitude), higher or lower image resolution was obtained. As a consequence, in few cases (see images at 3 mN/m and 10 mm/min and 2 mN/M and 80 mm/min, Figures 4 and 5), it was possible to resolve that the fringes of nanoparticles are inside channels with a depth of 1−2 nm with respect to the area between the fringes empty of nanoparticles. In the vertical patterning obtained at 3 mN and 10 mm/min, the depth of these channels is ∼1 nm. The stripes separating these channels are not continuous and contain numerous holes with the same depth without nanoparticles. This height difference is similar to

Figure 5. AFM image (40 μm × 40 μm) of DPPC/NiFe pattern on mica at 2 mN/m and 80 mm/min at 22 °C. 4529


Langmuir

Article

Figure 6. AFM image (2 μm × 2 μm) of DPPC_NiFe pattern on mica at 5 mN/m and 80 mm/min at 22 °C.

Figure 7. AFM image (40 μm × 40 μm) of DPPC/CsCoCr pattern on mica at 3 mN/m and 80 mm/min at 19 °C.

of a DPPC pattern, which would be obtained at the same conditions on pure water. On the contrary, LB films prepared on a 10−3 M subphase show an inhomogeneous surface mostly covered with islands of close packed nanoparticles (∼15 and 30 nm height). This result suggests the presence of bilayers in some zones of the film (see Figure SI4). Lower concentrations of the nanoparticles were not tried as this can give rise to the decomposition of the nanoparticles.17 A control experiment has been performed by immersing the substrate in a 10−4 M NiFe subphase without DPPC monolayer at the air−water interface. After a few minutes, the substrate was withdrawn at a constant speed. AFM images of the transferred film show that the surface is homogeneously covered with isolated groups with a few nanoparticles, and therefore, pattern is not obtained (see Figure SI5). This demonstrates that the adsorption of the negatively charged NiFe nanoparticles onto DPPC monolayer plays an important role in the pattern formation. Another factor that needs to be taken into account is the time. If transfer is performed not immediately after the target surface pressure has been reached (for instance, 90 min after), pattern is not achieved. In these cases, the substrate is covered homogeneously with NiFe nanoparticles (see Figure SI6). If this time is reduced to 20 min, pattern is obtained, but the zones between the fringes are also covered with nanoparticles (see Figure SI6). Finally, a pattern of nanoparticles separated by almost empty zones is obtained if the DPPC monolayer is transferred 5 min after the target surface pressure has been

reached (see Figure SI6). This indicates that the adsorption of the nanoparticles dissolved in the subphase onto the DPPC monolayer increases with time. This could explain that the width of the fringes and their separation do not change regularly with the surface pressure or transfer velocity as for patterns of DPPC in pure water. When a PBA subphase is present, the pattern formation is a more complicated process since DPPC monolayer changes with the time due to the increased adsorption of the nanoparticles. Patterns of Other PBA Nanoparticles. This method can be extended to other PBA nanoparticles of different size. Thus, it is possible to obtain patterns of larger nanoparticles such as CsCoCr, NiCr, and NiCo with average diameters of 35, 26, and 23 nm calculated by DLS or of smaller nanoparticles such as CsCoFe nanoparticles with an average diameter of 14 nm calculated by DLS. Some examples of these patterns are given below. Tilted fringes of CsCoCr nanoparticles with a height comprised between 25 and 35 nm, in agreement with DLS measurements, and a width of ∼3−5 μm, separated by empty zones of ∼10 μm can be obtained at a surface pressure of 3 mN/m and a transfer velocity of 80 mm/min at 19 °C (see Figure 7). A vertical or tilted pattern of NiCr nanoparticles with a height of 20−23 nm and a width of ∼1−2 μm, separated by empty zones of ∼25 μm depending on the zone of the substrate can be obtained at a surface pressure of 2 mN/m and a transfer velocity of 10 mm/min at 22 °C (see Figure SI7). Finally, in the case of NiCo nanoparticles, vertical fringes with a height 4530


Langmuir

Article

Figure 8. Height and phase AFM images (30 μm × 30 μm) of DPPC/NiCo pattern on mica at 6 mN/m and 80 mm/min.

mean of 15 nm separated by empty zones of ∼15 μm can be obtained at a surface pressure of 6 mN/m and a transfer velocity of 80 mm/min at 22 °C (see Figure 8). In this case, it was possible to resolve that the fringes of nanoparticles are inside channels with a depth of ∼1 nm with respect to the area between the fringes empty of nanoparticles. When the size of the nanoparticles is reduced with respect to NiFe nanoparticles, it is still possible to obtain a pattern. An example of a pattern obtained with CsCoFe nanoparticles at 2 mN/m and 30 mm/min at 19 °C is shown in Figure SI8. It is formed by branched horizontal fringes of the nanoparticles with a mean height of 13 nm. However, for much smaller nanoparticles such as the CsNiCr ones (average diameter of 8 nm calculated by DLS),14a,17 patterns could not be obtained. The adsorption of the CsNiCr nanoparticles on the DPPC monolayer seems to be higher than that of the other PBA nanoparticles as the mica substrate is completely covered in all images, even at concentrations lower than 10−4 M. In some cases, empty circles surrounded by close-packed CsNiCr nanoparticles were found (see image at 1.5 mN/m and 30 mm/min with a concentration of 10−5 M, Figure SI9). A possible explanation for this different behavior is that the smaller size of these nanoparticles enhances their adsorption onto the DPPC monolayer preventing the pattern formation. Mechanism. The presence of channels containing the nanoparticles with a depth consistent with that of DPPC

confirms that the mechanism of pattern formation is related to that of DPPC alone. Two possible mechanisms have been proposed for the vertical and horizontal patterns of DPPC. The horizontal pattern formation has been attributed to the substrate-mediated first-order phase transition of DPPC from the LE to the LC phases, whereas the vertical pattern has been attributed to meniscus oscillation during LB transfer.9,25 The determination of the precise mechanism of the pattern formation for NiFe nanoparticles is out of the scope of this work since more measurements with other techniques are needed. A comparison with the results obtained for DPPC on pure water indicates that the formation of horizontal patterns is only achieved at very special conditions for NiFe nanoparticles (high transfer velocity, intermediate surface pressures) and it is not perfect, in contrast to DPPC monolayers on pure water that give rise to horizontal stripes at a wider variety of surface pressures and transfer velocities. Furthermore, the distances between the fringes of nanoparticles on the horizontal patterns are much higher than those obtained between horizontal channels on pure DPPC patterns. It seems then that the presence of the PBA nanoparticles in the subphase favors the formation of vertical patterns of DPPC and hence that the mechanism for the patterning of these nanoparticles may be related to that proposed for the vertical pattern of DPPC (meniscus oscillation during LB transfer). This would explain the presence of channels, in which nanoparticles are 4531


Langmuir

Article

could be feasible as these substrates have already been used to build DPPC patterns with the LB technique.9

preferentially adsorbed during the transfer process, separated by stripes of DPPC in the LC state. This contrasts with the behavior observed at the air−water interface as BAM images suggest that adsorption of the nanoparticles is higher in the LC zone. A possible explanation of the opposite behavior observed in the transferred film is that during the transfer process the NiFe nanoparticles are expulsed from the condensed zones of DPPC monolayer, which are more hydrophobic than the expanded ones. Because of this, nanoparticles are placed preferentially in the lower and more hydrophilic zones of the transferred films that correspond to the naked substrate or DPPC molecules in the LE state. Finally, other mechanisms have also been suggested. Thus, stripe micropatterns also parallel to the transfer direction for lipid/lipopolymer monolayers have been obtained by Tanaka et al.26 In a recent publication these authors conclude from data obtained by imaging ellipsometry and fluorescence microscopy that at slow velocities the parallel stripe formation may be due to spinodal decomposition, whereas at relatively high velocities it may be a dominant effect of the shear force.27 In this case, the small contact angle determined by in situ ellipsometric imaging excludes the contribution of a local fluctuation of the meniscus in contrast to the mechanism proposed by Chi et al. for vertical DPPC patterns.9 On the other hand, Pignataro et al. have proposed a nonequilibrium model to explain perpendicular or parallel periodic patterns of phospholipids based on the stability of surfactant concentration and film thickness coupled fluctuations near the meniscus of a surfactant-covered receding thin film.24

■

ASSOCIATED CONTENT

S Supporting Information *

Compression isotherm of DPPC on a 10−3 M NiFe subphase; DLS of 10−3 and 10−4 M NiFe solutions; AFM image (103 μm × 103 μm) of DPPC/NiFe pattern on mica at 3 mN/m and 80 mm/min at 22 °C; AFM images of DPPC/NiFe on a 10−5 and 10−3 M NiFe subphases; reference AFM image without DPPC monolayer; NiFe pattern degradation with time after compression; and AFM images of DPPC/NiCr, DPPC/ CsCoFe, and DPPC/CsNiCr patterns. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (M.C.-L.); eugenio. [email protected] (E.C.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Financial support from the European Union (ERC Advanced grant SPINMOL, Project HINTS), the Spanish MICINN (Project Consolider-Ingenio in Molecular Nanoscience, CSD2007-00010, MAT2007-61584, CTQ-2008-06720 and CTQ-2011-26507), the Generalitat Valenciana (Prometeo Program), and the French programme ANR-blanc (project MS-MCNP no. 30615) is gratefully acknowledged. Eva Tormos is greatly acknowledged by AFM images of NiFe nanoparticles obtained at 10−3 and 10−5 M concentration.

■

CONCLUSION In this work, we report a novel method to prepare patterns of nanoparticles over large areas of the substrate based on the adsorption of the nanoparticles dispersed in an aqueous subphase onto a phospholipid monolayer at the air−water interface. The main advantage of this method is that it does not require a treatment of the substrate and that can be performed in a single step. Furthermore, it can be applied to negatively charged nanoparticles dispersible in water avoiding the formation of a monolayer of the nanoparticles at the air− water interface, which is not possible for many cases. The main problem of this method is that the control of the size is lower than that achieved by dewetting of monolayers of nanoparticles,2 which is due probably to the increasing adsorption with time of the nanoparticles under the phospholipid monolayer. The proposed method is general and could be extended to other negatively charged nanoparticles or molecules, enlarging their possible applications. In the case of the PBA, these patterns, consisting of densely packed nanoparticles, could display interesting optical, magnetic, or electrical properties depending on the PBA derivative. In particular, it could be useful for investigation of magnetic interactions and magnetic anisotropy in arrays. On the other hand, electrical conductivity at the nanoscale through this type of films opens up interesting perspectives for magneto-transport and magneto-optical studies. In fact, it has been proposed that this could be applied to fabricate nanoscale interconnects between semiconductor and magnetic materials, with possible applications in semiconductors industry, optoelectronics, clinical diagnosis, sensor applications, ion-selective properties, battery applications, and microelectro-mechanical systems.18 For this purpose, mica substrates should be replaced by silicon wafers, something that

■

REFERENCES

(1) (a) Martínez, R. V.; Martínez, J.; Chiesa, M.; García, R.; Coronado, E.; Pinilla-Cienfuegos, E.; Tatay, S. Large-scale Nanopatterning of Single Proteins used as Carriers of Magnetic Nanoparticles. Adv. Mater. 2009, 22, 588−591. (b) Coronado, E.; MartíGastaldo, C.; Galán-Mascarós, J. R.; Cavallini, M. Polymetallic OxalateBased 2D Magnets: Soluble Molecular Precursors for the Nanostructuration of Magnetic Oxides. J. Am. Chem. Soc. 2011, 132, 5456− 5468. (2) Huang, J.; Kim, F.; Tao, A. R.; Connor, S.; Yang, P. Spontaneous formation of nanoparticle stripe patterns through dewetting. Nature Mater. 2005, 4, 896−900. (3) (a) Ulman, A. An Introduction to Ultrathin Organic Films from Langmuir-Blodgett to Self-Assembly; Academic Press: San Diego, 1991. (b) Coronado, E.; Mingotaud, C. Hybrid organic/inorganic LangmuirBlodgett films. A supramolecular approach to ultrathin magnetic films. Adv. Mater. 1999, 11, 869−872. (4) Fried, T.; Shemer, G.; Markovich, G. Ordered two-dimensional arrays of ferrite nanoparticles. Adv. Mater. 2001, 13, 1158−1161. (5) Acharya, S.; Hill, J. P.; Ariga, K. Soft Langmuir−Blodgett Technique for Hard Nanomaterials. Adv. Mater. 2009, 21, 2959−2981. (6) Tao, A. R.; Huang, J.; Yang, P. Langmuir−Blodgettry of Nanocrystals and Nanowires. Acc. Chem. Res. 2008, 41, 1662−1673. (7) Gleiche, M.; Chi, L. F.; Fuchs, H. Nanoscopic channel lattices with controlled anisotropic wetting. Nature 2000, 403, 173−174. (8) Chen, X.; Hirtz, M.; Fuchs, H.; Chi, L. F. Self-organized patterning: Regular and spatially tunable luminescent submicrometer stripes over large areas. Adv. Mater. 2005, 17, 2881−2885. (9) Chen, X.; Lenhert, S.; Hirtz, M.; Lu, N.; Fuchs, H.; Chi, L. F. Langmuir−Blodgett Patterning: A Bottom−Up Way To Build Mesostructures over Large Areas. Acc. Chem. Res. 2007, 40, 393−401. 4532


Langmuir

Article

(10) Lu, N.; Chen, X. D.; Molenda, D.; Naber, A.; Fuchs, H.; Talapin, D. V.; Weller, H.; Muller, J.; Lupton, J. M.; Feldmann, J.; Rogach, A. L.; Chi, L. F. Lateral Patterning of Luminescent CdSe Nanocrystals by Selective Dewetting from Self-Assembled Organic Templates. Nano Lett. 2004, 4, 885−888. (11) Hsiao, F.-W.; Lee, Y.-L.; Chang, C.-H. On the characteristics of mixed Langmuir monolayer templates containing dipalmitoyl phosphatidylcholine for gold nanoparticle formation. Colloids Surf., B 2009, 73, 110−115. (12) Sarkar, R.; Pal, P.; Mahato, M.; Kamilya, T.; Chaudhuri, A.; Talapatra, G. B. On the origin of iron-oxide nanoparticle formation using phospholipid membrane template. Colloids Surf., B 2010, 79, 384−389. (13) Ferlay, S.; Mallah, T.; Ouahes, R.; Veillet, P.; Verdaguer., M. A room-temperature organometallic magnet based on Prussian blue. Nature 1995, 378, 701−703. (14) (a) Brinzei, D.; Catala, L.; Louvain, N.; Rogez, G.; Stéphan, O.; Gloter, A.; Mallah, T. Spontaneous stabilization and isolation of dispersible bimetallic coordination nanoparticles of CsxNi[Cr(CN)6]y. J. Mater. Chem. 2006, 16, 2593−2599. (b) Larionova, J.; Folch, B.; Guari, Y.; Molvinger, K.; Luna, C.; Sangregorio, C.; Innocenti, C.; Caneschi, A.; Guerin, C. Synthesis and studies of water-soluble Prussian Blue-type nanoparticles into chitosan beads. Phys. Chem. Chem. Phys. 2010, 12, 12760−12770. (c) Prado, Y.; Lisnard, L.; Heurtaux, D.; Rogez, G.; Gloter, A.; Stephan, O.; Dia, N.; Riviere, E.; Catala, L.; Mallah, T. Tailored coordination nanoparticles: assessing the magnetic single-domain critical size. Chem. Commun. 2011, 47, 1051−1053. (15) (a) Brinzei, D.; Catala, L.; Mathonière, C.; Wernsdorfer, W.; Gloter, A.; Stéphan, O.; Mallah, T. Photoinduced Superparamagnetism in Trimetallic Coordination Nanoparticles. J. Am. Chem. Soc. 2007, 129, 3778−3779. (b) Volatron, F.; Heurtaux, D.; Catala, L.; Mathonière, C.; Gloter, A.; Stéphan, O.; Repetto, D.; ClementeLeón, M.; Coronado, E.; Mallah, T. Photo-induced magnetic bistability in a controlled assembly of anisotropic coordination nanoparticles. Chem. Commun. 2011, 47, 1985−1987. (16) Catala, L.; Brinzei, D.; Prado, Y.; Gloter, A.; Stéphan, O.; Rogez, G.; Mallah, T. Core-multishell magnetic coordination nanoparticles: toward multifunctionality on the nanoscale. Angew. Chem., Int. Ed. 2009, 48, 183−187. (17) Clemente-León, M.; Coronado, E.; López-Muñoz, A.; Repetto, D.; Mingotaud, C.; Brinzei, D.; Catala, L.; Mallah, T. Magnetic Langmuir−Blodgett Films of Bimetallic Coordination Nanoparticles of Cs0.4Ni[Cr(CN)6]0.9. Chem. Mater. 2008, 20, 4642−4652. (18) (a) Song, Y. Y.; Jia, W. Z.; Li, Y.; Xia, X. H.; Wang, Q. J.; Zhao, J. W.; Yan, Y. D. Synthesis and Patterning of Prussian Blue Nanostructures on Silicon Wafer via Galvanic Displacement Reaction. Adv. Funct. Mater. 2007, 17, 2808−2814. (b) Cobo, S.; Molnár, G.; Carcenac, F.; Szilágyi, P. Á .; Salmon, L.; Vieu, C.; Bousseksou, A. Thin Films of Prussian Blue: Sequential Assembly, Patterning and Electron Transport Properties at the Nanometric Scale. J. Nanosci. Nanotechnol. 2010, 10, 5042−5050. (19) (a) Fleury, B.; Volatron, F.; Catala, L.; Brinzei, D.; Rivière, E.; Huc, V.; David, C.; Miserque, F.; Rogez, G.; Baraton, L.; Palacin, S.; Mallah, T. Grafting a Monolayer of Superparamagnetic CyanideBridged Coordination Nanoparticles on Si(100). Inorg. Chem. 2008, 47, 1898−1900. (b) Ghirri, A.; Candini, A.; Evangelisti, M.; Gazzadi, G. C.; Volatron, F.; Fleury, B.; Catala, L.; David, C.; Mallah, T.; Affronte, M. Magnetic imaging of cyanide-bridged co-ordination nanoparticles grafted on FIB-patterned Si substrates. Small 2008, 4, 2240−2246. (20) Horcas, I.; Fernández, R.; Gómez-Rodríguez, J. M.; Colchero, J.; Gómez-Herrero, J.; Baro, A. M. WSXM: a software for scanning probe microscopy and a tool for nanotechnology. Rev. Sci. Instrum. 2007, 78, 013705. (21) Brinzei, D.; Catala, L.; Rogez, G.; Gloter, A.; Mallah, T. Magnetic behaviour of negatively charged nickel(II) hexacyanoferrate(III) coordination nanoparticles. Inorg. Chim. Acta 2008, 361, 3931− 3936.

(22) (a) Clemente-Leon, M.; Agricole, B.; Mingotaud, C.; GómezGarcía, C. J.; Coronado, E.; Delhaes, P. Toward New Organic/ Inorganic Superlattices: Keggin Polyoxometalates in Langmuir and Langmuir−Blodgett Films. Langmuir 1997, 13, 2340−2347. (b) Clemente-León, M.; Coronado, E.; Gómez-García, C. J.; Mingotaud, C.; Delhaès, P.; Ravaine, S.; Romualdo-Torres, G. Polyoxometalate monolayers in langmuir-blodgett films. Chem.Eur. J. 2005, 11, 3979−3987. (23) Wang, Z.; Li, X.; Yang, S. Studies of Dipalmitoylphosphatidylcholine (DPPC) Monolayers Embedded with Endohedral Metallofullerene (Dy@C82). Langmuir 2009, 25, 12968−12973. (24) Raudino, A.; Pignataro, B. Switching Direction of Laterally Ordered Monolayers Induced by Transfer Instability. J. Phys. Chem. B 2007, 111, 9189−9192. (25) Lenhert, S.; Gleiche, M.; Fuchs, H.; Chi, L. F. Mechanism of regular pattern formation in reactive dewetting. ChemPhysChem 2005, 6, 2495−2498. (26) Purrocker, O.; Förtig, A.; Lüdtke, K.; Jordan, R.; Tanaka, M. Confinement of Transmembrane Cell Receptors in Tunable Stripe Micropatterns. J. Am. Chem. Soc. 2005, 127, 1258−1264. (27) Seitz, P. C.; Reif, M.; Yoshikawa, K.; Jordan, R.; Tanaka, M. Dissipative Structure Formation in Lipid/Lipopolymer Monolayers. J. Phys. Chem. B 2011, 115, 2256−2263.

4533


Patterning of Magnetic Bimetallic Coordination Nanoparticles of

Recommend Documents