betaFeedback

WorksPhysics and Astronomy

Anisotropic Effect of Dipolar Interaction in Ordered Ensembles of Nanoparticles

Abstract

We implement extensive computer simulations to investigate the hysteresis characteristics in the ordered arrays (lx×lyl^{}_x\times l^{}_y) of magnetic nanoparticles as a function of aspect ratio Ar=ly/lxA^{}_r=l^{}_y/l^{}_x, dipolar interaction strength hdh^{}_d, and external magnetic field directions. We have considered the aligned anisotropy case, αα is the orientational angle. It provides an elegant en route to unearth the explicit role of anisotropy and dipolar interaction on the hysteresis response in such a versatile system. The superparamagnetic character is dominant with weak dipolar interaction (hd≤0.2h^{}_d\leq0.2), resulting in the minimal hysteresis loop area. Remarkably, the double-loop hysteresis emerges even with moderate interaction strength (hd≈0.4h^{}_d\approx0.4), reminiscent of antiferromagnetic coupling. These features are strongly dependent on αα and ArA^{}_r. Interestingly, the hysteresis loop area increases with hdh^{}_d, provided ArA^{}_r is enormous, and the external magnetic field is along the yy-direction. The coercive field μoHcμ^{}_oH^{}_c, remanent magnetization MrM^{}_r, and the heat dissipation EHE^{}_H also depend strongly on these parameters. Irrespective of the external field direction and weak dipolar interaction (hd≤0.4h^{}_d\leq0.4), there is an increase in μoHcμ^{}_oH^{}_c with hdh^{}_d for a fixed αα and Ar≤4.0A^{}_r\leq4.0. The dipolar interaction also elevates MrM^{}_r as long as ArA^{}_r is huge and the field is along the yy-direction. EHE^{}_H is minimal for negligible and weak dipolar interaction, irrespective of ArA^{}_r, αα, and the field directions. Notably, the magnetic interaction enhances EHE^{}_H if ArA^{}_r is enormous and the magnetic field is along the long axis of the system. These results are beneficial in various applications of interest such as digital data storage, spintronics, etc.

I Introduction

Ordered arrays of magnetic nanoparticles (MNPs) have received significant attention due to their intriguing physics and numerous technological applications [1, 2, 3, 4, 5]. Such systems have unique physical and chemical properties that are utterly different from their corresponding bulk counterparts [6, 7]. They are also of immense importance in various applications such as spintronics, magnetic hyperthermia, drug delivery, biomedicine, etc. [8, 9, 10, 11, 12]. Therefore, understanding of various magnetic properties of these nanosystems is the need of time.

The physics of non-interacting nanoparticle arrays are fairly known [13]. However, MNPs interact because of the dipolar interaction in such a system [14]. The dipolar interaction affects the systematic properties significantly in such an ordered system due to the long-range and anisotropic nature [15, 16]. For instance, it induces spin-glass like characteristics in MNPs assembly with positional defects [17, 18]. On the other hand, it promotes ferromagnetic coupling in highly anisotropic systems, such as the linear and columnar arrangement of MNPs [19]. The dipolar interaction strongly affects the magnetic characteristics in a dense assembly [20, 21, 22, 23, 24, 25]. Torche et al. studied the local heat dissipation in a dipolar interacting nanoparticle assembly using transition state theory [21]. The dipolar interaction is found to enhances heat dissipation. Masunaga et al. investigated the dipolar interaction effect on the magnetic properties in Nickel nanoparticles ensembles [22]. The hysteresis characteristics such as blocking temperature are strongly affected by the dipolar interaction. Kechrakos et al. studied the magnetic and transport properties in an ordered assembly of MNPs using Monte Carlo methods [23]. The dipolar interaction induces anisotropic magnetic behaviour between the in-plane and out-of-plane directions of the system. Figueiredo et al. found an enhancement in the blocking temperature because of dipolar interaction [24]. Myrovali et al. observed an elevation in the area under the hysteresis curve with dipolar interaction [25]. The dipolar interaction also dictates the ground state spin morphologies [26, 27, 28, 29]. For example, it promotes ferromagnetic coupling in face-centred cubic while antiferromagnetic in the cubic assembly of MNPs [27]. The most favourable configuration is ferromagnetic in the triangular lattice [28]. In contrast, the minimum energy state is antiferromagnetic in the square array of nanoparticles [29].

Moreover, the dipolar interaction induces the anisotropic properties by creating an additional anisotropy termed as shape anisotropy in such a system [30, 31, 32]. Therefore, these ordered nanoparticles ensembles provide a rich theoretical framework to study the role of dipolar interaction and magnetic anisotropy on magnetic response. Much research efforts ranging from experimental to theoretical modelling have been devoted to unearthing the magnetic properties in such versatile systems [33, 34, 35, 36, 37, 38, 39]. Deng et al. presented a detailed and very informative overview regarding forming an anisotropic assembly of MNPs [33]. The dipolar interaction is one of the main reasons for such a self-assembled system. Li et al. analyzed the magnetic properties in dense arrays of magnetic nanoparticles using experiments [34]. They are found to possess enhanced magnetic properties along the array direction compared with the dispersed particles. Wen et al. devised procedures to manipulate the shape and magnetocrystalline anisotropy during the self-assembly process [35]. Maximum magnetic anisotropy is observed when the easy axis of the particle is aligned along the array axis of the sample. Allia et al. investigate the magnetic properties in an assembly of nanoparticles using the rate equation approach with collinear and randomly distributed easy axes [36]. The hysteresis loop area is significantly higher with aligned anisotropy in comparison with the randomly oriented case. Dyab et al. fabricated the anisotropic assembly of nanoparticles. [37]. Such systems have a very high coercive field and remanent magnetization, reminiscent of ferromagnetic character. Hoffelner et al. investigated the orientational alignment nanoparticles using a dynamical magnetic field [38]. The dipolar interaction promotes the collinear arrangement of easy axes in a highly anisotropic system. In recent work, we studied the magnetic relaxation characteristics in the ordered assembly of nanoparticles using kinetic Monte Carlo simulation (kMC) with aligned anisotropy axes [39]. There is a fastening or slowing down of relaxation depending on the orientation of anisotropy axes with sufficient dipolar coupling strength.

The above discussion indicates that the anisotropy axes orientations play a crucial role in determining the magnetic properties in the ordered arrays of dipolar interacting MNPs. However, a complete understanding of the effect of anisotropy axes orientations, dipolar interaction, system sizes, the direction of an external magnetic field is far from complete. Therefore, we investigate the magnetic properties in two-dimensional ordered arrays of MNPs as a function of these parameters using kinetic Monte Carlo simulations. These studies provide a deeper insight into the interplay between magnetic anisotropy and dipolar interactions on various magnetic characteristics in such a useful system. For instance, as dipolar interaction offers ferromagnetic coupling with aligned anisotropy in the columnar and linear array of magnetic nanoparticles, one can control the orientation of MNPs using an external magnetic field with ease in such cases. Such systems are advantageous in various applications such as magnetic hyperthermia, digital information storage [40, 41].

The rest of the article is organized as follows. We discuss the model used and various energy terms in Sec. II. We preset and discuss the simulation results in Sec. III. Finally, a summary of the present work is provided in § 4.

Theoretical Framework

We consider an ordered assembly of MNPs arranged in the two-dimensional arrays in the xyxy-plane as shown in the schematic Figure 1. The system dimension is lx×lyl_x \times l_y and aspect ratio Ar=ly/lxA_r = l_y/l_x. Let the nanoparticle diameter and the lattice constant be DD and aa, respectively. The particle has a magnetic moment μ=MsV\mu= M_sV; saturation magnetization is MsM_s, and V=πD3/6V = \pi D^3/6 is the nanoparticle volume. Let the particle has magnetocrystalline anisotropy K⃗=Keffk^\vec{K} = K_{\mathrm{eff}}\hat{k}, KeffK_{\mathrm{eff}} is the anisotropy strength, and k^\hat{k} is the unit vector along with the anisotropy or easy direction. We have considered the collinear anisotropy axes orientations, i.e. the anisotropy axes of all the MNPs are aligned, making an angle α\alpha with respect to the yy-axis of the system [please see the schematic Figure 1].

Image not extracted

Figure 1. Schematic of the two-dimensional ordered arrays of nanoparticles. Each nanoparticle has a diameter DD, and aa is the lattice constant. The blue dashed line denotes the direction of the anisotropy axis. All the MNPs are assumed to have the same anisotropy axes orientations, α\alpha is the orientation angle.

Each nanoparticle has the following energy because of magnetocrystalline anisotropy [13, 42]

EK=KeffVsin⁡2Φ(1)E_K = K_{\mathrm{eff}}V\sin^2\Phi \tag*{(1)}

Here Φ\Phi is the angle between the anisotropy vector and the magnetic moment. As nanoparticles primarily interact because of dipolar interaction, the corresponding interaction energy can be evaluated using the following relation [23, 43]

Edip=μoπMs2D6144a3∑j,j≠i[μ^i⋅μ^j(rij/a)3−3(μ^i⋅r^ij)(μ^j⋅r^ij)(rij/a)3].(2)E_{\mathrm{dip}} = \frac{\mu_o\pi M_s^2D^6}{144a^3}\sum_{j,j\ne i}\left[\frac{\hat{\mu}_i\cdot\hat{\mu}_j}{(r_{ij}/a)^3}-\frac{3(\hat{\mu}_i\cdot\hat{r}_{ij})(\hat{\mu}_j\cdot\hat{r}_{ij})}{(r_{ij}/a)^3}\right]. \tag*{(2)}

Here μo\mu_o is the permeability of free space; ithi^{th} and jthj^{th} nanoparticles have unit magnetic moment vectors μ^i\hat{\mu}_i and μ^j\hat{\mu}_j, respectively. The centre-to-centre distance between the ithi^{th} and jthj^{th} magnetic moments is rijr_{ij}, r^ij\hat{r}_{ij} is the unit vector associated with r⃗ij\vec{r}_{ij}. In such a case, the expression of the dipolar field can be written as [44]

μoH⃗dip=πμoMsD324a3∑j,j≠i3(μ^j⋅r^ij)r^ij−μ^j(rij/a)3.(3)\mu_o\vec{H}_{\mathrm{dip}} = \frac{\pi\mu_oM_sD^3}{24a^3}\sum_{j,j\ne i}\frac{3(\hat{\mu}_j\cdot\hat{r}_{ij})\hat{r}_{ij}-\hat{\mu}_j}{(r_{ij}/a)^3}. \tag*{(3)}

We evaluate this sum precisely without Ewald summation or a cutoff radius, similar to recent works [19, 32, 44]. We define a control parameter hd=D3/a3h_d = D^3/a^3 to model the variation of dipolar interaction strength. It correctly captures the physics of interaction strength variation as the dipolar field, and corresponding energy varies as 1/rij31/r_{ij}^3 [please see Eq. (2) and Eq. (3)]. The particles are at the closest approach with hd=1.0h_d = 1.0 because D=aD = a in this case. Therefore, hd=1.0h_d = 1.0 captures the physics of the most substantial dipolar interacting situation. In contrast, hd=0h_d = 0 represents the non-interacting situation.

We apply an oscillating magnetic field to investigate the magnetic hysteresis behaviour in the ordered arrays of dipolar coupled MNPs. It is given by [32]

μoH⃗=μoHocos⁡(2πνt)e^,(4)\mu_o \vec{H} = \mu_o H_o \cos(2\pi\nu t)\hat{e}, \tag*{(4)}

where μoHo\mu_o H_o and ν\nu are the magnitude and linear frequency of the external magnetic field, respectively, and tt is the time. e^\hat{e} is the unit vector along the applied oscillating magnetic field direction, which is x^\hat{x} (along the xx-axis) and y^\hat{y} (along the yy-axis) in the present work. We can then write the expression for total energy as [44, 45]

E=KeffV∑isin⁡2Φi+μoπMs2D6144a3∑i,  j,  j≠i[μ^i⋅μ^j−3(μ^i⋅r^ij)(μ^j⋅r^ij)(rij/a)3]−μoMsV∑iμ^i⋅H⃗(5)E = K_{\mathrm{eff}} V \sum_i \sin^2 \Phi_i + \frac{\mu_o \pi M_s^2 D^6}{144a^3} \sum_{i,\;j,\;j\ne i} \left[\frac{\hat{\mu}_i \cdot\hat{\mu}_j - 3(\hat{\mu}_i \cdot\hat{r}_{ij})(\hat{\mu}_j \cdot\hat{r}_{ij})}{(r_{ij}/a)^3}\right] - \mu_o M_s V \sum_i \hat{\mu}_i \cdot\vec{H} \tag*{(5)}

Here the ithi^{\mathrm{th}} magnetic is inclined an angle Φi\Phi_i with respect to the anisotropy axis.

We probed the hysteresis characteristics in the ordered assembly of MNPs using state of the art kinetic Monte Carlo (kMC) simulations technique. In particular, we investigate the hysteresis response as a function of aspect ratio Ar=ly/lxA_r = l_y/l_x, dipolar interaction strength hdh_d, anisotropy orientation angle α\alpha and direction of the applied oscillating magnetic field. The kMC algorithm implement in the present article is described in detail in the references [19, 32, 44]. Therefore, we do not restate it to avoid duplications. It is a well-known fact that heat is dissipated due to hysteresis. The amount of heat dissipation equals the hysteresis loop area EHE_H, which can be numerically evaluated using the following expression [11]

EH=∮M(H) dH,(6)E_H = \oint M(H)\,\mathrm{d}H, \tag*{(6)}

The above integral is calculated over one complete cycle of the external magnetic field. M(H)M(H) is the magnetization of system at magnetic field HH.

Simulations Results

We consider magnetite (Fe3_3O4_4) nanoparticles with the following values of system parameters: D=8D = 8 nm, Keff=13×103K_{\mathrm{eff}} = 13 \times10^{3} Jm−3^{-3}, and Ms=4.77×105M_s = 4.77 \times10^{5} Am−1^{-1}. We have considered five values of system sizes viz lx×ly=20×20l_x \times l_y = 20 \times20, 10×4010 \times40, 4×1004 \times100, 2×2002 \times200, and 1×4001 \times400. So, the total number of MNPs in the system is 400. The corresponding aspect ratio Ar=ly/lxA_r = l_y/l_x of the system is 1.0, 4.0, 25.0, 100, and 400, respectively. The control parameter hdh_d is varied from 0 to 1.0 to model the effect of the dipolar interaction strength. The anisotropy axis orientation angle α\alpha is varied between 0 to 90°. Therefore, it captures the physics of perfectly aligned to perpendicularly aligned anisotropy axes. All the simulations are performed at temperature T=300T = 300 K. We have applied the oscillating magnetic field along xx and yy-direction with respect to the sample. The magnetic field strength μoHo\mu_o H_o and ν\nu is taken as 0.10 T and 10510^{5} Hz, respectively.

First, we investigate the hysteresis behaviour in the square array of nanoparticles as a function of dipolar interaction strength, anisotropy angle and direction of the external magnetic field. Figure 2 shows the magnetic hysteresis curve as a function of α\alpha with four representative values of hd=0.2h_d = 0.2, 0.4, 0.6 and 1.0. The alternating magnetic field is applied along xx and yy-direction. The magnetic field (xx-axis) and magnetization axes (yy-axis) are rescaled by the single-particle anisotropy field HK=2Keff/MsH_K = 2K_{\mathrm{eff}}/M_s and MsM_s, respectively. We can infer the following observations from hysteresis curves with the magnetic field along xx-direction: (1) The hysteresis loop area decreases as α\alpha is varied from 0 to 90° with weak dipolar interaction (hd≤0.2h_d \leq0.2). (2) The double-loop hysteresis curve emerges with aligned anisotropy (α=0∘\alpha= 0^\circ) and moderate interaction strength hd=0.4h_d = 0.4, which is an indication of the antiferromagnetic coupling dominance. (3) The hysteresis loop area also increases with α\alpha for substantial hdh_d. (4) The hysteresis curve has all the signatures of antiferromagnetic coupling dominance with the most substantial dipolar interaction strength (hd=1.0h_d = 1.0), irrespective of α\alpha. Remarkably, the hysteresis behaviour with the field applied along yy-direction and α=0∘\alpha= 0^\circ is precisely similar to that of α=90∘\alpha= 90^\circ and μoH⃗=Hox^\mu_o \vec{H} = H_o \hat{x}, irrespective of hdh_d. Similar observations can also be made for other complementary angles α\alpha. The emergence of double loop hysteresis is a peculiar characteristic of antiferromagnetic interaction. It can be explained by the fact that dipolar interaction induces a biasing field in the plane of the ordered arrays. In contrast, the anisotropy field instigates the magnetization to align along a particular direction depending on the orientational angle α\alpha. Consequently, the magnetization follows the external magnetic field as long as its strength is less than the coercive field, i.e. μoH/HK<0.5\mu_o H/H_K < 0.5. The dipolar interaction dominates the hysteresis for μoH/HK>0.5\mu_o H/H_K > 0.5, resulting in a spontaneous transition in magnetization orientation. Yang et al. also observed similar double loop hysteresis in the thin-film [46]. Our observations are in excellent agreement with them. The observation of antiferromagnetic coupling dominance in the square arrangement of MNPs is also in excellent agreement with the work of Ewerlin et al. and Chen et al. [47, 48].

Image not extracted

Figure 2. Magnetic hysteresis in the square array of nanoparticles as a function of dipolar interaction strength and external field direction. We have taken into account four values of interaction strength: hd=0.2h_d = 0.2 [(a) and (e)], hd=0.4h_d = 0.4 [(b) and (f)], hd=0.6h_d = 0.6 [(c) and (g)], and hd=1.0h_d = 1.0 [(d) and (h)]. Double loop hysteresis emerges even with moderate hd=0.4h_d = 0.4, signature of antiferromagnetic coupling.

The anisotropy axis direction should affect the hysteresis mechanism in the rectangular array of MNPs. So, we now probe the hysteresis response in a rectangular arrangement of nanoparticles. We plot the hysteresis curve with aspect ratio Ar=4.0A_r = 4.0 in Figure 3. We have used same set of other parameters as that of Figure 2. Irrespective of the external magnetic field direction, the hysteresis loop area decreases with α\alpha for small hdh_d. Interestingly, the hysteresis curve’s shape changes from a line to the double-loop hysteresis as α\alpha and hdh_d are increased with the external alternating magnetic field along the xx-axis (shorter length) of the system. Remarkably, the hysteresis loop area increases with hdh_d and collinear anisotropy axis (α=0∘\alpha= 0^\circ), provided the magnetic field is applied along the yy-direction (long axis of the system). It is because ferromagnetic coupling gets enhanced with hdh_d in such an anisotropic system. Interestingly, double loop hysteresis tends to emerge even with moderate hd=0.4h_d = 0.4 as α\alpha is varied from 00 to 90∘90^\circ. In such a case, the nature of the dipolar interaction changes from ferromagnetic to antiferromagnetic as the orientation of the anisotropy axis varied from perfectly collinear to perpendicularly aligned [45]. Our results with α=0∘\alpha= 0^\circ and 90∘90^\circ are in perfect agreement with the work of Li et al. and Yoshida et al. [34, 49]. We could not compare for other α\alpha as they have shown results only with two extremum α\alpha values. These observations also agree well with the work of Sahoo et al. [50].

Image not extracted

Figure 3. Hysteresis response in a system with aspect ratio Ar=4.0A_r = 4.0 as a function of hdh_d and external field direction. We have taken four values of interaction strength: hd=0.2h_d = 0.2 [(a) and (e)], hd=0.4h_d = 0.4 [(b) and (f)], hd=0.6h_d = 0.6 [(c) and (g)], and hd=1.0h_d = 1.0 [(d) and (h)]. The hysteresis loop area is minimal with small hdh_d, independent of the external field direction. The antiferromagnetic coupling is dominant with appreciable hdh_d, resulting in double-loop hysteresis.

Next, we study the magnetic hysteresis behaviour in systems with very large aspect ratios. We plot the hysteresis curves with Ar=25.0A_r = 25.0 and 100.0100.0 in Figure 4 and Figure 5, respectively. One can draw the following vital conclusions from these curves: (i) In the case of aligned anisotropy (α=0∘\alpha= 0^\circ) and the external magnetic field along the xx-direction, the magnetization ceases to follow the external field, resulting in the non-hysteresis behaviour. (ii) The shape of the hysteresis curve also changes from a straight line to double loop hysteresis with α\alpha and hdh_d in such case. (iii) The hysteresis loop area also increases with hdh_d and aligned anisotropy axis (α=0∘\alpha= 0^\circ), provided the external field is along the yy-direction. (iv) There is also a decrease in the hysteresis loop area as α\alpha is varied from 0∘0^\circ to 90∘90^\circ. Moreover, the dominance of antiferromagnetic coupling increases with α\alpha and substantial hdh_d. Consequently, the double hysteresis loop emerges even with moderate interaction strength (hd≥0.4h_d \ge0.4). In recent work, Yuan et al. fabricated similar thin films of nanoparticles [51]. Similar MNPs ensembles are also found in biosystems, such as in the brain of migratory birds [51]. Our results could be extremely useful in assessing the hysteresis characteristics of such an assembly of MNPs. These observations are also in perfect qualitative agreement with the work of Alphandéry et al. [52].

Image not extracted

Figure 4. Hysteresis curves with aspect ratio Ar=25.0A_{\mathrm{r}} = 25.0 as a function of magnetic interaction strength hdh_{\mathrm{d}} and external field direction. We have taken four values of interaction strength: hd=0.2h_{\mathrm{d}} = 0.2 [(a) and (e)], hd=0.4h_{\mathrm{d}} = 0.4 [(b) and (f)], hd=0.6h_{\mathrm{d}} = 0.6 [(c) and (g)], and hd=1.0h_{\mathrm{d}} = 1.0 [(d) and (h)]. The double loop hysteresis emerges for appreciable hdh_{\mathrm{d}} with external field along xx-direction. The ferromagnetic coupling starts to dominant the hysteresis with external field applied along the yy-axis.

Image not extracted

Figure 5. Hysteresis curves in a system with very large aspect ratio Ar=100.0A_r = 100.0 as a function of hdh_d and external field direction. We have taken into account four values of interaction strength: hd=0.2h_d = 0.2 [(a) and (e)], hd=0.4h_d = 0.4 [(b) and (f)], hd=0.6h_d = 0.6 [(c) and (g)], and hd=1.0h_d = 1.0 [(d) and (h)]. Non-hysteresis is observed with external field along xx-axis and α≤45∘\alpha\le45^\circ. The dipolar interaction induces ferromagnetic interaction with external field along the yy-direction and α<90∘\alpha< 90^\circ.

The anisotropy axis orientation and external magnetic field direction should strongly affect the hysteresis in the highly anisotropic system. Therefore, we now analyze the hysteresis mechanism in a system with an extremely high value of aspect ratio, i.e. Ar=400A_r = 400 in Figure 6. It corresponds to a linear array of dipolar interacting MNPs. In the case of the external magnetic field along the shorter length (along the xx-axis) and moderate dipolar interaction strength hdh_d, the non-hysteresis is observed with aligned anisotropy (α=0∘\alpha= 0^\circ). The hysteresis properties also depend weakly on α\alpha and hdh_d in such a case. Interesting physics emerges with the external magnetic field along the long axis of the system, i.e. yy-direction. There is an increase in the ferromagnetic interaction with hdh_d, and collinear anisotropy axis (α=0∘\alpha= 0^\circ). Consequently, the hysteresis loop area is exceedingly large, and it increases with hdh_d. The hysteresis loop area also decreases with α\alpha, but there is a weak dependence on it. The results with external field along the MNPs array axis is in perfect agreement with our recent work [19]. These are also in excellent qualitative agreement with the work of Serantes et al. [53]. The observation of the anisotropic effect of dipolar interaction with the chain-like array of MNPs is also in agreement with the recent work of Valdés et al. [54].

Image not extracted

Figure 6. Magnetic hysteresis variation with highly anisotropic system, Ar=400.0A_r = 400.0, hdh_d and external magnetic field direction. We have taken four values of interaction strength: hd=0.2h_d = 0.2 [(a) and (e)], hd=0.4h_d = 0.4 [(b) and (f)], hd=0.6h_d = 0.6 [(c) and (g)], and hd=1.0h_d = 1.0 [(d) and (h)]. Non-hysteresis is observed with external field along x-direction and most substantial hd=1.0h_d = 1.0, independent of α\alpha. Ferromagnetic coupling increases with hdh_d and external field along the long axis of the system (y-direction), irrespective of α\alpha.

We now study the coercive field μoHc\mu_o H_c variation with dipolar interaction strength and anisotropy axis orientation to quantify the magnetic hysteresis. Figure 7 shows the variation of μoHc\mu_o H_c as a function of α\alpha and hdh_d. We have considered five values of Ar=1.0A_r = 1.0, 4.04.0, 25.025.0, 100100, and 400400. We have also taken into account two cases of magnetic field direction along the xx and yy-axis. The coercive field values are extracted from the corresponding hysteresis curve. In the case of weak dipolar interaction (hd≤0.4h_d \le0.4), there is an increase in μoHc\mu_o H_c with hdh_d for a given α\alpha and relatively smaller Ar≤4.0A_r \le4.0, irrespective of magnetic field direction. While for the field applied along the xx-direction, μoHc\mu_o H_c is minimal with weakly interacting MNPs. (hd≤0.2h_d \le0.2) and large ArA_r, independent of the anisotropy axis orientation α\alpha. In contrast, it is enormous even with moderate hdh_d and depends weakly on α\alpha and hdh_d in such a case. Remarkably, μoHc\mu_o H_c increases with hdh_d for a fixed α\alpha with the field applied along yy-direction and significant ArA_r. It is because the ferromagnetic coupling is enhanced with hdh_d. In other words, the dipolar interaction creates an anisotropy termed as shape anisotropy in such a case, resulting in an enhanced coercive field. The study of remanence variation can also provide a better understanding of hysteresis characteristics. Therefore, we now analyze the variation of remanent magnetization MrM_r as a function of α\alpha and hdh_d in Figure 8. MrM_r is extremely small with weak dipolar interaction, irrespective of ArA_r and α\alpha. In the case of field applied along the xx-direction, there is an increase in MrM_r with an increase in anisotropy angle α\alpha for a given hdh_d, irrespective of ArA_r. On the other hand, MrM_r increases with hdh_d for Ar=1.0A_r = 1.0 and field applied along yy-direction even with large dipolar interaction strength hd≤0.9h_d \le0.9, independent of anisotropy axis orientation. There is an increase in MrM_r with hdh_d for large ArA_r, irrespective of α\alpha because of an enhancement in the ferromagnetic coupling. Remarkably it reaches close to 1.0 with the most substantial dipolar interaction. It means that ferromagnetic coupling strength is the largest in such a case.

Image not extracted

Figure 7. The variation of coercive field μoHc\mu_o H_c (scaled by HKH_K) as a function of hdh_d and α\alpha with various ArA_r. The magnetic field is applied along x-axis [(a)-(e)] and y-axis [(f)-(j)]. Irrespective of the applied field direction and weak dipolar interaction (hd≤0.4h_d \le0.4), there is an increase in μoHc\mu_o H_c with hdh_d for a fixed α\alpha, provided the aspect ratio is not very large Ar≤4.0A_r \le4.0. Notably, μoHc\mu_o H_c increases with hdh_d for a given α\alpha and the external field along the yy-direction.

Image not extracted

Figure 8. The variation of remanent magnetization MrM_r as a function of hdh_d and α\alpha with various ArA_r. The alternating magnetic field is applied along xx [(a)-(e)] and yy axes [(f)-(j)]. MrM_r increases with α\alpha for a fixed hdh_d and magnetic field along the xx-axis, irrespective of ArA_r. There is also an increase in MrM_r with hdh_d for large ArA_r and field applied along the long axis of the system. It is because ferromagnetic coupling enhances with hdh_d in such a case.

Finally, we study the amount of heat dissipation EHE_H variation with hdh_d and α\alpha in Figure 9. The value of EHE_H is minimal for negligible and weak magnetic interaction strength. In the case of the field applied along the xx-axis, EHE_H increases with α\alpha for a moderate and fixed hdh_d. While for the field applied along the yy-direction, EHE_H increases with hdh_d for a fixed α\alpha and relatively smaller Ar≤4.0A_r \le4.0, provided hd≤0.8h_d \le0.8. It starts to diminish as hdh_d is further increased, indicating the dominance of antiferromagnetic coupling. Remarkably, the value of EHE_H increases with hdh_d and large ArA_r, irrespective of α\alpha. It is because the dipolar interaction promotes ferromagnetic coupling in such a system. Notably, EHE_H is exceedingly large with the highly anisotropic system Ar=400A_r = 400. The decrease of EHE_H with α\alpha for the external magnetic field along the yy-axis is in perfect qualitative agreement with the work of Serantes et al. [53]. It is also in perfect agreement with our recent work [19]. Conde-Leborán et al. also obtained similar results in a linear array of MNPs [55]. Our results also agree well with them.

Ten surface plots showing the variation of heat dissipated with $h_d$ and $\alpha$ for various $A_r$, with the external oscillating magnetic field applied along the x-axis and y-axis

Figure 9. The variation of the amount of heat dissipated EHE_H with hdh_d and α\alpha for various ArA_r. The external oscillating magnetic field is applied along x-axis [(a)–(e)] and y-axis [(f)–(j)]. EHE_H is minimal for negligible and weak dipolar interaction, independent of ArA_r, α\alpha, and external field direction. There is an increase in EHE_H with hdh_d and large ArA_r, provided the alternating magnetic field is along the long axis of the system, irrespective of α\alpha.

Summary and conclusion

We have investigated magnetic hysteresis in the two-dimensional (lx×lyl_x \times l_y) array of nanoparticles using numerical simulations with aligned anisotropy. In particular, we probed the hysteresis mechanism as a function of aspect ratio Ar=ly/lxA_r = l_y/l_x, anisotropy axis orientational angle α\alpha, magnetic interaction strength hdh_d, and the applied alternating magnetic field direction using kinetic Monte Carlo simulations. The assumption of an aligned anisotropy axis provides an elegant way to investigate the precise role of anisotropy and dipolar interaction on the hysteresis characteristics in such an ordered assembly of MNPs. The superparamagnetic character is dominant with small dipolar interaction (hd≤0.2h_d \le0.2), resulting in the minimal hysteresis loop area. These observations are also robust to the aspect ratio, anisotropy axis orientation, and the external magnetic field directions. Interestingly, the hysteresis behaviour with the external field along the xx-direction and relatively smaller ArA_r is precisely the same as with the field along the yy-direction for complementary orientational angle α\alpha. It implies that the hysteresis characteristics with α=0∘\alpha= 0^\circ and the field applied along the xx-direction are the same as that of α=90\alpha= 90 and external magnetic field along the yy-direction. One can draw similar conclusions for other complementary angles also. Remarkably, double loop hysteresis emerges even with moderate dipolar interaction strength hd≈0.4h_d \approx0.4, reminiscent of antiferromagnetic coupling. These features are strongly dependent on the α\alpha and aspect ratio of the system. The dominance of the antiferromagnetic coupling is the strongest with the most substantial dipolar interaction strength (hd=1.0h_d = 1.0), irrespective of α\alpha and magnetic field direction. Interestingly, non-hysteresis is observed in the highly anisotropic system (Ar=400A_r = 400) even with small dipolar interaction strength and external magnetic along the shorter axis of the sample (xx-direction). On the other hand, the dipolar interaction promotes the ferromagnetic coupling with the applied field along the yy-direction, the long axis of the sample in such a system. Consequently, the hysteresis loop area increases with hdh_d, and it has a weak dependence on the anisotropy axis orientation. Ordered arrays of MNPs with aligned anisotropy axes are frequently observed in experiments [56, 57, 58]. We believe that our results could be beneficial in predicting the hysteresis properties of such MNPs ensembles.

The study of coercive field μ0Hc\mu_0 H_c, remanent magnetization MrM_r, and the heat dissipation EHE_H also provides vital quantitative information regarding hysteresis. Irrespective of the applied field direction and weak dipolar interaction (hd≤0.4h_d \le0.4), there is an increase in μ0Hc\mu_0 H_c with hdh_d for a fixed α\alpha, provided the aspect ratio is not very large Ar≤4.0A_r \le4.0. Notably, μ0Hc\mu_0 H_c increases with hdh_d for a given α\alpha and the external field along the yy-direction. In such a case, the dipolar interaction creates an additional anisotropy, known as the shape anisotropy, resulting in an enhancement in the coercive field. MrM_r increases with α\alpha for a fixed hdh_d and magnetic field along the xx-axis, irrespective of ArA_r. There is also an increase in MrM_r with hdh_d for large ArA_r and field applied along the long axis of the system. It is because ferromagnetic coupling enhances with hdh_d in such a case. The amount of heat dissipation EHE_H due to the hysteresis also strongly depends on these parameters. EHE_H is minimal for negligible and weak dipolar interaction, irrespective of ArA_r, α\alpha, and external field direction. Conversely, EHE_H increases with α\alpha and hdh_d (hd≤0.4h_d \le0.4), provided the external field is along the xx-axis. Remarkably, there is an enhancement in EHE_H with hdh_d and large ArA_r and magnetic field applied along the long axis of the system, irrespective of α\alpha. In such a case, the dipolar interaction induces ferromagnetic coupling, resulting in increased heat dissipation.

In conclusion, we analyzed the hysteresis characteristics in the ordered arrays of nanoparticles with aligned anisotropy axes using extensive numerical simulations. We have also studied the hysteresis response as a function of dipolar interaction strength, the system’s aspect ratio, and the external magnetic field directions. The hysteresis properties depend strongly on these parameters. The assumption of common easy axes is one of the elegant routes to probe the precise role of anisotropy and dipolar interactions. Such systems are also realized in experiments and beneficial in diverse technological applications [59, 60]. We believe that our results are extremely useful in these contexts. Therefore, we believe that the present work could instigate combined efforts in experimental, analytical, and computational research for these precious and physics enriched systems.

ACKNOWLEDGMENTS

Most of the numerical simulations presented in this work have been performed in the Department of Physics, Indian Institute of Technology (IIT) Delhi, India. The author is thankful to Prof. Varsha Banerjee for providing the computational facilities at IIT Delhi.

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. [1]T. P. Bigioni, X.-M. Lin, T. T. Nguyen, E. I. Corwin, T. A. Witten, and H. M. Jaeger, Nature materials 5, 265 (2006).
  2. [2]L. Jiang, X. Chen, N. Lu, and L. Chi, Accounts of chemical research 47, 3009 (2014).
  3. [3]V. Håkonsen, G. Singh, J. A. De Toro, P. S. Normile, E. Wahlström, J. He, and Z. Zhang, Advanced Science 8, 2002683 (2021).
  4. [4]S. Polarz, Advanced Functional Materials 21, 3214 (2011).
  5. [5]S. Sacanna, D. J. Pine, and G.-R. Yi, Soft matter 9, 8096 (2013).
  6. [6]A. Böker, J. He, T. Emrick, and T. P. Russell, Soft matter 3, 1231 (2007).
  7. [7]A. Horechyy, N. E. Zafeiropoulos, B. Nandan, P. Formanek, F. Simon, A. Kiriy, and M. Stamm, Journal of Materials Chemistry 20, 7734 (2010).
  8. [8]A. C. Gandhi, R. Pradeep, Y.-C. Yeh, T.-Y. Li, C.-Y. Wang, Y. Hayakawa, and S. Y. Wu, ACS Applied Nano Materials 2, 278 (2018).
  9. [9]G. Reiss and A. Hütten, Nature materials 4, 725 (2005).
  10. [10]P. Couvreur, Advanced drug delivery reviews 65, 21 (2013).
  11. [11]M. Anand, J. Carrey, and V. Banerjee, Physical Review B 94, 094425 (2016).
  12. [12]Q. A. Pankhurst, J. Connolly, S. K. Jones, and J. Dobson, Journal of physics D: Applied physics 36, R167 (2003).
  13. [13]J. Carrey, B. Mehdaoui, and M. Respaud, Journal of Applied Physics 109, 083921 (2011).
  14. [14]P. Poddar, T. Telem-Shafir, T. Fried, and G. Markovich, Physical Review B 66, 060403 (2002).
  15. [15]M. Anand, Journal of Magnetism and Magnetic Materials 522, 167538 (2021).
  16. [16]J. G. Ovejero, D. Cabrera, J. Carrey, T. Valdivielso, G. Salas, and F. J. Teran, Physical Chemistry Chemical Physics 18, 10954 (2016).
  17. [17]E. Winkler, R. Zysler, M. V. Mansilla, D. Fiorani, D. Rinaldi, M. Vasilakaki, and K. Tрохidou, Nanotechnology 19, 185702 (2008).
  18. [18]S. Mørup, EPL (Europhysics Letters) 28, 671 (1994).
  19. [19]M. Anand, Journal of Applied Physics 128, 023903 (2020).
  20. [20]A. Kostopoulou, K. Brintakis, M. Vasilakaki, K. Trohidou, A. Douvalis, A. Lascialfari, L. Manna, and A. Lappas, Nanoscale 6, 3764 (2014).
  21. [21]P. Torche, C. Munoz-Menendez, D. Serantes, D. Baldomir, K. Livesey, O. Chubykalo-Fesenko, S. Ruta, R. Chantrell, and O. Hovorka, Physical Review B 101, 224429 (2020).
  22. [22]S. Masunaga, R. Jardim, and J. Rivas, Journal of Applied Physics 109, 07B521 (2011).
  23. [23]D. Kechrakos and K. Trohidou, Journal of nanoscience and nanotechnology 8, 2929 (2008).
  24. [24]W. Figueiredo and W. Schwarzacher, Journal of Physics: Condensed Matter 19, 276203 (2007).
  25. [25]E. Myrovali, N. Maniotis, A. Makridis, A. Terzopoulou, V. Ntomprougkidis, K. Simeonidis, D. Sakellari, O. Kalogirou, T. Samaras, R. Salikhov, et al., Scientific Reports 6, 1 (2016).
  26. [26]K. De’Bell, A. MacIsaac, I. Booth, and J. Whitehead, Physical Review B 55, 15108 (1997).
  27. [27]J. Luttinger and L. Tisza, Physical Review 70, 954 (1946).
  28. [28]P. Politi, M. G. Pini, and R. Stamps, Physical Review B 73, 020405 (2006).
  29. [29]A. MacIsaac, J. Whitehead, K. De’Bell, and P. Poole, Physical Review Letters 77, 739 (1996).
  30. [30]D. Schmool, R. Rocha, J. Sousa, J. Santos, G. Kakazei, J. Garitaonandia, and L. Lezama, Journal of applied physics 101, 103907 (2007).
  31. [31]C. Martinez-Boubeta, K. Simeonidis, A. Makridis, M. Angelakeris, O. Iglesias, P. Guardia, A. Cabot, L. Yedra, S. Estradé, F. Peiró, et al., Scientific reports 3, 1 (2013).
  32. [32]M. Anand, arXiv preprint arXiv:2104.02961 (2021).DOI
  33. [33]K. Deng, Z. Luo, L. Tan, and Z. Quan, Chemical Society Reviews 49, 6002 (2020).
  34. [34]Q. Li, C. W. Kartikowati, T. Iwaki, K. Okuyama, and T. Ogi, Royal Society Open Science 7, 191656 (2020).
  35. [35]T. Wen, Y. Li, D. Zhang, Q. Zhan, Q. Wen, Y. Liao, Y. Xie, H. Zhang, C. Liu, L. Jin, et al., Journal of colloid and interface science 497, 14 (2017).
  36. [36]P. Allia, G. Barrera, and P. Tiberto, Journal of Magnetism and Magnetic Materials 496, 165927 (2020).
  37. [37]A. K. Dyab, M. Ozmen, M. Ersoz, and V. N. Paunov, Journal of Materials Chemistry 19, 3475 (2009).
  38. [38]D. Hofflner, M. Kundt, A. M. Schmidt, E. Kentzinger, P. Bender, and S. Disch, Faraday discussions 181, 449 (2015).
  39. [39]M. Anand, arXiv preprint arXiv:2106.14271 (2021).arxiv.org/abs/2106.14271
  40. [40]B. Mehdauoi, R. Tan, A. Meffre, J. Carrey, S. Lachaize, B. Chaudret, and M. Respaud, Physical Review B 87, 174419 (2013).
  41. [41]V. Krishnamurthy, G. Mankey, B. He, M. Piao, J. Wiest, D. E. Nikles, L. Porcar, and J. Robertson, Physical Review E 77, 031403 (2008).
  42. [42]M. Anand, J. Carrey, and V. Banerjee, Journal of Magnetism and Magnetic Materials 454, 23 (2018).
  43. [43]N. Usov, O. Serebryakova, and V. Tarasov, Nanoscale Research Letters 12, 1 (2017).
  44. [44]R. Tan, J. Carrey, and M. Respaud, Physical Review B 90, 214421 (2014).
  45. [45]M. Anand, V. Banerjee, and J. Carrey, Physical Review B 99, 024402 (2019).
  46. [46]J.-S. Yang, C.-R. Chang, and C. Lai, Journal of magnetism and magnetic materials 239, 28 (2002).
  47. [47]M. Ewerlin, D. Demirbas, F. Brüssing, O. Petracic, A. A. Ünal, S. Valencia, F. Kronast, and H. Zabel, Physical Review Letters 110, 177209 (2013).
  48. [48]B. Chen, H. Xu, C. Ma, S. Mattauch, D. Lan, F. Jin, Z. Guo, S. Wan, P. Chen, G. Gao, et al., Science 357, 191 (2017).
  49. [49]T. Yoshida, Y. Matsugi, N. Tsujimura, T. Sasayama, K. Enpuku, T. Viereck, M. Schilling, and F. Ludwig, Journal of Magnetism and Magnetic Materials 427, 162 (2017).
  50. [50]Y. Sahoo, M. Cheon, S. Wang, H. Luo, E. Furlani, and P. Prasad, The Journal of Physical Chemistry B 108, 3380 (2004).
  51. [51]H. Yuan, I. J. Zvonkina, A. M. Al-Enizi, A. A. Elzatahry, J. Pyun, and A. Karim, ACS applied materials & interfaces 9, 11290 (2017).
  52. [52]E. Alphandéry, Y. Ding, A. Ngo, Z. Wang, L. Wu, and M. Pileni, Acs Nano 3, 1539 (2009).
  53. [53]D. Serantes, K. Simeonidis, M. Angelakeris, O. Chubykalo-Fesenko, M. Marciello, M. D. P. Morales, D. Baldomir, and C. Martinez-Boubeta, The Journal of Physical Chemistry C 118, 5927 (2014).
  54. [54]D. P. Valdés, E. Lima Jr, R. D. Zysler, and E. De Biasi, Physical Review Applied 14, 014023 (2020).
  55. [55]I. Conde-Leborán, D. Serantes, and D. Baldomir, Journal of Magnetism and Magnetic Materials 380, 321 (2015).
  56. [56]A. J. Amali, P. Saravanan, and R. K. Rana, Angewandte Chemie 123, 1354 (2011).
  57. [57]K. Hu, J. Sun, Z. Guo, P. Wang, Q. Chen, M. Ma, and N. Gu, Advanced Materials 27, 2507 (2015).
  58. [58]K. Jain, S. Pathak, and R. Pant, RSC advances 6, 70943 (2016).
  59. [59]D. Fragouli, R. Buonsanti, G. Bertoni, C. Sangregorio, C. Innocenti, A. Falqui, D. Gatteschi, P. D. Cozzoli, A. Athanassiou, and R. Cingolani, ACS nano 4, 1873 (2010).
  60. [60]D. Lisjak and A. Mertelj, Progress in Materials Science 95, 286 (2018).

Paper details

vlnc:W_ughbsf43
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents