Two-component magnetic structure of iron oxide nanoparticles mineralized in Listeria innocua protein cages

Share Embed


Descripción

JOURNAL OF APPLIED PHYSICS 107, 114703 共2010兲

Two-component magnetic structure of iron oxide nanoparticles mineralized in Listeria innocua protein cages Robert J. Usselman,1,a兲 Michael T. Klem,2,3 Stephen E. Russek,1 Mark Young,4,5 Trevor Douglas,6,5 and Ron B. Goldfarb1 1

National Institute of Standards and Technology, Boulder, Colorado 80305, USA Department of Chemistry and Geochemistry, Montana Tech of the University of Montana, Butte, Montana 59701, USA 3 Center for Advanced Supramolecular and Nano Systems, Montana Tech of the University of Montana, Butte, Montana 59701, USA 4 Department of Plant Sciences and Plant Pathology, Montana State University, Bozeman, Montana 59717, USA 5 Center for Bio-Inspired Nanomaterials, Montana State University, Bozeman, Montana 59717, USA 6 Department of Chemistry and Biochemistry, Montana State University, Bozeman, Montana 59717, USA 2

共Received 4 February 2010; accepted 19 March 2010; published online 7 June 2010; publisher error corrected 21 June 2010兲 Magnetometry was used to determine the magnetic properties of maghemite 共␥-Fe2O3兲 nanoparticles formed within Listeria innocua protein cage. The electron magnetic resonance spectrum shows the presence of at least two magnetization components. The magnetization curves are explained by a sum of two Langevin functions in which each filled protein cage contains both a large magnetic iron oxide core plus an amorphous surface consisting of small noncoupled iron oxide spin clusters. This model qualitatively explains the observed decrease in the temperature dependent saturation moment and removes an unrealistic temperature dependent increase in the particle moment often observed in nanoparticle magnetization measurements. © 2010 American Institute of Physics. 关doi:10.1063/1.3400033兴 I. INTRODUCTION

Magnetic nanoparticles are of interest in bionanotechnology with applications in fields such as diagnostics, magnetic resonance imaging, drug delivery, and cellular targeting. Nanoparticles that combine both physical properties amenable to imaging and chemical properties that facilitate specific desired interactions with biomolecules are important in biomedicine. Tailoring magnetic properties of nanoparticles requires precise control over the size, composition, magnetic moment, and anisotropy energy. Functional nanoparticles rely on accurate placement of relevant molecules on the surface of the nanoparticles. Biomimetic macromolecular structures are attractive for multimodal “smart” nanoagents. In particular, protein cages have been used as versatile platforms for biomineralization, protein coat engineering, and attachment of ligand groups.1–5 Protein cages provide a macromolecular architecture for preparing different metal compositions having unique magnetic characteristics with specific limiting sizes.6,7 Characterization of metal oxide cores mineralized within the protein cage is essential to understand how to tune the magnetic properties of nanoparticles for desired applications. Mineralization within protein cages produces nearly uniform nanoparticles with a maximum size limited by the inner diameter of the constraining vessel. Ferrimagnetic maghemite nanoparticle cores, with high surface-to-volume ratios, have a net magnetic moment resulting from unequal a兲

Author to whom correspondence should be addressed. Electronic mail: [email protected]. Tel.: 303-497-4975. FAX: 303-497-7364.

0021-8979/2010/107共11兲/114703/4/$30.00

opposing ion moment 共Fe2+ and Fe3+兲 sublattices. The physical mechanisms contributing to the magnetic properties of nanoparticles, which include the temperature and particle size dependences of the magnetic moments, are not well understood. Several investigations have reported that the magnetic moment increases with temperature8–11 while the saturation magnetization decreases for ferritin and small nanoparticles with diameters less than 20 nm. Mørop and Frandsen8 have suggested that a thermoinduced magnetization can explain such anomalies, whereas Silva et al.9 reported the importance of including the moment distribution in analyzing temperature dependent magnetic data. Contributions that affect the temperature dependent magnetic moment may include finite size effects, spin disorder, crystalline defects, and ligand-oxide surface bonding. Our simple Langevin analysis of maghemite nanoparticles in Listeria innocua Dps 共LDps兲 cages is consistent with these previous magnetometry observations. At high magnetic fields 共7 T兲, the magnetic moment versus field data shows a component that has a temperature dependent saturation and a component that does not saturate. The magnetization data can be well fit with a modified Langevin function that yields an effective magnetic moment per particle that increases with temperature, and a total saturation moment that decreases with temperature. An increase in the particle moment with a concurrent decrease in the total saturated moment does not make physical sense given a fixed number of particles. The magnetization curves can be better explained by the sum of two Langevin functions in which each filled protein cage contains both a large magnetic iron oxide core

107, 114703-1

© 2010 American Institute of Physics

114703-2

Usselman et al.

FIG. 1. X-band EMR of LDps iron oxide nanoparticles at 298 K.

and small noncoupled iron oxide clusters. This model explains the observed temperature dependent saturation moment and removes the unphysical increase in particle moment as a function of temperature. II. EXPERIMENTAL

In this study, maghemite 共␥-Fe2O3兲 nanoparticles were mineralized within the small quasispherical LDps protein cage growth chamber with a nominal inner diameter of 5 nm. Transmission electron microscopy images confirm the encapsulation of the iron oxide nanoparticles within the protein cages with average diameters of 4 ⫾ 0.9 nm. Details of the mineralization and characterization have been previously described.12 Electron magnetic resonance 共EMR兲 spectrum of a solution sample of 1 mg/mL LDps was recorded at room 298 K on a commercial x-band spectrometer with the following acquisition parameters: 10 mW power, 100 kHz modulation, 5 G modulation amplitude. The solution sample was freeze-dried and enclosed in a 1 cm long fused silica tube with an inner diameter of 0.3 mm. Magnetization was measured on a commercial SQUID magnetometer. III. RESULTS AND DISCUSSION

EMR studies of small nanoparticles as a function of temperature13 and frequency14 indicate the presence of at least two distinct magnetic components: a narrow linewidth at g = 2 and a broad linewidth. The narrow component was attributed to small fluctuating particles, whereas the broad linewidth is from large superparamagnetic nanoparticles. There does remain some uncertainty how to theoretically model the narrow species in EMR data, i.e., small particles15,16 or surface spins.17 Figure 1 is an x-band EMR spectrum of LDps maghemite iron oxide nanoparticle illustrating at least two magnetic components in the sample. The zero-field-cooled 共ZFC兲 and field-cooled 共FC兲 magnetization curve, measured in a small applied magnetic field of ␮0H = 20.0 mT, is shown in Fig. 2. Nanoparticles have a complex superparamagnetic behavior, with a blocking temperature Tb共d兲 ⬵ KaV / KB ln共␶ f 0兲 that increases with particle size. Here, Ka is the anisotropy energy density, V is the magnetic volume of the particle, ␶ is a characteristic measure-

J. Appl. Phys. 107, 114703 共2010兲

FIG. 2. 共Color online兲 Magnetization for FC and ZFC LDps iron oxide particles in 20.0 mT applied field as a function of temperature. The data is on a log scale to highlight the low temperature peak along with a model calculation that assumes a mean diameter of 3.6 nm, temperature independent anisotropy energy of 4 ⫻ 104 J / m3, and magnetization of 370 kA/m.

ment time, and f 0 is an attempt frequency that is on the order of the magnetic resonance frequency. The temperature dependent ZFC moment has a peak at low temperature whose position is indicative of a small superparamagnetic particle size. The peak is due to a transition from fixed moments at low temperatures to fluctuating moments at high temperatures. The peak width and ZFC-FC bifurcation point is a function of the size distribution of particles and the anisotropy energy density. The average blocking temperature determined by the position of the peak is 5 K, which corresponds to particle sizes with magnetic diameters of d = 3.5 nm. There was no observed shift in mean blocking temperature that is often associated with strong interparticle dipolar interactions.18–20 The model fits in Fig. 2 show the ZFC-FC data for LDps particles on a semilog plot to highlight the peak along with a model prediction. The model assumes that at low field the moment of the cage is due to a magnetic particle of diameter d which is paramagnetic for T ⬎ Tb共d兲, or a Stoner– Wohlfarth ferromagnet, for T ⬍ Tb共d兲. The FC curves show significant deviations from simple paramagnetic behavior assumed in the model which predicts that the moment should decrease as 1 / T for T Ⰷ Tb. The fall-off with temperature is significantly weaker than 1 / T. The low field data indicate that the cages cannot be modeled as a simple paramagnetic particle with a temperature independent moment. High-field magnetization curves were measured as a function of magnetic field for several temperatures ranging from 10 to 300 K. The moment versus field data for LDps are shown in Fig. 3. The magnetization behavior of these artificially mineralized cages is distinct from typical antiferromagnetic ferritin nanoparticle magnetization curves11,21–23 which show no saturation behavior at these field and temperature values. The data can be well fit 共the dashed lines in Fig. 3兲 to a modified Langevin function: M = M 0关coth共x兲 − 1 / x兴 + ␹0H, where the saturation magnetization M 0 = ␮ pN and x = ␮0␮ pH / kT, ␮ p is the average particle magnetic moment, N is the number of particles, and ␹0 is the susceptibility. Here, the susceptibility ␹0 is the sum of a paramagnetic component from the sample and a diamagnetic component from the sample holder. The normalized data do not lie on

114703-3

J. Appl. Phys. 107, 114703 共2010兲

Usselman et al.

FIG. 3. 共Color online兲 LDps moment vs field curves for the experimental data and both one-component 共dashed兲 and two-component 共solid兲 model data. The two-component model assumes a bimodal distribution of paramagnetic particles, each cage consists of one particle of mean diameter 3.6 nm and 200 particles with mean diameter of 0.7 nm 共8 ␮B兲. The parameters used in this calculation are identical to those used in Fig. 2. The inset is the temperature dependence of the moment for the 3.6 nm particle as determined from the model fits.

the same line in a universal Langevin curve 共data not shown兲. The curvature increases as a function of temperature which may imply interparticle dipolar interactions, however, the protein shell should minimize inter cage dipolar interactions. The protein cage thickness is approximately 2.5 nm, For a 1000 Bohr magneton core particle with a 1 nm magnetically disordered layer24 would produce a field of about 0.2 mT to its nearest neighbor. Despite possible weak dipolar interactions, the variation in the low field slopes indicates a temperature dependent ␮ p using a single Langevin model. The decrease in saturation magnetization and the increase in ␮ p with temperature can be better understood if it is hypothesized that the cages consist of multiple components22 including a large iron oxide particle with a diameter slightly less than the cage diameter, which saturates at fields on the order of 7 T, and numerous smaller iron oxide clusters that do not fully saturate.24 We model the system by assuming that the cage consists of one superparamagnetic particle of ⬃3.6 nm mean diameter, or ⬃1000 ␮B and 200 paramagnetic particles of 8 ␮B. The two-component model is simply the sum of two weighted Langevin functions, 1 for the core and 200 for the clusters with the total number of particles constant. The calculated diamagnetic component for the quartz tube was included in the simulations. The large particle volume was fit to a single value while the particle magnetization was allowed to vary as a function of temperature, Fig. 3 共inset兲. The inset shows the temperature dependence of the large magnetic moment, which decreases by ⬃20% at high temperatures, and is consistent with what is expected from bulk properties. This model can qualitatively explain the LDps data as shown by the solid lines in Fig. 3. The saturation moment per cage, as determined from fitting the curves in Fig. 3, is 2670 ␮B with 2136 Fe atoms/cage, which corresponds to a 4.8 nm sphere of ␥-Fe2O3. This model suggests the LDps nanoparticles have a superparamagnetic core of ⬃3.6 nm and a paramagnetic shell of ⬃1.2 nm. The bimodal distribution of particles qualitatively explains the apparent temperature dependent saturation magne-

tization, the linear field component that does not saturate, and does not invoke an increase in a particle magnetic moment as a function of temperature. The two-component model is consistent with the ZFC-FC data that show the existence of a noninteracting core region similar to that of the filled cage volume. The two-component model shows a decrease in the large particle moment at high temperature but does deviate from the experimental data at the low field fits in the magnetization data. Better fits of the variable temperature magnetization curves may be obtained for the associated low field slopes by including a distribution of large particles. Nonetheless, a simple two-component model simultaneously fits the five different temperatures and uses physically meaningful parameters including a conserved number of particles and a saturated moment that drops by 20 percent from 10 to 300 K. However, the two-component model does not explain the deviation from the universal Langevin curve. These effects may in fact be due to intracage interactions, which is beyond the scope of this paper. The significance of the two-component model is consistent with a core/shell nanoparticle structure with a welldefined core and an amorphous surface structure.20 Surface effects and the idea of surface spin clusters contribute significantly to qualitative trends observed in nanoparticle magnetization curves. These small clusters are in agreement with observed EMR data of narrow linewidths as small particles with high surface-to-volume ratios. Overall, the twocomponent model resolves the anomalous increase in magnetic moment as the saturation magnetization decreases observed in simple Langevin analysis of nanoparticle magnetization curves. IV. CONCLUSIONS

We have presented EMR, ZFC-FC magnetization data, and variable temperature magnetization curves for Listeria innocua protein cage mineralized with maghemite 共␥-Fe2O3兲. The blocking temperature gives superparamagnetic core diameters of 3.6 nm. A simple Langevin analysis of the moment versus field data yields unphysical parameters: the saturation moment decreases rapidly with temperature while the moment per cage, taken from the low field slope, strongly increases with temperature. A two-component model that includes a core region plus many uncoupled clusters, with a moment near 8 ␮B, qualitatively explains the observed temperature dependent saturated moment, predicts a temperature dependent moment which is consistent with maghemite bulk properties, and requires fewer fitting parameters. ACKNOWLEDGMENTS

This research was supported in part by grants from the National Science Foundation 共Grant No. CBET-0709358兲 and Office of Naval Research 共Grant No. N00014-03-10692兲. R.J.U. is a National Research Council postdoctoral associate. Work partially supported by U.S. government, not subject to U.S. copyright. 1

M. L. Flenniken, D. A. Willits, A. L. Harmsen, L. O. Liepold, A. G.

114703-4

Harmsen, M. J. Young, and T. Douglas, Chem. Biol. 13, 161 共2006兲. E. Gillitzer, D. Willits, M. Young, and T. Douglas, Chem. Commun. 共Cambridge兲 2002, 2390. 3 M. Uchida, M. L. Flenniken, M. Allen, D. A. Willits, B. E. Crowley, S. Brumfield, A. F. Willis, L. Jackiw, M. Jutila, M. J. Young, and T. Douglas, J. Am. Chem. Soc. 128, 16626 共2006兲. 4 M. Uchida, M. T. Klem, M. Allen, P. Suci, M. Flenniken, E. Gillitzer, Z. Varpness, L. O. Liepold, M. Young, and T. Douglas, Adv. Mater. 共Weinheim, Ger.兲 19, 1025 共2007兲. 5 T. Douglas and M. Young, Science 312, 873 共2006兲. 6 M. T. Klem, D. A. Resnick, K. Gilmore, M. Young, Y. U. Idzerda, and T. Douglas, J. Am. Chem. Soc. 129, 197 共2007兲. 7 M. T. Klem, J. Mosolf, M. Young, and T. Douglas, Inorg. Chem. 47, 2237 共2008兲. 8 S. Mørup and C. Frandsen, Phys. Rev. Lett. 92, 217201 共2004兲. 9 N. J. Silva, L. D. Carlos, and V. S. Amaral, Phys. Rev. B 71, 184408 共2005兲. 10 B. J. Kim, H. I. Lee, S. B. Cho, S. Yoon, B. J. Suh, Z. H. Jang, T. G. St Pierre, S. W. Kim, and K. S. Kim, J Appl. Phys. 97, 10M524 共2005兲. 11 J. G. E. Harris, J. E. Grimaldi, D. D. Awschalom, A. Chiolero, and D. Loss, Phys. Rev. B 60, 3453 共1999兲. 12 M. Allen, D. Willits, J. Mosolf, M. Young, and T. Douglas, Adv. Mater. 共Weinheim, Ger.兲 14, 1562 共2002兲. 2

J. Appl. Phys. 107, 114703 共2010兲

Usselman et al. 13

R. J. Usselman, M. Klem, T. Douglas, M. Young, Y. Idzerda, and D. Singel 共unpublished兲. 14 R. J. Usselman, M. Klem, M. Allen, E. D. Walter, K. Gilmore, T. Douglas, M. Young, Y. Idzerda, and D. J. Singel, J. Appl. Phys. 97, 10M523 共2005兲. 15 R. Berger, J. C. Bissey, and J. Kliava, J. Phys.: Condens. Matter 12, 9347 共2000兲. 16 R. Berger, J. C. Bissey, J. Kliava, H. Daubric, and C. Estournes, J. Magn. Magn. Mater. 234, 535 共2001兲. 17 N. Noginova, F. Chen, T. Weaver, E. P. Giannelis, A. B. Bourlinos, and V. A. Atsarkin, J. Phys.: Condens. Matter 19, 246208 共2007兲. 18 J. L. Dormann, D. Fiorani, and E. Tronc, Adv. Chem. Phys. 98, 283 共1997兲. 19 J. M. Vargas, W. C. Nunes, L. M. Socolovsky, M. Knobel, and D. Zanchet, Phys. Rev. B 72, 184428 共2005兲. 20 R. D. Zysler, C. A. Ramos, E. De Biasi, H. Romero, A. Ortega, and D. Fiorani, J. Magn. Magn. Mater. 221, 37 共2000兲. 21 S. A. Makhlouf, F. T. Parker, and A. E. Berkowitz, Phys. Rev. B 55, R14717 共1997兲. 22 F. Brem, G. Stamm, and A. M. Hirt, J. Appl. Phys. 99, 123906 共2006兲. 23 R. A. Brooks, J. Vymazal, R. B. Goldfarb, J. W. M. Bulte, and P. Aisen, Magn. Reson. Med. 40, 227 共1998兲. 24 A. Millan, A. Urtizberea, N. J. O. Silva, F. Palacio, V. S. Amaral, E. Snoeck, and V. Serin, J. Magn. Magn. Mater. 312, L5 共2007兲.

Lihat lebih banyak...

Comentarios

Copyright © 2017 DATOSPDF Inc.