Surface Engineering of Core/Shell Iron/Iron Oxide Nanoparticles from Microemulsions for Hyperthermia

Abstract

This paper describes the synthesis and surface engineering of core/shell-type iron/iron oxide nanoparticles for magnetic hyperthermia cancer therapy. Iron/iron oxide nanoparticles were synthesized from microemulsions of NaBH₄ and FeCl₃, followed by surface modification in which a thin hydrophobic hexamethyldisilazane layer - used to protect the iron core - replaced the CTAB coating on the particles. Phosphatidylcholine was then assembled on the nanoparticle surface. The resulting nanocomposite particles have a biocompatible surface and show good stability in both air and aqueous solution. Compared to iron oxide nanoparticles, the nanocomposites show much better heating in an alternating magnetic field. They are good candidates for both hyperthermia and magnetic resonance imaging applications.

1. Introduction

Magnetic particles have been used in many important applications such as data storage and recording, xerography and highly sensitive magnetic sensors. Currently, magnetic nanoparticles are attracting attention in both the medical and biological fields for applications including magnetic separation of biological entities, therapeutic drug delivery, hyperthermia for tumor therapy and contrast enhancement agents for magnetic resonance imaging (MRI) applications [1,2].

Hyperthermia is a promising approach to cancer therapy. This is based on the principle that under an alternating magnetic field (AMF), a magnetic particle can generate heat by hysteresis losses [3,4]. Generating temperatures of 42°C can damage and kill cancer cells, usually with minimal injury to normal tissues. By killing cancer cells and damaging proteins and structures within cells, hyperthermia may shrink tumors.

MRI using conventional contrast agents lacks sensitivity for detecting small tumors due to the weakness of site targeting capability. Thus, magnetic nanoparticles were developed as contrast agents for MRI because the nanostructure modified the proton relaxation time and, thus, enhance the diagnosis sensitivity of MRI.

Various magnetic nanoparticles have been used for biomedical applications, such as magnetite (Fe₃O₄), maghemite (γ-Fe₂O₃), ferrites, but their application is limited by their magnetic properties. Fe nanoparticles have very high magnetization, close to that of bulk Fe (218 emu/g) [5]. Meanwhile, the coercivity of the Fe nanoparticles is tunable. Increased magnetization provided by the Fe nanoparticles allows possibly more effective generation of local hyperthermia than iron oxide nanoparticles when subjected to an alternating magnetic field in a frequency range safe for human subjects (50kHz to a few hundred kHz). Furthermore, Fe nanoparticles decrease the T2 relaxation time [6, 7] compared to iron oxide nanoparticles and the R2* decay constant is less [8], suggesting that Fe nanoparticles can produce better MRI images.

Metallic iron nanoparticles can be synthesized via many methods, popularly by thermal decomposition of iron pentacarbonyl [9], microemulsion methods [8,10], or the reduction of iron salts in aqueous solution [11]. Nano-sized iron particles have very high specific surface area, so that they are not thermodynamically stable, resulting in extreme reactivity with oxidizing agents [5]. This may limit iron nanoparticles in both hyperthermia and MRI applications. Other important aspects of magnetic particles for biomedical applications are nontoxicity, biocompatiblilty, injectability, and a high-level accumulation in the target tissue or organ [12]. Modification of the surface chemistry of the nanoparticles provides a way of optimizing the particle’s properties. Surface chemical modifications should not compromise either the physical characteristics nor the chemical properties of the nanomaterial. This is critical for Fe nanoparticles due to their high reactivity and high surface-to-volume ratio.

In this work, Fe/Fe oxide nanoparticles, which have a metallic iron core and an iron oxide shell, are prepared using a previously developed microemulsion method [8,10]. Then the surfactant coating on the nanoparticle is replaced by a hydrophobic layer to prevent oxidization of the metallic core and improve the lifetime of the nanoparticle in aqueous environments. Finally, a phospholipid is assembled onto the composite nanoparticle surface to produce a biocompatible coating. It is shown that the resulting nanoparticles show much better heating in an AMF than iron oxide nanoparticles.

2. Experimental

Fe/Fe₃O₄ core-shell nanoparticles were synthesized through the reduction of aqueous FeCl₃ by aqueous NaBH₄ via combination of two water-in-oil microemulsion solutions [8,10]. Each micro-emulsion solution contains a surfactant: cetyl trimethyl ammonium bromide (CTAB); a co-surfactant: n-butanol; an oil phase: n-octane, and a water phase. Their water phases are 0.20M FeCl₃ and 0.80M NaBH₄, respectively. To prepare the microemulsions, CTAB was combined with n-octane and n-butanol in two flasks. FeCl₃ and NaBH₄ water solutions were prepared separately and then mixed with oil phase in the flasks. In each microemulsion solution, the molar ratio between the surfactants and oil phase was fixed as: CTAB : n-butanol : n-Octane = 1:4.1:11.7, and the volume ratio between oil phase and water phase (O/W) was varied from 7:1 to 1.3:1, for the synthesis of iron nanoparticles with different size and magnetic properties.

The micro-emulsion solutions were agitated for 45 minutes under Ar. Afterwards, the NaBH₄ micro-emulsion solution was added to the same volume of FeCl₃ micro-emulsion solution slowly over ~10 min under high speed agitation and then left to react for a further 10 min. The resulting nanoparticles were separated using either a centrifuge or a magnetic field, followed by washing with de-gassed de-ionized (DI) water (three times) and methanol (twice).

The next step was the passivation procedure to generate the core-shell structure of iron nanoparticles. The nanoparticles were dispersed in 0.5 wt.% trimethylamine N-oxide ((CH₃)₃NO) isopropyl alcohol solution, and sonicated for 30 minutes. After sonication, the nanoparticles were rinsed with methanol and dried under flowing Ar. Finally, the nanoparticles were placed in a desiccator that was filled with Ar for two days to enhance the protective oxide layer for passivation. (CH₃)₃NO works as a mild oxidant that can thicken the crystalline Fe₃O₄ shell on the iron core [9] or can totally oxidize the iron nanocrystallites to iron oxide nanoparticles [13]. Two days passivation of iron particles in Ar is also critical step to improve the stability of iron nanoparticles. This two-step oxidization procedure enables the iron nanoparticles to be produced and oxidized reproducibly, resulting in reproducible magnetic properties.

The subsequent surface engineering of the nanoparticles includes three procedures: CTAB coating removal; surface silanization with hydrophobic hexamethyldisilazane (HMDS); and finally modification with a biocompatible phospholipid coating. For removing the CTAB surfactant on the surface, 100 mg of nanoparticles were sonicated in 10 ml isopropanol and Tetramethyl ammonium hydroxide solution (volume ratio 3:1) for 25 minutes at room temperature. After washing using isopropanol and then methanol, the sample was dried under flowing Ar. Next, the iron nanoparticles were silanized using the HMDS. 100 mg of nanoparticles were dispersed in 10 mL 1.0 vol. % HMDS toluene solution in a glass vial which was filled with Ar. The sample was then discontinuously sonicated at 50°C for 4 hours. The nanoparticles were then separated and carefully dried at 90°C for 2 minutes under Ar.

To obtain a biocompatible coating on the nanoparticles surface, phosphatidylcholine (PC) was assembled onto the nanoparticles surface after the silanization. First the PC was diluted with chloroform to form a 50 mg/ml solution. Silane-modified nanoparticles were dispersed in the desired volume of PC solution and agitated at room temperature for 30 min, assisted by discontinuous sonication. Finally, the nanoparticle suspension was dried using flowing Ar to evaporate the chloroform solvent. The PC-coated nanoparticles were stored at 4°C in a refrigerator.

A FEI Tecnai F20 field emission gun transmission electron microscope (TEM) operated at 200 KV was used to examine the shape and size of the nanoparticles. X-ray diffraction (XRD) measurements were performed using a Rigaku D/MAX diffractometer with Cu-K_α radiation.

The quasi-static magnetic properties were characterized (saturation magnetization, M_s, and coercivity, H_c) from hysteresis loop measurements using a Lakeshore model 7300 vibrating sample magnetometer (VSM).

The surface coating was characterized using infrared spectra obtained using a Nicolet Avatar FT-IR 330 apparatus with an attenuated total reflection (ATR) unit.

Heating tests were performed using a Hafler power amplifier to drive a resonant network comprising a copper coil and capacitors used to achieve a real input impedance matched to the amplifier for maximum efficiency. A Tektronix 60 MHz AC probe was used with an Agilent Infinium digital oscilloscope to measure the current. Details of the heating set-up and measurements are described elsewhere [3, 10].

3. Results and Discussion

The microemulsion method is an important approach to obtain nanoparticles with a narrow size range and uniform chemical and physical properties. Many nanosized materials can be synthesized by either chemical reduction of metal ions or via co-precipitation reactions in microemulsions [14]. In our work, iron nanoparticles were produced by reducing ferric chloride with sodium borohydride in microemulsions. The reaction can be represented by equation (1):

$$2{\text{FeCl}}_3+6{\text{NaBH}}_4+18{\text{H}}_2\text{O}\to 2\text{Fe}\downarrow +21{\text{H}}_2\uparrow +6\text{B}{(\text{OH})}_3+6\text{NaCl}$$ (1)

Fig.1. shows TEM images of iron nanoparticle synthesized from microemulsions. From Fig.1.(a) and (b), we can see that nanoparticles obtained from the microemulsion with a O/W ratio 7:1 have an average diameter of ~8 nm, while nanoparticles obtained from the microemulsion with a O/W ratio 2.5:1 have an average diameter of ~16 nm. Fig.1.(c) is a higher magnification image showing the core-shell structure of the nanoparticles where, after passivation in an inert atmosphere, a 2–3 nm thick iron oxide (Fe₃O₄) layer is formed on the Fe particle surface. Fig.1.(d) is an electron diffraction pattern obtained from the nanoparticles. The diffraction pattern has rings corresponding to both bcc α-Fe and the inverse spinel-structured Fe₃O₄ - the (110), (200), (220) reflections of α-Fe and the (220), (311), (511) reflections of Fe₃O₄ can be clearly distinguished.

Fig. 1.

(a,b,c) TEM images of iron nanoparticles: (a) from a microemulsion with O/W = 7:1, (b) from a microemulsion with O/W = 2.5:1; (c) high magnification image showing the core-shell structure of the nanoparticle; and (d) electron diffraction pattern from the nanoparticles showing the (110), (200) and (220) reflections from α-Fe and the (220), (311) and (511) reflections from Fe₃O₄.

It is worth noting that the oxidation of iron is strongly dependent on experimental parameters such as temperature, passivation time, oxygen partial pressure, iron particle size and surface situation of iron. These parameters influence the thickness and composition of the oxide layer. Some differences have been reported concerning the crystal structure of the oxide shell. Typically, the oxide layer has been reported to be either magnetite (Fe₃O₄) [9] or a mixture of magnetite and maghemite (γ-Fe₂O₃)[15]. (Thick iron oxide layers may contain multiple oxide layers, i.e. Fe: FeO: Fe₃O₄: Fe₂O₃. [16]). Fe₃O₄ and γ-Fe₂O₃ and have a similar spinel crystal structure with only a small difference in the lattice constants. In an earlier paper [8] we showed that the newly-produced iron oxide shell on iron core nanoparticles that we produced was Fe₃O₄. However with extended exposure to water or after long storage times in air, the lattice parameter was closer to that of γ-Fe₂O₃. This suggests that the iron oxide shell may have a tendency to change from magnetite to maghemite given sufficient time. In this paper, since we examined the iron nanoparticles soon after the passivation, the composition of the iron shell was found to be Fe₃O₄, which is in line with our earlier observations [8].

Fig.2. shows the average particle size of the iron nanoparticles, measuring using the TEM, as a function of the O/W ratio in the microemulsions. Particle size decreases with increasing O/W ratio from 20 to 8 nm as the O/W ratio increases from ~1.3 to 7.0. When the O/W ratio is high, the molar ratio of water to surfactant is also increased, resulting in a high surface tension at the oil/water interface. This produces small water droplets and defines the iron nanoparticle size.

Fig. 2.

Iron particle size as a function of the O/W ratio of the microemulsions, in which the molar ratio of surfactant and oil is fixed at CTAB : n-butanol : n-Octane = 1:4.1:11.7.

Fig.3 shows the magnetic properties of the nanoparticles as a function of nanoparticle diameter. Both the saturation magnetization and coercivity increase with increasing iron particle size: M_s increased from 36 to 113 emu/g and the H_c increased from 14 to 185 Oe as the particle diameter increased from 8 to 20 nm. The iron composite nanoparticles are all less than 20 nm, and, thus, their inner iron cores can be considered to be single domain. The saturation magnetization arises from both the iron core (218 emu/g), and the iron oxide shell (for Fe₃O₄, 80~92 emu/g), based on the relative weight percentage of iron, iron oxide and the non-magnetic coatings on the particle surface. For particles having a similar shell thickness, the weight ratio of the iron core to the iron oxide shell is greater for large particles than for small particles. This is the reason for the higher M_s of the larger particles. (The very thin iron oxide shell may not contribute to M_s due to its superparamagnetic behavior.) Meanwhile, the specific surface areas are inversely proportional to the particle radius. Thus, the low M_s value of the small iron particles is also from the higher absorption of non-magnetic coating on their surface. The increasing coercivity with increasing particle diameter (for small particles) is expected based on the random anisotropy model [17]. However, when the size of Fe particles is less than ~8 nm, they probably become superparamagnetic producing no magnetic moment at room temperature.

Fig. 3.

Saturation magnetization and coercivity as a function of nanoparticle diameter.

CTAB is an ionic surfactant that has strong positive change and can be strongly absorbed onto the iron nanoparticle surface via its headgroup [18] during the synthesis process. It is a good protection layer for the iron nanoparticles preventing the oxidation of iron during its passivation and storage. But the CATB coating probably contains some residuals from the microemulsion and by-products from the reaction, such as B(OH)₃, octane and butanol. It cannot be considered a biocompatible coating, and, thus, limits the application of the nanoparticles. Usually, a surfactant can be removed by strong washing, sintering [19] or replacement by other reaction agents or surfactants [20,21]. In this work, the CTAB coating on the nanoparticles was removed using a mixture of TMAH and isopropanol. TMAH is a strong organic base. In photolithography, it is used as both the developer and the stripper that remove the photoresist and polymer residuals [22]. Also, it can improve the dispersion of the nanoparticles in aqueous solution [23,24]. HMDS silanization is a moderate method to obtain a hydrophobic monolayer on the sample surface and this modification process can be completed either in the vapor or liquid phases. After modified by HMDS silane, a thin hydrophobic layer covers the Fe/Fe₃O₄ nanoparticle surface. This thin layer enhances the linking between the nanoparticle surface and phospholipid, and also improves the stability of the iron nanoparticle in aqueous solution. The principle of HMDS silanization is shown in Fig.4(a). The phosphatidylcholine incorporates choline as a hydrophilic headgroup and the fatty acids as the hydrophobic tail in its structure. HMDS-coated Fe/Fe₃O₄ nanoparticle are further modified with PC through mechanism shown in Fig.4. (b). The hydrophobic Van Der Waals interaction between the hydrophobic tail of PC and the hydrophobic surface of nanoparticle forms a thermodynamically-defined interdigitated bilayer structures surrounding each nanoparticle [25]. PC coating provides a biocompatible surface for the iron nanoparticles and allows them to be well dispersed in aqueous solutions.

Fig. 4.

Schematic of surface engineering procedures for the Fe/Fe₃O₄ nanoparticles. (a) HMDS silanization on the nanoparticle; (b) PC assembly onto the Fe/Fe₃O₄ nanoparticle with a hydrophobic surface to form biocompatible coating.

The surface modification results in changes in the magnetic properties of the iron nanoparticles due to the oxidation of iron that occurs during the modification process, mostly during the CTAB removal step. The oxidation of the iron nanoparticles is reflected in their change in saturation magnetization. Fig.5 compares the magnetic properties of the particles after the CTAB removal and after HMDS silanization with the untreated nanoparticles. The original CTAB-coated iron particles have an M_s of 104 emu/g and H_c of 180 Oe, curve (a) on Figure 5. The same sample after CTAB removal by TMAH has M_s and H_c values of 73 emu/g and 119 Oe, respectively. In contrast, the sample coated with HMDS after the CTAB removal shows slightly higher M_s and H_c values of 83 emu/g and 152 Oe, respectively.

Fig. 5.

Magnetic properties of Fe/Fe₃O₄ nanoparticles with different surface conditions: (a) as-synthesized Fe/Fe₃O₄ nanoparticles; (b) particles after CTAB removal and with HMDS silanization;(c) particles after CTAB removal only. Sample (b) and (c) were measured one day after passivation in Ar. Insert figure shows details of the coercivity of the nanoparticles.

PC modification is a very moderate self-assembled process, and it is very helpful to keep the iron core intact inside the nanocomposites. It does not affect the coercivity of the iron composite nanoparticles. No change in the coercivity of the Fe/Fe₃O₄ nanoparticle was observed after PC modification. However, the M_s value decreased due to the extra mass added from the non-magnetic PC. Fig.6 shows the FTIR spectra of the PC coated iron composite nanoparticles. Bands observed around 817, 966, 1076 and 1243 cm⁻¹ are attributed to the −PO₄³⁻ group vibration mode. The bands observed at 1390 and 1460 cm⁻¹ are due to bending vibration of −CH₂- group. The two bands at 2853 and 2925 cm⁻¹ are related to vibrations of symmetric and asymmetric methylene group and methyl group.

Fig. 6.

FTIR spectrum of PC coated iron composite nanoparticles.

The stability of iron nanoparticles depends on their particle size, surface condition and environment. The Fe/Fe₃O₄ nanoparticles become chemically unstable in aqueous solution, especially when they are small and have a hydrophilic surface. Fe/Fe₃O₄ nanoparticles (with diameters large than 10 nm) synthesized from microemulsion after careful passivation can be stable in air for more than 3 months without obvious changes in their magnetic properties.

Fig. 7(a) and (b) show TEM images of the Fe/Fe₃O₄/HMDS/nanoparticle before and after dispersion in water for 1.5 hours. The water-exposed nanoparticles still maintains a metallic iron core.

Fig. 7.

TEM images of Fe/Fe₃O₄/HMDS/PC nanoparticles: (a) without exposure to water; (b) exposed to water for 1.5 hours.

Fig. 8 shows a plot of the temperature rise as a function of time for the Fe-based nanoparticles dispersed in DI water with a concentration of 6 mg/ml under an alternating magnetic field of 150 Oe at a frequency 250 kHz. The M_s and H_c values of Fe/Fe₃O₄/HMDS/PC nanoparticles for the measurement are 55 emu/g and 69 Oe, respectively. And in the nanoparticles, the weight ratio between the PC coating and the Fe/Fe₃O₄/HMDS core is 0.3:1.0. Fe-based nanoparticles show much greater heating compared to Dextran-coated Fe oxide particles. The heating effect, strongly depends on the magnetic particle’s properties, measurement conditions, particle size distribution and particle dispersion [26]. Meanwhile, the stability of the particles also plays an important role. The iron core in this composite nanoparticle provides high magnetic saturation and a large hysteresis loop resulting in a substantial amount of heat produced by the nanoparticles.

Fig. 8.

Temperature versus time for Fe/Fe₃O₄/HMDS/PC nanoparticles and Dextran-coated Fe oxide particles dispersed in water at a concentration of 6 mg/ml under an alternating magnetic field of 150 Oe at 250 kHz.

4. Conclusions

Fe/Fe₃O₄ composite nanoparticles coated with biocompatible coatings were prepared by a microemulsion method. By tuning the microemulsion conditions, nanoparticles with different particle sizes from 8 nm to 20 nm were obtained. Both the M_s and H_c of nanoparticles increase with the particle size. A hydrophobic HMDS protection layer and a biocompatible PC coating were coated onto the nanocomposite surface. By tuning the size and composition of each component of these nanocomposites, the nanostructures can be tailored for their physical and chemical properties. These composite nanoparticles present higher stability in aqueous solution while still maintaining good magnetic properties. With the thin hydrophobic silane layer protection, the oxidation iron is greatly reduced, giving iron nanoparticle enough life time in aqueous environment for biomedical applications. Compared to iron oxide nanoparticles, the iron core in this composite nanoparticle provides high magnetization and increased hysteresis losses, enhancing its use for hyperthermia.