Ferromagnetic Behavior of Ultrathin Manganese Nanosheets - The

ARTICLE pubs.acs.org/JPCC

Ferromagnetic Behavior of Ultrathin Manganese Nanosheets Sreemanta Mitra,†,‡ Amrita Mandal,†,‡ Anindya Datta,§ Sourish Banerjee,‡ and Dipankar Chakravorty*,† †

MLS Prof’s Unit, Indian Association for the Cultivation of Science, Kolkata-700032, India University School of Basic and Applied Science (USBAS), Guru Gobind Singh Indraprastha University, New Delhi-110075, India ‡ Department of Physics, University of Calcutta, Kolkata-700009, India §

ABSTRACT: Ferromagnetic behavior has been observed experimentally for the first time in nanostructured manganese. Ultrathin (∼0.6-nm) manganese nanosheets were synthesized inside the two-dimensional channels of sol
gel-derived Na-4 mica. The magnetic properties of the confined system were measured within the 2
300 K temperature range. The confined structure was found to show a ferromagnetic behavior with a nonzero coercivity value. The coercivity value remained nonzero throughout the entire temperature range of measurement. The experimental variation of the susceptibility as a function of temperature can be satisfactorily explained on the basis of a two-dimensional system with a Heisenberg Hamiltonian involving direct exchange interaction.

’ INTRODUCTION Manganese (Mn) is probably one of the most important dopants for semiconductors for creating a ferromagnetic response.1
3,6
9 In addition, dilute solutions of Mn-based systems exhibit a wide variety of magnetic properties.4,5 However, Mn by itself is not ferromagnetic, even though Mn is a d-shell-based transition metal in its bulk form. This fact gives rise to a fundamental question about Mn's ground state, which is still not very clearly understood. The unique properties of manganese among the first-row transition metal elements as atoms, clusters, or crystals probably arise because of its atomic configuration. Manganese has an exactly half-filled 3d orbital and a fully filled 4s orbital, with an electron configuration of 3d54s2. The energy required to change the electronic configuration from 3d54s2 to 3d64s1 is high enough (∼2.4 eV)10 to keep the former as its ground-state configuration. If two Mn atoms are brought closer together, the 3d and 4s orbitals split into bonding and antibonding states, so the magnetic ground state (ferro- or antiferromagnetic) depends on the energy of the splitting and the exchange interaction.10 Calculations have shown that the ferromagnetic ground state is not energetically stable for Mn.11 There is a great deal of debate among different theoretical studies, even for the simplest dimer molecule Mn2, with regard to ferromagnetic and antiferromagnetic ground states.12
20 Spin density functional calculations of manganese nanostructures such as nanowires or nanorods showed these structures to be in high-moment states, with magnetic moments per atom having values in the range of 2.96
3.79 μB depending on the morphology of the nanostructure.21 Experimentalists have shown that small manganese clusters exhibit complex magnetic behavior with the signature of superparamagnetism.22
25 According to theoretical understanding r 2011 American Chemical Society

and experimental findings, the magnetic property fluctuates between ferromagnetic and antiferromagnetic ground states as the cluster size changes from 40 to 80 atoms.39 Beyond a cluster size of 80 atoms, the structures slowly converge to that of the bulk, and the magnetic ordering becomes antiferromagnetic. However, some magnetic deflection experiments showed nonzero magnetic moments for manganese clusters of 11
99 atoms, indicating ferromagnetic ordering of atomic spins, which had a lower limit on the number of atoms in a cluster than that mentioned earlier.22 Thus, manganese exhibits contradictory magnetic ground states from the standpoint of theory and experiments. Induced ferromagnetism has been observed in manganese on clean ferromagnetic substrates,26
28 whereas for a nonmagnetic substrate, monolayer deposition of Mn resulted in antiferromagnetic ordering.29 One-thousand-atom Mn clusters with heights of 10 nm and diameters of 15
25 nm on Si(111) and Si(112) were found to exhibit temperature-dependent ferromagnetic-like behavior below 10 K, as a result of some surface orientation effect that might be correlated with the surface dangling-bond density or cluster shape.30 However, for Mn thin films, the reported ground state is antiferromagnetic in nature.31,40 Although two-dimensional Mn crystalline systems would be more stable compared to small clusters, their ferromagnetic response remains illusive to date. Hence, the search for magnetic behavior in nonferromagnetic transition metals has often focused on the effect of reduced dimensions.22 To investigate this effect, we have synthesized manganese nanosheets inside the two-dimensional crystal channels of Na-4 mica, with a Received: May 23, 2011 Revised: June 16, 2011 Published: July 11, 2011 14673

dx.doi.org/10.1021/jp204765e | J. Phys. Chem. C 2011, 115, 14673–14677

The Journal of Physical Chemistry C

ARTICLE

very simple synthesis method. In this article, we report on their ferromagnetic behavior.

’ SYNTHESIS AND CHARACTERIZATION Na-4 mica (Na4Mg6Al4Si4F4O20 3 xH2O) template was synthesized by the usual sol
gel technique, taking aluminum nitrate, magnesium nitrate (used as obtained from E Merck), tetraethylorthosilicate (TEOS), and ethanol as precursors. In brief, to prepare 1 g of Na-4 mica powder, 2.186 g of aluminum nitrate and 2.241 g of magnesium nitrate were dissolved in ethanol, and then the solution was stirred vigorously for 1 h to obtain a homogeneous mixture. To this mixture was added 1.291 cm3 of TEOS to achieve the target composition. Nitric acid (0.1 N) was also added as a catalyst. This solution was stirred for 3 h and placed in an air oven at 333 K for 3 days. The dried gel was crushed and calcined at 748 K for 12 h. Equal amounts of gel powder and crystalline sodium fluoride (NaF) were mixed thoroughly and heated at 1163 K for 18 h in a platinum crucible under ordinary atmosphere. This step was needed for the dissolution of the gel in NaF. The reaction product was washed thoroughly with saturated boric acid several times to remove the water-insoluble fluoride salts. The powder was then washed again with 1 M NaCl solution three times to completely saturate all exchange sites with Na+ ions. The resultant product was then washed with deionized water several times and dried at 333 K in an air oven to yield pure Na-4 mica powder. The unit cell of Na-4 mica has a layered structure with an interlayer spacing of 0.6 nm.32
34,38 A 2.36 g sample of Na-4 mica powder was then subjected to the ion-exchange reaction 2Na+ S Mn2+ by soaking it in a mixture of manganese nitrate [Mn(NO3)2] and dextrose (C6H12O6) in aqueous solution at an elevated temperature (368 K) and pressure inside a Tefloncoated autoclave cell for 5 days. The pH of the solution was kept neutral throughout the ion-exchange process. The latter could occur only in the case of ions that were mobile. As such, in the present case, ion-exchange reactions with the other species, namely, Al, Mg, and Si, are ruled out. The resultant powder was removed from the autoclave and washed thoroughly with deionized water several times to ensure that no manganese nitrate molecules were present on the surface of the Na-4 mica powder. This was confirmed by a simple chemical group test analysis: Sodium carbonate was added to the filtrate, and no white precipitate of manganese carbonate was obtained. The washed powder was then put in an alumina boat and placed in a muffle furnace at 675 K under ordinary atmosphere for 2 h. The carbon of the dextrose molecules reduced the Mn ions into Mn metal while generating carbon dioxide. X-ray diffraction of the material was performed using a Bruker D-8 SWAX X-ray diffractometer with a Cu KR monochromatic source of wavelength 0.15408 nm. To study the microstructure, Mn nanosheets were extracted from the mica channels by etching the composite sample with 10% HF aqueous solution and centrifuged in a SORVALL RC 90 ultracentrifuge at 30000 rpm for 30 min. The resultant residue was washed thoroughly with deionized water and then dispersed in acetone. From that dispersion, a drop was taken and investigated in a JEOL 2010 transmission electron microscope operated at 200 kV. The magnetic properties of the composite were studied with an MPMS SQUID magnetometer (Quantum Design) in the temperature range of 2
300 K.

Figure 1. XRD pattern of manganese and Na-4 mica composite.  Numbers indicate the corresponding interplanar spacings (in A).

Figure 2. (a) Transmission electron micrograph of manganese nanosheets, (b) enlargement of a portion of the image in panel a, (c) highresolution lattice image of one manganese nanosheet, and (d) selectedarea electron diffraction pattern of the region in panel b.

’ RESULTS AND DISCUSSION Figure 1 shows the X-ray diffraction pattern of the composite sample in the range 2θ = 5
80o. The presence of manganese was confirmed by the standard JCPDS value (file number 17-0910). Only the (200) plane of manganese was found to grow inside the two-dimensional mica nanochannels. The rest of the XRD lines originated from the basic structure of Na-4 mica.35,36 The c axis of Na-4 mica was found to be perpendicular to the (200) plane of Mn, allowing growth along this direction only. Figure 2a shows a transmission electron micrograph of the randomly assembled partially etched nanocomposites containing Mn nanosheets. Figure 2b shows an enlarged view of the same assembly. The nanosheets formed from manganese nanodisks of circular and rhombus-like structures. As the manganese nanosheets formed within the layers of Na-4 mica, their thickness was limited by the channel thickness (i.e., 0.6 nm). Figure 2c shows a high-resolution lattice image of one of the nanosheets in which the (200) plane of Mn can be observed along with the lattice planes corresponding to Na-4 mica. The interplanar spacing was found 14674

dx.doi.org/10.1021/jp204765e |J. Phys. Chem. C 2011, 115, 14673–14677

The Journal of Physical Chemistry C

ARTICLE

Table 1. Interplanar Spacings Estimated from Electron Diffraction Data and JCPDS Filea.) observed (nm)

Na-4 mica (nm)

manganese (nm)

0.30

0.303 (023)


b

0.26 0.20

0.263 (201) 0.202 (006)





0.18

0.186 (205)




0.17

0.173 (205)




0.16

0.166 (135)




0.141

0.143 (007)




0.133




0.1336 (200)

a

Numbers in parentheses describe the Miller indices of the corresponding lattice planes. b Indicates that no planes with these interplanar spacings exist in the corresponding phases.

Figure 4. Variation of magnetization (M) with temperature (T) under both field-cooled (FC) and zero-field-cooled (ZFC) conditions measured at 5 mT.

Figure 3. AFM images of manganese nanosheets and the corresponding height profile.

Figure 5. Variation of magnetization (M) with magnetic field (H) measured at 2 K. Inset: Enlargement of the region near zero magnetization.

to be 0.133 nm. As a result, the interatomic distance was 0.266 nm. From the TEM images, the sizes of the nanosheets were found to be around 70
150 nm. Figure 2d shows a selected-area electron diffraction (SAED) pattern of one of the nanosheets. As the sample was partially etched, some spots in the SAED pattern correspond to the Na-4 mica structure, whereas spots due to the (200) plane of manganese also arose, as expected. The results are summarized in Table 1. To determine the thickness of the manganese nanosheets synthesized in the present work, atomic force microscopy (Veeco model CP II microscope) was used. The height profiles of the nanosheets were obtained by etching the composite powder in 10% HF aqueous solution for 4 days and dispersing it on a freshly cleaved atomically flat mica (SPI Supplies, West Chester, PA) surface. A typical profile is shown in Figure 3. The heights measured were either 0.6 nm or an integral multiple thereof, which confirms that the manganese nanosheets were indeed grown within the nanochannels of the Na-4 mica structure. The AFM images gave a thickness value equal to the thickness of the interlayer space in Na-4 mica reported in the literature.32
34 These facts led us to conclude that the original films were no thicker than 0.6 nm and, hence, dissolution of manganese during etching did not take place. Magnetic measurements were carried out for the composite

powder in the temperature range of 2
300 K. Figure 4 shows the variation of the magnetization as a function of temperature measured at an applied magnetic field 5 mT under both zero-field-cooled (ZFC) and field-cooled (FC) conditions. The absence of any local maxima in the ZFC magnetization
temperature curve indicates ferromagnetic coupling between the spins. This is borne out by the magnetization
magnetic field hysteresis curve measured at 2 K, which is shown in Figure 5. In the inset of Figure 5, the enlargement of the hysteresis curve near zero magnetization shows a nonzero coercivity value (∼60 Oe). The magnetization
magnetic field hysteresis curve for the composite at 300 K is also shown in Figure 6. A finite coercivity present here (as observed from the inset) also indicates ferromagnetic behavior even at room temperature. The saturation magnetization was not achieved even at maximum field, indicating that all spins cannot become parallel to the magnetic field because of the thermal energy. Figure 7 shows the magnetization as a function of magnetic field in the case of pure Na-4 mica, which exhibits diamagnetic characteristics. This contribution was subtracted from the results presented herein. Regarding the Curie temperature of the ultrathin Mn films, we believe that it is above room temperature in view of the fact that magnetic hysteresis was observed at room temperature. The precise nature 14675

dx.doi.org/10.1021/jp204765e |J. Phys. Chem. C 2011, 115, 14673–14677

The Journal of Physical Chemistry C

Figure 6. Variation of magnetization (M) with magnetic field (H) measured at 300 K. Inset: Enlargement of the region near zero magnetization.

ARTICLE

Figure 8. EDAX analysis of the nanocomposite.

Figure 9. Variation of susceptibility (χ) with temperature (T) (open circles, experimental data; solid line, theoretically fitted curve). Figure 7. Variation of magnetization (M) with magnetic field (H) for Na-4 mica.

of the effect of size on the magnetic behavior cannot be inferred from our experimental data because the latter pertain to only one thickness of Mn nanosheet, namely, 0.6 nm. Also, the EDAX data (measured in a JEOL JSM-6700F field-emission scanning electron microscope) in Figure 8 show that no magnetic impurity is present in the nanocomposite under investigation. The EDAX data show that the amount of oxygen atoms present is just sufficient to fulfill the stoichiometric needs of the aluminum, magnesium, and silicon present in the Na-4 mica phase. Hence, it can be safely concluded that there is no possibility for the formation of oxides of manganese in our system. This is also consistent with the fact that the Mn nanosheets were formed by subjecting Mn to reduction treatment at 675 K. To determine the possible origin of this ferromagnetism, of the two exchange interactions possible (viz., direct and indirect), we ruled out the possibility of an indirect exchange interaction, as the hopping integral decays exponentially with the distance between magnetic centers, and considered only the direct exchange interaction.20,21 We applied the Heisenberg Hamiltonian X H ¼
2 Jij Si 3 Sj ð1Þ hiji

based variation of susceptibility with temperature for a twodimensional ferromagnet37 χ ≈ expðβ=TÞ

ð2Þ

with β ¼ ð4πJS2 Þ=kB

ð3Þ

where J is the exchange coupling constant, S is the spin quantum number, and kB is the Boltzmann constant. The calculation was performed by taking S = 5/2 for manganese and J/kB as a parameter. The fitting looks comprehensive, and the positive J value found indicates ferromagnetic coupling. Both the experimental data and the fitted curve are shown in Figure 9. A previous theoretical investigation did arrive at the conclusion that monolayers and bilayers of Mn grown on the tungsten(100) plane have a ferromagnetic ground state.41 Considering the thickness of 0.6 nm for our manganese films, our experimental results on their ferromagnetic behavior are consistent with the above-mentioned theoretical calculations. Also, theoretical predictions show that the obtained interatomic distance of the manganese atoms ensures a ferromagnetic response.15

’ CONCLUSIONS In summary, we have synthesized manganese nanosheets with a thickness of 0.6 nm in the nanochannels of layered Na-4 mica 14676

dx.doi.org/10.1021/jp204765e |J. Phys. Chem. C 2011, 115, 14673–14677

The Journal of Physical Chemistry C by a simple process. These nanosheets are made up of nanodisks with sizes of about 70
150 nm. The nanodisks consist of the (200) planes of Mn, aligned parallel to the nanochannel itself, as confirmed both by X-ray diffraction and selected-area electron diffraction experiments. This confined Mn structure exhibits ferromagnetic behavior in its ultrathin configuration, as is evident from the magnetization measurements. Calculation of the variation in magnetic susceptibility as a function of temperature on the basis of the Heisenberg ferromagnetic model involving direct exchange interaction matches the experimental data satisfactorily.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT The work was supported by a grant awarded by Nano Mission Council, Department of Science and Technology, New Delhi, India. S.M. and A.M. thank the University Grants Commission, New Delhi, India, for Junior Research Fellowships. D.C. thanks the Indian National Science Academy for awarding an Honorary Scientist’s position. S.M. thanks Dr. Molly De Raychaudhury for fruitful discussions. Support was partly derived from a grant received from Department of Science and Technology, New Delhi, India, under an Indo-Australian Project on Nanocomposites. ’ REFERENCES (1) Ohno, H.; Munekata, H.; Penney, T.; von Molnar, S.; Chang, L. L. Phys. Rev. Lett. 1992, 68, 2664. (2) Dietl, T.; Ohno, H.; Matsukura, F.; Cibert, J.; Ferrand, D. Science 2000, 287, 1019. (3) Sharma, P.; Gupta, A.; Rao, K. V.; Owens, F. J.; Sharma, R.; Ahuja, R.; Osorio Guillen, M.; Johansson, B.; Gehring, G. A. Nat.Mat. 2003, 2, 673. (4) Chu, D.; Kenning, G. G.; Orbach, R. Phys. Rev. Lett. 1994, 72, 3270. (5) Macdonald, A. H.; Schiffer, P.; Samrath, N. Nat. Mater. 2005, 4, 195. (6) Dong, X.; Osada, M.; Ueda, H.; Ebina, Y.; Kotani, Y.; Ono, K.; Ueda, S.; Kobayashi, K.; Takada, K.; Sasaki, T. Chem. Mater. 2009, 21, 4366. (7) Xu, C.; Chun, J.; Lee, H. J.; Jeong, Y. H.; Han, S. E.; Kim, J. J.; Kim, D. E. J. Phys. Chem. C 2007, 111, 1180. (8) Liu, E.; Zhao, N.; Li, J.; Du, X.; Shi, C. J. Phys. Chem. C 2011, 115 (8), 3368. (9) Norris, D. J.; Yao, N.; Charnock, F. T.; Kennedy, T. A. Nano Lett. 2001, 1, 3. (10) Nayak, S. K.; Rao, B. K.; Jena, P. J. Phys.
Condens. Matter 1998, 10, 10863. (11) Hobbs, D.; Hafner, J. J. Phys.
Condens. Matter 2001, 13, L681. (12) Nesbet, R. K. Phys. Rev. 1964, 135, A460. (13) Pederson, M. R.; Reuse, F.; Khanna, S. N. Phys. Rev. B 1998, 58, 5632. (14) Bobadova-Parvanova, P.; Jackson, K. A.; Srinivas, S.; Horoi, M. Phys. Rev. A 2003, 67, 061202(R). (15) Harris, J.; Jones, R. O. J. Chem. Phys. 1979, 70, 830. (16) Morisato, T.; Khanna, S. N.; Kawazoe, Y. Phys. Rev. B 2005, 72, 014435. (17) Bauschlicher, C. W., Jr. Chem. Phys. Lett. 1989, 156, 95. (18) Wang, B.; Chen, Z. Chem. Phys. Lett. 2004, 387, 395.

ARTICLE

(19) Yamamoto, S.; Tatewaki, H.; Moriyama, H.; Nakano, H. J. Chem. Phys. 2006, 124, 124302. (20) Paul, S.; Misra, A. J. Mol. Struct. (THEOCHEM) 2009, 907, 35. (21) Zelen
y , M.; Sob, M.; Hafner, J. Phys. Rev. B 2009, 80, 144414. (22) Knickelbein, M. B. Phys. Rev. Lett. 2001, 86, 5255. (23) Knickelbein, M. B. Phys. Rev. B 2004, 70, 014424. (24) Mertes, K. M.; Suzuki, Y.; Sarachik, M. P.; Shtrikman, H.; Zeldov, E.; Rumberger, E. M.; Hendrickson, D. N.; Christou, G. Solid State Commun. 2003, 127, 131. (25) Baumann, C. A.; Van Zee, R. J.; Bhat, S. V.; Weltner, W., Jr. J. Chem. Phys. 1983, 78, 190. (26) Dreysse, H.; Demangeat, C. Surf. Sci. Rep. 1997, 28, 65. (27) Yamada, T. M.; Bischoff, M. M. J.; Mizoguchi, T.; van Kempmen, H. Surf. Sci. 2002, 516, 179. (28) O’Brien, W. L.; Tonner, B. P. Phys. Rev. B 1994, 50, 2963. (29) Bode, M.; Heinze, S.; Kubetzka, A.; Pietzsch, O.; Hennefarth, M.; Getzlaff, M.; Wiesendanger, R.; Nie, X.; Bihlmayer, G.; Bl€ugel, S. Phys. Rev. B 2002, 66, 014425. (30) Prokes, S. M.; Glembocki, O. J.; Carlos, W. E.; Kennedy, T. A. Appl. Surf. Sci. 2003, 214, 103. (31) Eder, M.; Hafner, J.; Moroni, E. G. Surf. Sci. 1999, 423, 244. (32) Mukherjee, P. K.; Chatterjee, K.; Chakravorty, D. Phys. Rev. B 2006, 73, 035414. (33) Bhattacharya, S.; Datta, A.; Dhara, S.; Chakravorty, D. J. Phys. D: Appl. Phys. 2009, 42, 235404. (34) Paulus, W. J.; Komerneni, S.; Roy, R. Nature 1992, 357, 571. (35) Park, M.; Lee, D. H.; Choi, C. L.; Kim, S. S.; Kim, K. S.; Choi, J. Chem. Mater. 2002, 14, 2582. (36) Franklin, K. R.; Lee, E. J. Mater. Chem. 1996, 6, 109. (37) Takahashi, M. Phys. Rev. Lett. 1987, 58, 168. (38) Bhattacharya, S.; Datta, A.; Chakravorty, D. Appl. Phys. Lett. 2010, 96, 093109. (39) Mejía-L
opez, J.; Romero, A. H.; Garcia, M. E.; L.Mor
an-L
opez, J. Phys. Rev. B 2006, 74, 140405(R). (40) Bergman, A.; Nordstr€om, L.; Klautau, A. B.; Frota-Pess^oa, S.; Eriksson, O. Phys. Rev. B 2006, 73, 174434. (41) Ferriani, P.; Heinze, S.; Bihlmayer, G.; Bl€ugel, S. Phys. Rev. B 2005, 72, 024452.

14677

dx.doi.org/10.1021/jp204765e |J. Phys. Chem. C 2011, 115, 14673–14677