Core–shell nanostructures of carbon encapsulated iron nanoparticles (CEINPs) show unique properties and technological applications, because carbon shell provides extreme chemical stability and protects pure iron core against oxidation without impairing the possibility of functionalization of the carbon surface. Enhancing iron core magnetic properties and, in parallel, improving carbon shell sealing are the two major challenges in the synthesis of CEINPs. Here, we present the synthesis of both CEINPs and a new carbon encapsulated multi-iron nanoparticle by a new modified arc discharge reactor. The nanoparticle size, composition, and crystallinity and the magnetic properties have been studied. The morphological properties were observed by scanning electron microscopy and transmission electron microscopy. In order to evaluate carbon shell protection, the iron cores were characterized by selected area diffraction and fast Fourier transform techniques as well as by electron energy loss and energy dispersive X-ray spectroscopies. Afterward, the magnetic properties were investigated using a superconducting quantum interference device. As main results, spherical, oval, and multi-iron cores were controllably synthesized by this new modified arc discharge method. The carbon shell with high crystallinity exhibited sufficient protection against oxidation of pure iron cores. The presented results also provided new elements for understanding the growth mechanism of iron core and carbon shell.

1. Introduction

There has been an increasing interest in fabrication of magnetic based nanomaterials due to their potential applications in data storage [1], Li-ion battery [2], highly sensitive magnetic sensors [3], and spintronics devices [4]. Moreover, magnetic nanoparticles are attracting attention in both the medical and biological fields for applications including magnetic separation of biological entities [5], tissue engineering [6], food analysis [7], therapeutic drug delivery [8], hyperthermia for tumor therapy [9], contrast enhancement agents for magnetic resonance imaging applications [10, 11], water purification [12], and catalysis [13, 14]. Most of the studies in these areas focus on particles based on oxides since magnetic nanoparticles which are based on pure metallic materials are very sensitive to oxidation given their high specific surface area and reactivity [15].

The key point of using magnetic nanoparticles is the possibility and capability to control them by external magnetic field. Naked metallic nanoparticles are chemically highly active and are easily oxidized in air, resulting in negative effects on magnetic properties. To avoid such consequences, well developed graphitic carbon layers can provide an effective barrier to maintain magnetic properties. Coating metallic cores by silica [16], carbon [1720], and precious-metal [21] is an example of popular inorganic coating approaches. Compared to the polymer, silica, or Au layer, the carbon layer has many advantages, such as higher chemical and thermal stability, better conductivity, and higher biocompatibility [2224]. Moreover, carbon-coated nanoparticles are usually in the metallic state, and, compared to the corresponding oxides, they have a higher magnetic moment [25]. In particular, at present, a hot topic is magnetic carbon-coated metal (such as Fe, Co, Ni, and Mn) composites that are always metal phase as the nucleus of core–shell structural composite nanomaterials [26].

In this research study, carbon encapsulated iron nanoparticles (CEINPs) were designed and produced in a way to make them suitable nanocarriers for biomedical applications. So far, different methods including arc discharge [2729], hydrothermal reaction [30], chemical vapor deposition [31], detonation synthesis [32], chemical vapor condensation [33], magnetron and iron-beam cosputtering [34], laser ablation [35], and cocarbonization [36] are used to synthesize core–shell structures in which metallic nanoparticles are encapsulated by precious metals, polymers, silica, or carbon. However, each of these methods has particular advantages and disadvantages comparing to our method for synthesis of CEINPs. Until now, all reported studies have been only on monoiron nanoparticles and the synthesis of multi-iron core at carbon shell has not yet been reported [3747].

In current nanoparticle fabrication processes, control of the particle size and composition is still empirical, which means that a large number of experimental trials are required to optimize any given process [48]. Plasma properties such as size, gradient temperature, thermal conductivity, and cooling rate mainly depend on the used inert gases. Consequently, all parameters are kept constant to study the influences of helium and nitrogen on morphologies properties as well as magnetic properties. Here, we present the synthesis of both CEINPs and carbon encapsulated multi-iron nanoparticles. The morphological properties were studied by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). In order to verify the absence of iron oxide, the samples were analyzed by selected area diffraction (SAED), electron energy loss spectroscopy (EELS), and energy-dispersive X-ray spectroscopy (EDX) techniques as well as fast Fourier transform (FFT). In addition, carbon shell crystallinity was investigated by Raman spectroscopy. Afterward, the magnetic properties were obtained from superconducting quantum interference device (SQUID) characterization technique.

2. Experimental Details

A modified arc discharge reactor was utilized [4951]. Nitrogen, helium, and their mixtures were used to synthesize each sample, from 0% to 100% of N2. Accordingly, each nanoparticle sample was named based on the nitrogen content; that is, 75% N + 25% H is N75. Two carbon graphite rods (99.99%) with mm thickness were prepared as cathode and anode. The cathode electrode was sharpening at an angle near to 20°. Both electrodes were placed in perpendicular position, in order to generate plasma between them. A needle was placed on top of them to deliver the precursor droplets. A spherical glass covered the whole reactor. Then, the precursor was prepared by dissolving ferrocene (0.5 wt %) in isooctane. Initially, rotary vane pump was used for making vacuum inside the glass chamber up to a pressure of 1 Pa. Afterward, the gases were introduced inside the glass chamber (continuous flow, 1.6 l/min). The pressure was kept at near atmospheric pressure condition by a micro valve. An arc was generated between anode and cathode by a programmable constant power supply which was set at 40 A for each production run of 5 min. A syringe pump along with an injector regulates the droplets flow of precursor to the plasma zone (droplet flow rate 1 mL/min). Consequently, the synthesis process took place inside the glass chamber. Afterwards, by using ethanol, the nanoparticles were collected from the glass walls. Finally, nanoparticles were separated from ethanol through a magnetic filtration.

3. Results and Discussion

Influences of nitrogen and helium mixtures on CEINPs morphologies and magnetic properties were studied. As it is shown in Figure 1(a), SEM image reflects the CEINPs in spherical and agglomeration form. TEM images in Figures 1(b), 1(c), 1(d), 1(e), and 1(f) show geometry of each individual nanoparticle from N0, N25, N50, N75, and N100 samples, respectively. Spherical iron nanoparticles were observed with mean diameter of 7.5 nm in N0 samples. By adding 25% and 50% nitrogen, mainly oval shape iron nanoparticles with mean dimensions of 7.6 × 6.5 nm and 8 nm × 5.6 nm were obtained in N25 and N50 samples, respectively. Afterwards, by adding 25% more nitrogen, iron nanoparticles tended to form comparatively smaller oval iron core with mean dimensions of 7.8 nm × 4.7 nm in N75 sample. Using only nitrogen led to the formation of carbon encapsulated multi-iron nanoparticles (CEMINPs). Consequently, in case of N100 sample, the majority of nanoparticles have small multi-iron cores. CEMINPs mainly consisted of 1 nm and 2 nm spherical iron and one or two oval shaped 4.3 nm × 2.5 nm iron nanoparticles. Figures 2(a) and 2(b) represent TEM images from N100 samples. However, as it is shown in Figure 2(c), iron particles with single oval shape and dimensions of 3 nm × 9 nm were observed in N100 sample as well. In general, according to the TEM observations as we increased nitrogen gas percentage versus helium to generate plasma, the iron cores were forming oval shape. Based on TEM observations extremely small iron particles of 1 to 3 nm do not form a bigger single iron core and instead formed multi-iron core at carbon shell nanoparticle. Gutsch et al. discussed coalescence effect, which merges nanoparticles to form a bigger one in gas phase production of nanoparticles [52]. The condition of nucleation occurs in iron supersaturation vapor, which permits obtaining the particles with stable growth. Afterwards, coalescence occurs when the concentration is high enough. Once the carbon shell formation begins, the iron cores growth and coalescence are finished. Obtained results highlighted the coalescence effects on formation of iron nanoparticles from sample N100 to N0. The presence of oval iron nanoparticles and multi-iron cores shows the coalescence of the nucleus before starting carbon shell formation. The coalescence of only two iron particles was observed. This can be probably due to surpassing of the critical radius of carbon shell for the growth after the first iron coalescence.

The effect of nitrogen and helium in growth formation can be attributed to their differences in thermal conductivity and plasma temperature. Thermal conductivity of nitrogen is about 9 times smaller than helium. Thermal conductivity regulates the cooling rate and gradient of temperature. Thus comparatively nitrogen exhibits lower cooling rate and longer gradient of temperature over highly supersaturated vapor. Supersaturated vapor results in rapid production of numerous nanoparticles. In addition, relatively more carbon species are available in supersaturated vapor for carbon shell formation at higher temperature generated by nitrogen plasma rather than helium plasma. Accordingly, a decrease trend in carbon shell formation is obvious from sample N100 to N0 in Figure 1. It should be noted that the carbon shell formation limits the critical radius of iron. According to our observation when iron particles are small enough (1–3 nm) they are trapped into a single carbon shell.

Since nitrogen is diatomic gas, its plasma has higher energy contents for a given temperature than the atomic gases such as helium [53]. Because of high electronegativity of nitrogen, more energetic electrons are necessary to produce self-sustained arc discharge. Consequently, when nitrogen is used due to the high temperature the carbon electrodes were deformed during the experiment (Figure  S1 in Supplementary Material available online at http://dx.doi.org/10.1155/2015/450183). Regardless of inert gas type, in this arc discharge reactor the electron bombardment begins from cathode to anode and therefore the carbon from the anode electrode was consumed and deformed. This deformation caused changes in plasma shape, made it unstable and prevented continuous synthesis. Moreover, it has negative effect on formation and size distributions of CEINPs and CEMINPs. The size distributions of each sample are shown in Figure S2. N0 sample reveals the best iron size distribution comparing to the other samples, because by using only helium the carbon electrodes show smaller deformation. In spite of usefulness of presented method for CEINPs and CEMINPs synthesis, plasma instability and discontinuous synthesis are the chief drawbacks. A brief investigation on the deformed carbon electrode by SEM has been done. SEM image from carbon piece taken from anode electrode is presented in Figure 3 and it revealed formation and deposition of spherical carbon particles on top of each other. Spaces between particles are visible clearly; therefore possibly this technique can be used for the synthesis and formation of carbon porous materials. Structure of carbon piece on anode suggests a mechanism of growth, which is large cross section of nanoparticles that facilitates the attachment of free electrons to the nanoparticles in the plasma region, therefore becoming attracted by anode due to the electrostatic forces. This contributes to forming very structured carbon deposition on anode electrode in Figure 3. Thinner carbon deposition on cathode was observed probably due to the neutral nanoparticles and negative radicals of carbon. Moreover, the cathode suffers a significant ion bombardment (e.g., [54]). In addition, according to the given growth mechanism of CEINPs, the nitrogen plasma has higher temperature compared to helium plasma, which results in higher availability of carbon species. According to our observation, higher contribution of nitrogen shows an increase in deposition of carbon related structure on the electrodes.

Due to the importance of iron core sufficient protection and in order to show the efficiency of carbon shells against the oxidation of core, samples were evaluated by SAED and FFT techniques. SAED is shown in Figure 4(a); diffraction points of iron are indexed to α-Fe crystal structure as identified by diffraction rings corresponding to the (−1, 0, 1) and (0, −1, 1) planes. However, as it is shown in Figure 4(b) from FFT analysis, trace of iron carbide was detected in interface between core and shell. Most importantly, diffraction points of iron oxides were not detected in any of our samples. In addition, nanoparticles were analyzed by EELS and EDX analysis along with chemical mapping. Figure 5 represents an example of EELS analysis at nanoscale and EDX together with chemical mapping at micro scale. As a result, absences of oxygen are confirmed in iron cores; therefore carbon shell properly sealed the iron cores in all samples. In addition, the fewer defects in carbon shell provide higher protection efficiency. Accordingly, carbon shells order/disorder structure was analyzed by RAMAN spectroscopy. The D-band of metal/C nanoparticles is considered to be a disorder-induced feature owing to lattice distortion or amorphous carbon background signal [55]. The peak intensity ratio is often used to determine the extent of structural disorder in graphite and/or the size of the graphitic domains [56]. Thus, the decoupled are considered. and peaks are located near ~1350 and ~1600 cm−1, respectively. RAMAN spectra peak−height ratios were determined from Lorentzian fittings as the example spectra shown in Figure 6(a). The carbon shells crystallinities degree () was obtained for each sample accordingly and was less than one for all samples, indicating high carbon crystallinity. Comparison of of each sample in Figure 6(b) reveals slight decrease in ratio as more nitrogen is used. This means that the graphitization degree of the samples increases slightly with increasing nitrogen gas concentration. Zhao et al. studied carbon nanostructures production by AC arc discharge plasma process at atmospheric pressure and concluded that the value of decreases from 1.2 to 0.6 with the increasing arc discharge current [57]. In their study, the temperature has been increased by increasing arc discharge electric current, while in our study the electric current was kept constant and the temperature was regulated according to the gas mixture of N2/He at near atmospheric pressure. Nitrogen exhibits higher plasma temperature in comparison with helium plasma; the lowest corresponds to N100 sample ( ~ 0.81), when only nitrogen is used.

Afterwards, the magnetic properties were characterized by superconducting quantum interference device (SQUID). CEINPs and CEMINPs magnetic behaviors were investigated at 300 K. Figure 7(a) represents comparison of normalized hysteresis curve for N100, N75, N50, N25, and N0 samples. The saturation magnetization for the nanocrystalline ferrites, in general, is found to be lower compared to their bulk value, which is attributed to surface spin effects [58]. Saturation magnetization largely depends on size, shape, metal composition, and crystalline magnetic anisotropy energy, as well as coating by nonmagnetic materials [52, 59]. N0 sample exhibits higher saturation magnetization than samples generated in presence of nitrogen. Therefore, from morphological point of view, as the iron cores are reforming from spherical shape to smaller oval shape, the saturation magnetization is decreasing. Size dependency of saturation magnetization is reported in the literature [6062]. Interestingly, in spite of smaller iron cores, the multi-iron cores nanoparticles (N100 sample) exhibit higher saturation magnetization than single oval shape nanoparticles in N75 sample. Jafari et al. studied the effect of carbon shell on the structural and magnetic properties of Fe3O4 and concluded that the magnetization of Fe3O4 nanoparticles was reduced after coating with carbon [63]. Bittova et al. carried out a research study on the effects of coating on dipolar interparticle interactions and showed the reduction of saturation magnetization after coating as well [64]. In our samples, the single iron cores are mainly located in the center of a large carbon shell while the multi-iron cores are distributed in a carbon shell; therefore probably carbon shell has less negative magnetic effect to multi-iron cores comparatively due to higher average distance between center iron cores. Increasing the saturation magnetization of magnetic nanoparticles may permit more effective development of multifunctional agents for simultaneous targeted cell delivery, magnetic resonance imaging contrast enhancement, and targeted cancer therapy in the form of local hyperthermia [65]. Zero-field-cooling (ZFC) and field-cooling (FC) curves exhibit the main attribute of superparamagnetic system. ZFC and FC magnetization from N100 sample are plotted in Figure 7(b). All the samples were in superparamagnetic state at room temperature. Multi-iron cores nanoparticles blocking temperature () value (30 K) is lowest among the others, followed by N75, N50 43 K, N25, and N0 with 36 K, 43 K, 45 K, and 55 K, respectively. The values versus median size from size distribution analysis are plotted in Figure 7(c). The finding that decreases as the size of the nanoparticles decreases is consistent with conventional Stoner-Wohlfarth theory [66]: the energy barrier, over which the magnetization of the nanoparticles should be thermally activated, increases as the size of the nanoparticles increases. Consequently, the lowest value of CEMINPs confirms the existence of smaller iron nanoparticles comparing to other samples. The sum of multi-iron cores in each carbon shell is greater than a single oval shape in N75 sample but its value is lower. This is an interesting characteristic of CEMINPs. The result of low value opens the possibility of using superparamagnetic particles in applications that required high standard and needs to operate at very low temperatures such as sensors in aerospace industry. Moreover, magnetic properties results of presented CEMINPs and CEINPs are important because in principle interests in magnetic particles are all due to the possibility to governing, heating, and detecting them by external magnetic field for biomedical applications.

4. Conclusion

Mono- and multi-iron nanoparticles encapsulated in carbon shell were controllably synthesized using a new modified arc discharge reactor. Influence of nitrogen on plasma evidenced higher concentration of carbon species and changes the morphologies of CEINPs and CEMINPs. Moreover, the nature of gas has influences on yield of nanoparticles; in particular, nitrogen plasma provided highest yield in our study. The diffraction points of iron oxide were not detected from SAED and FFT techniques. In addition, EELS analysis at nanoscale and EDX analysis at microscale showed no trace of oxygen in iron cores; hence samples were well protected by carbon shells. Based on the Raman spectroscopy results, best carbon crystallinity is observed when only nitrogen is used. Sufficient iron core protection along with low defects of carbon shell structure promotes very small pores and therefore minimizes the possibility of iron oxidation, degradation, and/or possible toxicity; hence both CEINPs and CEMINPs are suitable for highly chemically active environment. All the particles show a superparamagnetic behavior at room temperature as determined by SQUID measurements. Lowest blocking temperature of 30 K evidenced the synthesis of smaller multicores than monocore nanoparticles. Interestingly, although multi-iron cores are smaller than iron core in CEINPs (N75), they exhibit higher saturation magnetization; one of the reasons can be the existence of a few iron cores in a single carbon shell and therefore carbon shell has lower negative effect on magnetic properties. Owing to the extremely close distance (around 2 nm) between iron cores in CEMINPs, there will be higher visibility comparing to CEINPs and therefore this type of nanoparticles has more priority to be used as contrast agent. On the other hand it is noteworthy to mention that, due to the carbon shell spherical shape and iron superparamagnetic behavior, their movement in body fluid can be smooth and controllable and hence the CEINPs can be potentially used in drug delivery. It is concluded that the gas nature of the reactor plasma used in the research project has significant effect on the morphological properties of CEINPs. Accordingly, by this method, CEINPs and CEMINPs can be synthesized based on the desired applications.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.


The authors acknowledge Centres Científics i Tecnològics of Universitat de Barcelona (CCIT-UB). This work was partially financed by the MICYN of Spanish Government, under Contract MAT2010 20468, and by AGAUR of Generalitat de Catalunya, under Contracts 2009SGR00185 and 2014SGR0948.

Supplementary Materials

By increasing nitrogen concentration over helium, carbon structure formed on anode electrode as can be seen in Figure S1. This deformation has prevented continuous synthesis of CEINPs. The SEM image of the carbon structure on anode is provided in Figure 3 and related growth mechanism is described in the manuscript. It should be noted that electrodes deformation effects on the formation and size distributions of CEINPs and CEMINPs. In order to evaluate the size distribution of oval shape iron cores, the equivalent volume of sphere radius was considered. Accordingly the size distribution for all samples are plotted in Figure S2.

  1. Supplementary Material