Magnetic studies of polylactic-co-glicolic acid nanocapsules loaded with selol and γ-Fe2O3 nanoparticles

Ewa Mosiniewicz-Szablewska *a, Antonio C. Tedesco b, Piotr Suchocki c and Paulo C. Morais de
aInstitute of Physics, Polish Academy of Sciences, Al. Lotników 32/46, 02-668 Warsaw, Poland. E-mail: mosin@ifpan.edu.pl
bDepartment of Chemistry, Center of Nanotechnology and Tissue Engineering – Photobiology and Photomedicine Research Group, Faculty of Philosophy, Sciences and Letters of Ribeirão Preto, University of São Paulo, Ribeirão Preto, São Paulo 14040-901, Brazil
cDepartment of Bioanalysis and Drug Analysis, Medical University of Warsaw, Warsaw, Poland
dGenomic Sciences and Biotechnology, Catholic University of Brasília, Brasília, DF 70790-160, Brazil
eInstitute of Physics, University of Brasília, Brasília, DF 70910-900, Brazil

Received 18th May 2020 , Accepted 24th August 2020

First published on 24th August 2020


The as-prepared (MSE-NCs sample) and lyophilized (LMSE-NCs sample) polylactic-co-glicolic acid (PLGA) nanocapsules loaded with maghemite (γ-Fe2O3) nanoparticles and selol (Se-based anticancer drug) were investigated by means of dc magnetization, ac susceptibility and electron spin resonance (ESR) measurements over the temperature range of 4–300 K. The magnetic data of the as-synthesized nanocapsules containing only maghemite nanoparticles (M-NCs sample) or selol (SE-NCs sample) were also collected for comparison. The magnetic nanocapsules reveal perfect superparamagnetic (SPM) behavior only around room temperature; at temperatures lower than 200 K the SPM scaling is not observed and all samples behave as interacting superparamagnetic (ISPM) materials. The evolution from the ISPM to the SPM regime is marked by a steady decrease in the hysteretic properties of all samples, with the temperature dependence of the coercivity decreasing slower than the T1/2 behavior predicted for non-interacting SPM particles. The SPM character of the samples is also confirmed by the occurrence of a maximum in the temperature dependence of both real χ′(T) and imaginary χ′′(T) components of the ac magnetic susceptibility, which shifts towards higher temperatures with increasing frequency. Moreover, upon decreasing the temperature the ESR signal shifts to lower fields and gradually broadens, following closely the predictions for the ESR of SPM particles. Additionally, an unusual giant diamagnetic response is observed at low temperatures. The ZFC magnetization is found to reverse its direction and becomes diamagnetic, whereas the FC branch remains positive. Even when compared with usual superconductors, the order of the diamagnetic susceptibility for the lyophilized sample (−10−2 emu g−1 Oe−1) is quite considerable. The nanocapsules herein reported and the presented analysis of their magnetic properties we envisage can support the engineering of magnetic nanocapsules for applications in magnetic drug delivery systems and as magnetic hyperthermia inductors in antitumor therapy.


Introduction

The development of a magnetic drug delivery system (DDS) which is able to deliver cytotoxic materials selectively to the neoplastic cells without affecting normal cells is currently one of the most active areas of cancer research. Among the many DDSs, polymeric magnetic nanocapsules (NCs) are proposed as systems containing both an active drug and magnetic nanoparticles (NPs).1 When magnetic NCs enter the bloodstream, external high-gradient magnetic fields can be used to concentrate them at the target tumor site within the body.2 Additionally, the activation by ac magnetic fields generates heat due to the Néel and/or Brown relaxation mechanism3,4 of magnetic NPs. Small NPs (10–20 nm) are used for magnetohyperthermia because of their ability to produce a significant level of heating in relatively low-amplitude ac magnetic fields.5 Magnetic NPs used for biomedical applications are mainly iron oxide NPs, such as magnetite (Fe3O4) or its more stable oxidized form of maghemite (γ-Fe2O3).6

It has been known for a long time that selenium (Se) plays an important role as an antioxidant and antitumor agent.7 In particular, selenium compounds containing Se at its 4+ oxidation state, such as in selenites, present the highest antioxidant and anticancer activities.8 However, Se(IV)-containing compounds generally present high systemic toxicity, limiting their clinical applications. In this context, a selenite-containing compound labelled selol was first synthesized from sunflower oil at the Medical University of Warsaw, Poland.9 Selol is a mixture of selenitetriglycerides, a semi-synthetic group of compounds containing Se(IV), and shows antitumor activity and low systemic toxicity.10,11 It appears to act primarily through the induction of oxidative stress in cancer cells.10 Interestingly, selol was shown to sensitize leukemia cells to the cytotoxicity of vincristine and doxorubicin, insomuch that it was suggested that selol could be used in combination with other drugs in chemotherapeutic protocols.11

Inspired by these findings, we directed our attention to the possibility of introducing selol into polymeric magnetic NCs in order to use them as a potential magnetic DDS for cancer therapy. We used polylactic-co-glicolic acid (PLGA), maghemite (γ-Fe2O3) NPs and selol to fabricate selol-loaded PLGA magnetic NCs (MSE-NCs sample).12 The in vitro antitumor activity of the MSE-NCs sample was evidenced using neoplastic murine melanoma (B16-F10),12,13 oral squamous carcinoma (OSCC),13 murine (4T1) and human (MCF-7) breast,14 and pulmonary adenocarcinoma (A549)15 cell lines. Further exposure of these cell lines to an alternating magnetic field increased the antitumor effect of the MSE-NCs. Moreover, the MSE-NCs sample presented the antitumor effect without affecting normal cells.14 Further investigations showed that the association of the MSE-NCs surface with folic acid increased its cell uptake (by specific tumors cell receptors) and toxicity against cancer cells.16 Additionally, MSE-NCs with the shell conjugated to doxorubicin showed a significantly increased cytotoxicity against breast adenocarcinoma (4T1) cells which is suggestive of additive or synergic effects of selol and doxorubicin.17,18

The results mentioned above suggest that the MSE-NCs sample should be considered as an effective material system for magnetic drug delivery and magnetic hyperthermia inductor in antitumor therapy. For such an application, the MSE-NCs sample should show not only antitumor effects but also SPM properties, thus allowing easy remote manipulation (employing the dc magnetic field gradient) and heating (employing an ac magnetic field), simultaneously increasing the functionality of the nanoplatform upon loading an anti-tumor cargo (such as doxorubicin as the chemotherapy agent). Considering that the pioneering synthesis of this type of magnetic nanocapsule was done by our team, their magnetic behavior may turn out to be equally interesting for designing future nanomaterials. Therefore, the present study reports on the magnetic properties of PLGA NCs loaded with maghemite NPs and selol (MSE-NCs sample) assessed by means of dc magnetization, ac susceptibility and ESR measurements. These magnetic studies are of considerable interest for not only the development of new and very effective methods of cancer treatment but also as new hybrid nanomaterials showing giant diamagnetism at low temperatures.

Experimental details

Preparation of the samples

Selol (5% of selenium) was provided by the Department of Bioanalysis and Drug Analysis from the Medical University of Warsaw, Poland.9

Polylactic-co-glycolic acid (PLGA) NCs, loaded with selol only (SE-NCs), maghemite (γ-Fe2O3) NPs only (M-NCs) and maghemite NPs and selol (MSE-NCs), were successfully prepared by the nanoprecipitation method as described in ref. 12. Maghemite NPs were incorporated into the nanocapsule using a highly-stable ionic magnetic fluid sample. The obtained NCs contain 10 mg mL−1 of selol and/or 0.5 × 1016 maghemite NPs per mL.

The microstructure of the samples was investigated by means of a scanning electron microscope (SEM) and a transmission electron microscope (TEM).12–14 The obtained MSE-NCs sample showed no agglomeration, and a negative surface charge while revealing a narrow monomodal size distribution with an average diameter 〈d〉 of 236 nm. The MSE-NCs sample presented an electron-dense core of maghemite NPs localized inside and also on the NCs’ surface. The maghemite NPs used to fabricate the MSE-NCs sample had a mean diameter of 10 nm. Unlike the MSE-NCs sample, NCs from the M-NCs sample were organized in clusters with maghemite NPs mainly on their surface.14 As for the SE-NCs sample, as in MSE-NCs, the nanocapsules presented a spherical shape and were individually distributed.14 All the NCs formulations exhibited good physical stability at 4 °C during the three-month storage period.

Magnetic characterization

All magnetic measurements were performed on the PLGA-based NCs containing selol only (SE-NCs sample), maghemite NPs only (M-NCs sample) or maghemite NPs and selol before (MSE-NCs sample) and after lyophilization (LMSE-NCs sample). Dc magnetization and ac magnetic susceptibility data were collected using a Quantum Design PPMS extraction magnetometer in a wide temperature range (4–300 K).

Thermal dependencies of dc magnetization in the zero field cooled–field cooled (ZFC–FC) regime were recorded with the applied magnetic field of 10, 50 and 100 Oe. The ZFC curves were obtained by first cooling the samples in a zero magnetic field from 300 to 4 K. Then the magnetic field H = 10 Oe, 50 Oe and 100 Oe was applied and the magnetization was measured with increasing temperature. The FC curves were obtained in a similar manner except that the samples were cooled in the same measuring field H = 10 Oe, 50 Oe and 100 Oe.

Magnetic hysteresis loops measurements were performed at selected temperatures in the applied magnetic field of ±20 kOe.

Temperature dependencies of ac magnetic susceptibility were recorded using an excitation field of 10 Oe and different driving frequencies in the range of 10–10[thin space (1/6-em)]000 Hz.

ESR spectra were collected by means of a standard X-band spectrometer (Bruker EMX – 10/12) operating around 9.46 GHz with 100 kHz field modulation. Resonance absorption was measured as a first derivative of the absorbed microwave power versus magnetic field.

Results and discussion

Dc magnetization measurements

Fig. 1 shows the ZFC and FC magnetization as a function of temperature under the applied dc magnetic fields (H) of 50 Oe and 100 Oe for the SE-NCs, M-NCs, MSE-NCs and LMSE-NCs samples. The SE-NCs sample shows a diamagnetic behavior, as expected. For the M-NCs, MSE-NCs and LMSE-NCs samples, the ZFC-FC curves present the typical profile of superparamagnetic (SPM) NPs, i.e. the splitting of the ZFC and FC curves below the irreversibility temperature around Tirr = 150 K and a well-defined maximum in the ZFC curve at the blocking temperature around TB = 126 K and 106 K for H = 50 Oe and 100 Oe, respectively. More or less the same TB values for all the samples indicate that the effective magnetic volume is identical for all these NCs. Tirr larger than TB is a fingerprint of inter-particle magnetic disorder and/or broad particle size distribution.19
image file: d0cp02706e-f1.tif
Fig. 1 Temperature dependencies of magnetization under the ZFC–FC protocol for the as-prepared PLGA NCs containing selol (SE-NCs sample), maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample). Insets show the dc susceptibility versus temperature measured for the applied magnetic fields (50 and 100 Oe).

The maximum of the ZFC curve is very broad and a clear Curie–Weiss law behavior is not observed above TB for all M-NCs, MSE-NCs and LMSE-NCs samples (see Fig. 1). This indicates the existence of the non-negligible magnetic dipole–dipole interaction among the maghemite NPs, which tends to increase the energy barriers of individual NPs and, therefore, enters the SPM regime at different temperatures. An increase in H leads to a shift of the ZFC peak to lower temperatures, accompanied by a slight decrease in the magnitude of the dc susceptibility (see insets in Fig. 1), whereas for non-interacting SPM particles, the magnitude should be field independent.

Below Tirr, the FC curve splits from the ZFC curve and increases very slowly with decreasing temperature, not following the Curie–Weiss law, which also reveals the presence of the non-negligible dipole–dipole interaction among the NPs.

Around 280 K a kink is seen in both MZFC(T) and MFC(T) curves for the M-NCs and MSE-NCs samples which is associated with the melting point of the suspension.

The effect of the inter-particle interaction can be investigated by means of an ordering temperature T0 calculated from the high temperature SPM regime. In the case of non-interacting SPM particles, it is expected that in this region the system should obey the Curie law χ = C/T. For an assembly of interacting SPM particles, the low field susceptibility is expected to be of the form:20

 
image file: d0cp02706e-t1.tif(1)
where μmean is the mean magnetic moment per particle, kB is the Boltzman constant and T0 is the effective temperature that is usually interpreted as a measure of the strength of the interparticle interaction in the system. In accordance with this equation, the reciprocal of the FC susceptibility shows a linear behavior well above TB when it is multiplied by mS2(T) = MS2(T)/MS2(0) in order to correct for the temperature dependence of μmean (see Fig. 2). The value of T0 is obtained by extrapolating this linear behavior to the temperature axis and is found to be −(262 ± 10) K, −(258 ± 10) K and −(302 ± 10) K for the M-NCs, MSE-NCs and LMSE-NCs samples, respectively. This fact further supports the existence of the strong dipolar inter-particle interaction with demagnetizing character. This interaction is stronger in the LMSE-NCs sample than in the M-NCs and MSE-NCs samples.


image file: d0cp02706e-f2.tif
Fig. 2 Details of the reciprocal of the FC magnetization as a function of temperature for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample). The vertical axis has been multiplied by mS2(T) in order to correct the temperature dependence of μmean in eqn (1).

At low temperature, an unusual diamagnetic behavior in the ZFC branch is observed in all M-NCs, MSE-NCs and LMSE-NCs samples (see Fig. 1). The initial magnetization starts with a significant negative value and increases monotonically upon heating the sample up to the compensation temperature (Tcomp) at which the magnetization direction flips positive. Tcomp decreases with increasing H from 38 K (H = 50 Oe) to 12 K (H = 100 Oe). Even compared with usual superconductors, the order of the diamagnetic susceptibility (−10−3 emu g−1 Oe−1 for the M-NCs and MSE-NCs samples and −10−2 emu g−1 Oe−1 for the LMSE-NCs sample) is quite remarkable.

Fig. 3 sums up the ZFC magnetization values at 4 K for all samples under an increasing magnetic field of H = 10 Oe, 50 Oe, and 100 Oe. The negative magnetization decreases with increasing H and finally disappears around H = 122 Oe and 133 Oe for the M-NCs plus MSE-NCs samples and for the LMSE-NCs sample, respectively. Above these values the ZFC process shows normal SPM behavior of magnetic NPs. It is worth noting that the negative magnetization is one order of magnitude higher for the LMSE-NCs sample than that for the M-NCs and MSE-NCs samples.


image file: d0cp02706e-f3.tif
Fig. 3 Negative ZFC magnetization at 4 K versus applied dc magnetic field for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample).

This giant diamagnetism observed at low temperatures and at low magnetic fields can be attributed to the antiferromagnetic dipolar interaction among maghemite NPs mentioned above. The dipolar interaction acts like an additional internal field (Hi) opposite to the positive applied field (H) during the measurements of the ZFC curve and hinders the response of the magnetic moments to the H field. By slightly increasing H, the magnetization becomes zero exactly at H opposite and equal in magnitude to Hi. Hence, the value of Hi, giving a direct quantitative information about the strength of dipolar interaction among NPs in the sample, can be assumed as equal to H requested for compensating the initial negative magnetization induced by the ZFC procedure. The ZFC magnetization curves shown in Fig. 3 lead to values for Hi of about 122 Oe in the case of the M-NCs and MSE-NCs samples and about 133 Oe in the case of the LMSE-NCs sample. These results clearly support the presence of the dipolar interaction of antiferromagnetic type which is stronger in the LMSE-NCs sample as compared to the M-NCs and MSE-NCs samples.

Fig. 4 shows the field dependent magnetization curves of the SE-NCs, M-NCs, MSE-NCs and LMSE-NCs samples obtained at room temperature. The magnetization curve of the SE-NCs sample shows a typical diamagnetic behavior, as expected. The shapes of the remaining magnetization curves, with zero value of both the remanence and coercivity, are typical of particles in the SPM state and can be described by the first-order Langevin function.21 To take into account the effects of size dispersion always present in any real nanomaterial system, the magnetization is better fitted to a weighted sum of Langevin functions.21 Additionally, in the case of the M-NCs, MSE-NCs and LMSE-NCs samples, this function has to be modified to include a diamagnetic contribution coming from selol and PLGA NCs. The magnetization curve of the system can therefore be described by:21

 
image file: d0cp02706e-t2.tif(2)
where T is the temperature, H is the applied magnetic field, a represents the diamagnetic contribution, kB is the Boltzman constant, μ = πMPSD3/6 is the magnetic moment of each particle, MPS is the particle magnetization density, MSS is the saturation magnetization of the sample. Additionally, f(D) in eqn (2) is the log-normal distribution function which, for spherical particles with diameter D, has the form:21
 
image file: d0cp02706e-t3.tif(3)
where 〈D〉 is the mean diameter and σ is the standard deviation of D. The value of the saturation magnetization MSS is extracted from the fitting of M versus 1/H and extrapolation of the magnetization to infinite field.


image file: d0cp02706e-f4.tif
Fig. 4 Magnetization curves measured at room temperature for the as-prepared PLGA NCs containing selol (SE-NCs sample), maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample). Symbols correspond to the experimental data, the solid red-lines represent the fits according to eqn (2). The insets show the resulting log-normal distribution functions of particle diameters.

Solid red-lines in Fig. 4 represent the best fits for the room-temperature magnetization data obtained with the following parameters: 〈D〉 = (9.8 ± 0.1) nm, σ = 0.35 ± 0.01, MPS = (396 ± 4) emu cm−3 and a = −(7.3 ± 0.1) × 10−11 emu Oe−1 for the M-NCs sample, 〈D〉 = (10.2 ± 0.1) nm, σ = 0.31 ± 0.1, MPS = (412 ± 4) emu cm−3 and a = −(7.1 ± 0.1) × 10−10 emu Oe−1 for the MSE-NCs sample, and 〈D〉 = (9.7 ± 0.1) nm, σ = 0.40 ± 0.01, MPS = (392 ± 4) emu cm−3 and a = (1.1 ± 0.1) × 10−8 emu Oe−1 for the LMSE-NCs sample. The estimated mean magnetic moment of the particle (μmean) is around (19[thin space (1/6-em)]505 ± 20) μB, (22[thin space (1/6-em)]881 ± 23) μB and (18[thin space (1/6-em)]723 ± 19) μB for the M-NCs, MSE-NCs and LMSE-NCs samples, respectively.

The resulting log-normal distribution of NPs’ diameters for the MSE-NCs sample is narrower than that for the M-NCs sample (see insets in Fig. 4), whereas the mean nanoparticle diameter is larger (about 4%). Moreover, the mean magnetic moment of the maghemite NPs extracted from the MSE-NCs sample is larger than the value extracted from the M-NCs sample. This is likely due to the partial clustering of maghemite NPs into chain-like structures within the MSE-NCs sample as opposed to more dispersed NPs inside the M-NCs sample. Furthermore, the abnormally lower (about 10%) diamagnetic contribution (parameter a) found in the MSE-NCs sample indicates that the chains of maghemite NPs settle down inside the PLGA NCs containing selol as opposed to the M-NCs sample where dispersed maghemite NPs are present mainly on the surface of the selol-free PLGA NCs. These conclusions are in good agreement with the TEM measurements presented earlier,14 showing that PLGA NCs from the MSE-NCs sample are individually distributed with maghemite NPs localized inside as well as on the NCs’ surface whereas the NCs from the M-NCs sample are organized in clusters with maghemite NPs dispersed mainly on their surface.

In the case of the LMSE-NCs sample, the resulting log-normal distribution of NPs’ diameters is broader than these for the M-NCs and MSE-NCs samples. Moreover, the mean magnetic moment of the maghemite NPs extracted from the LMSE-NCs sample is smaller than the values extracted from the M-NCs and MSE-NCs samples. This is likely due to the stronger dipolar interaction among maghemite NPs caused by shortening the particle–particle distance after lyophilization. As mentioned above, one-dimensional chains of maghemite NPs are formed within the MSE-NCs sample due to the magnetic dipolar interaction. The density of chains increases with the lyophilization of the PLGA magnetic NCs. Therefore, in the LMSE-NCs sample, where the chain density is higher, some particles in different chains might attract each other due to the dipolar interaction. Thus, the increasing number of chains may entangle with each other to form ribbon-like structures. However, the growth of such ribbon-like structures would result in an increase in the magnetostatic energy due to the repulsive force among parallel magnetic dipoles. Therefore, the competition between dipolar attraction and parallel dipole repulsion among neighboring chains may result in the formation of domain-like structures (mediated by the dipolar interaction), which vary their magnetic moment direction to maintain a total low energy in analogy to traditional magnetic domains mediated by exchange coupling. Thus, the maghemite NPs within the LMSE-NCs sample behave as disordered magnetic NPs whose net magnetic moments results from the superposition of magnetization vectors of different domain-like structures. The frustrated interactions between different domain-like structures give rise to a reduction of the mean magnetic moment of the particle observed in the LMSE-NCs sample. Furthermore, the abnormally large diamagnetic contribution (parameter a) found in the LMSE-NCs sample indicates that the domain-like structures interact with the applied magnetic field as SPM moments increasing the total magnetization.

Magnetic hysteresis measurements, at selected temperatures, were performed in the temperature range of 10–300 K and selected hysteresis loops recorded at 10 K and 300 K are shown in Fig. 5 (left panels). In order to assess the development of both coercivity and remanence while decreasing the measuring temperature, the hysteresis loops at lower fields were magnified (see Fig. 5 – right panels).


image file: d0cp02706e-f5.tif
Fig. 5 Typical hysteresis loops recorded at 10 and 300 K (left panels) for the as-prepared PLGA NC containing selol (SE-NCs sample), maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample). The right panels show the magnification of curves in lower field range (±2 kOe).

The field-dependent magnetization curves for the SE-NCs sample show typical diamagnetic behavior, as expected (see Fig. 5 – upper left panel). The magnetization curves for the M-NCs, MSE-NCs and LMSE-NCs samples show typical SPM behavior, meaning the magnetic order contribution below the blocking temperature (TB) and no hysteresis above TB (see Fig. 5 – right panels). However, the saturation magnetization of the LMSE-NCs sample is 40 times larger than that of the M-NCs and MSE-NCs samples indicating that the dipolar interaction is beneficial for the alignments of the magnetic moments. The dipolar interaction acts like an additional internal field to align and assemble the magnetic NPs.

For the magnetic hysteresis measuring cycles, which started at low temperatures, a continuous reduction of both coercivity and remanence is observed for all M-NCs, MSE-NCs and LMSE-NCs samples while increasing the temperature (see Fig. 6 and 7), although both parameters do not vanish to zero even at room temperature. It means that after a series of measurements the pure SPM state of all the magnetic NPs at 300 K is not achieved. This confirms the existence of the magnetic dipolar interaction among maghemite NPs, which introduces an additional energy barrier for the rotation of the magnetic NPs’ magnetic moment, leading to finite coercivity and remanence above TB.22


image file: d0cp02706e-f6.tif
Fig. 6 Temperature dependence of the coercivity (HC) for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample). The solid blue-lines in the main figures show the fitting according to the modified Kneller's law, described by eqn (5). The straight solid red-lines in the insets show the T1/2 dependency.

image file: d0cp02706e-f7.tif
Fig. 7 Temperature dependence of the remanence (MR) for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample). The insets show the temperature dependence of the remanence-to-saturation ratio MR/MS.

For an assembly of non-interacting single domain magnetic NPs with uniaxial anisotropy, the temperature-dependence of the coercivity (HC) below TB can be written in the form of a simple model based on thermal activation of the magnetic NPs’ moments over the anisotropy barrier (Kneller's law) as:23

 
image file: d0cp02706e-t4.tif(4)
where HC0 is the coercivity at T = 0 K and can be estimated by extrapolating the HCversus T curve towards the field axis. The solid red-lines in the insets of Fig. 6 present the linear fits according to eqn (4). However, it is seen from the main plots in Fig. 6 that the temperature dependence of the coercivity cannot be fitted (see the solid blue-lines) using eqn (4), clearly showing that the assumption of non-interacting SPM NPs is not valid in the case of the M-NCs, MSE-NCs, and LMSE-NCs samples. Instead, the temperature-dependence of HC for all three samples (see Fig. 6) is found to follow closely the modified Kneller's law given below:24
 
image file: d0cp02706e-t5.tif(5)
The obtained fitting parameters according to eqn (5) are: HC0 = (327 ± 32) Oe, TB = (60 ± 6) K and α = 0.42 ± 0.03 for the M-NCs sample, HC0 = (317 ± 31) Oe, TB = (63 ± 6) K and α = 0.41 ± 0.03 for the MSE-NCs sample and HC0 = (323 ± 32) Oe, TB = (59 ± 6) K and α = 0.42 ± 0.03 for the LMSE-NCs sample. Actually, the fact that the HC0versus T data of all samples (M-NCs, MSE-NCs and LMSE-NCs) do not follow the non-interacting SPM behavior predicted by eqn (4) but instead present a slower (α < 0.5) temperature dependence given by eqn (5) is in agreement with predictions for the case of a collection of magnetic NPs with the magnetic dipolar interaction.24

The temperature-variations of both remanence (MR) and remanence-to-saturation ratio (MR/MS) are shown in Fig. 7. At T = 4 K, MR/MS = 0.31/0.30/0.29 for the M-NCs/MSE-NCs/LMSE-NCs samples is smaller than the expected value (MR/MS = 0.5) for non-interacting, randomly oriented single-domain particles with uniaxial symmetry predicted by the Stoner–Wohlfarth theory.25 This finding is an additional confirmation for the existence of the inter-particle magnetic dipole–dipole interaction. It has been suggested that MR/MS > 0.5 and MR/MS < 0.5 values should be expected for systems with ferro- and antiferromagnetic interactions, respectively.26 Thus, the present MR/MS = 0.31/0.30/0.29 values indicate that in all samples the inter-particle interaction is of antiferromagnetic nature.

The temperature dependence of the saturation magnetization (MS) for all M-NCs, MSE-NCs and LMSE-NCs samples is shown in Fig. 8. The value of MS was evaluated for each temperature by plotting the magnetization obtained from the initial magnetization curve versus the reciprocal of the applied field and extrapolating to zero. From Fig. 8, it is evident that MS increases while decreasing T for all samples following a Bloch-type law:27

 
image file: d0cp02706e-t6.tif(6)
where MS(0) is the saturation magnetization at 0 K, B is the Bloch's constant and β is the Bloch's exponent.


image file: d0cp02706e-f8.tif
Fig. 8 Temperature dependence of the saturation magnetization for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample) – open symbols correspond to the experimental data, and the solid red-lines represent fits according to eqn (6).

Fittings of the data (open circles) shown in Fig. 8 using eqn (6) (shown by solid red-lines) result in the following parameter values: MS(0) = (1.74 ± 0.01) × 10−1 emu g−1, B = (7.4 ± 0.1) × 10−8 K−2, β = 1.93 ± 0.04 for the M-NCs sample, MS(0) = (1.57 ± 0.01) × 10−1 emu g−1, B = (6.4 ± 0.1) × 10−8 K−2, β = 1.96 ± 0.04 for the MSE-NCs sample and MS(0) = (6.96 ± 0.01) emu g−1, B = (1.4 ± 0.1) × 10−4 K−2, β = 1.90 ± 0.04 for the LMSE-NCs sample. It is seen that the Bloch's exponents (β) for all samples are not equal to 3/2, as it is predicted for bulk ferromagnets,28 but instead close to 2. Certain deviations from the Bloch's law were observed in the case of magnetic NPs and they were mainly related to the increasing surface-to-volume ratio as well as to the local symmetry and exchange breaking between atoms located at the NPs’ surface. In the case of particles and clusters, some theoretical calculations have shown that Bloch's exponent higher than 3/2 and close to 2 can be explained as a consequence of particles’ size reduction.28 The magnetization decreases faster at higher temperatures in the NPs than in the bulk material, due to the lack of full coordination at the NPs’ surface, which leads to larger spin deviations in the NPs’ shell than in the NPs’ core.

Fig. 9 shows the temperature dependence of the magnetization difference (ΔM = MFCMZFC) compared with the temperature dependence of the coercivity (HC) for the M-NCs, MSE-NCs and LMSE-NCs samples. The observed shift of the ΔM curve towards higher temperatures in all samples can be explained by the demagnetizing role played by the dipolar interaction,29 stronger in the lower temperature range.


image file: d0cp02706e-f9.tif
Fig. 9 Temperature dependence of the magnetization difference (ΔM = MFCMZFC) compared with the temperature dependence of coercivity (HC) for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample).

As it can be seen from the above considerations, the simple picture of a system of non-interacting magnetic NPs behaving superparamagnetically, which undergoes single particle blocking at low temperatures, is not enough to fully describe the magnetic behavior of the M-NCs, MSE-NCs and LMSE-NCs samples.

A system comprising magnetic NPs is considered to be SPM at high temperatures when its anhysteretic magnetization curve can be described by the Langevin function (as shown in Fig. 4) and the reduced magnetization M/MS obeys a scaling law when plotted against H/T.30Fig. 10 presents the M/MSversus H/T curves in the high temperature region (200–300 K) for the M-NCs, MSE-NCs and LMSE-NCs samples. The excellent degree of superposition of the experimental data (open symbols) shown in Fig. 10 indicates that the classical SPM scaling holds for all samples in the above-mentioned temperature region.


image file: d0cp02706e-f10.tif
Fig. 10 Reduced magnetization M/MSversus H/T measured at different temperatures for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample).

On the other hand, at lower temperatures (below 200 K), the M/MSversus H/T curves no longer superpose; instead M/MS scales with H/MS (see open symbols in Fig. 11). This finding indicates the gradual emergence of the interacting superparamagnetic (ISPM) regime of magnetic nanoparticulate systems,30 in which the effect of the long range magnetic dipolar interaction among isolated magnetic moments is no longer negligible.


image file: d0cp02706e-f11.tif
Fig. 11 Reduced magnetization M/MSversus H/MS at different temperatures for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample).

Dynamic magnetic properties

In order to study the effect of interparticle interaction on the dynamics of the blocking process, the temperature dependence of the in-phase (real) component χ′(T) and the out-of-phase (imaginary) component χ′′(T) of the ac magnetic susceptibility for different frequencies f of the excitation field has been measured, as shown in Fig. 12 (solid symbols). The experimental data of all magnetic samples (M-NCs, MSE-NCs, and LMSE-NCs) exhibit the expected behavior of a SPM system, i.e. the occurrence of a maximum in both components χ′(T) and χ′′(T) at different Tmax temperatures (Tmax′ and Tmax′′) which shifts towards higher temperatures with increasing frequency f.31
image file: d0cp02706e-f12.tif
Fig. 12 Temperature dependence of the real (χ′) and imaginary (χ′′) components of the magnetic susceptibility at different excitation frequencies for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample). Arrows indicate increasing frequencies. The data were taken with an external magnetic field of Hac = 10 Oe.

The real component of the ac susceptibility of all samples shows a value not equal to zero for T approaching zero (see left panels in Fig. 12). It may be due to the presence of small chain-like agglomerates of particles coupled by the magnetic dipolar interaction. This interaction modifies the magnetic behavior of the system, introducing a collective component which has the influence on the low temperature magnetic relaxation. It has been shown31 that the magnetic relaxation of an interacting magnetic NPs system at low temperatures is extended towards longer time scales as compared to the relaxation of a non-interacting magnetic NPs system. Another indication of the influence of the magnetic dipole–dipole interaction on the dynamics of the samples comes from the increase of the height of the peak in χ′′(T) with increasing frequency (see right panels in Fig. 12), whereas it is almost constant with frequency for a non-interacting magnetic NPs system.31

At about 280 K a kink is seen in both χ′(T) and χ′′(T) curves for the as-prepared M-NCs and MSE-NCs samples, which is associated with the melting point of the suspension.

The empirical parameter Φ,32 representing the relative shift of Tmax′′ per interval of frequency, namely:

 
image file: d0cp02706e-t7.tif(7)
calculated from the frequency dependence of the maximum in the imaginary χ′′(T) part of the ac susceptibility is equal to 0.10, 0.10 and 0.09 for the M-NCs, MSE-NCs and LMSE-NCs samples, respectively. These values are close to the 0.10–0.13 range found for SPM systems.32 The slightly lower value for the LMSE-NCs sample may originate from the stronger interparticle magnetic dipole–dipole interaction,32 in agreement with the concentrated nature of the lyophilized sample.

For a thermally activated blocking process, the maxima of both χ′(T) and χ′′(T) show an exponential dependence of the applied frequency given by the Néel–Arrhenius law:33

 
f = f0[thin space (1/6-em)]exp(−Ea/kBT),(8)
where f0 is an attempt frequency in the range of 109–1011 Hz for SPM systems, kB is the Boltzmann constant and Ea is the magnetic anisotropy energy barrier. Fig. 13 shows the observed frequency dependence of the maximum in the imaginary χ′′(T) part of the ac susceptibility for the M-NCs, MSE-NCs and LMSE-NCs samples. The observed linear dependence of ln(f) versus 1/Tmax′′ indicates that the Néel–Arrhenius model correctly fits the data. The fitted parameters related to the solid red-lines are f0 = (1.8 ± 0.5) × 1011 Hz and Ea/kB = (2280 ± 38) K for the M-NCs sample, f0 = (1.6 ± 0.5) × 1011 Hz and Ea/kB = (2195 ± 52) K for the MSE-NCs sample, and f0 = (3.5 ± 0.5) × 1011 Hz and Ea/kB = (2593 ± 58) K for the LMSE-NCs sample.


image file: d0cp02706e-f13.tif
Fig. 13 Logarithmic frequency (ln[thin space (1/6-em)]f) versus inverse of the blocking temperature (1/Tmax′′) obtained from the imaginary component χ′′(T) for the as-prepared PLGA NCs containing maghemite NPs (M-NCs sample), maghemite NPs and selol (MSE-NCs sample), and lyophilized PLGA NCs containing maghemite NPs and selol (LMSE-NCs sample). Solid red-lines are the best fits using eqn (8).

From the fitted values of the energy barrier it is possible to estimate the effective magnetic anisotropy constant Keff, which is given by Ea = KeffV, where V is the average magnetic NP volume. Using the assessed average particle diameters 〈D〉 = 9.8 nm, 10.2 nm and 9.7 nm, the following values of Keff are obtained to be 64 kJ m−3, 55 kJ m−3 and 75 kJ m−3 for the M-NCs, MSE-NCs and LMSE-NCs samples, respectively.

The resulting effective anisotropy constant is approximately an order of magnitude larger than the first-order magnetocrystalline anisotropy constant of bulk maghemite (Kbulk1 = 4.7 kJ m−3),34 indicating an additional source of magnetic anisotropy to the single-particle energy barrier. Such enhancement is actually observed in magnetic NPs and attributed to the dipolar interaction (for concentrated systems) and/or surface effects (in diluted systems).35,36 Once the TEM images show that the employed maghemite NPs are nearly spherical,14 no major contribution from surface anisotropy should be expected, since a perfect spherical particle should not provide any contribution from surface anisotropy, as previously demonstrated from symmetry arguments.37

However, for magnetic NPs in the form of an elongated ellipsoid, with axes c and a = b, the possible contribution from shape anisotropy is given by:

 
image file: d0cp02706e-t8.tif(9)
where image file: d0cp02706e-t9.tif and image file: d0cp02706e-t10.tif are the demagnetizing factors determined by the shape of the NP, image file: d0cp02706e-t11.tif, image file: d0cp02706e-t12.tif and MS is the saturation magnetization of the sample. Assuming m = 2, one obtains Kshape = 7.1 J m−3, 5.3 J m−3 and 1.1 kJ m−3 for the M-NCs, MSE-NCs and LMSE-NCs samples, respectively. It gives a contribution of 0.15%, 0.12% and 23% of Kbulk1 to the resulting effective anisotropy constant. Therefore, the contribution from the shape anisotropy does not cause a substantial increase in the effective magnetic anisotropy constant, even for very elongated ellipsoids, which are not observed in the TEM micrographs. Consequently, the observed enhancement of the effective magnetic anisotropy constant should be mainly related to the effect of the magnetic dipolar interaction among the NPs.

ESR measurements

The selected ESR spectra of the MSE-NCs and LMSE-NCs samples, recorded at various temperatures, are presented in Fig. 14. The room temperature spectra of both samples show well-defined single broad signals with an effective gyromagnetic value of about g = 2.07 and 2.06 and the peak-to-peak line width ΔHpp = 792 Oe and 985 Oe for the MSE-NCs and LMSE-NCs samples, respectively. Upon decreasing the temperature, both signals shift to lower fields and gradually broaden, closely following the predictions for the ESR of SPM NPs’ systems.38 Therefore, it may be expected that the broadening and shift of the resonance signal are associated with the blocking of the magnetization in maghemite NPs. The peak-to-peak line width of both signals considerably exceeds the magnetocrystalline-anisotropy-determined minimum value, ΔHpp = 400 Oe, for non-interacting single domain maghemite NPs.39 This confirms the existence of the non-negligible dipole–dipole interaction among the maghemite NPs, which is stronger in the lyophilized sample (LMSE-NCs).
image file: d0cp02706e-f14.tif
Fig. 14 Experimental ESR spectra of the as-prepared (MSE-NCs) and lyophilized (LMSE-NCs) PLGA NCs containing maghemite NPs and selol recorded at various temperatures.

The ESR lines were fitted using a Gaussian line shape function and the results for T = 300 K are shown in Fig. 15. The fitting could be regarded as satisfactory for both samples. The observed slight asymmetry of both ESR lines suggests the presence of the dipolar interaction among maghemite NPs which is stronger in the LMSE-NCs sample.


image file: d0cp02706e-f15.tif
Fig. 15 Experimental ESR spectra (symbols) of the as-prepared (MSE-NCs) and lyophilized (LMSE-NCs) PLGA NCs containing maghemite NPs and selol recorded at T = 300 K. Solid red-lines represent the best fit using a Gaussian line shape function.

Fig. 16 shows the temperature dependence of the spectral parameters calculated from the ESR signals of both samples: resonance field Hres, peak-to-peak line width ΔHpp, peak-to-peak signal amplitude App, integrated intensity Iint, reciprocal of integrated intensity Iint−1, and product of integrated intensity and temperature Iint × Tμeff2. The integrated intensity (herein taken as Iint = App × ΔHpp2) is proportional to the magnetic susceptibility of the spin system. All parameters depend strongly upon temperature. The results in Fig. 16a show that Hres shifts to lower magnetic fields with decreasing temperature and the shift is smaller for the LMSE-NCs sample than for the MSE-NCs sample. It may be due to the stronger antiferromagnetic dipolar interaction among NPs in the lyophilized sample. Maghemite NPs hosted by polymeric NCs containing non-magnetic selol could generate an internal magnetic field (Hint) that might change the resonance condition as follows:

 
= B(Happ + Hint),(10)
where h is the Planck's constant, ν is the microwave frequency, μB is the Bohr's magneton, and Happ is the applied external magnetic field. The internal magnetic field is defined as:
 
Hint = Hdem + Hdip + Hdip′,(11)
where Hdem is the demagnetization field, Hdip is the internal magnetic field from the dipole–dipole interaction among maghemite NPs, Hdip′ is the internal magnetic field from the dipole–dipole interaction between agglomerates. The internal magnetic field acting on the magnetic NPs strongly depends on the concentration and size of agglomerates. Large-size agglomerates of maghemite NPs, as in the lyophilized LMSE-NCs sample, could produce a stronger average internal magnetic field and generate the stronger dipole–dipole magnetic interaction in comparison to smaller agglomerates likely present in the as-prepared MSE-NCs sample. The antiferromagnetic coupling of agglomerates could cause a decrease in the internal magnetic field which in turn shifts the resonance field in opposite direction and thus larger magnetic agglomerates in the LMSE-NCs sample could create a smaller resonance field than that observed in the MSE-NCs sample.


image file: d0cp02706e-f16.tif
Fig. 16 Temperature dependence of the ESR parameters calculated from the spectra of the as-prepared (MSE-NCs) and lyophilized (LMSE-NCs) PLGA NCs containing maghemite NPs and selol: (a) resonance field (Hres), (b) peak-to-peak line width (ΔHpp), (c) peak-to-peak amplitude (App), (d) integrated intensity (Iint), (e) reciprocal of integrated intensity (Iint−1), and (f) product of integrated intensity and temperature Iint × T.

The line width ΔHpp for both investigated samples increases linearly with decreasing temperature (see Fig. 16b). It may be explained in the frame of the core–shell model40 in which the structure of the magnetic NP should be considered as being made of a single-domain ordered core and a surface shell of disordered spins, the latter interacting with each other in a spin-glass-like state and with the particle core. The surface spin fluctuation slows down with decreasing temperature, leading at low temperatures to a state of frozen disordered shell spins. The degree of spin frustration is expected to increase with decreasing the NP size. It is worth mentioning that for a typical NP with 10 nm diameter approximately one third of the atoms lie within two atomic layers of the surface. From the ESR point of view, at a particular temperature, the surface shell is transiently brought into a spin-glass-like state for times larger than the Larmor period. The distribution of canting angles of frustrated spins at the magnetic NP's shell gives rise to a wide spread of internal fields, thus increasing the ESR line width at low temperatures which is observed in many magnetic NPs systems. In the case of the MNSE-NCs and LMSE-NCs samples the strong inter-particle dipolar interaction remarkably affects the surface spins,40 quenching their contribution to the line width broadening. Therefore, the fast low-temperature increase in the ESR line width is not herein observed. Instead, ΔHpp for both investigated samples (MNSE-NCs and LMSE-NCs) increases linearly with decreasing temperature in the all temperature range. Such a linear temperature dependence of the ESR line width was also observed in other NPs systems.41

The integrated intensity Iint (and the reciprocal of the integrated intensity Iint−1) for both samples (MNSE-NCs and LMSE-NCs) shows a Curie–Weiss type behavior in the low temperature range (see Fig. 16d and e). The negative sign of the Curie–Weiss temperature (−565 K and −638 K for the MSE-NCs and LMSE-NCs samples, respectively) indicates the existence of the antiferromagnetic dipolar interaction which is stronger in the LMSE-NCs sample, in agreement with the dc and ac magnetization measurements described above.

Fig. 16f presents the temperature dependence of the product of integrated intensity and temperature for the MSE-NCs and LMSE-NCs samples. This product is proportional to the square of the effective magnetic moment of the maghemite NPs producing the ESR spectrum: Iint × Tμeff2. Therefore, the magnetic moment decreases as the temperature is lowered, indicating that the effective antiferromagnetic interaction is prevailing in both samples.

In order to reveal different possible temperature ranges of relaxation processes, a semilogarithmic plot of the shift of resonance field (δHres = Hres(T) − Hres(∞)) versus the line width (ΔHpp) is shown in Fig. 17. Hres(∞) is the resonance field in the limit of high temperatures. According to the theory of Nagata and Ishihara,38 the simple power relationship between the resonance field shift (δHres) and the line width (ΔHpp) for a SPM system of particles having a statistical distribution of shapes and sizes can be expressed as:

 
δHres ∼ (ΔHpp)n,(12)
where n = 2 (3) is predicted for partially (randomly) oriented particles. Any detected change of slope on this type of plot should be a sign of a relaxation type variation in the studied samples. A closer inspection of Fig. 17 reveals the existence of only one temperature range for both samples. The assessed slope of the δHresversus ΔHpp curves provide n = 4.5 and 5.4 for the MNSE-NCs and LMNSE-NCs samples, respectively. Thus, both ESR data are shown to be due to the ISPM regime behavior of maghemite NPs coupling by dipolar interaction, which is stronger in the lyophilized sample.


image file: d0cp02706e-f17.tif
Fig. 17 Relationship between the resonance field shift (δHres) and peak-to-peak line width (ΔHpp) for the ESR signal of the as-prepared (MSE-NCs) and lyophilized (LMSE-NCs) PLGA NCs containing maghemite NPs and selol – solid symbols correspond to the experimental data whereas solid lines represent the linear fits according to eqn (12).

Conclusions

In summary, we prepared magnetic polylactic-co-glicolic acid (PLGA) nanocapsules (NCs) loaded with maghemite (γ-Fe2O3) nanoparticles (NPs) and selol (a new Se-based anticancer drug). As shown in previous studies,12–18 the prepared material presents antitumor activity on the neoplastic cells which increases after further exposure of these cells to an alternating magnetic field. Moreover, the loaded NCs show the antitumor effect without affecting normal cells. Additionally, the association of NCs’ surface with folic acid increases its cell uptake (by specific tumor cell receptors) and toxicity of the maghemite-selol-loaded NCs to cancer cells. Also, selol-loaded NCs with the shell conjugated to doxorubicin show a significantly increased cytotoxicity against cancer cells, which is suggestive of additive or synergic effects of selol and doxorubicin.

The magnetic properties of the as-fabricated (MSE-NCs sample) and lyophylized (LMSE-NCs sample) PLGA NCs were investigated by means of dc magnetization, ac susceptibility and ESR measurements over the 4 K to 300 K temperature range. Although the mean diameter of the maghemite NPs is in the limit of superparamagnetism (SPM), the magnetic properties of the NCs do not exhibit the classical behavior of SPM particles. The magnetic NCs show a perfect SPM response at temperatures above 200 K, manifested by the split between ZFC/FC magnetization curves, the absence of hysteresis in the field dependent magnetization curve above the blocking temperature and the scaling of the reduced magnetization to the H/T ratio. However, at temperatures below 200 K, the SPM scaling is no longer observed and the magnetic NCs behave as interacting superparamagnetic (ISPM) materials. The coercivity of the magnetic NCs exhibits a temperature dependence which is quenched compared to the case of non-interacting NPs, in agreement with predictions for the case of a collection of NPs with the magnetic dipolar interaction. We found that this interaction is antiferromagnetic in the as-fabricated magnetic NCs (MSE-NCs sample), where the ordering temperature arising from the inter-particle interaction is T0 = −258 K. After lyophylization (LMSE-NCs sample) the antiferromagnetic interaction becomes stronger (T0 = −302 K), which is due to an enhanced clustering of maghemite NPs into chain-like structures.

The ac susceptibility measurements confirm the SPM behavior of the as-fabricated magnetic NCs manifested by the occurrence of a maximum in the temperature dependence of both real χ′(T) and imaginary χ′′(T) components of the ac magnetic susceptibility, which shifts towards higher temperatures with increasing frequency. Both χ′(T) and χ′′(T) maxima follow the predictions of the thermally activated Néel–Arrhenius model. The effective magnetic anisotropy constant Keff inferred from the χ′′(T) versus f data for the as-fabricated and lyophylized magnetic NCs is approximately an order of magnitude larger that the first-order magnetocrystalline anisotropy constant of bulk maghemite Kbulk1, what may be attributed to the presence of the strong dipolar interaction among maghemite NPs.

The room temperature ESR spectrum of the as-fabricated and lyophylized samples (MSE-NCs and LMSE-NCs) shows a well-defined single broad signal which shifts to lower fields and gradually broadens upon decreasing the temperature, closely following the predictions for the ESR of SPM NPs systems. However, the temperature dependence of the spectral parameters calculated from the ESR signal of both samples, i.e. resonance field Hres, peak-to-peak line width ΔHpp and integrated intensity Iint, do confirm the existence of the non-negligible dipole–dipole interaction among maghemite NPs, which is stronger in the lyophilized sample (LMSE-NCs).

Indeed, the performed investigation shows that the PLGA NCs loaded with maghemite NPs and selol contain chain-like magnetic agglomerates coupled by the antiferromagnetic dipolar interaction. However, these agglomerates are sufficiently small to show SPM behavior at room temperature under static conditions. It means that selol-loaded magnetic NCs can thus be a new promising magnetic drug delivery system and a magnetohyperthermia inductor in antitumor therapy.

Moreover, the magnetic NCs show giant diamagnetism at low temperatures, which may be also attributed to the antiferromagnetic dipolar interaction among maghemite NPs. Even compared with the usual superconductors, the order of the diamagnetic susceptibility (−10−3 and −10−2 emu × g−1 × Oe−1 in the MSE-NCs and LMSE-NCs samples, respectively) is quite considerable. This effect is much stronger in the lyophilized NCs (LMSE-NCs sample) due to the increasing role of the antiferromagnetic dipole–dipole interaction in this sample. The giant diamagnetism has not been observed earlier in this type of materials. From this point of view, the magnetic NCs herein investigated are also an interesting research material.

Conflicts of interest

There are no conflicts to declare.

References

  1. S. Khoee and M. Yaghoobian, Eur. J. Med. Chem., 2008, 44, 2392 CrossRef.
  2. Q. A. Punkhurst, J. Connolly, S. K. Jones and J. Dobson, J. Phys. D: Appl. Phys., 2003, 36, R167 CrossRef.
  3. R. Hergt, W. Andra, C. G. D’ambly, I. Hilger, W. A. Kaiser, U. Richter and H. Schmidt, IEEE Trans. Magn., 1998, 34, 3745 CrossRef CAS.
  4. R. E. Rosensweig, J. Magn. Magn. Mater., 2002, 252, 370 CrossRef CAS.
  5. Y. Lv, Z. Deng and J. Liu, IEEE Trans. Nanobiosci., 2005, 4, 284 CrossRef.
  6. P. Tartaj, M. del Puerto Morales, S. Veintemillas-Verdaguer, T. Gonzáles-Carreño and C. J. Serna, J. Phys. D: Appl. Phys., 2003, 36, 182 CrossRef.
  7. P. Suchocki, I. Misiewicz, K. Skupińska, K. Wacławek, Z. Fijałek and T. Kasprzycka-Guttman, Oncol. Rep., 2007, 18, 893 CAS.
  8. E. Drake, Med. Hypotheses, 2006, 67, 318 CrossRef CAS.
  9. B. Fitak, M. Grabowski and P. Suchocki, Patent Pol., Pl176530 (Cl. A61K31/095), Warsaw, Poland, 1999 Search PubMed.
  10. P. Suchocki, I. Misiewicz-Krzemińska, K. Skupińska, K. Niedźwiedzka, K. Lubelska, Z. Fijałek and T. Kasprzycka-Guttman, Pharmacol. Rep., 2010, 62, 352 CrossRef CAS.
  11. P. Suchocki, I. Misiewicz, K. Skupińska, K. Wacławek, Z. Fijałek and T. Kasprzycka-Guttman, Oncol. Rep., 2007, 18, 893 CAS.
  12. A. M. Falqueiro, F. L. Primo, P. C. Morais, E. Mosiniewicz-Szablewska, P. Suchocki and A. C. Tedesco, J. Appl. Phys., 2011, 109, 07B306 CrossRef.
  13. A. M. Falqueiro, M. P. Siqueira-Moura, D. R. Jardim, F. L. Primo, P. C. Morais, E. Mosiniewicz-Szablewska, P. Suchocki and A. C. Tedesco, J. Appl. Phys., 2012, 111, 07B335 CrossRef.
  14. L. L. C. Estevanato, J. R. Da Silva, A. M. Falqueiro, E. Mosiniewicz-Szablewska, P. Suchocki, A. C. Tedesco, P. C. Morais and Z. G. M. Lacava, Int. J. Nanomed., 2012, 7, 5287 CAS.
  15. L. R. Souza, L. A. Muehlmann, M. S. Costa dos Santos, R. Ganassin, R. Simón-Vázquez, G. A. Joanitti, E. Mosiniewicz-Szablewska, P. Suchocki, P. C. Morais, A. Gonzáles-Fernández, R. B. Azevedo and S. N. Báo, J. Nanobiotechnol., 2014, 12, 32 CrossRef.
  16. R. Ganassin, L. R. Souza, K. R. Py-Daniel, J. P. F. Longo, J. M. Coehlo, M. C. Rodrigues, Ch-S Jiang, J. Gu, P. C. Morais, E. Mosiniewicz-Szablewska, P. Suchocki, S. N. Báo, R. B. Azevedo and L. A. Muehlmann, J. Nanosci. Nanotechnol., 2018, 18, 522 CrossRef CAS.
  17. R. Ganassin, C. Merker, M. C. Rodrigues, N. F. Guimarães, C. S. C. Sodré, Q. Silva Ferreira, S. W. Silva, A. S. Ombredane, G. A. Joanitti, K. R. Py-Daniel, J. Zhang, C.-S. Jiang, P. C. Morais, E. Mosiniewicz-Szablewska, P. Suchocki, J. P. F. Longo, J. Meijer, I. Estrela-Lopis, R. B. Azevedo and L. A. Muehlmann, Artif. Cells, Nanomed., Biotechnol., 2018, 46, 2002 CAS.
  18. R. Ganassin, F. H. Horst, N. S. Camargo, S. B. Chaves, P. C. Morais, E. Mosiniewicz-Szablewska, P. Suchocki, J. P. F. Longo, J. Meijer, R. B. Azevedo and L. A. Muehlmann, Artif. Cells, Nanomed., Biotechnol., 2018, 46(Suppl. 2), 1046 CrossRef CAS.
  19. H. Khusrhid, Z. N. Porshokouh, M.-H. Phan, P. Mukherjee and H. Srikanth, J. Appl. Phys., 2014, 115, 17E131 CrossRef.
  20. S. K. Sharma, R. Kumar, S. Kumar, M. Knobel, C. T. Meneses, V. V. Siva Kumar, V. R. Reddy, M. Singh and C. G. Lee, J. Phys.: Condens. Matter, 2008, 20, 235214 CrossRef CAS.
  21. R. Kaiser and G. Miskolczyk, J. Appl. Phys., 1970, 41, 1064 CrossRef CAS.
  22. S. Morup, F. Bodker, P. V. Hendriksen and S. Linderoth, Phys. Rev. B: Condens. Matter Mater. Phys., 1995, 52, 287 CrossRef.
  23. E. F. Kneller and F. E. Luborsky, J. Appl. Phys., 1963, 34, 656 CrossRef CAS.
  24. X. Batlle, M. Garcia del Muro, J. Tejada, H. Pfeiffer, P. Goand and E. Sinn, J. Appl. Phys., 1993, 74, 3333 CrossRef CAS.
  25. C. R. Denham, R. P. Blackemore and R. B. Frankel, IEEE Trans. Magn., 1980, 16, 1006 CrossRef.
  26. G. Hadjipanayis, D. J. Sellmyer and B. Brandt, Phys. Rev. B: Condens. Matter Mater. Phys., 1981, 23, 3349 CrossRef CAS.
  27. D. Ortega, E. Vélez-Fort, D. A. García, R. García, R. Litrán, C. Barrera-Solano, M. Ramírez-del-Solar and M. Domínguez, Philos. Trans. R. Soc., A, 2010, 368, 4407 CrossRef CAS.
  28. P. V. Hendriksen, S. Linderoth and P. A. Lindgard, Phys. Rev. B: Condens. Matter Mater. Phys., 1993, 48, 7259 CrossRef CAS.
  29. W. C. Nunes, F. Cebollada, M. Knobel and D. Zanchet, J. Appl. Phys., 2006, 99, 08N705 CrossRef.
  30. P. Allia and P. Tiberto, J. Nanopart. Res., 2011, 13, 7277 CrossRef CAS.
  31. T. Jonsson, P. Norblad and P. Svedlindh, Phys. Rev. B: Condens. Matter Mater. Phys., 1998, 57, 497 CrossRef CAS.
  32. J. L. Dormann, D. Fiorani and E. Tronc, Adv. Chem. Phys., 1997, 98, 326 Search PubMed.
  33. L. F. Gamarra, G. E. S. Brito, W. M. Pontuschka, E. Amaro, A. H. C. Parma and G. F. Goya, J. Magn. Magn. Mater., 2005, 289, 439 CrossRef CAS.
  34. G. Bate, in Magnetic Oxides, ed. D. J. Craik, Wiley, London, 1975, p. 689 Search PubMed.
  35. G. F. Goya, IEEE Trans. Magn., 2002, 38, 2610 CrossRef CAS.
  36. F. Luis, F. Petroff, J. M. Torres, L. M. García, J. Bartolomé, J. Carrey and A. Vaurès, Phys. Rev. Lett., 2002, 88, 217205 CrossRef CAS.
  37. F. Bødker, S. Morup and S. Linderoth, Phys. Rev. Lett., 1994, 72, 282 CrossRef.
  38. K. Nagata and A. Ishihara, J. Magn. Magn. Mater., 1992, 104–107, 1571 CrossRef CAS.
  39. D. J. Dunlop and O. Ozdemir, Rock Magnetism: Fundamentals and Frontiers, Cambridge University Press, New York, 1997 Search PubMed.
  40. E. Mosiniewicz-Szablewska, A. R. Clavijo, A. P. O. R. Castilho, L. G. Paterno, M. A. Pereira-da-Silva, J. Więckowski, M. A. G. Soler and P. C. Morais, Phys. Chem. Chem. Phys., 2018, 20, 26696 RSC.
  41. Y. Köseoglu and H. Kavas, J. Nanosci. Nanotechnol., 2008, 8, 584 CrossRef.

This journal is © the Owner Societies 2020