Simple Synthesis of Versatile AkaganÃ©ite-Silica CoreâShell Rods

Article pubs.acs.org/cm

Simple Synthesis of Versatile Akaganéite-Silica Core−Shell Rods Niek Hijnen* and Paul S. Clegg School of Physics and Astronomy, The University of Edinburgh, Mayfield Road, Edinburgh EH9 3JZ, United Kingdom ABSTRACT: We present a simple method for preparing large quantities of micrometer-sized core−shell silica rods, via needle-like akaganéite (β-FeOOH) template particles. The preparation of the akaganéite needles by aging ferric chloride solutions at elevated temperature is explored in order to find improved conditions for the formation of well-dispersed high-aspect-ratio needles. The influence of reaction conditions on irreversible bundling of the needles, yield, and the particle size distribution are reported and discussed. Subsequent coating of the template particles with silica leads to core−shell rods with an aspect ratio (A = L/D) that decreases with shell thickness. This facilitates control over the final aspect ratio and is accompanied by a decrease in the polydispersity. The composition of these core−shell particles offers various possibilities in terms of modifying the particle properties, which we demonstrate by preparing fluorescent particles and hollow silica rods while outlining other interesting options. We also include some optical microscopy images illustrating particle behavior and ordering under gravity, in a homogeneous magnetic field and at the interfaces of phase-separated binary liquids. KEYWORDS: rod-like colloids, core−shell particles, adjustable properties, akaganéite, hollow silica

■

INTRODUCTION Colloidal particles with tunable behavior are needed as the building blocks of new self-assembled materials. Tuning can be achieved by modifying shape or interactions,1−4 by confinement or by applying one or more external fields.5 Using these routes, it has been possible to create empty liquids using clays,6 bicontinuous gels with jammed interfaces (bijels),7−10 and colloidal membranes assembled via rotating magnetic fields.11 To unleash a host of new possibilities we want to combine two or more of these. For example, we would like to trap rodlike particles on a percolating liquid−liquid interface to create a bijel with tunable morphology. Subsequently, we aim to make this structure switchable using applied electric or magnetic fields. The first step is to prepare building blocks that can fulfill all of these demands. Suspensions of rodlike colloids were among the first anisotropic particles to be the subject of extensive investigations. One point of interest is that, at high concentration, suspensions of rods show a rich phase behavior, including nematic and smectic phases. 12−14 Fractal clusters and homogeneous networks of rods induced by attractive interactions also have curious properties.15 Furthermore, when confined on liquid−liquid interfaces, rodlike particles show behavior distinctly different from that of spherical particles.16 This originates from strong capillary interactions and frustration.17−19 Finally, rodlike particles can be aligned using external electric fields and external magnetic fields if the particles exhibit, or can be induced to have, a magnetic moment. Reorienting rods with a field may allow liquid−liquid interfaces to be reorganized. For studying particles at liquid interfaces, silica particles have proven to be very useful, because of the potential for easily varying the surface chemistry, which determines the contact © 2012 American Chemical Society

angle. This is an important parameter for particle-stabilized emulsions and bijels,9,20 and for the capillary interactions between particles of a specific anisotropic shape.21 Previously, silica rods22 and ellipsoids23 have been prepared by adding a silica coating onto anisotropic inorganic particles, and recently methods for directly preparing silica ellipsoids and rods have been reported.24−26 Silica generally is an attractive material for colloidal particles for various reasons, such as its optical properties and easy chemical modification. It is therefore also very convenient as a coating material for colloidal particles to create multifunctional core−shell colloids with specific properties,27 which is important for many technological applications. In this paper, we report an easy route to prepare micrometersized core−shell rods on a large scale, by taking a specific type of iron oxide particles as the template for growing rod-shaped silica. For the iron oxide cores, akaganéite (β-FeOOH) particles are prepared via the forced hydrolysis of a ferric chloride solution, which is a simple method known to produce elongated particles.28 In general, suspensions of small akaganéite particles with relatively low aspect ratios ( 0.1 M and [FeCl3] > 1 M. Saturation concentrations of Fe(III) ions are much higher at low pH,35 so the concentration quickly drops below the saturation value, stopping further precipitation. Another factor possibly involved in the observed trends may be dissolution of the needles (evidence for this is reported later), which could be related to the nonmonotonic decrease in yield with [FeCl3]. Recovery of the yield at higher [HCl] for 0.6 M [FeCl3] appears a genuine observation and could be due to the balance between dissolution and (re)precipitation of akaganéite.

Figure 1. Powder X-ray diffraction (XRD) patterns of sample S5 (see Table 1) dried, ground, and well-packed in a capillary (bottom) and deposited onto a glass slip by drying a concentrated suspension (top). Arrows indicate missing diffractions due to the preferred orientation of the needles parallel to the substrate in the latter case. The reference data from the powder diffraction file are shown in red. 3451

dx.doi.org/10.1021/cm301772p | Chem. Mater. 2012, 24, 3449−3457

Chemistry of Materials

Article

Figure 3. Representative TEM image of (a) akaganéite needles (sample S1), which were (b) subsequently coated with a 100-nm silica shell. (c) A zoomed-in image reveals the core−shell structure; (d) corresponding aspect ratio distributions. The bottom row shows (e) smaller akaganeite needles (sample S5) prepared by aging in an oil bath, (f) coated with a 55-nm silica shell, and (g) coated with a 90-nm silica shell.

Table 1. Effect of Heating Method, Volume, and Time on Resulting Particle Yield, Average Particle Dimensions (with Polydispersities s), and Aspect Ratio for Aging a 0.6 M [FeCl3]/0.07 M [HCl] Solution sample

heating method

volume (mL)

time (h)

particle yield (mg/mL)

S1 S2 S3 S4 S5 S6

oven oven oven oil bath oil bath oil bath

25 500 500 500 500 500

24 24 16 24 16 8

1.03 1.22 0.39 1.28 1.51 0.95

L (sL) (μm) 2.71 3.53 3.22 1.46 1.48 1.92

(0.44) (0.30) (0.30) (0.46) (0.37) (0.31)

D (sD) (μm) 0.11 0.12 0.12 0.06 0.06 0.08

(0.31) (0.34) (0.39) (0.46) (0.38) (0.28)

A (L/D) 24.1 29.5 26.6 23.4 26.0 22.7

conceivable that some particles may become so strongly bound that they are not redispersed by the employed methods. We find bundling to be minimized by increasing both ferric chloride and hydrochloric acid concentrations (Figure 3a) of the aged solutions. Here, the high Fe(III) concentration is probably responsible for the increased particle size. A decrease in the presence of aggregates with increasing ionic strength is somewhat counterintuitive. However, in all samples, the ionic strength is very high, so a further increase of the concentrations will not affect the possibility of aggregation. We expect the decreased bundling to be related to the increased Cl− concentration, because of its adsorption onto the particle surface, making it less likely for particles to aggregate irreversibly. There might also be an effect of the lower pH of the solutions, but this was not checked independently. Agitating the solution during aging increases bundling (Figure 3d, discussed in more detail later), probably because of orthokinetic aggregation. This might suggest that the presence of more Cl− does actually introduce some barrier preventing the irreversible aggregation, which is overcome more easily by providing some kinetic energy (by stirring) to the particles. Ideally, the irreversible bundling of the needles is minimized while still obtaining a good yield. To meet these criteria, samples prepared at higher concentrations are of interest: 0.45 M > [FeCl3] > 1 and 0.05 M > [HCl] > 0.09 M. TEM images (see Figure 3a for a representative image) show that products

A second effect observed in some of our earlier samples is the strong, or even irreversible, aggregation into dense bundles of more or less aligned needles. This needed to be minimized to obtain suitable template particles for preparing well-defined core−shell rods. Bundling becomes evident after silica coating (discussed later in this section), when not individual needles, but instead mainly bundles of needles are coated (Figure 2b, contrast with Figure 3b). More-thorough attempts to disperse and stabilize particles prior to, and during, coating by sonication or changing the pH of the aqueous PVP solution appeared futile. This bundling is likely to be similar to parallel aggregation of akaganéite rods into rafts, encountered in similar experiments, which act as templates for the formation of hematite cubes.35,38 At any point during the formation of the particles, the ionic strength is very high, so there will be no barrier preventing the otherwise charge-stabilized particles from aggregating. Like most iron oxide particles, the akaganéite needles are expected to have a high Hamaker constant in water (α-Fe2O3: Ah ≈ 6 × 10−20 J,44 γ-FeOOH: Ah ≈ 5.5 × 10−20 J45), resulting in strong van der Waals forces. Particles are most strongly bound when the contact area is large, which probably explains why, in the coated aggregates, the long axes of the needles are parallel, while the rods being rectangular rather than cylindrical28 also aid in having a large contact area. Therefore, the occurrence of aggregation in the formation of the particles is far from surprising, and, according to this reasoning, it is 3452



Article

are very similar with no significant differences in terms of bundling. Average particle lengths are ∼3 μm and average particle diameters are ∼100 nm, giving average aspect ratios of ∼30. The spread in particle length, and therefore that of the aspect ratio, is large and polydispersities of ∼40% are found. Surprisingly, the aspect ratio distributions in many of the analyzed samples are found to be bimodal (Figure 3d, “bare”). This bimodality (discussed in more detail later) is more pronounced in some samples than others and originates from the length distribution, which shows the same feature. Scaling Up and Modifying the Aging Procedure. To scale up the yield, the synthesis was performed at a larger volume (500 mL), and some aspects of the aging procedure were varied for fixed concentrations and temperature. Concentrations of a sample from our previous experiments which resulted in little bundling and a good yield (0.6 M FeCl3, 0.07 M HCl) were identified, and the effect of varying the aging procedure on the product was investigated (see Table 1). The samples aged in an oven show significant differences in both particle length and yield (see Table 1). The primary effect of the increased volume will be the decreased heating rate, resulting in the solution effectively aging for a shorter time at the final temperature. With the previously reported dissolution−reprecipitation mechanism35 in mind, this would suggest that the higher yield and larger particle length of sample S2 is related to the dissolution of akaganéite having progressed less far, whereas for sample S3, most of the nucleation and growth still needs to take place. The effective dissolution at later times, as evidenced by the decreasing yield and particle size, would be a result of the Fe(III) concentration dropping and causing the dissolution rate to overtake the precipitation rate. A curious feature of the length distributions that underlies the observed changes in average particle length is the emergence of a bimodality (Figure 4a). Increasing the aging time from 16 h (sample S3) to 24 h (sample S2) results in the distribution changing from a monomodal distribution to a distribution that slightly hints at bimodality. The sample with the smallest volume (sample S1), and effectively the longest aging time, shows a clear bimodal distribution. While the time evolution of the length distribution is correlated with trends in yield, we do not yet understand the relationship between the two. Since aging solutions in an oven without agitation allows particles to settle into a dense sediment, and will not be a very homogeneous process, in terms of temperature and solute concentrations, the effect of agitating the solution during aging was investigated. Solutions were aged in an oil bath to allow homogenization by stirring at low speeds, which, in addition, increases the heating rate. This results in higher yields at shorter times and smaller particles (Table 1). TEM images (Figure 3e) reveal a product looking more “fibrous” and, as discussed earlier, appearing more bundled. Again, the yield is found to decrease upon prolonged aging, occurring when the aging is extended from 16 h to 24 h, indicating the effective dissolution of akaganéite. Particle length distributions are bimodal (Figure 4b), confirming that this is not just an effect of the particles ending up in a dense sediment when heating solutions in an oven. These results clearly show that the aging time has a pronounced effect on the yield and the size distributions of the obtained particles. The change in yield can be understood in terms of the changing balance between dissolution and precipitation. What exactly is the mechanism behind the

Figure 4. Comparison of particle length distributions of samples in Table 1: (a) samples S1, S2, and S3 and (b) samples S4 and S6.

evolution of the length distribution, which could be slightly different for the two heating methods, remains unclear. It is unknown whether the emergence and growth of a peak at smaller lengths, which accompanies the decreasing average length and yield, is exclusively the result of the dissolution of larger particles, while growth maintains part of the population at larger length, or if nucleation is also involved. Such mechanisms are most likely to result in a continuous distribution of lengths, in contrast to our observations. Nucleation at a later stage seems unlikely since the Fe(III) concentration can be expected to have dropped sufficiently to suppress it completely. Therefore, a nonuniform dissolution and growth of particles seems to be the most sensible explanation at this point, which also fits in with the dissolution−reprecipitation mechanism, but the involvement of nucleation cannot be ruled out. Silica Coating. The main challenge in coating iron oxide particles with a silica shell via the Stöber process46 is preventing aggregation before or during the shell growth.47 Similarly, the akaganéite needles appear stable in ethanol, but are destabilized and flocculate rapidly upon the addition of ammonia. An adsorbed layer of PVP39 as steric stabilization was previously found to be an excellent solution to prevent aggregation.23 This is also found to be true here, and we were able to grow a silica shell of fairly homogeneous thickness around template particles, 3453



Article

particles are discarded when transferring the PVP-stabilized particles into ethanol. The observed decrease in the standard deviation of the aspect ratio is expected, because, as L/D decreases, its standard deviation becomes increasingly dominated by that of D, which is significantly smaller than that of L. However, the observed decrease will be positively influenced by the fractionation mentioned previously. The slight increase of the standard deviation of D with silica coating is remarkable, but is probably due to a slight increase in measured bundles since individual bundles become easier to identify after silica coating. Trends in the standard deviations and average dimensions combined result in the polydispersities decreasing by ∼10% after the addition of a shell of 90−100 nm resulting in sA ≈ 30% (sample S1) and sA ≈ 24% (sample S5). The porous silica shells offer ways to modify the particle properties. Exploiting the porosity of the silica layer, it is possible to dissolve the akaganéite cores, leaving hollow silica rods (Figure 6a). The layers are quite thick in the case of larger

resulting in core−shell rods (see Figures 3b and 3c). The bundling was not completely removed and still occurs to a small extent, but it does not have a strong effect on the rodlike morphology of the final core−shell particles. Adding a shell onto the template particles prepared by aging in an oil bath results in more irregular rods for thinner shells (Figure 3f) due to the increased bundling, while increasing shell thickness nicely smooths out irregularities (see Figure 3g). A striking effect of the added silica coating is that it significantly increases particle stabilization. This is confirmed by bright-field microscopy of particles in dispersion (images not shown), and also is evident from TEM images where coated particles are generally better separated compared to bare particles. Detailed characterization of the particle dimensions from TEM images reveals how these are affected by the silica shell. As expected, most significant is the effect of the silica coating on the aspect ratio (see Figures 5 and 3d). The relative increase of

Figure 6. Hollow silica particles (a) obtained after dissolving the akaganeite cores of core−shell particles of sample S1 (shown in Figure 3b) and a CLSM snapshot of revived hollow FITC-labeled particles dispersed in an index matching solvent (dimethylsulfoxide, DMSO).

particles, and, as suggested by Sacanna et al., the silica becomes less porous as the thickness increases.23 While no additional PVP was added after starting the coating process to increase porosity, it is still possible to rapidly dissolve the iron oxide cores out of thick shells by dispersing them in fuming hydrochloric acid (lower concentrations worked extremely slowly, as no dissolution of the cores was observed within a day). The resulting hollow particles can be somewhat fragile, depending on shell thickness, probably because of their large size and aspect ratio, as indicated by the occasional broken rod in TEM images. Breaking of particles most likely occurs in the washing steps, where particles are pressed into a dense sediment by centrifugal forces. This is especially significant for particles with thin silica shells (see, for instance, Figure 2b), and, therefore, usually a shell of 75−100 nm (in the case of rods of L ≈ 3 μm) is grown before removing the core, somewhat limiting the aspect ratio of the hollow particles. Alternatively, dispersions of hollow particles could be cleaned by dialysis to circumvent centrifugation steps to finally obtain intact hollow particles, even for thin silica shells. The silica shell can also easily be chemically modified; for instance, a fluorescent dye can be incorporated into the shell during its growth. Hollowing out fluorescently labeled shells is disastrous for their fluorescent properties, as is demonstrated by a very noisy and faint fluorescence signal.23 However, we find that, afterward, FITC can be linked to the free amine groups (from excess APTES used in the synthesis) on the silica surface throughout the

Figure 5. Aspect ratio of samples S1 and S5 determined from TEM images (symbols, with the standard deviations as error bars), as a function of silica thickness, with the evolution of the length and diameter of sample S1 with shell thickness as inset, where solid lines are expected trends from a homogeneous coating of a rod (eq 1) with the same dimensions as the average values found for the bare needles.

the particle diameter is much larger than that of the particle length, resulting in a rapid decrease of the aspect ratio with shell thickness. Since the shell thickness is fairly homogeneous around the particles, the aspect ratio of particles is expected to roughly decrease as

A=

L + 2x D + 2x

(1)

where x is the shell thickness. Figure 5 shows that the aspect ratio obtained from TEM images agrees quite well with this model. However, the decrease of the aspect ratio is consistently slower than expected. The inset graph in Figure 5 show that this is due to the average length increasing (∼13%) more than expected, while the particle diameter nicely follows the expected trend. While the thickness of the shells at the tips of the particles is much less consistent compared to that at the sides, the observed increase in average particle length will primarily originate from fractionation. This most likely occurs when spinning down particles from the aqueous PVP solution, where the supernatant is often still turbid, indicating that small 3454



Article

∼5 μm. While this clearly indicates the presence of local smectic ordering and fractionation, with the latter probably aiding the former, it can very well be that the bulk ordering of the sediment is nematic. An interesting feature of the iron oxide (β-FeOOH) cores are their magnetic properties. Akaganéite is an antiferromagnetic material, and some recent reports have shown it to exhibit superparamagnetic behavior when in the form of nanoparticles.33,48,49 The magnetic properties of our particles might well be different from that of the previously studied nanoparticles, because of their much larger size49 and possible effects of the silica coating.50 Instead of studying these in detail, we simply looked at the orientational behavior of these particles in a homogeneous magnetic field. Figure 8 shows an aqueous

porous shells. This “revival” of the fluorescent labeling results in nicely fluorescent particles that give clear CLSM images (Figure 6b). These particles are well-stabilized in ethanol or water; moreover, the addition of a (thin) nonfluorescent shell around these particles is straightforward and can have various advantages (e.g., distinguishing closely packed particles). Particle Ordering and Manipulation. To demonstrate the versatility of these silica rods, we selected one of the synthesized template particles (sample S2). We prepared core− shell rods with plain and fluorescent silica, and hollow particles by removing cores from the former. These were then used to illustrate structuring in sediments and at an interface and particle alignment in a magnetic field. Suspensions sediment quite quickly into a dense region within which the particles exhibit liquid crystalline ordering. To demonstrate this we left an aqueous suspension of hollow silica rods to form a dense sediment over the course of a few days. Figure 7a shows a photograph of the sediment taken through

Figure 8. Microscope images showing the effect of a magnetic field on particle orientation. Insets show Fourier transforms of the images.

dispersion of core−shell particles sedimented onto the bottom glass surface of a capillary, where their orientational and translational freedom is somewhat constrained by the wall. The particles were still completely mobile, because of the electrostatic repulsions between particles and glass, and were able to rotate their long axes perpendicular to the glass wall, as observed by occasional “blinking” in the phase contrast imaging. Figure 8 shows the effect of switching on a homogeneous magnetic field, with the arising anisotropy being highlighted by Fourier transforms of the images (insets). The short axes align parallel to the magnetic field direction, while the long axes become increasingly restricted to the plane perpendicular to the magnetic field. This is observed clearly due to the latter preferentially being parallel to the glass wall. Similar behavior is reported for hematite ellipsoids,51 while goethite (α-FeOOH) rods are found to first align with their long axes parallel to the magnetic field before the induced magnetic moments along the short axes take over at higher fields, and the effect becomes more similar to what is observed here.52,53 A very interesting feature of akaganéite is its transformibility into hematite and magnetite, which changes the magnetic properties.32−34,41 Both this, as well as the high aspect ratio and control over particle dimensions, make these particles a rich system for studying the dynamics of rods in a magnetic field.51 Fluorescent core−shell rods are very suitable for fluorescence confocal imaging in samples containing lower particle concentrations. Figure 9 shows images of partially wetting particles in phase-separated water−lutidine mixtures.54 At an off-critical liquid composition (Figure 9a), nonspherical droplets are formed and individual particles can be distinguished at higher magnification (inset). In a sample where the liquids have separated via spinodal decomposition (Figure 9b), large liquid domains are stabilized, suggesting that

Figure 7. (a) Capillary holding a sediment of hollow silica particles dispersed in water viewed through crossed polarizers, and (b) a polarized light microscopy image of the same sediment showing evidence for the presence of small smectic domains. The top inset is an enlargement of the main image, and the bottom inset is a Fourier transform of the main image.

crossed polarizers, revealing the birefringence of the sample. A closer look at this sediment using polarization microscopy (Figure 7b) shows the birefringence textures in more detail. Dark regions indicate where particles orient perpendicular to the plane of view. The layering observed in Figure 7a is therefore alternating regions where particles stack either parallel or perpendicular to the glass walls. What is striking is the presence of striped regions in the colored domains (shown as an enlargement in the top right inset of Figure 7b). A Fourier transform of the main image (bottom right inset) reveals bands confirming the existence of some typical length scale. Using radial averaging, a peak is found at a q value corresponding to 4.0 μm (red ring in bottom right inset) corresponding well to the average length (L = 4.1 μm) of these particles. The image was taken in the top region of the sample, which might explain the broadening of the peaks toward higher q. Some fractionation will probably have taken place over the long distance from which the particles had to sediment at the top of the sample, resulting in the top region of the sediment containing mainly smaller particles. This might also have aided the observed smectic ordering, as, locally, the size distribution is narrower compared to the original dispersion. Further evidence for this effect is that, closer to the bottom of the sediment, the broadening toward higher q disappears and the q value of the peaks obtained from radial averaging the Fourier transforms of images decreases up to values corresponding to a length scale of 3455



Article

hematite or magnetite, which allows altering of the magnetic properties of the particles in a very straightforward fashion. Their tunability was demonstrated by using optical microscopy to investigate particles with different properties in various situations. Sediments formed under gravity are birefringent, because of liquid crystalline ordering and in suspensions of hollow silica rods in water evidence is found of fractionation and smectic ordering. The magnetic properties of the core−shell particles are evidenced by observing the particles orienting their long axes perpendicular to a magnetic field. Finally, fluorescent core−shell particles are shown to be suitable for studying particles at a liquid interface, and the images reveal details that encourage further investigation of similar systems. All in all, these easy-to-prepare core−shell rods have readily adjustable properties, offering interesting possibilities in many directions.

Figure 9. FITC-labeled core−shell rods (green) in a phase-separated mixture of water and 2,6-lutidine (red) at (a) 20 wt % and (b) 28 wt % (critical composition) of 2,6-lutidine.

the particles are fairly close to neutral wetting. True bicontinuity found in bijels,7,10 however, is not observed here. Jamming of rods at the interface and their interplay with interfacial curvature are expected to introduce some curious behavior. We note in Figure 9b, for instance, that the image reveals a roughness of the interface on the scale of the particle length, possibly indicating that particles flip out of their preferred planar configuration with respect to the interface.19 This could be related to jamming of rods and frustrated packings at the curved interfaces. Another striking and regular feature of this sample are extremely thin fluid bridges that refuse to snap off (inset of Figure 9b), something that is not observed in similar experiments where spheres are used.7,10 These findings indicate a significant difference in the structures formed with rodlike rather than spherical particles, and this will be investigated in future work.

■

CONCLUSIONS We have described a simple method for preparing a new type of micrometer-sized core−shell rods, composed of an akaganéite (β-FeOOH) core and a silica shell. The synthesis of the akaganéite template particles was explored in some detail, reporting trends in yield and particle size distributions with concentrations and time, discussing these in the context of previous reports on the preparation of akaganéite particles. Irreversible aggregation into compact bundles was found to be the biggest obstacle for the purpose of coating individual needles, and this is almost completely removed by adjusting the concentrations. However, some further improvements could possibly be made by enhancing particle stabilization (e.g., steric) during preparation. The subsequent growth of a smooth silica shell around the template particles reduces the aspect ratio in a controlled way (with the final aspect ratio depending on the shell thickness), while it also decreases the polydispersity. The silica coating strongly increases colloidal stability, and offers many routes to varying the particle properties, thereby greatly enhancing the potential uses of the system. Chemical modification of the silica shell is easily achieved, giving control over surface chemistry and functionality. In addition, particle properties can be further tuned through the iron oxide cores. First of all, the iron oxide core can be removed, exploiting the porosity of the silica shell, to obtain much less dense and less optically absorbing hollow silica rods. Another interesting prospect is offered specifically by the use of akaganéite as template particles. One of the current interests in akaganéite is namely due to its potential to be transformed to

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Stephen Mitchell for assistance with TEM, Gary Nichol for XRD measurements, and Mathias Reufer and Vincent Martinez for taking images of the magnetic field alignment. Andy Schofield kindly commented on the manuscript, and we are grateful to Job Thijssen and Joe Tavacoli for helpful advice and discussions. This work was supported by the Marie Curie Training Network ITN-COMPLOIDS No. 234810.

■

REFERENCES

(1) Glotzer, S. C.; Solomon, M. J. Nat. Mater. 2007, 6, 557. (2) Solomon, M. J. Curr. Opin. Colloid Interface Sci. 2011, 16, 158. (3) Sacanna, S.; Pine, D. J. Curr. Opin. Colloid Interface Sci. 2011, 16, 96. (4) Lee, K. J.; Yoon, J.; Lahann, J. Curr. Opin. Colloid Interface Sci. 2011, 16, 195. (5) Grzelczak, M.; Vermant, J.; Furst, E. M.; Liz-Marzan, L. M. ACS Nano 2010, 4, 3591. (6) Ruzicka, B.; Zaccarelli, E.; Zulian, L.; Angelini, R.; Sztucki, M.; Moussaid, A.; Narayanan, T.; Sciortino, F. Nat. Mater. 2011, 10, 56. (7) Herzig, E. M.; White, K. A.; Schofield, A. B.; Poon, W. C. K.; Clegg, P. S. Nat. Mater. 2007, 6, 966. (8) Sanz, E.; White, K. A.; Clegg, P. S.; Cates, M. E. Phys. Rev. Lett. 2009, 103, 255502. (9) Tavacoli, J. W.; Thijssen, J. H. J.; Schofield, A. B.; Clegg, P. S. Adv. Funct. Mater. 2011, 21, 2020. (10) Lee, M. N.; Mohraz, A. Adv. Mater. 2010, 22, 4836. (11) Osterman, N.; Poberaj, I.; Dobnikar, J.; Frenkel, D.; Ziherl, P.; Babić, D. Phys. Rev. Lett. 2009, 103, 228301. (12) Vroege, G. J.; Lekkerkerker, H. N. W. Rep. Prog. Phys. 1992, 55, 1241. (13) Bolhuis, P.; Frenkel, D. J. Chem. Phys. 1997, 106, 666. (14) Dogic, Z.; Fraden, S. Curr. Opin. Colloid Interface Sci. 2006, 11, 47. (15) Solomon, M. J.; Spicer, P. T. Soft Matter 2010, 6, 1391. (16) Madivala, B.; Vandebril, S.; Fransaer, J.; Vermant, J. Soft Matter 2009, 5, 1717. (17) Loudet, J. C.; Alsayed, A. M.; Zhang, J.; Yodh, A. G. Phys. Rev. Lett. 2005, 94, 018301. (18) Madivala, B.; Fransaer, J.; Vermant, J. Langmuir 2009, 25, 2718. 3456



Article

(19) Basavaraj, M. G.; Fuller, G. G.; Fransaer, J.; Vermant, J. Langmuir 2006, 22, 6605. (20) Aveyard, R.; Binks, B. P.; Clint, J. H. Adv. Colloid Interface Sci. 2003, 100, 503. (21) Lehle, H.; Noruzifar, E.; Oettel, M. Eur. Phys. J. E 2008, 26, 151. (22) van Bruggen, M. P. B. Langmuir 1998, 14, 2245. (23) Sacanna, S.; Rossi, L.; Kuipers, B. W. M.; Philipse, A. P. Langmuir 2006, 22, 1822. (24) Zhang, H.; Bandosz, T. J.; Akins, D. L. Chem. Commun. 2011, 47, 7791. (25) Shen, S.; Gu, T.; Mao, D.; Xiao, X.; Yuan, P.; Yu, M.; Xia, L.; Ji, Q.; Meng, L.; Song, W.; Yu, C.; Lu, G. Chem. Mater. 2012, 24, 230. (26) Kuijk, A.; van Blaaderen, A.; Imhof, A. J. Am. Chem. Soc. 2011, 133, 2346. (27) Guerrero-Martinez, A.; Perez-Juste, J.; Liz-Marzan, L. M. Adv. Mater. 2010, 22, 1182. (28) Matijevic, E.; Scheiner, P. J. Colloid Interface Sci. 1978, 63, 509. (29) Maeda, H.; Maeda, Y. Langmuir 1996, 12, 1446. (30) Maeda, H.; Maeda, Y. Phys. Rev. Lett. 2003, 90, 018303. (31) Fang, X.-L.; Li, Y.; Chen, C.; Kuang, Q.; Gao, X.-Z.; Xie, Z.-X.; Xie, S.-Y.; Huang, R.-B.; Zheng, L.-S. Langmuir 2010, 26, 2745. (32) Yue, J.; Jiang, X.; Yu, A. J. Nanopart. Res. 2011, 13, 3961. (33) Milosevic, I.; Jouni, H.; David, C.; Warmont, F.; Bonnin, D.; Motte, L. J. Phys. Chem. C 2011, 115, 18999. (34) Chaudhari, N. K.; Yu, J.-S. J. Phys. Chem. C 2008, 112, 19957. (35) Bailey, J. K.; Brinker, C. J.; Mecartney, M. L. J. Colloid Interface Sci. 1993, 157, 1. (36) Blesa, M. A.; Matijevic, E. Adv. Colloid Interface Sci. 1989, 29, 173. (37) Music, S.; Krehula, S.; Popovic, S. Mater. Lett. 2004, 58, 2640. (38) Pu, Z.; Cao, M.; Yang, J.; Huang, K.; Hu, C. Nanotechnology 2006, 17, 799. (39) Graf, C.; Vossen, D. L. J.; Imhof, A.; van Blaaderen, A. Langmuir 2003, 19, 6693. (40) Sacanna, S.; Rossi, L.; Wouterse, A.; Philipse, A. P. J. Phys.: Condens. Matter 2007, 19, 376108. (41) Piao, Y.; Kim, J.; Bin Na, H.; Kim, D.; Baek, J. S.; Ko, M. K.; Lee, J. H.; Shokouhimehr, M.; Hyeon, T. Nat. Mater. 2008, 7, 242. (42) Rasband, W. ImageJ 1.44p, http://imagej.nih.gov/ij/. (43) Matijevic, E.; Kitazawa, Y. Colloid Polym. Sci. 1983, 261, 527. (44) Schudel, M.; Behrens, S. H.; Holthoff, H.; Kretzschmar, R.; Borkovec, M. J. Colloid Interface Sci. 1997, 196, 241. (45) Blakey, B. C.; James, D. F. Colloids Surf. A 2003, 231, 19. (46) Stöber, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26, 62. (47) Thies-Weesie, D. M. E.; Philipse, A. P.; Kluijtmans, S. G. J. M. J. Colloid Interface Sci. 1995, 174, 211. (48) Song, Y.; Bac, B. H.; Lee, Y.-B.; Kim, M. H.; Kang, I. M. CrystEngComm 2011, 13, 287. (49) Urtizberea, A.; Luis, F.; Millan, A.; Natividad, E.; Palacio, F.; Kampert, E.; Zeitler, U. Phys. Rev. B 2011, 83, 214426. (50) Reufer, M.; Dietsch, H.; Gasser, U.; Grobety, B.; Hirt, A. M.; Malik, V. K.; Schurtenberger, P. J. Phys.: Condens. Matter 2011, 23, 065102. (51) Reufer, M.; Martinez, V. A.; Schurtenberger, P.; Poon, W. C. K. Langmuir 2012, 28, 4618. (52) Lemaire, B. J.; Davidson, P.; Ferre, J.; Jamet, J. P.; Petermann, D.; Panine, P.; Dozov, I.; Jolivet, J. P. Eur. Phys. J. E 2004, 13, 291. (53) Thies-Weesie, D. M. E.; de Hoog, J. P.; Hernandez Mendiola, M. H.; Petukhov, A. V.; Vroege, G. J. Chem. Mater. 2007, 19, 5538. (54) White, K. A.; Schofield, A. B.; Wormald, P.; Tavacoli, J. W.; Binks, B. P.; Clegg, P. S. J. Colloid Interface Sci. 2011, 359, 126.

3457


Simple Synthesis of Versatile AkaganÃ©ite-Silica CoreâShell Rods

Recommend Documents

Simple Synthesis of Versatile AkaganÃ©ite-Silica CoreâShell Rods