Enhanced Collective Magnetic Properties in 2D Monolayers of Iron

Jan 25, 2016 - Esterase-Cleavable 2D Assemblies of Magnetic Iron Oxide Nanocubes: Exploiting ... Nanoparticle Superlattices: The Roles of Soft Ligands...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/Langmuir

Enhanced Collective Magnetic Properties in 2D Monolayers of Iron Oxide Nanoparticles Favored by Local Order and Local 1D Shape Anisotropy Delphine Toulemon,† Yu Liu,† Xavier Cattoen̈ ,‡ Cédric Leuvrey,† Sylvie Bégin-Colin,† and Benoit P. Pichon*,† †

Institut de Physique et Chimie des Matériaux de Strasbourg (IPCMS, UMR 7504 UdS/ECPM CNRS), 23 rue du Loess, BP 43, 67037, Strasbourg, France ‡ Institut Néel, CNRS and Univ. Grenoble-Alpes, UPR 2940, 25 rue des Martyrs, 38042 Grenoble, France S Supporting Information *

ABSTRACT: Magnetic nanoparticle arrays represent a very attractive research field because their collective properties can be efficiently modulated as a function of the structure of the assembly. Nevertheless, understanding the way dipolar interactions influence the intrinsic magnetic properties of nanoparticles still remains a great challenge. In this study, we report on the preparation of 2D assemblies of iron oxide nanoparticles as monolayers deposited onto substrates. Assemblies have been prepared by using the Langmuir− Blodgett technique and the SAM assisted assembling technique combined to CuAAC “click” reaction. These techniques afford to control the formation of well-defined monolayers of nanoparticles on large areas. The LB technique controls local ordering of nanoparticles, while adjusting the kinetics of CuAAC “click” reaction strongly affects the spatial arrangement of nanoparticles in monolayers. Fast kinetics favor disordered assemblies while slow kinetics favor the formation of chain-like structures. Such anisotropic assemblies are induced by dipolar interactions between nanoparticles as no magnetic field is applied and no solvent evaporation is performed. The collective magnetic properties of monolayers are studied as a function of average interparticle distance, local order and local shape anisotropy. We demonstrate that local control on spatial arrangement of nanoparticles in monolayers significantly strengthens dipolar interactions which enhances collective properties and results in possible super ferromagnetic order.

■

INTRODUCTION Owing to their unique properties depending on their environment, nanoparticles are of particular interest for the development of the next generation of magnetic devices.1−4 Below a critical size, nanoparticles consist in single nano magnets which can interact together through dipolar interactions.5 Therefore, their intrinsic magnetic properties can be strongly modulated as a function of the structure of their assemblies and differ significantly to those of isolated nanocrystals.6−8 Therefore, the fine modulation of collective properties of assemblies as a function of the spatial arrangement of nanoparticles became very attractive for the development of advanced applications.9,10 In the case of sensors, strong dipolar interactions between nanoparticles enhance the efficiency for nano detection of physical (magnetic field, temperature) or chemical (adsorption of (bio)molecules, pH) external stimuli.9,11,12 In this domain, spinel iron oxide nanoparticles are of particular interest because of their physical properties (relatively high magnetization, magnetic anisotropy) and other advantages (low-cost, abundance, low cytotoxicity, and so forth)13 For instance, tightly packed assemblies of © 2016 American Chemical Society

conductive Fe3‑δO4 nanoparticles have been integrated into magneto-resistive devices and were shown to support a cotunelling charge transfer process which was significantly modulated by a magnetic field.14 Nevertheless, the forward integration of such nanoparticle arrays into technological devices requires at first a better comprehension of the collective properties. Highly structured assemblies of nanoparticles supported onto substrates can be prepared following bottom-up strategies.9,10,15 These techniques are mainly based on the stability of non aggregated nanoparticles in solvents as well as the design of surface functionalities on both substrates and nanoparticles. The Langmuir−Blodgett (LB) technique is certainly one of the most popular way to prepare well-defined assemblies of magnetic nanoparticles as monolayers and multilayers.16−20 The precise control on the dimensionality of assemblies has highlighted the strong influence of 2D shape anisotropy on the collective magnetic properties.17,18,21,22 The orientation of Received: November 10, 2015 Revised: January 22, 2016 Published: January 25, 2016 1621

DOI: 10.1021/acs.langmuir.5b04145 Langmuir 2016, 32, 1621−1628

Article

Langmuir

Figure 1. Schematic illustration of the LB, MW-click, and T-click assembling techniques.

sample preparation induces self-assembly of nanoparticles driven by solvent evaporation.39 However, the spontaneous assembly of nanoparticles through dipolar interactions without applying any external stimuli has never been reported. Herein we report on the study of the collective properties of monolayers of 20 nm sized iron oxide nanoparticles. Monolayers have been prepared by the Langmuir−Blodgett technique and by SAM-assisted assembling combined to “click” chemistry with activation either by external heating or by microwave irradiation (Figure 1). Although all samples consist in well-defined monolayers, the variation of experimental conditions upon the assembling process results in different spatial arrangements of nanoparticles. Such a modulation of the assembly structure strongly affects dipolar interactions thus the collective magnetic properties. This systematic study brings important information on the structure-properties relationship of nanoparticle monolayers.

magnetic moments is favored in-plane which enhances dipolar interactions in comparison to 3D random assemblies such as powders. Furthermore, anisotropic and strong dipolar interactions in monolayer raised up collective properties to superferromagnetism.17 The modulation of the size of ligands grafted at the surface of nanoparticles also enabled us to tune dipolar interactions as a function of the interparticle distance.23 However, multilayer assemblies which involve a larger number of interacting nanoparticles favor stronger dipolar interactions and better stability against external magnetic fields.17,19,20,24 Another approach, the Layer-by-Layer technique was also used to prepare multilayered hierarchical assemblies.10,25 The increase of the interlayer distance favored in-plane dipolar interactions within layers and their antiparallel coupling.26 More recently, unidirectional shape anisotropy has been also reported in pyramidal assemblies of cobalt ferrite nanoparticles.27 Alternatively, chemically assisted assembling based on selfassembled monolayers (SAMs) of organic molecules has become very attractive due to its high versatility to functionalize substrates and to control selectively the assembly of nanoparticles.9,10,28,29 Taking into account that dipolar interactions are long-range interactions, the CuAAC “click” reaction30−32 allowed us preparing monolayers characterized by interparticle distances which could be modulated up to 100 nm and to study isolated nanoparticles.31 A similar strategy involving mixed SAMs has been also reported to control the formation of sizable domains of nanoparticles with tunable dipolar interactions depending on the number of constituting nanoparticles.33 The assembling of nanoparticles into chains has also been reported.34−37 The unidimensionality of these assemblies results in even much stronger dipolar interactions which is commonly ascribed to the colinearity of magnetic moments along the chain. While assembling is usually promoted by an external magnetic field, spontaneous assembly promoted by dipolar interactions under zero field has been rarely reported.38,39 In this case, a minimum dipolar energy, i.e., a critical nanoparticle size is required to overcome thermal fluctuations.40−42 For instance, 21 nm sized Fe3O4 nanoparticles resulted in linear chains and flux-closure rings while smaller sizes resulted in non interacting nanoparticles.38 Nevertheless, one also has to take into account that nanoparticle assemblies are usually studied by TEM which

■

EXPERIMENTAL SECTION

Chemicals. Reactants and solvents were purchased from different companies: tetrahydrofuran (THF), methanol and ethanol (Carlo Erba), iron stearate (Strem Chemicals), oleic acid (99%, Alpha Aesar), triethylamine (99.5%, Fluka), and docosene (99%, TCI). Synthesis and Functionalization of Iron Oxide Nanoparticles. Synthesis of NP@OA. The synthesis of iron oxide nanoparticles coated with oleic acid has been reported previously by our group.43 Briefely, iron stearate (1.38 g, 2.22 mmol) and oleic acid (1.24 g, 4.44 mmol) were added to 20 mL of docosene (bp 364 °C). The mixture was first heated at 100 °C and kept under stirring for 1 h. The reaction medium was subsequently heated under air atmosphere to reflux for 2 h (heating rate of 5 °C min−1). The nanoparticles were then washed several times with a CHCl3/EtOH mixture by performing centrifugation in order to remove docosene, excess in oleic acid and side products. Finally, nanoparticles were obtained as a black suspension after dispersion in CHCl3. Functionalization of NP@N3. Nanoparticles were functionalized by (12-azidododecyl)phosphonic acid (AP12N3) according to the procedure which was reported previously.32 10 mL THF solutions of AP12N3 (15 mg) and NP@OA (16.7 mg) were mixed. The medium was stirred for 15 h at room temperature. Excess of azido phosphonic acid and desorbed oleic acid were removed by washing 3 times with 20 mL of THF by using an ultrafiltration device with a 30 kDa regenerated cellulose membrane (Millipore). NP@N3 were obtained as a stable suspension in THF. Nanoparticle Assembling. Langmuir−Blodgett Monolayer Preparation. Langmuir−Blodgett monolayer has been prepared following a similar procedure as we have already reported.23,44 Fresh 1622

DOI: 10.1021/acs.langmuir.5b04145 Langmuir 2016, 32, 1621−1628

Article

Langmuir

Figure 2. SEM images of monolayers of iron oxide nanoparticles prepared by a) LB technique, b) T-click technique, and c) MW-click technique. Insets are autocorrelograms. silicon wafers were soaked at room temperature in a piranha solution (3:1 H2SO4:H2O2) for 10 min and extensively rinsed with water before drying under a N2 stream. The substrate was dipped partially in the water subphase prepared onto a Langmuir trough (KSV 5000, 576 × 150 mm2). A 5 mg.mL−1 NP@OA suspension in CHCl3 (150−200 μL) was spread onto the water subphase at room temperature. After 10 min stabilization, barriers located at both extremities of the trough were moved (compression rate: 5 mm min−1) to the center to form a nanoparticle monolayer by mechanical compression. The surface pressure−area isotherm was recorded during the film compression using a Wilhelmy plate. The monolayer formed at the air/water interface was transferred onto a silicon wafer after pressure of 30 mN m−1 was reached and stabilization for 10 min. Film transfer occurs by the concomitant pulling out of the silicon wafer from the water subphase at a rate of 1 mm min−1 and by moving the barriers with the aim to maintain the surface pressure at 30 mN m−1. Click Monolayer Preparation. T-click and MW-click monolayers have been prepared following the exact procedures we have already reported in refs 31 and 32, respectively. First, SAM-CC self-assembled monolayers were prepared by dipping freshly cleaned gold substrates (5 × 5 mm2) by H2/O2 plasma into an ethanolic solution of 11mercaptoundec-1-yne (AT11) (10 mM) for 24 h. Then a 5 mL suspension of NP@N3 (5 mg/mL) in a THF/NEt3 mixture (v:v/99:1, 5 mg.mL−1) was prepared, and the CuBr(PPh3)3 catalyst (6.5 mg, 6.7.10−3 mmol) was added. SAM-CC was dipped in the suspension. Tclick CuAAC reaction was performed by heating the reaction medium to reflux for 48 h. MW-click CuAAC reaction was performed under exposure to microwave irradiation for 1 h using a microwave device (Anton-Paar Monowave 300 working at Tmax 120 °C, 50 W, and 2.45 GHz) in a 10 mL closed vial. Characterization Techniques. Nanoparticle assemblies were characterized by using a Scanning Electron Microscope (SEM) JEOL 6700F equipped with a field emission gun (FEG) and operating at 3 kV. Image analysis of SEM micrographs was performed by using Digital micrograph (Gatan) and ImageJ softwares following the method we have reported earlier.17 Magnetic properties have been characterized by using a Superconducting Quantum Interfererence Device (SQUID) magnetometer (Quantum Design MPMS SQUIDVSM dc magnetometer). Measurements were performed on 5 × 5 mm2 substrates by applying an in-plane magnetic field. Temperature magnetization curves (M(T)) were recorded by applying a magnetic field of 75 G upon heating from 5 to 400 K after cooling down without applying a magnetic field (zero field cooled (ZFC) curve) and after cooling down under a magnetic field of 75 G (field cooled (FC) curve). A demagnetization procedure was performed before recording M(H) curves. A degauss field was applied at 400 K before cooling down. Magnetization curves were recorded at 300 and 5 K by applying a magnetic field from +7 T to −7 T and further from −7 T to +7 T.

shape with facets and a narrow size distribution centered around 20 ± 1.8 nm. Well-defined shape, narrow size distribution, and no aggregation are essential requirements to control precisely the intrinsic magnetic properties of individual nanoparticles thus to study their collective properties as a function of the structure of assemblies. Oleic acid (OA) being used as stabilizing agent during the synthesis, it remains adsorbed at the nanoparticle surface and allows preparing highly stable suspensions of NP@OA in chloroform, ready to use for the LB technique. In a second step, oleic acid was replaced by the (12-azidododecyl)phosphonic acid (AP12) ligands in order to prepare azide terminated nanoparticles (NP@N3) by strictly following the procedure we have also reported recently.31 Ligand exchange is driven by the stronger interaction with the iron oxide surface of the phosphonic acid groups of AP12 compared with the carboxylic acid groups of OA. A highly stable suspension of NP@N3 was also obtained in THF/triethylamine mixture (v:v/99:1) which corresponds to the reaction medium for “click” reactions. The amount of triethylamine was reduced in comparison to our previous study on 10 nm sized nanoparticles31 because of the lower colloidal stability of 20 nm sized nanoparticles, certainly due to their higher magnetization. 2D monolayers of iron oxide nanoparticles have been prepared following two different approaches (Figure 1). First, NP@OA have been assembled by the Langmuir−Blodgett technique.23,44 A suspension of NP@OA in chloroform was spread onto a water subphase in order to generate a monolayer at the air/water interface. Two barriers were moved from the extremities of the trough to the center in order to reduce the available surface area and to get particles tightly packed. This film was directly transferred onto a silicon wafer by pulling it out from the water subphase. SEM images show the homogeneous coverage of the substrate by nanoparticles (Figure 2a). The density of 2 200 ± 30 NP/μm2 is very close to the maximum theoretical value corresponding to hexagonal close packing (2 475 NP/μm2). It represents 91% of the maximum coverage which agrees with a dense and continuous monolayer of tightly packed nanoparticles. Second, NP@N3 have been assembled by performing copper-catalyzed azide−alkyne cycloaddition (CuAAC) “click” reactions onto an alkyne-terminated substrate (SAM@CC). A self-assembled monolayer (SAM) of 11-mercaptoundec-1-yne (AT11) molecules was prepared on a gold substrate (SAM@ CC).31,32 The self-assembly of molecules driven by van der Waals interactions between undecyl chains upon interaction of thiol groups with gold surface results in alkyne groups being readily available at the SAM surface to react with azide groups located at the surface of nanoparticles. The assembly of

■

RESULTS AND DISCUSSION Assemblies of Nanoparticles. Single crystal iron oxide nanoparticles were synthesized by thermal decomposition of iron stearate in docosene (bp 364 °C) as we have reported earlier.43,45 Nanoparticles feature a uniform, almost spherical 1623

DOI: 10.1021/acs.langmuir.5b04145 Langmuir 2016, 32, 1621−1628

Article

Langmuir

Brownian motion increases the kinetics of the reaction. Moreover, the high ability of iron oxide to absorb microwave irradiation combined with the high reflectance of the gold substrate favors the local increase in temperature at the nanoparticule/substrate interface. Therefore, nanoparticles assemble very fast onto the SAM-CC in a random fashion. In contrast, under the T-click conditions nanoparticles assemble much slower in chain-like structures which suggests a different pathway. The formation of chains can be assimilated to a nucleation/growth process. Pristine nanoparticles are assembled randomly onto the SAM-CC surface with large interparticle distances as shown by SEM after 1 h of reaction (Figure S1), which means they can be considered not to be interacting with each other. Nevertheless, these pristine nanoparticles induce local magnetic fields and attract nanoparticles in suspension which are in their vicinity. Dimers can thus be formed, which are expected to induce sufficiently high integrative dipole moment to direct spontaneously and simultaneously the growth of linear assemblies as reported for gold nanoparticles.48,49 Such a mechanism is possible because the T-CuAAC reaction is slow, contributing to freeze the assemblies formed on the gold surface. Although linear chains grow in different directions because of the isotropic shape of the nanoparticles and of the absence of external magnetic field, the locally higher integrative dipole moment favors the growth of longer chains. Nevertheless, the increase in nanoparticle density limits the growth of linear assemblies or provokes deviation from linearity. The non strict linearity may also be ascribed to strains resulting from the faceting of nanoparticles and misalignment of easy magnetization axis.50 In contrast to previous studies which reported on the formation of chains of magnetic nanoparticles in solution,40,42 these observations suggest that the formation of linear chains is directed by the dipolar interactions at the SAM surface, followed by click reaction. Furthermore, the spontaneous formation of chains in solution (no magnetic field) reported in other studies51,52 can also be ruled out since granulometry measurements do not show any hydrodynamic diameter larger than the nanoparticle diameter measured from TEM (not shown). Interestingly, although such 20 nm sized Fe3‑δO4 nanoparticles are superparamagnetic under the reaction conditions (T = 66 °C),53 magnetostatic forces are sufficient to induce linear assemblies though under zero field.38,40−42 Such a behavior is usually correlated to a high dipolar parameter λ (>3) which corresponds to the ratio of magnetic dipolar to thermal energies. It is worth noting that smaller Fe3‑δO4 nanoparticles of 10 nm led to the formation of shorter chains of 3−4 nanoparticles although it has not been mentioned in our previous work.31 This observation is contrary to previous studies38 and may be ascribed to the irreversible immobilization of nanoparticles onto SAM which reduces Brownian motion contrarily to what happens in solution. Indeed substrate affinity has been demonstrated to be of primary importance to favor the formation of linear assemblies.39 It is also well-known that van der Waals interactions also promote the assembling in the case of nanoparticles featuring low magnetic anisotropy.42 Therefore, the spontaneous formation of chains without applying an external magnetic field is directly dependent on the size of nanoparticles as reported earlier for superparamagnetic iron nanoparticles.38,40 A higher magnetization of nanoparticles (82 emu/g for 20 nm vs 61 emu/g for 11 nm)43 favors higher dipolar energy54 and a higher dipolar

nanoparticles was performed very simply by dipping the SAMCC into the NP@N3 suspension in a THF/triethylamine mixture. The cycloaddition reaction between azide groups and alkyne groups was catalyzed by CuBr(PPh3)3 and resulted in the irreversible formation of covalent linkages (triazole groups). It was thermally activated in two different ways: (i) classical heating of the external medium up to reflux of the solvent mixture (T-click)31 or (ii) microwave irradiation which fastens dramatically the reaction kinetic by taking advantage of the dielectric properties of solvents and reactants (MW-click).32 Experimental conditions were strictly identical with the exception of the heating mode and duration, and resulted in the homogeneous coverage of SAM-CC by NP@N3 (Figure 2b,c). However, it is noteworthy that monolayer formation reaches maximal conversion after a very long reaction time up to 48 h in the case of T-click conditions. By contrast, the reaction time is dramatically shortened down to 1 h in the case of MW-click conditions. Both T-click and MW-click monolayers display similar densities in nanoparticles of 1 400 ± 30 NP/μm2 and 1 470 ± 60 NP/μm2, respectively. Although maximum conversion is reached, these density values are much lower than the one calculated for the LB monolayer. This feature can be understood as a matter of the assembling process. The LB technique proceeds through the compression of nanoparticles at the air/water interface which favors tightly packed nanoparticles. In contrast, the “click” chemistry approach is based on Brownian motion and on the probability that individual nanoparticles will reach uncovered areas of SAMs. During the reaction, free surface area decreases and becomes small enough to enable the assembly of nanoparticles within reasonable reaction time. It results in uncovered areas which represents significant proportions up to 40% in comparison to the LB monolayer (9%). The spatial distribution of nanoparticles was also carefully studied by image processing.44,46 Autocorrelation spectra were calculated, and the corresponding radial profiles were fitted to calculate coherence lengths ζ. The LB monolayer exhibits a ζ value of 90 ± 10 nm which agrees with local order. It also reveals an interparticle distance of 1−2 nm which corresponds to oleic acid being adsorbed at the nanoparticle surface as a single layer. In contrast, such image analysis was not possible to perform for the MW-click monolayer which is highly disordered (Figure 2c). Nevertheless, nanoparticles seem to be assembled in domains of 4−6 entities. In contrast, the Tclick monolayer clearly displays relatively linear assemblies of nanoparticles with different orientations (Figure 2b). Chains are usually constituted of 6 to 8 nanoparticles separated by interparticle distances of about 1−2 nm which is correlated to tight packing of nanoparticles as previously mentioned for the LB monolayer. While the tight structure of the LB monolayer can be understood as resulting from the mechanical compression which brings particles closer, the different structure of MWclick and T-click monolayers can be explained by the different kinetics of the click reaction between azide and alkyne groups. In the case of MW-click, the CuAAC click reaction is dramatically accelerated by microwave dielectric heating.47 The reaction medium undergoes superheating up to 120 °C within few minutes although this temperature is much higher than the boiling temperature of THF (Teb 66 °C) in the case of T-click.32 The assembly process is kinetically controlled by the reaction between azide and alkyne groups at the nanoparticles and substrate surfaces, respectively. Therefore, favoring the 1624

DOI: 10.1021/acs.langmuir.5b04145 Langmuir 2016, 32, 1621−1628

Article

Langmuir

Figure 3. Magnetic measurements performed between +2 T and −2 T on monolayers prepared by LB, T-click, and MW-click techniques. Magnetization curves recorded against an applied magnetic field a) at 300 K and b) at 5 K.

The coercive field is directly related to the energy barrier required for aligning magnetic moments upon reversal of the magnetic field.7 Magnetization curves clearly show it is strongly affected by dipolar interactions between nanoparticles. The lowest HC value is observed for the MW-click monolayer which is correlated to disorder and low nanoparticle density, i.e., larger interparticle distance, in comparison to the LB monolayer. Therefore, the lowest HC in MW-click monolayer is correlated to lowest dipolar interactions.55 Nevertheless, the monolayer is constituted of domains of tightly packed nanoparticles, which number is not sufficient to enhance in-plane dipolar interactions in comparison to the LB monolayer.28 In contrast, the higher density combined to local ordering in the LB monolayer results in stronger dipolar interactions than in disordered MW-click monolayer. 2D shape anisotropy favors the preferential orientation of magnetic moments in the plane of the monolayer.17,18,56 The highest HC value is exhibited by the T-click monolayer despite being characterized by a similar nanoparticle density to the MW-click monolayer. It clearly shows that the formation of chains of nanoparticles results in 1D local shape anisotropy which significantly enhances dipolar interactions in comparison to 2D local order and shape anisotropy.57,58 Lowering locally the dimensionality of the assembly favors dipolar interactions with the closest neighbors and reduces magnetic frustrations although the chains constitute a 2D monolayer. Furthermore, these results show unambiguously that the anisotropy of assemblies predominates over the average interparticle distance. In contrast to HC, all three samples exhibit the same MR/MS ratio (0.47). The MR/MS ratio has been reported in several studies to be strongly influenced by the interparticle distance55,59 and dimensionality of the assembly.17,57,58 In this study, all assemblies being assimilated to monolayers, the 2D

parameter (λ ≈ 7 for 20 nm and 0.3 for 10 nm).38 Therefore, longer chains may be obtained with larger sizes of nanoparticles. Nevertheless, the too high increase in dipolar interactions will result in strong aggregation of nanoparticles in suspension. Furthermore, the spontaneous assembling of nanoparticles under zero field usually results either in chain like or ring-flux closure.41 The fact that the latter is not observed in our case shows that our approach based on “click” chemsitry is selective. Collective Magnetic Properties. Such differences in the local spatial arrangement of nanoparticles within monolayers have a significant effect on their collective magnetic properties. Monolayers prepared by the LB, T-click, and MW-click techniques have been studied by performing SQUID magnetometry. Magnetization curves recorded against a magnetic field (M(H)) at 300 K show perfect overlap of curves recorded from +7 T to −7 T and from −7 T to +7 T which corresponds to superparamagnetic behavior (Figure 3). In contrast, M(H) curves recorded at 5 K exhibit hysteresis characterized by their respective coercive field (HC) and MR/MS ratio which agree with a ferrimagnetic state (Table 1). Table 1. Magnetic Characteristics of 20 nm Sized Nanoparticles Assembled as LB, T-Click, MW-Click Monolayers and in the Powder State monolayer

MR/MS

HC (Oe)

TMAX (K)

LB T-click MW-click powder

0.47 0.47 0.47 0.35

347 434 325 309

301 and 361 >400 282 250

Figure 4. Magnetization curves recorded for the T-click monolayer a) at 5 K and b) at 300 K by applying a magnetic field in the plane of the assembly and in the perpendicular direction. 1625

DOI: 10.1021/acs.langmuir.5b04145 Langmuir 2016, 32, 1621−1628

Article

Langmuir

higher temperatures than the latter. This curve is very similar to the one we have reported earlier for a LB monolayer of 16 nm sized nanoparticles.17 The contribution at the lowest temperature (301 K) is correlated to the blocking temperature of the assembly. The second one (361 K) may correspond to superferromagnetism which results from larger dipolar energy than magnetic anisotropy energy.5,17,61,62 Both values being larger than those reported for 16 nm sized nanoparticles (215 and 350 K), it agrees with higher dipolar energy vs. magnetic anisotropy energy. Therefore, local ordering combined to shorter interparticle interactions in the LB monolayer enhances significantly dipolar interactions in comparison to the MW-click monolayer. While LB and MW-click monolayers show clearly TMAX below 400 K, the T-click monolayer does not exhibit any maximum in this temperature range, which means that TMAX is shifted to a temperature higher than 400 K. This result shows that the magnetic anisotropy of nanoparticles is much more stable against thermal fluctuations in the T-click monolayer. Such a behavior is ascribed to stronger dipolar interactions which are clearly favored by the local anisotropy induced by the formation of chains as observed by Gross et al. for cobalt nanoparticles.58 Local lowering of the dimensionality of the assembly reduces magnetic frustrations and superspin glass state.63 Dipolar energy is significantly enhanced against magnetic anisotropy and may be sufficient to overcome superspinglass behavior. Superspins are correlated in a ferromagnetic-like fashion, the so-called super ferromagnetism. Finally, field cooled (FC) curves exhibit relatively constant magnetization below TMAX confirming the strong dipolar interactions within all monolayers.

dimensionality seems to predominate over the variation of local order and shape anisotropy. Demagnetizing interactions are not sufficiently modulated to induce the variation of the MR/MS ratio. Furthermore, MR/MS and HC for all three monolayers are clearly higher than the ones recorded for nanoparticles in the powder state (HC = 309 Oe and MR/MS = 0.35) (Figure S3). The assembling of nanoparticles in 2D monolayers has a significant effect on the magnetic properties in comparison to 3D random assembly in the powder state.17 Shape anisotropy induced by bi-dimensionility has been investigated for the Tclick monolayer by recording magnetization curves by applying a magnetic field in the monolayer plane and in the direction perpendicular (Figure 4). The largest hysteresis cycle in the case of an in-plane field confirms the effect of shape anisotropy which favors stronger coupling of magnetic moments and lowers demagnetizing interactions within the plane of the monolayer. Such a behavior is confirmed by M(H) curves recorded at 300 K, which display a faster increase of magnetization when an in-plane magnetic field was applied. Magnetic moments of nanoparticles coupled in-plane require less energy to align in the direction of the magnetic field. Temperature dependent magnetization curves are also significantly modulated by the structure of monolayers (Figure 5). Zero field cooled (ZFC) curves were recorded from 5 to

■

CONCLUSIONS Monolayers of nanoparticles have been prepared by different approaches: the Langmuir−Blodgett technique and click chemistry activated by conventional heating or microwaves. The assembling technique has a strong influence on the structure of the resulting nanoparticle assemblies and on the collective magnetic properties. The LB technique favors local ordering of nanoparticles upon compression of nanoparticles monolayer before transfer onto a substrate whereas the “click” chemistry approach favors random assembling which can be modulated by the participation of dipolar interactions. In the case of “click” chemistry approaches, the competition between dipolar and thermal energies can be easily tuned by changing the assembling temperature (ca. 70 °C for T-click vs locally ≫120 °C for MW-click). Disorder in the MW-click monolayer alters the collective properties and demonstrates that shape anisotropy resulting from 2D monolayer structuration is not sufficient and has to be combined to local order to increase dipolar interactions energy. In contrast, local ordering in the LB monolayers enhances shape anisotropy which contributes to the marked enhancement of dipolar energy resulting in superferromagnetism. The strongest effect arises from T-click monolayer which exhibits local order and enhanced shape anisotropy due to the formation of chain-like structures. Under the experimental conditions, chains are formed spontaneously through dipolar interactions, no magnetic field being applied. The T-click monolayer may also exhibit superferromagnetism with blocking and Curie temperatures clearly higher than 400 K. The relative linearity of chains increases in-plane dipolar interactions in selective directions within the monolayer. One may notice that such a behavior is observable from chains

Figure 5. Temperature dependent magnetization (FC and ZFC) curves of MW-click, T-click, and LB films.

400 K by applying a magnetic field of 75 Oe after cooling down under zero field. Iron oxide nanoparticles being superparamagnetic at room temperature below a size of 25 nm,60 ZFC curves usually exhibit a maximum of magnetization at a critical temperature (TMAX) which is usually assimilated to the blocking temperature (TB) at which the transition between ferrimagnetic and superparamagnetic states occurs. Therefore, TMAX is directly influenced by dipolar interactions which control the balance between thermal energy and the magnetic anisotropy barrier of nanoparticles.7 First of all, nanoparticles in the powder state display the lowest TMAX value (Table 1), which is correlated to weaker dipolar interactions resulting from 3D random assembly. Among the 2D assemblies reported here, the MW-monolayer exhibits the lowest TMAX value which is correlated to the weakest dipolar interactions as a result of the largest average interparticle distances and smallest shape anisotropy owing to smaller nanoparticle domains.23,31,55 Although magnetization increases at similar temperatures for LB and MW-click monolayers, one may notice that the ZFC curve of the LB monolayer displays two contributions, both at 1626

DOI: 10.1021/acs.langmuir.5b04145 Langmuir 2016, 32, 1621−1628

Article

Langmuir

(12) Jeong, U.; Teng, X.; Wang, Y.; Yang, H.; Xia, Y. Superparamagnetic Colloids: Controlled Synthesis and Niche Applications. Adv. Mater. 2007, 19 (1), 33−60. (13) Tartaj, P.; Morales, M. P.; Gonzalez-Carreño, T.; VeintemillasVerdaguer, S.; Serna, C. J. The Iron Oxides Strike Back: From Biomedical Applications to Energy Storage Devices and Photoelectrochemical Water Splitting. Adv. Mater. 2011, 23 (44), 5243− 5249. (14) Pauly, M.; Dayen, J.-F.; Golubev, D.; Beaufrand, J.-B.; Pichon, B. P.; Doudin, B.; Bégin-Colin, S. Co-tunneling Enhancement of the Electrical Response of Nanoparticle Networks. Small 2012, 8 (1), 108−115. (15) Neouze, M.-A. Nanoparticle assemblies: main synthesis pathways and brief overview on some important applications. J. Mater. Sci. 2013, 48 (21), 7321−7349. (16) Acharya, S.; Hill, J. P.; Ariga, K. Soft Langmuir−Blodgett Technique for Hard Nanomaterials. Adv. Mater. 2009, 21 (29), 2959− 2981. (17) Pauly, M.; Pichon, B. P.; Panissod, P.; Fleutot, S.; Rodriguez, P.; Drillon, M.; Begin-Colin, S. Size dependent dipolar interactions in iron oxide nanoparticle monolayer and multilayer Langmuir-Blodgett films. J. Mater. Chem. 2012, 22 (13), 6343−6350. (18) Poddar, P.; Telem-Shafir, T.; Fried, T.; Markovich, G. Dipolar interactions in two- and three-dimensional magnetic nanoparticle arrays. Phys. Rev. B: Condens. Matter Mater. Phys. 2002, 66 (6), 060403. (19) Zhang, H.; Bao, N.; Yuan, D.; Ding, J. Nanomagnetism study of highly-ordered iron oxide nanocrystal assemblies fabricated by the Langmuir-Blodgett technique. Phys. Chem. Chem. Phys. 2013, 15 (35), 14689−14695. (20) Braun, K. F.; Sievers, S.; Albrecht, M.; Siegner, U.; Landfester, K.; Holzapfel, V. Stability of the magnetic domain structure of nanoparticle thin films against external fields. J. Magn. Magn. Mater. 2009, 321 (22), 3719−3725. (21) Faure, B.; Wetterskog, E.; Gunnarsson, K.; Josten, E.; Hermann, R. P.; Bruckel, T.; Andreasen, J. W.; Meneau, F.; Meyer, M.; Lyubartsev, A.; Bergstrom, L.; Salazar-Alvarez, G.; Svedlindh, P. 2D to 3D crossover of the magnetic properties in ordered arrays of iron oxide nanocrystals. Nanoscale 2013, 5 (3), 953−960. (22) Margaris, G.; Trohidou, K.; Kachkachi, H. Surface effects on the magnetic behavior of nanoparticle assemblies. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 85 (2), 024419. (23) Fleutot, S.; Nealon, G. L.; Pauly, M.; Pichon, B. P.; Leuvrey, C.; Drillon, M.; Gallani, J.-L.; Guillon, D.; Donnio, B.; Begin-Colin, S. Spacing-dependent dipolar interactions in dendronized magnetic iron oxide nanoparticle 2D arrays and powders. Nanoscale 2013, 5 (4), 1507−1516. (24) Pichon, B. P.; Leuvrey, C.; Ihawakrim, D.; Bernard, P.; Schmerber, G.; Begin-Colin, S. Magnetic Properties of Mono- and Multilayer Assemblies of Iron Oxide Nanoparticles Promoted by SAMs. J. Phys. Chem. C 2014, 118 (7), 3828−3837. (25) Decher, G. Fuzzy Nanoassemblies: Toward Layered Polymeric Multicomposites. Science 1997, 277 (5330), 1232−1237. (26) Pichon, B. P.; Louet, P.; Felix, O.; Drillon, M.; Begin-Colin, S.; Decher, G. Magnetotunable Hybrid Films of Stratified Iron Oxide Nanoparticles Assembled by the Layer-by-Layer Technique. Chem. Mater. 2011, 23 (16), 3668−3675. (27) Wen, T.; Zhang, D.; Wen, Q.; Zhang, H.; Liao, Y.; Li, Q.; Yang, Q.; Bai, F.; Zhong, Z. Magnetic nanoparticle assembly arrays prepared by hierarchical self-assembly on a patterned surface. Nanoscale 2015, 7 (11), 4906−4911. (28) Pichon, B. P.; Demortiere, A.; Pauly, M.; Mougin, K.; Derory, A.; Begin-Colin, S. 2D Assembling of Magnetic Iron Oxide Nanoparticles Promoted by SAMs Used as Well-Addressed Surfaces. J. Phys. Chem. C 2010, 114 (19), 9041−9048. (29) Guardingo, M.; Bellido, E.; Miralles-Llumà, R.; Faraudo, J.; Sedó, J.; Tatay, S.; Verdaguer, A.; Busqué, F.; Ruiz-Molina, D. Bioinspired Catechol-Terminated Self-Assembled Monolayers with Enhanced Adhesion Properties. Small 2014, 10 (8), 1594−1602.

constituted of a very limited number of nanoparticles and with random orientations. Therefore, longer chains aligned in the same direction are expected to increase the strength of collective properties. We are currently working on this aspect by improving the collective anisotropy of chains within the monolayer. Finally, these results on the structure-properties relationship of magnetic nanoparticles assemblies are very interesting with regard to potential applications such as magnetic and magnetoresitive sensors.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.5b04145. SEM image and magnetic characterizations of monolayer samples (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Tel: 0033 (0)3 88 10 71 33. Fax: 0033 (0)3 88 10 72 47. Email: [email protected] (B.P.P.). Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We are grateful to the Direction Générale de l’Armement (DGA) and Région Alsace for financial support. REFERENCES

(1) Cheong, S.; Kim, Y.; Kwon, T.; Kim, B. J.; Cho, J. Inorganic nanoparticle multilayers using photo-crosslinking layer-by-layer assembly and their applications in nonvolatile memory devices. Nanoscale 2013, 5 (24), 12356−12364. (2) Sun, S.; Murray, C. B.; Weller, D.; Folks, L.; Moser, A. Monodisperse FePt Nanoparticles and Ferromagnetic FePt Nanocrystal Superlattices. Science 2000, 287 (5460), 1989−1992. (3) Gutfleisch, O.; Willard, M. A.; Brück, E.; Chen, C. H.; Sankar, S. G.; Liu, J. P. Magnetic Materials and Devices for the 21st Century: Stronger, Lighter, and More Energy Efficient. Adv. Mater. 2011, 23 (7), 821−842. (4) Terris, B. D.; Thomson, T. Nanofabricated and self-assembled magnetic structures as data storage media. J. Phys. D: Appl. Phys. 2005, 38 (12), R199−R222. (5) Bedanta, S.; Kleemann, W. Supermagnetism. J. Phys. D: Appl. Phys. 2009, 42 (1), 013001. (6) Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Synthesis and characterization of monodisperse nanocrystals and close-packed nanocrystal assemblies. Annu. Rev. Mater. Sci. 2000, 30 (1), 545−610. (7) Majetich, S. A.; Sachan, M. Magnetostatic interactions in magnetic nanoparticle assemblies: energy, time and length scales. J. Phys. D: Appl. Phys. 2006, 39 (21), R407−R422. (8) Vanmaekelbergh, D. Self-assembly of colloidal nanocrystals as route to novel classes of nanostructured materials. Nano Today 2011, 6 (4), 419−437. (9) Bellido, E.; Domingo, N.; Ojea-Jiménez, I.; Ruiz-Molina, D. Structuration and Integration of Magnetic Nanoparticles on Surfaces and Devices. Small 2012, 8 (10), 1465−1491. (10) Kinge, S.; Crego-Calama, M.; Reinhoudt, D. N. Self-Assembling Nanoparticles at Surfaces and Interfaces. ChemPhysChem 2008, 9 (1), 20−42. (11) Latham, A. H.; Williams, M. E. Controlling Transport and Chemical Functionality of Magnetic Nanoparticles. Acc. Chem. Res. 2008, 41 (3), 411−420. 1627

DOI: 10.1021/acs.langmuir.5b04145 Langmuir 2016, 32, 1621−1628

Article

Langmuir (30) Kinge, S.; Gang, T.; Naber, W. J. M.; van der Wiel, W. G.; Reinhoudt, D. N. Magnetic Nanoparticle Assembly on Surfaces Using Click Chemistry. Langmuir 2011, 27 (2), 570−574. (31) Toulemon, D.; Pichon, B. P.; Cattoen, X.; Wong Chi Man, M.; Begin-Colin, S. 2D assembly of non-interacting magnetic iron oxide nanoparticles via ″click″ chemistry. Chem. Commun. 2011, 47 (43), 11954−11956. (32) Toulemon, D.; Pichon, B. P.; Leuvrey, C.; Zafeiratos, S.; Papaefthimiou, V.; Cattoen, X.; Begin-Colin, S. Fast Assembling of Magnetic Iron Oxide Nanoparticles by Microwave-Assisted Copper(I) Catalyzed Alkyne-Azide Cycloaddition (CuAAC). Chem. Mater. 2013, 25 (14), 2849−2854. (33) Pichon, B. P.; Pauly, M.; Marie, P.; Leuvrey, C.; Begin-Colin, S. Tunable Magnetic Properties of Nanoparticles 2D Assemblies Addressed by Mixed SAMs. Langmuir 2011, 27, 6235−6243. (34) Singamaneni, S.; Bliznyuk, V. N.; Binek, C.; Tsymbal, E. Y. Magnetic nanoparticles: recent advances in synthesis, self-assembly and applications. J. Mater. Chem. 2011, 21 (42), 16819−16819. (35) Tang, Z.; Kotov, N. A. One-Dimensional Assemblies of Nanoparticles: Preparation, Properties, and Promise. Adv. Mater. 2005, 17 (8), 951−962. (36) Wang, H. U. I.; Yu, Y.; Sun, Y.; Chen, Q. Magnetic nanochains: A review. Nano 2011, 06 (01), 1−17. (37) Kitching, H.; Shiers, M. J.; Kenyon, A. J.; Parkin, I. P. Selfassembly of metallic nanoparticles into one dimensional arrays. J. Mater. Chem. A 2013, 1 (24), 6985−699. (38) Klokkenburg, M.; Vonk, C.; Claesson, E. M.; Meeldijk, J. D.; Erne, B. H.; Philipse, A. P. Direct Imaging of Zero-Field Dipolar Structures in Colloidal Dispersions of Synthetic Magnetite. J. Am. Chem. Soc. 2004, 126 (51), 16706−16707. (39) Varon, M.; Pena, L.; Balcells, L.; Skumryev, V.; Martinez, B.; Puntes, V. Dipolar Driven Spontaneous Self Assembly of Superparamagnetic Co Nanoparticles into Micrometric Rice-Grain like Structures. Langmuir 2010, 26 (1), 109−116. (40) Butter, K.; Bomans, P. H. H.; Frederik, P. M.; Vroege, G. J.; Philipse, A. P. Direct observation of dipolar chains in iron ferrofluids by cryogenic electron microscopy. Nat. Mater. 2003, 2 (2), 88−91. (41) Chantrell, R. W.; Bradbury, A.; Popplewell, J.; Charles, S. W. Particle cluster configuration in magnetic fluids. J. Phys. D: Appl. Phys. 1980, 13 (7), L119−22. (42) Lalatonne, Y.; Richardi, J.; Pileni, M. P. Van der Waals versus dipolar forces controlling mesoscopic organizations of magnetic nanocrystals. Nat. Mater. 2004, 3 (2), 121−125. (43) Baaziz, W.; Pichon, B. P.; Fleutot, S.; Liu, Y.; Lefevre, C.; Greneche, J.-M.; Toumi, M.; Mhiri, T.; Begin-Colin, S. Magnetic Iron Oxide Nanoparticles: Reproducible Tuning of the Size and NanosizedDependent Composition, Defects, and Spin Canting. J. Phys. Chem. C 2014, 118 (7), 3795−3810. (44) Pauly, M.; Pichon, B. P.; Albouy, P.-A.; Fleutot, S.; Leuvrey, C.; Trassin, M.; Gallani, J.-L.; Begin-Colin, S. Monolayer and multilayer assemblies of spherically and cubic-shaped iron oxide nanoparticles. J. Mater. Chem. 2011, 21 (40), 16018−16027. (45) Demortiere, A.; Panissod, P.; Pichon, B. P.; Pourroy, G.; Guillon, D.; Donnio, B.; Begin-Colin, S. Size-dependent properties of magnetic iron oxide nanocrystals. Nanoscale 2011, 3 (1), 225−232. (46) Wu, C.-K.; Hultman, K. L.; O’Brien, S.; Koberstein, J. T. Functional Oligomers for the Control and Fixation of Spatial Organization in Nanoparticle Assemblies. J. Am. Chem. Soc. 2008, 130 (11), 3516−3520. (47) Kappe, O. C. Microwave dielectric heating in synthetic organic chemistry. Chem. Soc. Rev. 2008, 37 (6), 1127−1139. (48) Shipway, A. N.; Lahav, M.; Gabai, R.; Willner, I. Investigations into the Electrostatically Induced Aggregation of Au Nanoparticles. Langmuir 2000, 16 (23), 8789−8795. (49) Liao, J.; Zhang, Y.; Yu, W.; Xu, L.; Ge, C.; Liu, J.; Gu, N. Linear aggregation of gold nanoparticles in ethanol. Colloids Surf., A 2003, 223, 177−183. (50) Varon, M.; Beleggia, M.; Kasama, T.; Harrison, R. J.; DuninBorkowski, R. E.; Puntes, V. F.; Frandsen, C. Dipolar Magnetism in

Ordered and Disordered Low-Dimensional Nanoparticle Assemblies. Sci. Rep. 2013, 3.10.1038/srep01234 (51) Gao, M.-R.; Zhang, S.-R.; Jiang, J.; Zheng, Y.-R.; Tao, D.-Q.; Yu, S.-H. One-pot synthesis of hierarchical magnetite nanochain assemblies with complex building units and their application for water treatment. J. Mater. Chem. 2011, 21 (42), 16888−16892. (52) Zhang, F.; Wang, C.-C. Fabrication of One-Dimensional Iron Oxide/Silica Nanostructures with High Magnetic Sensitivity by Dipole-Directed Self-Assembly. J. Phys. Chem. C 2008, 112 (39), 15151−15156. (53) Baaziz, W.; Liu, X.; Florea, I.; Begin-Colin, S.; Pichon, B. P.; Ulhaq, C.; Ersen, O.; Soria-Sanchez, M.; Zafeiratos, S.; Janowska, I.; Begin, D.; Pham-Huu, C. Carbon nanotube channels selectively filled with monodispersed Fe3‑xO4 nanoparticles. J. Mater. Chem. A 2013, 1 (44), 13853−13861. (54) Panissod, P.; Drillon, M. Magnetic Ordering due to Dipolar Interaction in Low Dimensional Materials. In Magnetism: Molecules to Materials IV; Miller, J. S., Drillon, M., Eds.; Wiley VCH: Weinheim, 2003; pp 233−270. (55) Frankamp, B. L.; Boal, A. K.; Tuominen, M. T.; Rotello, V. M. Direct Control of the Magnetic Interaction between Iron Oxide Nanoparticles through Dendrimer-Mediated Self-Assembly. J. Am. Chem. Soc. 2005, 127 (27), 9731−9735. (56) Petit, C.; Russier, V.; Pileni, M. P. Effect of the Structure of Cobalt Nanocrystal Organization on the Collective Magnetic Properties. J. Phys. Chem. B 2003, 107 (38), 10333−10336. (57) Nakata, K.; Hu, Y.; Uzun, O.; Bakr, O.; Stellacci, F. Chains of Superparamagnetic Nanoparticles. Adv. Mater. 2008, 20 (22), 4294− 4299. (58) Gross, A. F.; Diehl, M. R.; Beverly, K. C.; Richman, E. K.; Tolbert, S. H. Controlling Magnetic Coupling between Cobalt Nanoparticles through Nanoscale Confinement in Hexagonal Mesoporous Silica. J. Phys. Chem. B 2003, 107 (23), 5475−5482. (59) Held, G. A.; Grinstein, G.; Doyle, H.; Sun, S.; Murray, C. B. Competing interactions in dispersions of superparamagnetic nanoparticles. Phys. Rev. B: Condens. Matter Mater. Phys. 2001, 64 (1), 012408. (60) Krishnan, K. M.; Pakhomov, A. B.; Bao, Y.; Blomqvist, P.; Chun, Y.; Gonzales, M.; Griffin, K.; Ji, X.; Roberts, B. K. Nanomagnetism and spin electronics: materials, microstructure and novel properties. J. Mater. Sci. 2006, 41 (3), 793−815. (61) Bedanta, S.; Eimuller, T.; Kleemann, W.; Rhensius, J.; Stromberg, F.; Amaladass, E.; Cardoso, S.; Freitas, P. P. Overcoming the Dipolar Disorder in Dense CoFe Nanoparticle Ensembles: Superferromagnetism. Phys. Rev. Lett. 2007, 98 (17), 176601. (62) Yamamoto, K.; Hogg, C. R.; Yamamuro, S.; Hirayama, T.; Majetich, S. A. Dipolar ferromagnetic phase transition in Fe3O4 nanoparticle arrays observed by Lorentz microscopy and electron holography. Appl. Phys. Lett. 2011, 98 (7), 072509. (63) Zelenakova, A.; Zelenak, V.; Mat’ko, I.; Streckova, M.; Hrubovcak, P.; Kovac, J. Superferromagnetism in chain-like Fe@ SiO2 nanoparticle ensembles. J. Appl. Phys. 2014, 116 (3), 033907.

1628

DOI: 10.1021/acs.langmuir.5b04145 Langmuir 2016, 32, 1621−1628