Supramolecular mechanisms in the synthesis of mesoporous magnetic nanospheres for hyperthermia

Abstract

The present work deals with the preparation of magnetic mesoporous nanocomposites with potential application for cancer treatment. The supramolecular mechanisms that govern the incorporation of γ-Fe₂O₃ nanoparticles into mesoporous silica spheres have been deeply analyzed. The modification of γ-Fe₂O₃ nanoparticles during the encapsulation into mesoporous SiO₂ has been studied. The alkaline conditions of these processes lead to an enlargement of maghemite nanoparticles. We hypothesize that this particle enlargement results from Fe³⁺ cations present in solution. The results presented in this work indicate the importance of the appropriate surface functionalization to incorporate nanosystems into mesoporous silica materials, as well as the modifications that magnetic nanoparticles undergo during the process. Finally, the ability to produce magnetic hyperthermia makes this material a very promising candidate for multifunctional thermoseeds for cancer treatment.

---

Introduction

In the last decade, mesoporous magnetic nanocomposites (MMNs) have awakened a great scientific and technological interest.^1–6 These systems combine the properties of magnetic nanoparticles (MNPs)^7,8 with the outstanding textural properties of mesoporous materials.^9–13 In this sense, MMNs exhibit high potential in the field of catalysis^14,15 but also for biomedical and biotechnological applications, such as drug/gene delivery systems,^16–19 NMR imaging,^20,21 enzyme immobilization,²² cellular uptake²³ and as thermoseeds for cancer treatment by hyperthermia.²⁴

Some authors have developed strategies based on impregnation of the mesoporous particles with the magnetic precursor.^25,26 However, the most common strategy to prepare mesoporous magnetic nanocomposites encompasses the previous synthesis of MNPs like Fe₃O₄, γ-Fe₂O₃, etc. Thereafter, these MNPs are encapsulated into mesoporous particles (commonly silica) with controlled size and shape. For this purpose, C. Sanchez et al.^27,28 and our research group^29–31 have independently applied aerosol assisted techniques to obtain encapsulated NP into mesoporous silica microspheres. Although MMNs prepared by this method have demonstrated high potential for some biomedical applications, the microspheres obtained exhibit some degree of polydispersity, thus ranging in size between 200 and 4000 nm.

The most used strategies to obtain nanometrical and monodisperse MMNs are those based on the precipitation in alkaline media.¹ This procedure is commonly based on a sol–gel process involving alkoxysilanes in the presence of a cationic structure directing agent.³² Commonly, the efficient incorporation of the MNPs requires the preparation of reverse microemulsions^33,34 or, as an alternative, a phase-transfer process from an organic medium to an aqueous solution prior to the encapsulation within the mesoporous silica nanospheres.^22,35–37 These procedures have resulted in MMNs with excellent magnetic properties and mesoporous ordering.

In this work, we have studied the mechanisms followed by functionalized γ-Fe₂O₃ nanoparticles to be encapsulated into silica mesoporous nanospheres. These nanocomposites have been designed to be used as thermoseeds for the cancer treatment by hyperthermia. In addition, the mesoporous SiO₂ spheres could provide an excellent matrix for loading and release of antitumoral drugs.^10–13 One of the questions most commonly considered when preparing MMNs is the extent of the defects, which are undergone by the mesoporous structure due to the incorporation of high amounts of MNPs. In this work we raise the opposite question, that is, to what extent MNPs are affected by the encapsulation process. The results obtained in this topic demonstrate that MNPs can undergo significant changes, which should be taken into account when preparing MMNs.

Materials and methods

γ-Fe₂O₃ magnetic nanoparticles have been synthesized following the Massart method.³⁸ The solutions have been prepared by coprecipitation of ferric chloride and ferrous chloride (molar ratio 2 : 1) in ammonium hydroxide. After removing the supernatant, the oxidation process has been performed with nitric acid 2 M followed by a ferric nitrate solution (0.33 M) heated at 100 °C with a vigorous stirring during 30 minutes. The obtained product was washed with acetone several times and resuspended in water to obtain a ferrofluid with 100 mg ml⁻¹ of γ-Fe₂O₃ magnetic nanoparticles.

Coating process

Three different series of magnetic mesoporous spheres were obtained through a coating/encapsulation 2-stage process (Scheme 1). Firstly, encapsulation of pure maghemite was attempted without any coating (sample γ-Fe₂O₃-0).

Scheme 1 Schematic diagram of the magnetic nanoparticles coating and encapsulation process.

The second series of maghemite nanoparticles (γ-Fe₂O₃-1) was coated with aminopropyltriethoxysilane (APTES), following a method previously described by Mornet et al.³⁹

Another set of magnetic nanoparticles was coated with silica by using tetraethylorthosilicate (TEOS) hydrolysis in basic media (γ-Fe₂O₃-2).⁴⁰

Encapsulation process

The preparation of mesoporous silica microspheres from solubilised silica precursors implies the use of cationic surfactants in basic medium (pH = 11). This procedure was developed by Grün et al.³² and our aim is to combine it with ferrofluids, thus supplying magnetic properties to these materials. For this purpose, the corresponding magnetic ferrofluid (γ-Fe₂O₃-0, γ-Fe₂O₃-1 or γ-Fe₂O₃-2), hexadecyltrimethylammonium bromide (CTAB), ethanol, ammonium hydroxide and distilled water were stirred for 15 min with a molar ratio of 1 : 3.75 : 710 : 140 : 1700. Then, TEOS (17.9 mmol) was added. After 2 hours of additional vigorous stirring, the precipitate was centrifuged and washed with distilled water. The dried powder was calcined at 425 °C in air for 3 hours with a heating rate of 5 °C min⁻¹ to remove the template. All the samples were dried in order to be used for the powder studies at 60 °C for 3 days.

For simplicity, the encapsulated pure maghemite samples were denoted as γ-Fe₂O₃-0@ and the previously coated samples were denoted as γ-Fe₂O₃-1@ and γ-Fe₂O₃-2@ for APTES and silica coated samples, respectively.

Characterization of materials

The size and the zeta potential determined by Dynamic Light Scattering and electrophoretic mobility, respectively, were carried out in a Zetasizer Nano ZS (Malvern Instruments) with a 633 nm laser. The ferrofluid was added to 10 ml of distilled water (used as the supporting electrolyte). The pH was adjusted by the multipurpose titrator MPT-2 adding volumes of 0.01 M and 0.05 M NaOH solutions.

Transmission electron microscopy (TEM) was carried out with a JEOL JEM 2000 FX microscope operated at 200 kV. Particles were dried and suspended in acetone. A few drops were placed onto the copper grid and allowed to dry at room temperature.

A scanning electron microscope (JSM 74001F, Jeol, Japan) equipped with an energy dispersive X-ray spectrometer (EDS-INCA system, Oxford Instruments) was used to investigate the chemical composition of mesoporous magnetic nanocomposites. In order to confirm the MNPs encapsulation, cross-section polishing (CP) using an Argon ion beam was carried out on the materials.

The X-ray diffraction (XRD) patterns were recorded with a Philips X'Pert diffractometer using the Cu Kα radiation (wavelength 1.5406 Å). Rietveld refinements were carried out using the atomic position set and the space group of the maghemite structure, Fd3m, by means of the FullProf 2000 computer program.

Hysteresis loops at room temperature (RT) were obtained with a vibrating sample magnetometer (VSM) LakeShore 7304 in applied magnetic field between ±1.5 × 10⁴ Oe. Magnetization measurements M(T,H) in applied magnetic fields between ±5 × 10⁴ Oe at 4.2 K were performed in powders conditioned in gelatine capsules with a Quantum Design SQUID magnetometer. DC magnetization was measured after cooling the sample both in Zero-Field-Cooled (ZFC) and Field-Cooled (FC) magnetic conditions, being the measuring field 50 Oe and the cooling field 50 kOe.

The heating behaviour of magnetic mesoporous particles was studied at an AC magnetic field of 15 kA m⁻¹ with a frequency of 100 kHz. The mesoporous magnetic particles were dispersed in distilled water under ultrasonication to create a ferrofluid with a concentration of 75 mg ml⁻¹. Thermally insulated Petri dish containing 2 ml of ferrofluid was placed within a water-cooled copper coil driven by a CELES generator. A Luxtron I652 Fluoroptic® thermometer unit connected to a computer was used to measure the ferrofluid temperature.

Results

Fig. 1 shows the variations of zeta potential as a function of pH on MNPs with different coatings. The uncoated nanoparticles, γ-Fe₂O₃-0, show the characteristic neutral isoelectric point (IP) of maghemite at pH = 7.3. The amino groups present in the surface of γ-Fe₂O₃-1 produce an IP shift towards higher pH values (pH = 9.1). On the other hand, the silica layer in γ-Fe₂O₃-2 sample induces an IP shifted towards lower values of pH (4.8). The studies by DLS showed single-modal distributions of particle sizes in all cases. As can be seen in Table 1, the hydrodynamic size of maghemite nanoparticles clearly increases with the coating process, especially when coating with SiO₂ (γ-Fe₂O₃-2). These results point out the efficiency of the coating process with these agents.

Fig. 1 Zeta potential measurements as a function of pH of maghemite nanoparticles (γ-Fe₂O₃-0), APTES coated maghemite nanoparticles (γ-Fe₂O₃-1) and silica coated maghemite nanoparticles (γ-Fe₂O₃-2).

Table 1 Isoelectric point (IP) and diameter of the magnetic nanoparticles (γ-Fe₂O₃-n) before the coating process

Sample	Coating	Isoelectric point (IP)	Ø p /nm
a Diameter of the particles by Dynamic Light Scattering.
γ-Fe2O3-0	—	7.3	8
γ-Fe2O3-1	NH2	9.1	18
γ-Fe2O3-2	SiO2	4.8	38

---

After the assessment of γ-Fe₂O₃ nanoparticles coating, we proceeded with the attempts to encapsulate them into mesoporous silica as described in the Materials and methods section and in ref. 32. Fig. 2 shows the TEM images obtained for the three attempts. In some images the particles appear as aggregated as a consequence of the sample preparation for the experiment. Fig. 2a and b correspond to γ-Fe₂O₃-0@. γ-Fe₂O₃ nanoparticles (darker contrast) appear as clearly segregated phases to the SiO₂ mesoporous particles, which appear as larger mesoporous spheres at the right side. Fig. 2c and d correspond to γ-Fe₂O₃-1@ sample. In this case, mesoporous particles containing nanoparticles are observed, demonstrating the successful encapsulation of γ-Fe₂O₃-1 nanoparticles. The darker contrast can be distinguished within the mesoporous particles. Finally, Fig. 2e and f show the attempts for encapsulating γ-Fe₂O₃-2 nanoparticles, i.e. those previously coated with an inorganic SiO₂ layer. The images are representative of the general scenario for this attempt and exhibit a complete segregation of γ-Fe₂O₃-2 nanoparticles with respect to silica mesoporous spheres.

Fig. 2 TEM images of the microspheres after the encapsulation process of γ-Fe₂O₃-0@ (a and b), γ-Fe₂O₃-1@ (c and d) and γ-Fe₂O₃-2@ (e and f).

Since TEM images indicate that only γ-Fe₂O₃-1@ is successfully incorporated into the SiO₂ mesoporous particles, we decided to carry out deeper studies on this sample. HR-TEM study provides valuable information about the encapsulation process as can be seen in Fig. 3. The size of these mesoporous particles could be estimated to be between 50 and 150 nm and, in some cases, γ-Fe₂O₃-1 nanoparticles are individually encapsulated into a SiO₂ mesoporous sphere as observed in Fig. 3a. Fig. 3b and c display the characteristics of the γ-Fe₂O₃ nanocrystals of around 10–15 nm in diameter. It should be noticed that the nanoparticles contrast at the HREM images are very poor (Fig. 3b and c) since they are embedded in a mesoporous SiO₂ matrix. In spite of it, it can be observed that the particles are crystalline and the interatomic distances are in agreement with the γ-Fe₂O₃ unit cell. Even more, due to the magnetic properties of the samples they have been embedded in a polymer in order to avoid the damage of the microscope lens and then making difficult the observation.

Fig. 3 (a) TEM image obtained for γ-Fe₂O₃-1@ with individual MNPs coated by SiO₂. (b and c) HRTEM images show γ-Fe₂O₃ nanoparticles exhibiting an ordered contrast with a periodicity of 0.3 nm. The FFT showed in the inset of (c) is in agreement with a γ-Fe₂O₃ cell along [001] zone axis.

Fig. 4 shows the study of the particle size and morphology for γ-Fe₂O₃-1@ mesoporous spheres. Fig. 4a shows a set of well defined spherical particles with a diameter ranging between 100 and 200 nm. DLS studies (Fig. 4b) confirm this observation, since a single modal size distribution centred at 150 nm is observed. It must be highlighted that no distribution appears around 18 nm, which would correspond to non-encapsulated γ-Fe₂O₃-1 MNPs.

Fig. 4 (a) SEM micrograph obtained for Fe₂O₃-1@ microspheres. (b) DLS analysis for Fe₂O₃-1@ demonstrating a single modal distribution centered around 200 nm.

Fig. 5 (top) shows a HR-SEM micrograph of a γ-Fe₂O₃-1@ mesoporous sphere. In order to observe the incorporation of γ-Fe₂O₃ nanoparticles, cross-section polishing (CP) using an Argon ion beam was carried out on the silica sphere. Small agglomerates of nanoparticles can be observed inside the spheres. The EDX spectra taken from the nanoparticle agglomerates demonstrate the presence of iron, in a different way than the smooth areas where only silicon is observed. These results completely agree with the EDX experiments carried out during HRTEM observations (Fig. 5, bottom).

Fig. 5 (Top) HRSEM image obtained for a Fe₂O₃-1@ microsphere after treatment with Ar ion beam. The image of the resulting hemisphere demonstrates the encapsulation of γ-Fe₂O₃ as confirmed by the corresponding EDX analysis. (Bottom) TEM image of a γ-Fe₂O₃ nanoparticle successfully encapsulated into mesoporous SiO₂ as confirmed by the corresponding EDX analysis.

Fig. 6 shows the wide angle XRD patterns and Rietveld refinements for pure maghemite nanoparticles γ-Fe₂O₃-0 and the pattern for the encapsulated sample γ-Fe₂O₃-1@. All the diffraction maxima can be assigned to a spinel-like phase as corresponding for maghemite. In addition a wide scattering signal is observed at low angle in the case of γ-Fe₂O₃-1@, as a consequence of the amorphous silica having part in this material.

Fig. 6 XRD patterns and Rietveld refinements of non-treated maghemite nanoparticles (γ-Fe₂O₃-0) and encapsulated nanoparticles into mesoporous silica Fe₂O₃-1@.

Lattice parameters, (Fe)–O–[Fe] distances and angles as well as the chemical formula calculated from occupancy factors are shown in Table 2. These data are in agreement with a γ-Fe₂O₃ phase, exhibiting additional vacancies when compared with the theoretical occupancy considered for stoichiometric γ-Fe₂O₃,⁴¹ that is (Fe)[Fe_1.67□_0.33]O₄. From the occupancy factor refinements, the vacancies in samples Fe₂O₃-0 and γ-Fe₂O₃-1@ are distributed in both tetrahedral and octahedral sites.

Table 2 Lattice parameters, (Fe)–O–[Fe] distances and angles, and calculated formula parameters obtained from Rietveld refinements for maghemite phase. (Fe) and [Fe] indicate Fe³⁺ cations in tetrahedral and octahedral sub-lattices, respectively, and □ means vacancies

Sample	a/Å	(Fe)–O–[Fe] distance/Å	(Fe)–O–[Fe] angle	Formula^a
a Calculated from refined occupancy factors. (Fe)[Fe1.67□0.33] O4 is considered as stoichiometric formula for maghemite.
γ-Fe2O3-0	8.3552 (8)	3.464 (0)	123.9 (2)	(Fe0.83□0.17) [Fe1.55□0.45]O4
γ-Fe2O3-1@	8.3486 (11)	3.461 (0)	124.3 (2)	(Fe0.90□0.10) [Fe1.50□0.50]O4

---

A detailed study of the crystallite size and strains of γ-Fe₂O₃ nanoparticles was carried out. Crystal size, maximum strains and agreement factors for the refinements are displayed in Table 3. The encapsulated magnetic nanoparticles, γ-Fe₂O₃-1@, undergo a significant size and strain increase with respect to non-treated maghemite. This fact was also observed when Rietveld refinements were made on γ-Fe₂O₃-0@ and also for γ-Fe₂O₃-2@ (see ESI†), indicating that γ-Fe₂O₃ nanoparticles grow during the coating/encapsulation process.

Table 3 Microstructural parameters (crystal size and strain) and agreement factors obtained from Rietveld refinements for maghemite phase

Sample	Crystal size^a/nm	Maximum strain^a (%%)	R p	R wp	R Bragg	χ ^2
a The standard deviations appearing in the global average apparent size and strain are calculated using the different reciprocal lattice directions. It is a measure of the degree of anisotropy, not of the estimated error.
γ-Fe2O3-0	8.12 (1)	16.72 (1)	9.75	14.5	5.12	1.27
γ-Fe2O3-1@	9.82 (1)	39.43 (3)	8.17	11.7	5.61	1.16

---

Fig. 7a shows the magnetization curves at room temperature (RT), as a function of the applied magnetic field for γ-Fe₂O₃-0 and γ-Fe₂O₃-1@ magnetic mesoporous microparticles. The saturation magnetization (M_s) of γ-Fe₂O₃-0 is 61.2 emu g⁻¹ which is a typical value for maghemite nanoparticles.⁴² The magnetization curve corresponding to γ-Fe₂O₃-1@ sample has been normalized taking into account the % weight of maghemite determined by ICP spectroscopy (in this case 15.02% in weight). In this way, the magnetization curve displayed in emu g⁻¹ refers only to maghemite encapsulated, discarding the SiO₂ present in this sample. In this way, the possible differences between γ-Fe₂O₃ nanoparticles before and after encapsulation can be determined. Both curves indicate a superparamagnetic behaviour as can be deduced for the nearly zero coercive forces. In addition, a slight decrease of M_s value can be observed for γ-Fe₂O₃-1@ (57.4 emu g⁻¹) with respect to γ-Fe₂O₃-0 (61.23 emu g⁻¹). Since both values are expressed in terms of maghemite mass, the lower M_s value of γ-Fe₂O₃-1@ indicates that γ-Fe₂O₃ nanoparticles exhibit some degree of magnetic disorder in this sample, with respect to γ-Fe₂O₃ nanoparticles before being encapsulated into mesoporous silica.

Fig. 7 (a) Room temperature magnetic hysteresis loops of pure maghemite (γ-Fe₂O₃-0), and γ-Fe₂O₃-1@ obtained by VSM. (b) ZFC and FC magnetizations for γ-Fe₂O₃-1@ measured as a function of temperature.

Magnetization measurements M(T,H) in applied magnetic fields between ±5 × 10⁴ Oe at 4.2 K were then performed. The magnetic parameters are displayed in Table 4, together with those obtained at 298 K (RT). In contrast to M_s values obtained at RT for γ-Fe₂O₃-1@ and γ-Fe₂O₃-0, the M_s values at 4.2 K are identical in both samples. These results indicate a misalignment of the antiferromagnetic sub-lattices at RT in γ-Fe₂O₃ nanoparticles after the coating/encapsulation process. This result would agree with the highest crystal strains calculated for γ-Fe₂O₃-1@ by the Rietveld refinements. Finally, the coercive field differences for samples γ-Fe₂O₃-1@ and γ-Fe₂O₃-0, although small, indicate a larger amount of pinning centres (defects) in the maghemite cores after the successful coating/encapsulation process.

Table 4 Magnetic parameters extracted from the hysteresis loops at 4.2 K and 298 K

Sample	H c/Oe, 4.2 K	M s, 4.2 K	H c/Oe, 298 K	M s, 298 K
a M s values in emu g^−1 of maghemite.
γ-Fe2O3-0	551	66	16	61.23
γ-Fe2O3-1@	619	66^a	42	57.40^a

---

Fig. 7b illustrates the ZFC/FC behaviour of the magnetic nanoparticles incorporated into the mesoporous matrix γ-Fe₂O₃-1@. The blocking temperature of this sample around 200 K gives an estimation of the size of the magnetic core around 5.3 nm using the relationship T_B = KV/25k_B, where K is the anisotropy constant and k_B the Boltzmann constant. This size refers to the magnetic effective size of the encapsulated nanoparticles, pointing out that the magnetic volume is smaller than the crystallite size determined by XRD (9.82 nm as shown in Table 3). The broad peaks of ZFC curves also reveal the existence of aggregated and magnetic interactions between magnetic nanoparticles.

The in vitro hyperthermia capability of γ-Fe₂O₃-1@ was studied applying an alternating magnetic field of 15 kA m⁻¹ and 100 kHz (Fig. 8). For comparison purposes, the temperature evolution for the ferrofluid of bare γ-Fe₂O₃ nanoparticles is also shown. The ferrofluid made of γ-Fe₂O₃ nanoparticles reaches the temperature equilibrium at 55 °C, out of the hyperthermia range. Encapsulated maghemite nanoparticles of γ-Fe₂O₃-1@ are able to heat the environment to 45 °C in 30 min, using concentrations of 75 mg ml⁻¹. This temperature is in the hyperthermia range for the treatment of solid tumours, i.e. between 42 and 47 °C. The specific absorption rate (SAR) is used as the power of heating of a material per gram. SAR was defined as follows:

where C is the specific heat capacity of the heated medium, in this case water, and ΔT/Δt is the initial slope of the heating curve. The encapsulated nanoparticles exhibited the heating rate of 0.04 K s⁻¹ and SAR of 2.3 W g⁻¹.

Fig. 8 Temperature evolution as a function of time of hyperthermia experiment with the γ-Fe₂O₃-1@ sample. For comparison purposes, plot for bare γ-Fe₂O₃ nanoparticles (100 mg ml⁻¹) is also shown.

Discussion

MMNs have been prepared through the encapsulation of γ-Fe₂O₃ nanoparticles within mesoporous silica spheres. The encapsulation process involves alkaline pH values and the presence of a cationic surfactant such as C₁₆TAB. The results described in this work evidence that previous surface functionalization is required for incorporating the magnetic nanoparticles into the mesoporous spheres.

TEM, HRTEM and HRSEM together with EDX spectroscopy demonstrate that only those maghemite nanoparticles functionalized with APTS can be encapsulated into mesoporous silica microspheres. In contrast, pristine γ-Fe₂O₃ nanoparticles and SiO₂ coated nanoparticles are mostly segregated out of the mesoporous microspheres.

In order to understand the mechanisms involved in the success or failure of encapsulation, we must consider the set of supramolecular interactions taking place between the different components during the synthesis. These interactions are supramolecular forces involving hydrophobic and electrostatic forces. The first ones refer to the attractive interactions established between the hydrophobic component of the surfactant (C₁₆TAB) and the γ-Fe₂O₃ nanoparticles. This interaction is likely to occur if γ-Fe₂O₃ nanoparticles are previously coated with an agent containing a hydrophobic alkane component, such as APTS. Lin et al. demonstrated that supramolecular interactions can occur between propyl groups of APTS and the C₁₆ hydrophobic chains of the hexadecyltrimethyl ammonium used as surfactant.⁴³ This surfactant–magnetic nanoparticles interaction would facilitate the encapsulation of the magnetic nanoparticles into the silica matrix as represented in Scheme 2. As an aspect to improve, we must highlight that this interaction can lead to defective mesoporous structures. Actually, APTS coated nanoparticles interact during the mesophase formation with the surfactant through the alkane chain. Consequently, the cooperative liquid crystal mechanism that governs these processes^44,45 is significantly affected. This fact would explain the lower textural properties observed for γ-Fe₂O₃-1@ compared to Fe₂O₃-0@ and Fe₂O₃-2@ (see ESI†), as well as the defective mesoporous structure observed by TEM and HRTEM.

Scheme 2 Supramolecular interactions that explain the formation of the mesoporous magnetic particles. (a) Electrostatic repulsion between the negatively charged nanoparticles and the soluble silica species (samples γ-Fe₂O₃-0, γ-Fe₂O₃-2). (b) Aminopropyl groups interact with hydrophobic organic chains of CTAB. In addition, the electrostatic repulsion is weaker due to the NH₂ groups allowing γ-Fe₂O₃ incorporation in the system (case of γ-Fe₂O₃-1).

The electrostatic forces refer to the electric potential at the γ-Fe₂O₃ nanoparticles surface and the electric charge of the hydrolyzed silica species at the pH of synthesis. In this case, the silica spheres are precipitated in basic media (pH around 11) and hydrolyzed silica species are negatively charged. In this way, γ-Fe₂O₃ nanoparticles with negative surface potential will be strongly repelled by the negatively charged silica species, avoiding their inclusion into the mesoporous spheres. This is the case for γ-Fe₂O₃-0 and γ-Fe₂O₃-2, which exhibit electrical surface potentials of around −30 mV. In contrast, γ-Fe₂O₃-1 nanoparticles undergo much lower electrostatic repulsion due to the primary NH₂ groups of the APTS coating (around 10 mV at the synthesis pH) and are not enough to overcome the attractive supramolecular hydrophobic forces.

Microstructural analyses demonstrate that γ-Fe₂O₃ nanoparticles undergo significant changes during the coating/encapsulation process. In his pioneering work, Wu et al. hypothesized about the possibility of magnetic particles degradation during the coating with mesoporous silica.⁴⁶ However, in our case, maghemite nanoparticles inside γ-Fe₂O₃-1@ are significantly larger than uncoated maghemite in the raw ferrofluid.

Before any treatment, γ-Fe₂O₃-0 nanoparticles exhibit entire volume size almost identical than coherent diffraction domain size (i.e. 8.0 nm and 8.1 nm determined by DLS and XRD, respectively), thus indicating that γ-Fe₂O₃-1@ nanoparticles constitute single nanocrystalline domains. However, after the coating/encapsulation process, both the entire volume and coherent diffraction domain increase. Actually, Rietveld refinements reveal a significant increase of the coherent diffraction domains for maghemite nanoparticles in γ-Fe₂O₃-1@ with respect to γ-Fe₂O₃-0. In addition, HRTEM images point out that some γ-Fe₂O₃ nanoparticles increased in size exhibiting diameters up to 15 nm (Fig. 3b and c). It must be highlighted that previous HRTEM studies of γ-Fe₂O₃-0 revealed nanoparticles ranging in size between 5 and 8 nm.³⁰ This fact could be explained in terms of a further precipitation of γ-Fe₂O₃ onto pre-existing nanoparticles, from iron cations present in the raw ferrofluid. In fact, Fe³⁺ cations were detected in the raw ferrofluid by titration, without previous nanoparticles dissolution treatment. The presence of Fe³⁺ cations in dissolution is justified by the strategy followed to prepare maghemite ferrofluids, through the previous formation of magnetite nanoparticles and described by:

Fe²⁺ + 2Fe³⁺ + 8OH⁻ = Fe₃O₄ + 4H₂O

The subsequent treatment with nitric acid solution to form maghemite leads to

Fe₃O₄ + 2H⁺ = γ-Fe₂O₃ + Fe²⁺ + H₂O

Finally, the addition of nitrates as of iron(III) nitrate would oxidize Fe²⁺ to Fe³⁺ cations.

During the encapsulation with mesoporous silica, pH increases until pH = 11 forming Fe(OH)₃ from Fe³⁺ in solution. The hydroxide would be incorporated to nanoparticles during surfactant calcination:

2Fe(OH)₃ + ΔQ = γ-Fe₂O₃ + 3H₂O

The newly grown γ-Fe₂O₃ seems to be structurally defective, resulting in larger particles with increased global strains as clearly demonstrated by the data collected in Table 3.

All the magnetic particles exhibit very small coercive forces and magnetic remanence at room temperature. S-like shape of the M(H) curve indicates a superparamagnetic dominating contribution for the majority of the particles, in accordance with the small size of the magnetic nanoparticles. After the coating/encapsulation, the saturation magnetization of γ-Fe₂O₃ nanoparticles slightly decreases at RT, which is associated with the existence of a surface layer where the iron cations possess unsaturated coordination spheres.⁴⁷ This fact is in agreement with the defective maghemite phase growth during the coating/encapsulation. In addition, the interfaces between the SiO₂ matrix and the magnetic nanoparticles are pinning sites that increase the coercive field (see Table 4).

For the magnetic mesoporous spheres the ZFC/FC curves suggest a wide size distribution of aggregated particles, which would be in agreement with the system consisting of groups of magnetic nanoparticles incorporated into mesoporous silica spheres, as observed by TEM. DLS results displayed in Table 1, indicated that γ-Fe₂O₃-0 and γ-Fe₂O₃-1 nanoparticles exhibit hydrodynamic size diameters expected for non-aggregated particles. Therefore, the aggregation must occur at the last stage of the synthesis, that is, during the encapsulation process. In the case of γ-Fe₂O₃-2 (coated with SiO₂), the aggregation seems to occur also during the coating process (before encapsulation), showing hydrodynamic size diameters of 38 nm.

The large irreversibility observed in the explored temperature range points towards the existence of defects in the ferromagnetic lattice at the interfaces between nanoparticles and the mesoporous matrix. Moreover, comparing the magnetic size calculated from blocking temperature (5.3 nm) and the microstructural size calculated by XRD (9.82 nm) indicates that the magnetically defective shell could extend around 4 nm from the surface towards the core. It must be taken into account that the deposition of new γ-Fe₂O₃ takes place simultaneously with the interaction of soluble silica species with the nanoparticles. Even more, in the case of γ-Fe₂O₃-1 sample, the nanoparticles are previously coated with aminopropylsilane, justifying that the crystal growth over the nanoparticles was structurally defective.

HRTEM images evidenced that MNPs could be even larger, exhibiting diameters between 10 and 15 nm. This fact evidences that the entire volume size is also larger than structural coherence, having consequences over magnetic properties as has been demonstrated by Di Marco et al. for γ-Fe₂O₃ ferrofluids.⁴⁸ Summarizing, the coating/encapsulation process leads to MNPs that exhibit significant changes when considering different size parameters, in such a way that entire volume > structure coherence > magnetic coherence.

In the hyperthermia experiment, the SAR value and the self-heating temperature indicate that the magnetic mesoporous spheres are suitable to be used as thermoseeds for hyperthermia treatment in the range of 42–45 °C. This temperature interval falls in the moderate hyperthermia range, which is characterized for its efficiency in cancer cell destruction without non-reversible damage on healthy cells.⁴⁹ As demonstrated in this work, the preparation of magnetic mesoporous spheres requires the chemical modification of the raw ferrofluid. This strategy leads to a slight decrease of the magnetic properties, by means of the enlargement of the magnetically disordered shell at RT in nanoparticles. Anyway, the magnetic mesoporous spheres exhibit enough SAR to act as potential thermoseeds in cancer treatment by hyperthermia.

Conclusions

In this work, mesoporous magnetic nanocomposites have been designed to be used as thermoseeds for the cancer treatment by hyperthermia. The supramolecular mechanisms that govern the encapsulation of γ-Fe₂O₃ nanoparticles into silica mesoporous nanospheres have been proposed. Besides, the structural and magnetic modifications undergone by γ-Fe₂O₃ nanoparticles have been for the first time determined.

The incorporation of γ-Fe₂O₃ nanoparticles into silica is determined by supramolecular forces, the γ-Fe₂O₃ surface electrical charge and the hydrophobic interaction between the γ-Fe₂O₃ coating and surfactant molecules.

The successful encapsulation of γ-Fe₂O₃ nanoparticles results in significant particle size changes as well as structure and magnetic coherence. After the coating/encapsulation process, a new defective γ-Fe₂O₃ phase is formed. Consequently, encapsulated γ-Fe₂O₃ nanoparticles exhibit entire volume > structure coherence > magnetic coherence.

Preliminary hyperthermia tests have shown that the magnetic mesoporous spheres reach the hyperthermia temperature range. This work opens up the possibility of synthesizing multifunctional mesoporous magnetic particles for hyperthermia treatment and drug delivery purposes, based on the superparamagnetic behaviour and the mesoporous structure.

Acknowledgements

This work was supported by the Spanish CICYT through the project MAT2008-00736 and by the CAM through project S2009/MAT-1472. We thank C. Aroca (ISOM, Universidad Politécnica de Madrid) for their technical assistance in the magnetic measurements and Prof. Osamu Terasaki for his valuable contribution in HRSEM studies.

Notes and references

1. J. Liu, S. Z. Qiao, Q. H. Hu and G. Q. Lu, Small, 2011, 7, 425–443.
2. M. Vallet-Regí, M. Colilla and B. González, Chem. Soc. Rev., 2011, 40, 596–607.
3. J. L. Vivero-Escoto, I. I. Slowing, B. G. Trewyn, V. S.-Y. Lin. Small, 6, 1952–1967.
4. E. Prouzet, S. Ravaine, C. Sanchez and R. Backov, New J. Chem., 2008, 32, 1284–1299.
5. C. Sanchez, L. Rozes, F. Ribot, C. Laberty-Robert, D. Grosso, C. Sassoye, C. Boissiere and L. Nicole, C. R. Chim., 2010, 13, 3–39.
6. Y. H. Deng, Y. Cai, Z. K. Sun and D. Y. Zhao, Chem. Phys. Lett., 2011, 510, 1–13.
7. A. H. Lu, E. L. Sabalas and F. Schüth, Angew. Chem., Int. Ed., 2007, 46, 1222–1244.
8. B. González, E. Ruiz-Hernández, M. J. Feito, C. López de Laorden, D. Arcos, C. Rámirez-Santillan, C. Matesanz, M. T. Portoles and M. Vallet-Regí, J. Mater. Chem., 2011, 21, 4598–4604.
9. I. I. Slowing, J. L. Vivero-Escoto, C. W. Wu and V. S.-Y. Lin, Adv. Drug Delivery Rev., 2008, 60, 1278–1288.
10. M. Vallet-Regí, A. Rámila, R. P. del Real and J. Pérez-Pariente, Chem. Mater., 2001, 13, 308–311.
11. M. Vallet-Regí, F. Balas and D. Arcos, Angew. Chem., Int. Ed., 2007, 46, 7548–7558.
12. M. Vallet-Regí, J. Intern. Med., 2010, 267, 22–43.
13. F. Balas, M. Manzano, P. Horcajada and M. Vallet-Regí, J. Am. Chem. Soc., 2006, 128, 8116–8117.
14. J. Ge, Q. Zhang, T. Zhang and Y. D. Yin, Angew. Chem., Int. Ed., 2008, 47, 8924–8928.
15. Y. H. Deng, Y. Cai, Z. K. Sun, J. Liu, C. Liu, J. Wei, W. Li, C. Liu, Y. Wang and D. Y. Zhao, J. Am. Chem. Soc., 2010, 1332, 8466–8473.
16. S. Giri, B. G. Trewyn, M. P. Stellmaker and V. S. Y. Lin, Angew. Chem., Int. Ed., 2005, 44, 5038–5044.
17. D. Castanotto and J. J. Rossi, Nature, 2009, 457, 426–433.
18. F. Torney, B. G. Trewyn, V. S.-Y. Lin and K. Wang, Nat. Nanotechnol., 2007, 2, 295–300.
19. E. Ruiz-Hernández, A. Baeza and M. Vallet-Regí, ACS Nano, 2011, 5, 1259–1266.
20. H. M. Liu, S. H. Wu, C. W. Lu, M. Yao, J. K. Hsiao, Y. Hung, Y. S. Lin, C. Y. Mou, C. S. Yang, D. M. Huang and Y. C. Chen, Small, 2008, 4, 619–626.
21. J. H. Lee, Y. M. Huh, Y. W. Jun, J. W. Seo, J. T. Jang, H. T. Song, S. Kim, E. J. Cho, H. G. Yoon, J. S. Suh and J. Cheon, Nat. Med., 2007, 13, 95–99.
22. L. Zhang, S. Z. Qiao, Y. G. Jin, H. G. Yang, S. Budihartono, F. Stahr, Z. F. Yan, X. L. Wang, Z. P. Hao and G. Q. Lu, Adv. Funct. Mater., 2008, 18, 3203–3212.
23. K. R. Lee, S. Kim, D. H. Kang, J. I. Lee, Y. J. Lee, W. S. Kim, D. H. Cho, H. B. Lim, J. Kim and N. H. Hur, Chem. Mater., 2008, 20, 6738–6742.
24. F. Martín-Saavedra, E. Ruíz-Hernández, A. Boré, D. Arcos, M. Vallet-Regí and N. Vilaboa, Acta Biomater., 2010, 6, 4522–4561.
25. J. Kim, Y. Piao, N. Lee, Y. Park, I. H. Lee, J. H. Lee, S. R. Paik and T. Hyeon, Adv. Mater., 2010, 22, 57–60.
26. T. Coradin, J. Larionova, A. A. Smith, G. Rogez, R. Clerac, C. Guerin, G. Blondin, R. E. P. Winpenny, C. Sanchez and T. Malla, Adv. Mater., 2002, 14, 896–898.
27. B. Julian-López, C. Boissiere, C. Chaneac, D. Grosso, S. Vasseur, S. Miraux, E. Duguet and C. Sanchez, J. Mater. Chem., 2007, 17, 1563–1569.
28. C. Boissiere, D. Grosso, A. Chaumonnot, L. Nicole and C. Sanchez, Adv. Mater., 2011, 23, 599–623.
29. E. Ruiz-Hernández, A. López-Noriega, D. Arcos, I. Izquierdo-Barba, O. Terasaki and M. Vallet-Regí, Chem. Mater., 2007, 19, 3455–3463.
30. E. Ruiz-Hernández, A. López-Noriega, D. Arcos and M. Vallet-Regí, Solid State Sci., 2008, 10, 421–426.
31. A. López-Noriega, E. Ruiz-Hernández, S. M. Stevens, D. Arcos, M. W. Anderson, O. Terasaki and M. Vallet-Regí, Chem. Mater., 2009, 21, 18–20.
32. M. Grün, I. Lauer and K. K. Unger, Adv. Mater., 1997, 9, 254–257.
33. D. K. Yi, S. S. Lee, G. C. Papaefthymiou and J. Y. Ying, Chem. Mater., 2006, 18, 614–619.
34. Y. S. Lin, S. H. Hu, Y. Hung, Y. H. Chou, C. Chang, M. L. Lin, C. P. Tsai and C. Y. Mou, Chem. Mater., 2006, 18, 5170–5172.
35. J. Kim, J. E. Lee, J. Lee, J. H. Yu, B. C. Kim, K. An, Y. Hwang, C. H. Shin, J. G. Park and T. Hyeon, J. Am. Chem. Soc., 2006, 128, 688–689.
36. J. Kim, H. S. Kim, N. Lee, T. Kim, H. Kim, T. Yu, I. C. Song, W. K. Moon and T. Hyeon, Angew. Chem., Int. Ed., 2008, 47, 8438–8441.
37. L. Zhang, S. Z. Quiao, Y. G. Jin, Z. G. Chen, H. C. Gu and G. Q. Lu, Adv. Mater., 2008, 20, 805–810.
38. R. Massart, IEEE Trans. Magn., 1981, 17, 1247–1248.
39. S. Mornet, J. Portier and E. Duguet, J. Magn. Magn. Mater., 2005, 293, 127–134.
40. W. Stöber, A. Fink and E. Bohn, J. Colloid Interface Sci., 1968, 26, 62–69.
41. Z. Somgyvári, E. Sváb, G. Mészáros, K. Krezhov, I. Nedkov, I. Sajó and F. Bourée, Appl. Phys. A: Mater. Sci. Process., 2002, 74, S1077–S1079.
42. F. Šulek, M. Drofenik, M. Habulin and Ž Knez, J. Magn. Magn. Mater., 2010, 322, 179–185.
43. S. Huh, J. W. Wiench, J. C. Yoo, M. Pruski and V. S. Lin, Chem. Mater., 2003, 15, 4247–4256.
44. F. Hoffman, M. Cornelius, J. Morell and M. Fröba, Angew. Chem., Int. Ed., 2006, 45, 3216–3251.
45. G. J. A. A. Soler-Illia, C. Sanchez, B. Lebeau and J. Patarin, Chem. Rev., 2002, 102, 4093–4138.
46. P. G. Wu, J. H. Zhu and C. G. Xu, Adv. Funct. Mater., 2004, 14, 345–351.
47. A. Millan, A. Urtizberea, N. J. O. Silva, F. Palacio, V. S. Amaral, E. Snoeck and V. Serin, J. Magn. Magn. Mater., 2007, 312, L5–L9.
48. M. Di Marco, M. Port, P. Couvreur, C. Dubernet, P. Ballirano and C. Sadun, J. Am. Chem. Soc., 2006, 128, 10054–10059.
49. J. Overgaard, Cancer, 1977, 2637–2646.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed synthesis method, FTIR characterization of γ-Fe₂O₃ nanoparticles, XRD studies on magnetic mesoporous silica spheres, N₂ adsorption porosimetry studies on magnetic mesoporous silica spheres, inductive coupled plasma (ICP) elemental analysis, Rietveld refinement results. See DOI: 10.1039/c1jm13102h

---