Intra- and interparticle magnetism of cobalt-doped iron-oxide nanoparticles encapsulated in a synthetic ferritin cage

We present an in-depth examination of the composition and magnetism of cobalt $\left({\mathrm{Co}}^{2+}\right)$-doped iron-oxide nanoparticles encapsulated in Pyrococcus furiosus ferritin shells. We show that the ${\mathrm{Co}}^{2+}$ dopant ions were incorporated into the $\gamma \text{−}{\mathrm{Fe}}_2{\mathrm{O}}_3/{\mathrm{Fe}}_3{\mathrm{O}}_4$ core, with small paramagnetic-like clusters likely residing on the surface of the nanoparticle that were observed for all cobalt-doped samples. In addition, element-specific characterization using Mössbauer spectroscopy and polarized x-ray absorption indicated that ${\mathrm{Co}}^{2+}$ was incorporated exclusively into the octahedral B sites of the spinel-oxide nanoparticle. Comparable superparamagnetic blocking temperatures, coercivities, and effective anisotropies were obtained for 7%, 10%, and 12% cobalt-doped nanoparticles, and were only slightly reduced for 3% cobalt, indicating a strong effect of cobalt incorporation, with a lesser effect of cobalt content. Due to the regular particle size and separation that result from the use of the ferritin cage, a comparison of the effects of interparticle interactions on the disordered assembly of nanoparticles was also obtained that indicated significantly different behaviors between undoped and cobalt-doped nanoparticles.

Figure 1

A typical low-resolution TEM image of Fe-based ferritin nanoparticles.

Figure 2

(a) X-ray diffraction pattern for undoped Fe-oxide sample, with the solid line indicating the result of the Rietveld refinement, and (inset) a selected $2\theta$ range of the background subtracted pattern for all samples. Results of the analysis of the XRD patterns showing including (b) the lattice parameter $a$, (c) the octahedral site occupancy, obtained from a Rietveld refinement of the patterns, and (d) $I_{220}/I_{400}$ variation with cobalt content.

Figure 3

(a) Mössbauer spectrum collected at 10 K for the undoped nanoparticle sample, with the results of a component fit, described in the text, and (b) distribution of hyperfine fields $\left(B_{hf}\right)$ observed for the Mössbauer spectra for the undoped and cobalt-doped nanoparticles that were consistent with results using individual components to represent the A and B sites of the Fe-oxide-based core [9].

Figure 4

Fe (a) and Co (b) XAS (top) and XMCD (bottom) spectra for the 0% (red), 3% (blue), 10% (green), and 12% (black) cobalt-doped nanoparticles measured at 10 K and 5 T.

Figure 5

Fe and Co magnetization measured from the XMCD spectra for 0% (black circles), 3% (red squares), and 12% (blue diamonds) as a function of field (a) and temperature (b). Note that the same scale was used for all Fe sites, and the Co scales are $4\times$ the Fe scale.

Figure 6

Field-dependent XMCD signal measured at 10 K for the Co (cyan $▿)$ and Fe $B_1$ (red $\square )$, $B_2$ (green $\diamond )$, and A sites (blue $▵)$ are shown, normalized to the value at 5 T. The first quarter of the hysteresis loops measured at 10 K are also shown (black curves).

Figure 7

Representative XMCD spectra collected over the $L_{2,3}$ edges of Fe (a) and Co (b). The dashed lines indicate the integrated XMCD signal, and $p$ and $q$ were used to determine $m_l/m_s$ using sum-rule analysis.

Figure 8

Ratio of the orbital and spin moments for ${\mathrm{Co}}^{2+}$ in the 3% and 12% cobalt-doped nanoparticles, determined using sum-rule analysis, as a function of (a) applied field and (b) temperature.

Figure 9

In-phase $\left({\chi }^{\prime }\right)$ (top) and out-of-phase $\left({\chi }^{\prime \prime }\right)$ (bottom) ac-susceptibility for undoped and cobalt-doped nanoparticles. Note the change in scale between samples.

Figure 10

(a) Frequency dependence of the maximum of the in-phase ac-susceptibility curves, with the lines indicating a fit to the Vogel-Fulcher law. Note the break in the scale between the 0% and cobalt-doped curves. (b) The effective anisotropies and (c) values of $T_0$, obtained from the Vogel-Fulcher fits, relative to the temperature of the maximum of ${\chi }^{\prime }\left(T\right)$ for 10 Hz.

Figure 11

Zero-field-cooled (ZFC) (red $\square )$ and field-cooled (FC) (black $◯)$ dc-susceptibility (right) and difference between the FC and ZFC magnetizations (right). The inset shows irreversibility persisting to high temperatures for the undoped nanoparticles.

Figure 12

(a) In-phase $\left({\chi }^{\prime }\right)$ ac-susceptibility curves measured using 10 Hz. (b) Zero-field-cooled magnetization curves.

Figure 13

Inverse dc-susceptibility for the undoped nanoparticles, with the line indicating a fit to a Curie-Weiss-like temperature dependence. The inset shows the temperature intercept $\theta$ that reflects the effects of interparticle interactions.

Figure 14

Hysteresis loops at (a) 2 K and (b) 100 K measured after cooling the samples in an applied field of 5 T.

Figure 15

Temperature dependence of the coercivity $\left(H_C\right)$ with the temperature rescaled to the onset temperature of the coercivity $T_{B,Hc}$. The lines are fits described in the text that indicate a $T^{1/2}$ dependence. The inset shows the anisotropy constant $\left(K\right)$ determined from the temperature dependence of $H_C$.

Figure 16

$\left(M-{\chi }_{\mathrm{HF}}H\right)$ vs $H/T$ curves measured for $T>T_B$ for (a) undoped and (b) 12% cobalt-doped nanoparticles.

Figure 17

Temperature dependence of the high-field susceptibility ${\chi }_{\mathrm{HF}}\left(T\right)$. The dashed lines indicate the contribution due to a surface anisotropy, which is independent of temperature.

Figure 18

Temperature dependence of the inverse high-field susceptibility ${\chi }_{\mathrm{HF}}\left(T\right)$, where the constant offset due to the surface anisotropy $\mathrm{[}{\chi }_{\mathrm{surf}}\left(T\right)\text{]}$ has been subtracted.

Figure 19

Temperature dependence of the saturation magnetization $\mathrm{[}{M}_S\left(T\right)\text{]}$. The lines indicate a fit described in the text.