Highly stable silica-coated manganese ferrite nanoparticles as high-efficacy T2contrast agents for magnetic resonance imagingAshfaq Ahmad, Hongsub Bae, and Ilsu Rhee

Citation: AIP Advances 8, 055019 (2018); doi: 10.1063/1.5027898View online: https://doi.org/10.1063/1.5027898View Table of Contents: http://aip.scitation.org/toc/adv/8/5Published by the American Institute of Physics

AIP ADVANCES 8, 055019 (2018)

Highly stable silica-coated manganese ferritenanoparticles as high-efficacy T2 contrastagents for magnetic resonance imaging

Ashfaq Ahmad, Hongsub Bae, and Ilsu Rheea

Department of Physics, Kyungpook National University, Daegu 41566, Korea

(Received 7 March 2018; accepted 9 May 2018; published online 18 May 2018)

Highly stable silica-coated manganese ferrite nanoparticles were fabricated for appli-cation as magnetic resonance imagining (MRI) contrast agents. The manganese ferritenanoparticles were synthesized using a hydrothermal technique and coated with sil-ica. The particle size was investigated using transmission electron microscopy andwas found to be 40–60 nm. The presence of the silica coating on the particle surfacewas confirmed by Fourier transform infrared spectroscopy. The crystalline structurewas investigated by X-ray diffraction, and the particles were revealed to have aninverse spinel structure. Superparamagnetism was confirmed by the magnetic hys-teresis curves obtained using a vibrating sample magnetometer. The efficiency ofthe MRI contrast agents was investigated by using aqueous solutions of the par-ticles in a 4.7 T MRI scanner. The T1 and T2 relaxivities of the particles were1.42 and 60.65 s-1 mM-1, respectively, in water. The ratio r2/r1 was 48.91, con-firming that the silica-coated manganese ferrite nanoparticles were suitable high-efficacy T2 contrast agents. © 2018 Author(s). All article content, except whereotherwise noted, is licensed under a Creative Commons Attribution (CC BY) license(http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1063/1.5027898

I. INTRODUCTION

Magnetic resonance imaging (MRI) is an accurate diagnostic imaging technique that can generatethree-dimensional anatomical images of the human body without using any harmful radiation likethat used in X-ray and other techniques.1–3 MR images are generated by the excitation and energyrelease of hydrogen protons inside the body in a strong magnetic field. The nuclear spins (hydrogenprotons) have different relaxation times depending on the water content and surroundings of theimaging. Two different relaxation times, T1 and T2, are used in MRI.4 T1 mainly depends on therate of energy transfer from the nuclear spins to the neighboring molecules, and T2 depends on thedephasing process of the nuclear spins caused by the neighboring magnetic inhomogeneity. High-contrast MRI is achieved by introducing contrast agents to decrease the relaxation times of the nuclearspins in tissues. In this process, the tissue-dependent uptake of the contrast agents is considered.5,6

Owing to the lack of reticuloendothelial systems (for example, Kupffer cells in the liver) in cancertissues, these tissues cannot capture the contrast agents, for example, magnetic nanoparticles, andrecognize them as external invaders. Thus, a T1 contrast agent accelerates T1 relaxation at a normaltissue site, which strengthens the signal from that site, thus providing a brighter image than thatof the cancer site. On the other hand, a T2 contrast agent accelerates T2 relaxation in a normaltissue, which weakens the MR signal at that site, whereas the MR signal at the cancer site remainsunaltered. Therefore, a T2 contrast agent can be used to distinguish between cancer and normaltissues.

Gadolinium-based contrast agents, such as Gd-diethylenetriaminepentacetate (DTPA), havebeen widely used since MRI was first applied as a clinical diagnostic technique. Gadolinium is

ailrhee@knu.ac.kr

2158-3226/2018/8(5)/055019/9 8, 055019-1 © Author(s) 2018

055019-2 Ahmad, Bae, and Rhee AIP Advances 8, 055019 (2018)

a paramagnetic material that accelerates the relaxation of nuclear spins.7,8 However, it is toxicand therefore chelated with DTPA to eliminate the toxicity. Gadolinium-based contrast agents arewidely used as T1 contrast agents to increase the contrast of the bright images, indicating nor-mal tissues, in MRI. Paramagnetic manganese in the form of Mn-dipyridoxyl diphosphate (DPDP)and Mn-ethylenediaminetetraacetic acid (EDTA)-PP is also used as a contrast agent.9,10 In Mn-DPDP, toxic manganese is chelated with DPDP, while manganese is inserted into liposomes inMn-EDTA-PP.

Since the late 1990s, iron oxide-based nanoparticle contrast agents have been investigated andclinically used as T2 contrasts agents. They compose of magnetic nanoparticle core and biocom-patible coating material, which prevents aggregation and sedimentation, and thereby allows highbiological tolerance.11 The magnetic particles are toxic and therefore must be coated with biocom-patible materials. Various coating materials are used for commercial nanoparticle contrast agents,such as dextran (Feridex, Resovist, Sinerem, Feraheme), polyethylene glycol (PEG; Clariscan), andsilica (GastroMARK). The coated particles have a diameter of a few tens of nanometers (see Table Ibelow). Moreover, studies for developing new T2 magnetic nanoparticle contrast agents coated withvarious biocompatible material have been conducted to achieve better contrast performance withrelatively low doses.12–20

Recently, chemical exchange saturation transfer (CEST) contrast agents have been activelyresearched.21–23 These contrast agents contain a material that provides exchangeable protons. Thesaturation magnetization of the proton can be transferred to the neighboring water molecules if aradio frequency (RF) wave absorbable by the proton is applied. Due to this magnetization trans-fer, T2 relaxation of the nuclear spins is accelerated. One such contrast agent, PARACEST, iscomposed of a paramagnetic element (lanthanide element) and a CEST (sugar or amino acid)material. In this combination, the paramagnetic element provides the chemical shift and the CESTmaterial provides the exchangeable proton. The main advantages of this contrast agent are that itallows on-off function of the contrast enhancement because the proton exchange is initiated by RFwaves, and that the RF wavelength can be controlled by choosing various paramagnetic and CESTmaterials.

Recently, researchers have focused on the use of manganese ferrite nanoparticles as MRI contrastagents due to their high magnetic susceptibility.24,25 Several groups have investigated manganese-based nanoparticles as an alternative to gadolinium for reducing the risk of toxicity.26 Manganesealso possesses a high spin quantum number and proton exchange kinetics. As mentioned above, somemanganese-based complexes such as Mn-DPDP have already been approved for clinical use.27

In the present study, manganese ferrite nanoparticles were synthesized by a hydrothermal route.Silica was employed as a coating material to ensure high stability and biocompatibility of the man-ganese ferrite nanoparticles. The morphological and crystalline structure and the magnetic propertiesof the nanoparticles were investigated using various analytical tools. The applicability of theseparticles as T2 contrast agents was also confirmed.

TABLE I. Relaxivities of prepared nanoparticles and commercial MRI contrast agents.

Company orContrast agent Coating material Particle size [nm] r1 [s-1 mM-1] r2 [s-1 mM-1] r2/r1 reference

Mn-ferrite Silica 40–60 1.24 (4.7 T) 60.65 (4.7 T) 48.91 This workMn-ferritin protein 14 10 (3 T) 74 (3 T) 7.4 45

Mn-ferrite 2.2 6.61 (4.7 T) 35.92 (4.7 T) 5.43 25

Mn-ferrite TEG 7 189.3 (0.5 T) 46

consideringonly iron (mM)

Mn-ferrite PEG 12 7.4 (1.5 T) 174 (1.5 T) 23.51 47

Gd-DTPA Gd-chelated 4.8 (4.7 T) 5.3 (4.7 T) 1.10 Bayer-Schering25

Mn-DPDP Mn-chelated 2.3 (1.5 T) 4.0 (1.5 T) 1.74 GE HealthcareFeridex (iron oxide) Dextran 80–150 7 (1.5 T) 66 (1.5 T) 9.43 AMAG Pharm.Resovist (iron oxide) Carboxydextran 45–60 8.7 (1.5 T) 61 (1.5 T) 7.01 Bayer Schering

055019-3 Ahmad, Bae, and Rhee AIP Advances 8, 055019 (2018)

II. EXPERIMENTAL

MnCl2·6H2O, FeCl3·6H2O and NaOH were purchased from Sigma Aldrich and were used asreceived without further modification. Deionized water was used throughout the reaction. The man-ganese ferrite nanoparticles were synthesized using a one-step hydrothermal technique in a 100-mLTeflon-lined stainless-steel reactor with a 60% filling degree. In brief, stoichiometric amounts ofmetal chloride precursors were dissolved with 20 mL deionized water in a 100-mL beaker using amagnetic stirrer, followed by the addition of 0.5 M NaOH solution until the pH reached 11, at whichpoint the color of the mixture turned brown. The solution was continuously stirred for 1 h at roomtemperature under argon atmosphere to obtain a homogenous suspension. Then, this homogeneousmixture was transferred into a Teflon-lined stainless-steel reactor and kept in a vacuum oven (JEIOTech OV-11-120) at 160 ◦C for 12 h. The solution was then naturally cooled to ambient temperaturein a vacuum oven. The obtained sample was magnetically decanted using a permanent magnet toseparate the unreacted precursors. The washing cycle was repeated several times by dispersing thenanoparticles in water and ethanol and using a magnet to separate the product from the solutionand was continued until complete removal of the unreacted products. Finally, the nanoparticles weredispersed in 50 ml water for silica coating.

Silica coating was performed using a previously reported technique.28 In brief, 60 mg of themanganese ferrite nanoparticles was dispersed in water and placed in an ultrasonic bath for 20 minto obtain a good dispersion. Subsequently, the particles were dispersed in ice-cold 0.1 M nitricacid in an ultrasonic bath followed by separation of the nanoparticles by centrifugation. Then, theparticles were re-dispersed in a 1 M citric acid solution. The particles were centrifuged again andalkalized by introducing a few drops of ammonia until pH reached 11, and then dispersed in water.The water suspension of the particles was added to a mixture of ethanol (150 mL), water (35 mL),and ammonia (7 mL) followed by addition of 0.6 mL of tetraethyl orthosilicate while stirring andcontinuously stirred at 40 ◦C for 6 h. A highly stable suspension of the silica-coated manganeseferrite nanoparticles was obtained and separated by high-speed (13500 rpm) centrifugation and thendispersed in water for further characterization.

The crystalline structure of the nanoparticles was investigated by X-ray diffraction (XRD; X’pertPRO, PANalytical). The shape and size of the nanoparticles were examined using transmission elec-tron microscopy (TEM; HT 7700, Hitachi Ltd). The chemical composition analysis was performedusing inductively coupled plasma (ICP) spectroscopy (IRISAP, Thermo Jarrell Ash). The concen-tration of the nanoparticles in the aqueous solution was also obtained by ICP. A vibrating samplemagnetometer (VSM; MPMS, Quantum Design) was used to measure the magnetic properties ofthe nanoparticles. Fourier transform infrared (FTIR; Nicolet 380, Thermo Scientific USA) measure-ments were performed to evaluate the status of the coating. MRI contrast effects were observed using a4.7 T MRI system (Bruker Biospec 47/40).

III. RESULTS AND DISCUSSION

The TEM images of the silica-coated manganese ferrite nanoparticles are shown in Fig. 1. Wecan see that cores consist of multiple manganese ferrite particles, and measure less than 15 nm indiameter. Average diameter of coated particles was about 40 nm as seen in the inset of Fig. 1, whichis the histogram of 100 particles obtained from TEM images.

XRD patterns of the bare particles are shown in Fig. 2. The (220), (311), (222), (400), (422),(511), and (440) crystal planes in the XRD patterns matched with those of the inverse spinel ferritecrystalline structure [JCPDS card number 10-0319].

FTIR spectra of the silica-coated and bare manganese ferrite nanoparticles are shown inFig. 3. The broad band at 3400 cm-1 corresponded to the O–H group stretching, while the band at1620 cm-1 was attributed to the vibration of the O–H group present on the surface of the samples.29,30

In addition, the absorption band at 1100 cm−1 was the characteristic peak of the anti-symmetricstretching vibrational mode of Si–O–Si siloxane bridges. The absorption peak at 950 cm-1 was dueto the contribution of the Si–O–H stretching vibration,31 while the band at 800 cm-1 was due tothe SiO4 ring vibration.32 The band at about 600 cm-1 corresponded to the Fe–O stretching in the

055019-4 Ahmad, Bae, and Rhee AIP Advances 8, 055019 (2018)

FIG. 1. TEM images of silica-coated manganese ferrite nanoparticles. Scale bar: (a) 500 nm and (b) 100 nm.

FIG. 2. XRD patterns of manganese ferrite nanoparticles. The indices of the crystal planes in the figure match with those ofan inverse spinel ferrite structure.

FIG. 3. FTIR spectra of bare and silica-coated manganese ferrite nanoparticles. The absorption bands correspond to thestretching and vibration modes of specific chemical bonds.

055019-5 Ahmad, Bae, and Rhee AIP Advances 8, 055019 (2018)

Fe–O–Si bond.33 In summary, the FTIR spectra confirmed the bonding of the silica to the surface ofthe manganese ferrite nanoparticles.

The magnetic hysteresis curves of the silica-coated and bare manganese ferrite nanoparticles atroom temperature are shown in Fig. 4. The saturation magnetization values for the coated and bareparticles were 10 and 7.5 emu/g, respectively, which were much lower than the corresponding bulkvalues of 80 emu/g.34,35 On the other hand, the coercive forces for the coated and bare particles were20 and 35 Oe, respectively, showing the superparamagnetic behavior. The coercive force also showeda symmetric behavior, indicating no exchange bias. The small values of saturation magnetizationfor the nanoparticles could be explained by some important factors, such as cationic distribution,impurity phase, and surface spin structure, determining the magnetic properties.34,36 In manganeseferrite, the cations Mn and Fe were distributed at the tetrahedral A sites and octahedral B sitesaccording to the inverse spinel crystalline structure. Some authors suggested that the redistribution ofmanganese and iron in the octahedral and tetrahedral sites upon annealing strengthened the exchangeinteraction, thus increasing the saturation magnetization.37 Researchers analyzed the manganeseferrite nanoparticles by neutron diffraction experiments and Mossbauer spectroscopy and found thatthe cationic distribution was related to the magnetic properties.38–40 They observed that particle sizeplayed an important role in the cationic distribution at the lattice sites of manganese ferrite.41 Lowsaturation magnetization similar to that observed in the present study was previously reported.37,38,42

The surface effects and the magnetic dead layer on the nanoparticle surface due to the large surface-to-volume ratio in the nanoparticle systems were found to be responsible for the saturation magnetizationof the nanoparticles being lower than the bulk value.

In the silica-coated nanoparticles, the saturation magnetization was lower than that of the bareparticles, whereas the coercive force was higher. This behavior was expected, since the silica–metalbond on the surface of the nanoparticles reduced their magnetic moment, leading to a decrease in themagnetization and an increase in the coercive force.43

For relaxation time measurements, six aqueous samples with different particle concentrationsranging from 0.13 to 0.92 mM were prepared. The amount of iron and manganese in the liquid wasdetermined using ICP spectroscopy. The liquid sample retained its colloidal suspension over a fewweeks before the relaxation measurements without any precipitates. T1 and T2 of the nuclear spinsin the aqueous solutions of the magnetic particles were measured using an MR scanner.

For T1 measurements, the inversion recovery pulse sequence was used. In this sequence, thesignal intensity is expressed as a function of T1 as follows:

I ∼M0(1 − 2e−t

T1 ). (1)

FIG. 4. Magnetic moments of bare and silica-coated manganese ferrite nanoparticles as a function of the applied magneticfield. The particles exhibit superparamagnetic behavior with negligible coercive force.

055019-6 Ahmad, Bae, and Rhee AIP Advances 8, 055019 (2018)

FIG. 5. T1 for three representative samples. It can be observed that T1 relaxation is faster for the sample with a larger particleconcentration.

By comparing the MR intensities with Eq. 1, we could obtain the T1 values for samples with threerepresentative particle concentrations, as shown in Fig. 5.

The Carr–Purcell–Meiboon–Gill multiple spin-echo pulse sequence was used for the T2

measurements. Signal intensity as function of T2 is expressed as follows:

I ∼Moe−t

T2 (2)

This relationship was used to determine the T2 values. Fig. 6 shows the T2 values for samples withthree representative particle concentrations.

The dependence of the MR image intensities on the particle concentration for T1 and T2 relax-ations is shown in Fig. 7. The particle concentration increased as we moved from left to right. ForT1 relaxation, the MR images became brighter with increasing particle concentration. On the otherhand, for T2 relaxation, the MR images become darker with increasing particle concentration.

Relaxivity is a measure of the ability of an MRI contrast agent to increase the relaxation ofthe surrounding nuclear spins and can be used to improve the contrast of MR images. Relaxivity isexpressed in the units of s-1 per mM of magnetic components (iron and manganese) in the liquid.

FIG. 6. T2 relaxation for three representative samples. It can be observed that T2 relaxation is faster for the sample with alarger particle concentration.

055019-7 Ahmad, Bae, and Rhee AIP Advances 8, 055019 (2018)

FIG. 7. Concentration dependence of the MR image intensity at a fixed imaging time. It can be observed that both T1 and T2relaxations are faster for the sample with a higher particle concentration.

The relaxivities (1/Tim) of the nuclear spins in an aqueous solution of the magnetic nanoparticlescan be expressed as44

1Tim=

1Ti

+ riC, (3)

where i = 1 or 2, 1/T i represents the relaxivity of the nuclear spins without the nanoparticle contrastagent, ri is the relaxivity of the nuclear spins per mM of magnetic components in the liquid, and Crepresents the concentration of nanoparticles in the aqueous solution.

The plots of 1/T1 versus particle concentration is shown in Fig. 8. In this figure, the slope givesthe T1 relaxivity, which is denoted as r1. The value of r1 was determined to be 1.24 s-1 mM-1. Theplots of 1/T2 versus particle concentration is shown in Fig. 9. The value of r2, the T2 relaxivity,was 60.65 s-1 mM-1. A contrast agent with a large relaxivity can give the same contrast effect witha lower dose than that with a small relaxivity. Also, the ratio r2/r1 is an indicator of the efficiencyof T2 contrast agents. A higher value of this ratio represents better T2 contrast efficacy. The r2/r1

value calculated for our particles was 48.91, which was much larger than that of previously reportedparticles, as shown in Table I.

For comparison, the relaxivities of the manganese-based nanoparticles and commercial contrastagents are summarized in Table I. Sana et al.45 synthesized manganese-ferritin nanocomposites andmeasured the relaxivities using a 3 T MRI scanner. The r1 and r2 values of their particles were 10and 74 s-1 mM-1, respectively. Thus, the ratio r2/r1 was 7.4, indicating that these particles couldbe applicable as T2 contrast agents. Li et al.25 reported ultrasmall manganese ferrite nanoparticleswith r1 and r2 values of 6.61 and 35.92 s-1 mM-1, respectively, at 4.7 T. The particle size of these

FIG. 8. Plot of 1/T1 as a function of particle concentration. The slope of the straight line gives T1 relaxivity of the nanoparticles.

055019-8 Ahmad, Bae, and Rhee AIP Advances 8, 055019 (2018)

FIG. 9. Plot of 1/T2 as a function of particle concentration. The slope of the straight line gives T2 relaxivity of the nanoparticles.

nanoparticles was 2.2 nm. Yang et al.46 synthesized tetraethylene glycol (TEG)-coated manganeseferrite nanoparticles with an average diameter of 7 nm. The T2 relaxivity of the particles dispersedin water at 0.5 T was 189.3 s-1 mM-1, while the T1 relaxivity of the particles in the cells was36.8 s-1 mM-1. The value of T2 relaxivity of the particles in water was much larger than that of otherreported particles. This was because only the amount of iron was considered while calculating therelaxivity. Leal et al.47 studied the particle size and magnetic field dependence of the relaxivities.They synthesized PEG-coated manganese ferrite nanoparticle with various diameters of 6, 7.5, 9, 12,and 14 nm and observed the T1 and T2 relaxivities using 1.5 T and 9 T MRI scanners. They foundthat both r1 and r2 strongly depended on the particle size and magnetic field. The largest r2/r1 valuewas found to be 152 for the 9-nm particles using the 9 T scanner. L. Yang et al. observed effect ofmanganese content on magnetization and relaxivity of manganese ferrite particles. Magnetization ofthe particles increased as manganese content increased up to 0.43; however, magnetization decreasedupon further manganese doping. They also measured relaxivity at two different fields of 0.5 and7 T. At 7 T, T2 relaxivity increased sharply, demonstrating the effect of high magnetic moment ofmanganese on T2 relaxivity.18 In Table I, the relaxivities of iron oxide-based commercial T2 contrastagents are also provided. Their r2/r1 values were lower than 10, but this value is much larger than thatof Gd-DTPA and Mn-DPDP contrast agents. Thus, commercial iron oxide-based contrast agents areprimarily used as T2 contrast agents. Comparing the relaxivity data of our manganese particles withprevious results, we concluded that our particles can be used as high-efficacy T2 contrast agents.

IV. CONCLUSION

We prepared highly stable silica-coated manganese ferrite nanoparticles and revealed their MRIcontrast abilities. The biocompatible silica coating resulted in highly stable aqueous solutions ofmanganese ferrite nanoparticles without any precipitation for long times. The particles showed super-paramagnetic behavior suitable for MRI contrast agents. The particles in the aqueous solution revealedthe strong effects for accelerating the T2 relaxation of nuclear spins, indicating that these particlesare suitable as T2 contrast agents for MRI. Furthermore, the r2/r1 value for the prepared particleswas much higher (49.81) than that of other reported nanoparticles. These results confirmed that oursilica-coated manganese ferrite nanoparticles could be used as high-efficacy T2 contrast agents.

ACKNOWLEDGMENTS

This research was supported by the National Research Foundation of Korea (2016R1A2B1006449).

055019-9 Ahmad, Bae, and Rhee AIP Advances 8, 055019 (2018)

1 E. R. Andrew, Nuclear magnetic resonance (Cambridge University Press, Cambridge, UK, 2009).2 P. C. Lauterbur, Nature 242, 190 (1973).3 W. P. Aue, E. Bartholdi, and R. R. Ernst, J. Chem. Phys. 64, 2229 (1976).4 I. Rhee, New Physics: Sae Mulli 65, 411 (2015).5 X. Liu, Y. Chen, H. Li, N. Huang, Q. Jin, K. Ren, and J. Ji, ACS Nano 7, 6244 (2013).6 H. Gao, Z. Yang, S. Zhang, S. Cao, S. Shen, Z. Pang, and X. Jiang, Sci. Rep. 3, 2534 (2013).7 D. H. Carr, J. Brown, G. M. Bydder, R. E. Steiner, H. J. Weinmann, U. Speck, A. S. Hall, and I. R. Young, Am. J. Roentgenol.

143, 215 (1984).8 J. P. Stack, N. M. Antoun, J. P. Jenkins, R. Metcalfe, and I. Isherwood, Neuroradiology 30, 145 (1998).9 C. Wang, Acta. Radiol. Suppl. 415, 1 (1998).

10 S. G. Silva, J. A. Nobrega, and F. R. P. Rocha, Microchem. J. 98, 109 (2011).11 Contrast Media: Safety Issues and ESUR Guidelines, edited by H. S. Thomsen and J. A. W. Webb (Springer Science &

Business Media, 2009).12 A. Senpan, S. D. Caruthers, I. Rhee, N. A. Mauro, D. Pan, N. A. Mauro, G. Hu, M. J. Scott, R. W. Fuhrhop, P. J. Gaffney,

S. A. Wickline, and G. M. Lanza, ACS Nano 3, 3917 (2009).13 H. Bae, T. Ahmad, I. Rhee, Y. Chang, S. Jin, and S. Hong, Nano. Res. Lett. 7, 44 (2012).14 T. Ahmad, H. Bae, Y. Iqbal, I. Rhee, S. Hong, Y. Chang, J. Lee, and D. Sohn, J. Magn. Magn. Mater. 381, 151 (2015).15 D. E. Sosnovik, M. Nahrendorf, and R. Weissleder, Basic Res. Cardiol. 103, 122 (2008).16 J. Estelrich, M. J. Sanchez-Martın, and M. A. Busquets, Int. J. Nanomedicine 10, 1727 (2015).17 T. Macher, J. Totenhagen, J. Sherwood, Y. Qin, D. Gurler, M. S. Bolding, and Y. Bao, Adv. Funct. Mater. 25, 490 (2015).18 L. Yang, L. Ma, J. Xin, A. Li, C. Sun, R. Wei, B. W. Ren, Z. Chen, H. Lin, and J. Gao, Chem. Mater. 29, 3038 (2017).19 P. Sharmiladevi, V. Haribabu, K. Girigoswami, A. S. Farook, and A. Girigoswami, Scientific Reports 7, 011179 (2017).20 Q. L. Vuong, J. F. Berret, J. Fresnais, Y. Gossuin, and O. Sandre, Adv. Healthc. Mater. 1, 512 (2012).21 K. M. Ward, A. H. Aletras, and R. S. Balaban, J. Magn. Reson. 143, 79 (2000).22 A. D. Sherry and M. Woods, Annual Rev. Biomed. Eng. 10, 391 (2008).23 A. N. Dula, S. A. Smith, and J. C. Gore, J. Neuroimaging 23, 526 (2013).24 J. Lu, S. Ma, J. Sun, C. Xia, C. Liu, Z. Wang, X. Zhao, F. Gao, Q. Gong, and B. Song, Biomaterials 30, 2919 (2009).25 Z. Li, S. X. Wang, Q. Sun, H. L. Zhao, H. Lei, M. B. Lan, Z. X. Cheng, X. L. Wang, and S. X. Dou, Adv. Healthc. Mater.

2, 958 (2013).26 E. M. Gale, H.-Y. Wey, I. Ramsay, Y.-F. Yen, D. E. Sosnovik, and P. Caravan, Radiology 286, 865 (2018).27 E. M. Gale, I. P. Atanasova, F. Blasi, I. Ay, and P. Caravan, J. Am. Chem. Soc. 137, 15548 (2015).28 A. Ahmad, H. Bae, and I. Rhee, AIP Advances 8, 015108 (2018).29 L. Chen, Z. Xu, H. Dai, and S. Zhang, J. Alloys Compd. 497, 221 (2010).30 S. Bruni, F. Cariati, M. Casu, A. Lai, A. Musinu, G. Piccaluga, and S. Solinas, Nanostructured Mater. 11, 573 (1999).31 B. Mojic, K. P. Giannakopoulos, Z. Cvejic, and V. V. Srdic, Ceram. Int. 38, 6635 (2012).32 J. Lewandowska-Łancucka, M. Staszewska, M. Szuwarzynski, M. Kepczynski, M. Romek, W. Tokarz, A. Szpak, G. Kania,

and M. Nowakowska, J. Alloys Compd. 586, 45 (2014).33 M. K. Yoo, I. Y. Kim, E. M. Kim, H. J. Jeong, C. M. Lee, Y. Y. Jeong, T. Akaike, and C. S. Cho, J. Biomed. Biotechnol.

2007, 94740.34 J. Chen, C. Sorensen, K. Klabunde, G. Hadjipanayis, E. Devlin, and A. Kostikas, Phys. Rev. B 54, 9288 (1996).35 M. Muroi, R. Street, P. McCormick, and J. Amighian, Phys. Rev. B 63, 184414 (2001).36 M. Goodarz Naseri, E. B. Saion, and A. Kamali, ISRN Nanotechnology 2012, 604241.37 M. Zheng, X. Wu, B. Zou, and Y. Wang, J. Magn. Magn. Mater. 183, 152 (1998).38 V. Becyte, K. Mazeika, T. Rakickas, and V. Pakstas, Mater. Chem. Phys. 172, 6 (2016).39 J. Li, H. Yuan, G. Li, Y. Liu, and J. Leng, J. Magn. Magn. Mater. 322, 3396 (2010).40 C. Liu, B. Zou, A. J. Rondinone, and Z. J. Zhang, J. Phys. Chem. B 104, 1141 (2000).41 A. Morrish, Z. Li, and X. Zhou, Le Journal de Physique IV 7, C1–513 (1997).42 N. M. Martins, L. M. Martins, C. O. Amorim, V. S. Amaral, and A. J. Pombeiro, Catalysts 7, 222 (2017).43 H. Ding, Y. Zhang, S. Wang, J. Xu, S. Xu, and G. Li, Chem. Mater. 24, 4572 (2012).44 Y. Okuhata, Adv. Drug Delivery Rev. 37, 121 (1999).45 B. Sana, C. L. Poh, and S. Lim, Chem. Commun. 48, 862 (2012).46 H. Yang, C. Zhang, X. Shi, H. Hu, X. Du, Y. Fang, Y. Ma, H. Wu, and S. Yang, Biomaterials 31, 3667 (2010).47 M. P. Leal, S. Rivera-Fernandez, J. M. Franco, D. Pozo, M. Jesus, and M. L. Garcıa-Martın, Nanoscale 7, 2050 (2015).