Magnetic properties of hybrid nanostructures 3

Chapter Room Temperature Ferromagnetism at SelfAssembled Monolayer Modified Ag Nanocluster-ZnO Nanowire Interface Magnetic characterization of ZnO nanowires (NW) decorated with thiol-capped Ag nanoclusters (NCs) reveals spontaneous field-dependent magnetization and hysteresis at room temperature. The saturation magnetization is temperature-independent for the 13 nm thiol-capped Ag NCs but unexpectedly increases with temperatures for the nm thiol-capped Ag NCs. The high magnetic moment results from the efficient dispersal of Ag NCs on the ZnO NW scaffold and charge transfer interaction between Ag and ZnO. The anomalous magnetic behavior in nm NCs may be due to spin reorientation of Ag-S dipoles mediated by Zn-S dipoles. 105 5.1 Introduction The metal nanoparticle-semiconductor nanowire interface represents a new class of integrated nanosystem useful in plasmonics, photocatalytic, sensors or solar cell systems. Metal nanoparticles such as silver or gold possess localized surface plasmons that have resonance absorption of light in the visible region, and can act as efficient antennae for the capture of solar energy on ZnO or TiO2 nanorods. , The metal nanoparticles can also act as electron sink, promoting interfacial charge transfer during charge recombination, thus promoting the photocatalytic activities of ZnO or TiO2. One interesting possibility is the enhancement of ferromagnetism across the metal nanoparticle-semiconductor interface due to interfacial charge transfer. Both Ag and undoped ZnO are non-ferromagnetic but recent reports suggest that interfacial ferromagnetism can arise when modified by a monolayer of thiolated organic molecule. For undoped ZnO, ferromagnetic-like behavior has been reported by Garcia et al. at room temperature for thiol-capped ZnO nanoparticles. Intrinsic defects such as oxygen vacancies, occupations of oxygen sites by Zn and Zn interstitials (Zni+) in the ZnO lattice can also trigger ferromagnetism. 9, 10 Lorenza Suber et al. found that dithiol-capped Ag nanoparticles exhibited permanent magnetism with a finite coercivity up to room temperature. 11 In order to give rise to ferromagnetism, the Ag d orbitals have to be localized near the top of a d band in order to produce a large density of states near the Fermi level N(Ef). However Ag has a full 4d10 shell and one 5s1 electron with the d bands far from Ef. 12 Given the low density of states at the Fermi level and diamagnetic character of bulk Ag, the occurrence of magnetism in thiol-capped Ag is surprising. Perhaps room temperature ferromagnetism in thiol-capped Ag NCs has a different origin 106 to the well-known theory of magnetism based on the unfilled character of 3d or 4f electron energy levels. 13 Fundamentally, it is interesting to consider the possibility of obtaining room temperature ferromagnetism from metal-semiconductor hybrid nanostructure since metallic decoration can tune the Fermi level of semiconductors and promote the electron transfer to induce ferromagnetic ordering. For example, the transfer of electrons from Ag nanoparticles to semiconductor Zn0.9Co0.1O nanorods shifts the Fermi level such that magnetic interactions between the localized Co2+ spins are enhanced to give rise to room temperature ferromagnetism. 14 To prepare the nanometal-semiconductor hybrid nanostructure, Ag nanoclusters were deposited on the ZnO nanowire (NW) array using a nanocluster beam (NC) deposition source. The versatility of the NC beam is that it is capable of producing well dispersed NCs and the process is cleaner (e.g. surfactant free) compared to the wet solution method. Ag NPs produced by wet chemical methods are usually capped by surfactants to improve the dispersion properties in solution. To study the magnetic properties of these materials require the recovery of sub-grams quantities of these in powder form, with inevitable coagulation of the nanomaterials. In contrast, using the NC beam deposition method, the magnetic properties can be directly measured on the substrate. To study the effect of Ag NCs on the interfacial ferromagnetism of ZnO NW, Ag NCs of two different sizes – 13 nm and nm – were deposited on ZnO NW arrays. A self-assembled monolayer (SAM) was formed on the Ag NC-ZnO NW using different thiols and it was found that interface magnetism could be observed for all the thiolcapped Ag NCs on ZnO NW, with small differences in the magnetic properties which can 107 be attributed to variation in work function induced by different thiols. Unless specified otherwise, the thiol in this work refers to 1,9-nonanedithiol since we find that the effect of this thiol is most representative. Magnetic characterization of ZnO NW decorated with thiol-capped Ag NCs reveals spontaneous field-dependent magnetization and hysteresis at room temperature, with ordering temperatures above 300K. The high magnetic moment results from the efficient dispersal of Ag NCs on the ZnO NW scaffold and possible charge transfer interaction between Ag and ZnO. The unique magnetic properties of these thiol-capped Ag NCs show that these are a new class of stable magnets with interfacial ferromagnetic properties triggered by the cooperative effect of charge transfer between organic molecules and a metal-semiconductor hybrid. 108 5.2 Experimental Section 5.2.1 Fabrication of thiol-capped Ag NCs-ZnO NWs (b) Scheme 5.1 Schematic showing the fabrication of thiol-capped Ag NCs on ZnO NWs: (a) deposition of Ag NCs by the nanocluster beam source on ZnO nanowire array, followed by (b) coating with thiol molecules (1,9-nonanedithiol). Inset: Illustration showing the creation of giant orbital moment by the effect of spin-orbit coupling on quasi-free electrons. A spin-orbit coupling field H* acting between the induced orbital moment and localized spin. ZnO NWs were grown on Si substrate using the hydrothermal method. 15 An equimolar mixture of 0.2M hexamethylenetetramine and zinc nitrate was sealed in a Teflon bottle and baked at 92°C for days. The precursor solution was changed at three hour intervals to ensure that the ZnO nanowires grow to give a uniform diameter. At the end of the growth process, the length of the ZnO NWs attained was ~5 µm. 109 After rinsing the ZnO/Si sample with ultrapure water and drying the substrates at 150°C, the sample was loaded into a high vacuum chamber (10-8 Torr.) equipped with a NC200U nanocluster beam deposition source from Oxford Applied Research. 16 Ag NCs were deposited on the ZnO/Si sample using the following conditions: pressure of helium aggregation gas: mTorr; flow rate: 30 sccm; power of 40 W. By changing the aggregation length, the Ag NC size can be tuned to 13 nm and nm. Five different thiols – 1,9nonanedithiol (referred to as C9-dithiol), Octadecanethiol (referred to as C18-thiol), 1H,1H,2H,2H-Perfluoro-1-hexanethiol and Adamantanethiol (Sigma-aldrich) were immobilized on different Ag/ZnO substrates by self-assembly in a 1mM absolute ethanolic solution in a glovebox for 48 h. For adamantanethiol, it was made up to 10mM in ethanol. 5.2.2 Magnetic characterization Magnetization curves, zero field-cooled and field-cooled (ZFC and FC) curves were recorded using superconducting quantum interference devices (SQUID, QUANTUM DESIGN). During the ZFC measurement, the sample was first cooled without an applied field to K, a field of 500 Oe was applied to measure the magnetic moment as a function of temperature to 350 K. For a FC curve, the sample was first cooled to K in the presence of an applied field of 500 Oe. The magnetic moment versus temperature curve was recorded to 350 K. The field was applied parallel to the substrate surface. The magnetic moment versus temperature was done by measuring hysteresis loops at different temperatures. 110 5.2.3 Electronic Structure Characterization X-ray photoemission spectroscopy (XPS) and ultraviolet photoemission (UPS) spectroscopy was performed at the Singapore Synchronton Light Source. 17 The valence band measurement was performed with photon energy of 60eV. All signals were taken at normal emission geometry. 111 5.3 Results and Discussion 5.3.1 Characterization of Ag NC-ZnO NWs by SEM, TEM and XPS Scheme 5.1 outlines the processes used in the preparation of thiol-capped NCs on the ZnO NWs. The Ag NCs were generated by the gas-phase aggregation of sputtered Ag atoms after collision with helium carrier gas, these were allowed to expand into a low pressure region through an orifice to form a NC beam which deposits directly onto a substrate (Figure 5.1a). A monolayer of thiol molecules was assembled on the surface of the top layer of Ag NCs by self-assembly in an ethanolic thiol solution under inert conditions (Figure 5.1b). Figure 5.2 (a) SEM of Ag NCs deposited on hydrothermally grown ZnO NWs. (b) magnified view showing the Ag NC-coated ZnO NWs. TEM image of a ZnO NW coated with 13 nm (c) and nm (d) Ag NCs capped by C9 dithiol. (e) High resolution TEM images of the Ag-ZnO interface. Inset shows the SAED of the ZnO NW. 112 Figure 5.3 XPS S2p and Zn2p level of single crystalline ZnO arrays of (i) Ag NC (ii) Ag NC modified with 1,9-nonanedithiol and (iii) Ag NC modified with octadecanethiol. Figure 5.2a and b show the low and high magnification SEM images of hydrothermally grown ZnO NW template decorated with Ag NCs. By changing the aggregation length and power, the Ag NCs can be tuned to average sizes of 13 nm and nm. The TEM images of 13 nm Ag NC and nm Ag NC capped with C9-dithiol on ZnO NW are shown in Figure 5.2c and d respectively. The high-resolution TEM image of the Ag NC-ZnO NW interface reveals near monolayer coverage of face-centered cubic (FCC) Ag NCs on hexagonal close-packed (HCP)-ZnO surface (Figure 5.2e). The regular spacing of lattice planes for the Ag nanoparticle is 0.232 nm, which is consistent with that of Ag-(111) planes. The single crystalline diffraction spots in the SAED pattern (inset in Figure 5.2e) can be indexed to hcp-ZnO, which corresponds to growth in the [0001] direction. The chemical shifts in Ag 3d5/2 core-level after thiol-bonding and 113 presence of thiolate groups at ~162 eV from monochromatic XPS measurements of S 2p confirm the formation of thiolate-bound monolayers (figure 5.3). 5.3.2 Effect of Ag NC size on Magnetic Properties of Thiol-capped Ag NC-ZnO NWs -2 M (μemu cm ) 80 40 a) (i) (ii) -40 -80 -6 Ag-ZnO C9thiol-Ag-ZnO -3 H (kOe) b) C9-Ag/ZnO 350K 50 -2 -50 4nm 13nm -100 -2 H (kOe) c) C9-Ag/ZnO 5K 50 100 M (μemu cm ) -2 M (μemu cm ) 100 -50 4nm 13nm -100 -2 H (kOe) Figure 5.4 (a) M-H plot for 13 nm Ag NC-ZnO substrate at 350K (i) before and (ii) after binding to C9dithiol. A ferromagnetic signal can be seen superimposed on a diamagnetic background. M-H plot of Ag NC-ZnO capped by 1,9nonanedithiol at (b) 350K and (c) 5K. The magnetization curves (M–H plots) measured at 350K for the Ag NC-ZnO NW before and after binding with C9dithiol are shown in Fig. 5.4a. The Ag NC-ZnO NW exhibits diamagnetic behavior at all temperatures. This is expected since both bulk Ag and ZnO are diamagnetic with a susceptibility χAg = -1.8 × 10-7 emu/g Oe and χZnO = -1.2 × 10-6 emu/g Oe, respectively. Following the assembly of the thiols on ZnO NW, Ag NC 114 Figure 6.10 (a) and (b) Representative hysteresis loops for thiol-capped single-crystalline hydrothermally grown NWs and NTs arrays of diameters ~150 nm and ~300 nm respectively, at 350K and 40K, after substraction of diamagnetic background signal. Temperature dependence of magnetization (Ms) and coercivity (Hc) measured for thiolcapped (c) NWs and (d) NTs arrays. 6.3.6 Temperature dependence of magnetization of thiol-capped ZnO NWs/NTs The variation of magnetization (Ms) and coercivity (Hc) with temperature for the thiol-capped NWs and NTs under ZFC conditions are shown in parts c and d of Figures 6.10, respectively. The persistence of an almost invariant Ms and Hc, over a broad interval between and 350 K in both the NTs and NWs is a clear signature of interfacial magnetism. The magnetic moment is as a result of orbital momentum at the surface conduction electrons induced by a localized charge and/or spin through spin-orbit interaction. The strong effective spin-orbit field, H*, acting between the localized spin 149 and the induced orbital momentum of the conduction electrons, leads to extremely high local magnetic anisotropy that blocks the moments from switching, and is responsible for the absence of thermal fluctuations effect.33 In addition, the thermally independent magnetization also rules out magnetic impurities as the origin of the magnetic properties of ZnO NTs as magnetic impurities dispersed in a nonmagnetic matrix results in the magnetization to decrease with temperature. The finite Hc, Mr and saturating magnetic signals as seen from the M-H curves measured at 300 K confirmed the room temperature ferromagnetism of ZnO NTs. From ZFC-FC curves of the thiol-capped ZnO NTs array in Figure 6.11, a curie temperature close to 350 K can be obtained for the thiol-capped NTs. 2.6 2.5 2.4 M (μemu) 2.3 2.2 2.1 2.0 ZFC FC 1.9 1.8 1.7 50 100 150 200 250 300 350 400 Temperature (K) Figure 6.11 Thermal dependence of magnetization of thiol-capped ZnO NWs array measured under ZFC and FC conditions under a field of 1000 Oe (magnetic moment not normalized to area). 150 6.3.7 DFT calculations of magnetic moment of thiol-capped ZnO (a) (b) (c) S (002) S S Zn Zn S Zn Zn S Zn S S S S Zn Zn S Zn S S Zn Zn S Zn Zn Zn S Zn Zn S S S (-100) S Zn Zn (0-10) S (-110) S Zn Zn S Zn Zn S Zn S S S Zn SZn Zn S Zn Zn Zn S Zn S S Zn S S Zn Zn S Zn Zn Zn S S Zn Zn S S Zn Zn S Zn Zn Zn S S S S Zn Zn S Zn S Zn Zn S Zn S Zn Zn S S S S S (1-10) (100) S S Zn SZn Zn S Zn Zn Zn S Zn S Zn Zn S Zn S S (010) S S S Zn Zn S Zn S S Zn Zn S Zn Zn S S Zn Zn S (00-2) Zn S Zn Zn S S Zn S Zn Zn S S Zn Zn S Zn S Zn Zn S S (d) 128° Figure 6.12 Schematics of the distribution of dipole moments on the lateral side walls and inner wall for thiol-capped single-crystalline (a) NW (b) NT, and (c) polycrystalline NT, in the presence of a field applied in-plane (perpendicular to NT axis). An extra layer of dipole moments is present on the inner surface of the hollow NT. Inset of a shows the zoom-in view of the interface of thiol-ZnO showing that –SH bonded to the (100) side plane of the hexagonal ZnO NW/NT at an angle of 128°. Grey atoms represent Zn; red atoms: O; yellow atoms: S; white atoms: H. Unaligned spins lead to randomly distributed dipole moments on polycrystalline facets in (c). (d) Isosurface of spin distribution of SCH3 residue on ZnO(200) at the level of 0.00002/Å. Zn is in grey, O in red, S in yellow, C in chocolate and H in white. When thiols interact strongly with ZnO to form close-packed self-assembled monolayers (SAMs),34 surface electrons orbit around the nanometric to micrometric selfassembled domains to induce giant anisotropic orbital momenta, with long-range 151 ordering of the magnetic dipoles possible on the aligned NTs/NWs array.35 Parts a-c of Figure 6. 12 is a schematic of the distribution of dipole moments on the lateral side walls and inner walls for the thiol-capped single-crystalline and polycrystalline ZnO NWs and NTs. The optimized structure shows that the Zn-S bond forms an angle of ~128° with the (100) plane for the thiol-capped single-crystalline ZnO NW and NT as shown in the inset of part a of Figure 6.10. To explore the possible origin of the observed magnetism, we performed ab initio density functional calculation on the S-CH3 residue adsorbed to the ZnO (100) surface using VASP code.36 Six (100) layers were used to model the infinite plane with in-plane periodicity of 12.997×10411 Å (from theoretical optimized equilibrium geometry). Atoms in the two layers at the bottom were fixed while all other atoms were relaxed with a force tolerance of 0.05eV/Å. Ultrasoft-pseudopotential was used to represent the ion-electrons (or spins) interaction. The generalized gradient approximation37 was adopted for exchange-correlation function. The wavefunction was expanded in plane waves with a cutoff energy of 520 eV and the Brillouin zone is sampled by (0,0,0) k-point. The result suggested a finite spin polarization with net spin localized on the S atom of S-CH3 residue, as indicated in part d of Figure 6.10. The total spin was 0.67 μB, with 0.62 μB from S atom and 0.07 μB from an O atom in vicinity to Zn. The relaxed S-Zn and S-C bond length is 2.32 and 1.82 Å, respectively. Judging from the shape of the spin distribution, the net spin is mainly from the 3p electrons of S atom with a magnetic moment oriented perpendicular to the (200) surface of ZnO. The calculation result indicates a plausible origin of the observed room-temperature ferromagnetism. Furthermore, as the magnetism originates from S sites, the result also suggests that the magnetic properties would be strongly affected by the Zn-S bond geometry and their assembly. 152 Figure 6.13 Normalized in-plane (perpendicular to tube axis) and out-of-plane (parallel to tube) hysteresis curves for thiol-capped (a) VPT-grown single-crystalline ZnO NWs (b) hydrothermally grown single-crystalline NWs (c) NTs fabricated by etching hydrothermally grown NWs, and (d) polycrystalline ALD NTs arrays. In and out-ofplane directions are defined with respect to the plane being the substrate on which the NW/NTs were grown (‘in’ refers to applied magnetic field in the direction perpendicular to tube axis while ‘out’ refers to applied magnetic field in direction parallel to tube axis). Zoom-in views of the hysteresis curves near the origins are provided in the insets. 6.3.8 Effect of crystallinity on magnetic anisotropy In order to understand the possible magnetic anisotropy in thiol-capped ZnO NWs and NTs arrays system induced by long-range ordering of the dipoles, in-plane and outof-plane magnetic measurements of the thiol-capped NTs and NWs were carried out by 153 alternating gradient magnetometry (AGM). The normalized magnetization curves for single-crystalline VPT-grown 80 nm NWs, hydrothermally grown 150 nm NWs, 300 nm NTs and polycrystalline 80 nm NTs are shown in parts a-d of Figure 6.13, after subtracting the diamagnetic background. The insets in part c of Figure 6.13 show the directions of applied magnetic field with respect to the NT geometry for in-plane and outof-plane measurements. The values of remanance (Mr/Ms), magnetization (Ms) and coercivity (Hc) in the in-plane and out-of-plane directions for the thiol-capped NWs and NTs are summarized in Table 6.2. The coercivities (Hc) of the NWs/NTs in two directions, both parallel and perpendicular to the tube or wire axis, occur in the range of 50-91 Oe, and not show significant morphology or crystallinity dependence. As for the single-crystalline ZnO NWs and NTs, the magnetization curves obtained exhibited a similar easy axis in the in-plane direction. This is different from the observations in ferromagnetic Co NWs/NTs, in which the Co NW exhibits a sharper magnetization reversal than the Co NT when applied field is parallel to the direction of the tube axis, attributed to more rapid domain nucleation and propagation in the case of Co NWs than Co NTs.38 Since the spins are located at the surface, the core of the thiol-capped ZnO NW does not participate in the magnetization process. In addition, magnetization reversal of ZnO NT should exclude the transverse mode where a net magnetization component appears inside the wall which is applicable for bulk ferromagnetic NTs.39 Hence, the resemblance between the magnetization curves for thiol-capped single-crystalline ZnO NTs and NWs is not unexpected as the ferromagnetic signal for the thiol-capped ZnO NTs/NWs only arises at the interface. 154 Table 6.2 Summary of Mr/Ms, Ms and Hc for the thiol-capped ZnO NWs and NTs. Mr/Ms,⊥ Mr/Ms,// Ms/µemu cm-2 Hc,⊥ (Oe) VPT NW (6.3μm) 0.12 0.07 25 80.0 83.2 Hydrothermal NW 0.11 0.07 36 64.8 56.5 Hydrothermal NT 0.21 0.15 166 72.3 90.9 ALD NT 0.07 0.14 154 50.2 70.1 Hc,// (Oe) However, the polycrystalline ZnO NTs exhibit an easy axis in the out-of-plane direction different from the single-crystalline NWs and NTs. According to the normalized magnetization curves for the single-crystalline and polycrystalline NWs and NTs, the saturation field of the single-crystalline ZnO NWs/NTs is lower for applied field perpendicular to the tube axis than in the parallel direction. In contrast, the saturation field of the polycrystalline ZnO NTs is lower for applied field parallel to the tube axis than in the perpendicular direction, and the in-plane remanant squareness (Mr,Ms,⊥) is only 0.07, much lower than 0.21 for the single-crystalline 300 nm NTs, while out-ofplane remanant squareness (Mr,Ms,//) of the polycrystalline NTs is 0.14, higher than the Mr,Ms,⊥ of 0.07. This change of magnetic anisotropy from a single-crystalline to a polycrystalline ZnO NT array should be mainly due to the change of the anisotropic orientation of the Zn-S bonds. The harder-to-saturate magnetic signal in the longitudinal direction for the single-crystalline NWs or NTs is consistent with the nature of orbital magnetism in this system: the rotation of the angular momenta with an applied external magnetic field would require the rotation of the electric charge distribution linked to the highly oriented Zn-S bond at the (200) surface of the NTs.11 Since a mismatch of the spin direction with the magnetic field makes it more difficult to saturate the signal in the 155 longitudinal direction, the preferential magnetization direction will be perpendicular to the NW/NT axis. This experimental observation matches the results of the DFT calculation. However, this long-range ordering of dipoles is intrinsically difficult in the case of thiol-capped ZnO polycrystalline NTs due to its rough surface and the disordered bond orientations, as shown in part c of Figure 6.12. Consequently, the polycrystalline ALD NTs does not possess long–range Zn-S dipole orientation on the lateral side walls (part c of Figure 6.12), thus shape anisotropy of the NTs plays a more important contribution to the magnetic anisotropy. In addition, the effect of coupling between NWs/NTs via dipole-dipole interaction on magnetic anisotropic properties, which may be present in a collection of wires or tubes with inherent easy axis along their long axis and in close proximity to one another, can be ruled out. Due to the low magnetic moment and large spacing between the ZnO NWs/NTs, such coupling is relatively small, and magnetostatic interaction between the NWs and NTs in the array may be negligible. Furthermore, the observed dependence of the anisotropy on the orientation of the organic molecule relative to the surface normal is similar to that of PbS nanoparticles linked to GaAs surface by 1,8-octanedithiol and 1,4-benzenedimethanethiol14 as well as thiolcapped gold surface.33 This convergence of experimental observations highlights an interesting phenomenon i.e. well-defined crystal faces play an important role in the magnetic anisotropy contributed by interfacial spins at a SAM modified interface. Although the magnetic anisotropy for different thiol-capped ZnO NWs/NTs array determined by remanant squareness is small, the results imply the possibility that by implementing these interfaces on 1D nanostructures, the defined magnetization direction with respect to the 1D nanostructure axis may allow tuning of magnetic anisotropy depending on the alignment, dimensions and crystallinity of the 1D nanostructures. 156 157 6.4 Conclusions In summary, we have investigated magnetic properties of thiol-capped 3D singlecrystalline ZnO NWs and NTs as well as polycrystalline ZnO NTs arrays. Room temperature ferromagnetism has been observed in these thiol-capped NWs and NTs array. A density functional calculation has been performed and the result indicates that the magnetism in such system mainly originated from the spin polarized 3p electrons in S sites. The dependence of Ms of ZnO NWs arrays on the NW height can be ascribed to a higher number of Zn-S bond spins formed at the ZnO-thiol interface. Similarly, the NTs exhibited a higher Ms than the NWs due to its larger surface-to-volume ratio and thus higher density of Zn-S bond spins. The observed dependence of the magnetic anisotropy on crystallinity of ZnO NT, suggests that the change in the ordering of Zn-S spins on the crystal facets can affect the extent of magnetic ordering, and alter the preferred magnetization direction. The magnetic properties in aligned thiol-capped ZnO NWs/NTs systems is very complex and many factors such as geometric parameters of ZnO NWs/NTs (length, diameter, wall thickness, neighbor-to-neighbor distances), chemical effect of thiolate bond formation can affect the observed magnetic properties. However, the results indicate that the rational control of the magnetic properties by tuning the morphology and crystallinity of these 1D ZnO nanostructures can be achieved. The ability to tune the magnetic properties demonstrated here offers opportunities for the utilization of these ferromagnetic ZnO NWs/NTs arrays as nanoscale building blocks in optoelectronic and magneto-optical devices. 158 References Wolf, S. A.; Awschalom, D. D.; Buhrman, R. A.; Daughton, J. M.; von Molnár, S.; Roukes, M. L.; Chtchelkanova, A. Y.; Treger, D. M. Science 2001, 294, 1488. Piramanayagam, S. N.; Aung, K. O.; Deng, S.; Sbiaa, R. J. Appl. Phys. 2009, 105, 07C118. Lee, H.; Yu, M. K.; Park, S.; Moon, S.; Min, J. J.; Jeong, Y. Y.; Kang, H. W.; Jon, S. J. Am. Chem. Soc. 2007, 129, 12739. Marianna, C.; Vincenzo, G.; Elvio, C.; Cinzia, G.; Fabia, G.; Enrique, F. P.; Miguel, A. G.; Liberato, M.; Roberto, C.; Pantaleo, D. C. Nano Lett. 2007, 7, 1386. Schoenhalz, A. L.; Arantes, J.T.; Fazzio, A. Appl. Phy. Lett. 2009, 94, 162503. Hong, N. H.; Sakai, J.; Brize, V. J. Phys. Condens. Matter 2007, 19, 036219. Wang, Q.; Sun, Q.; Chen, G.; Kawazoe, Y.; Jena, P. Phys. Rev. B 2008, 77, 205411. Venkatesan, M.; Fitzgerald, C. B.; Lunney, J. G.; Coey, J. M. D. Phys. Rev. Lett. 2004, 93, 177206. MacManus-Driscoll, J. L.; Khare, N.; Liu, Y.; Vickers, M. E Adv. Mater. 2007, 19, 2925. 10 He, Y.; Sharma, P.; Biswas, K.; Liu, E. Z.; Ohtsu, N.; Inoue, A.; Inada, Y.; Nomura, M.; Tse, J. S.; Yin, S.; Jiang, J. Z. Phys. Rev. B 2008, 78, 155202. 11 Garitaonandia, J. S.; Insausti, M.; Goikolea, E.; Suzuki, M.; Cashion, J. D.; Kawamura, N.; Ohsawa, H.; de Muro, I. G.; Suzuki, K.; Plazaola, F.; Rojo, T. Nano Lett. 2008, 8, 661-667. 12 López-Cartés, C.; Rojas, T. C.; Litrán, R.; Martínez-Martínez, D.; de la Fuente, J. M.; Penadés, S.; Fernández, A. J. Phys. Chem. B 2005, 109, 8761-8799. 159 13 Garcia, M. A.; Merino, J. M.; Pinel, E. F.; Quesada, A.; de la Venta, J.; González, M. L. R.; Castro, G. R.; Crespo, P.; Llopis, J.; González-Calbet, J. M. et al. Nano Lett. 2007, 7, 1489. 14 Seehra, M. S.; Dutta, P.; Neeleshwar, S.; Chen, Y. Y.; Chen, C. L.; Chou, S. W.; Chen, C. C.; Dong, C. L.; Chang. C. L Adv. Mater. 2008, 20, 1656. 15 Alexander, Z.; Gregory, L.; Sidney, R. C.; Ron, N. Adv. Mater. 2008, 20, 2552. 16 Wang, Q.; Sun, Q.; Jena, P. J. Chem. Phy. 2008, 129, 164714. 17 Deng, S.; Loh, K. P.; Yi, J. B.; Ding, J.; Tan, H. R.; Lin, M.; Foo, Y. L.; Zheng, M.; Sow, C. H. App. Phy. Lett. 2008, 93, 193111. 18 Qiu, X.; Li, L.; Tang, C.; Li, G. J. Am. Chem. Soc. 2007, 129, 11908. 19 Anna, M.; Teresa, V. S.; Li, G.; Tsang, S. C. J. Am. Chem. Soc. 2009, 131, 12540– 12541. 20 Li, Q. H.; Gao, T.; Wang, Y. G.; Wang, T. H. Appl. Phys. Lett. 2005, 86, 123117. 21 Wang, Q.; Lin, C. L.; Yu, X. B.; Wang, T. H. Appl. Phys. Lett. 2004, 84, 124. 22 Li, L.; Pan, S. S.; Dou, X.; Zhu, Y.; Huang, X.; Yang, Y.; Li, G.; Zhang, L. Direct J. Phys. Chem. C 2007, 111, 7288. 23 Greyson, E. C.; Babayan, Y.; Odom, T. W. Adv. Mater. 2004, 16, 1348. 24 Liu, R. B.; Pan, A. L.; Fan, H. M.; Wang, F.F.; Shen, Z. X.; Yang, G. Z.; Xie, S. S.; Zou, B. S. J. Phy. Condensed Matter 2007, 19, 136206. 25 Liu, J. F.; Liu, E. Z.; Wang, H.; Su, N. H.; Qi, J.; Jiang, J. Z. Nanotechnology 2009, 20, 165702. 26 Sui, Y. C.; Skomski, R.; Sorge, K. D.; Sellmyer, D. J. Appl. Phys. Lett. 2004, 84, 1525. 27 Zhang, P.; Sham, T. K. Phys. Rev. Lett. 2003, 90, 245502. 160 28 Liu, H.; Mun, B. S.; Thomton, G.; Isaacs, S. R.; Shon, Y. S.; Ogletree, D. F.; Salmeron, M. Phys. Rev. B 2005, 72, 155430. 29 Nobusada, K. J. Phys. Chem. B 2004, 108, 11904. 30 Xing, Y. J.; Xi, Z. H.; Zhang, X. D.; Song, J. H.; Wang, R. M.; Xu, J.; Xue, Z. Q.; Yu, D. P. Solid State Commun. 2004, 129, 671. 31 Dutta, P.; Pal, S.; Seehra, M. S.; Anand, A.; Roberts, C. B. Appl. Phys. Lett. 2007, 90, 213102. 32 Hori, H.; Teranishi, T.; Nakae, Y.; Seino, Y.; Miyake, M.; Yamada, S. Phys. Lett. A 1999, 263, 406. 33 Hernando, A.; Crespo, P.; García, M. A.; Pinel, E. F.; de la Venta, J.; Fernández, A.; Penadés, S. Phy. Rev. B 2006, 74, 052403. 34 Ulman, A. Chem. Rev. 1996, 96, 1533. 35 Jiang, G.; Wang, L.; Chen, T.; Yu, H.; Chen, C. Mater. Chem. Phys. 2006, 98, 76-82. 36 Kresse, G.; Hafner, H. Phys. Rev. B 1993, 47, 558. 37 Langreth, D. C.; Perdew, J. P. Phys. Rev. B 1980, 21, 5469. 38 Li, D.; Richard, S. T.; Gerd, B.; Lu, J. G. Adv. Mater. 2008, 20, 4575. 39 Escrig, J.; Bachmann, J.; Jing, J.; Daub, M.; Altbir, D.; Nielsch, K. Phy. Rev. B 2008, 77, 214421. 40 Greene, L. E.; Law, M.; Goldberger, J.; Kim, F.; Johnson, J. C.; Zhang, Y.; Saykally, R. J.; Yang, P. Angew. Chem. Int. Ed. 2003, 42, 3031. 41 Wang, M.; Gao, H.; Carl, V. T.; Lee, K. T; Chua, S. J. MRS 2008 Fall Symposium B. 2008. 42 Masuda, H.; Satoh, M. Jpn, J. Apl. Phy. 1996, 35, L126. 161 Chapter Conclusions We have taken a journey through the topic of opportunities in nanomagnetism with the emphasis on the use of nanomagnetism as a platform to illustrate some of the interests in the emerging field of nanoscience. Thus, the scope ranges from bulk ferromagnetic materials like FePt and CoCrPt to interfacial magnetism in thiol-capped ZnO. The large range of the scope, however, does not imply that they are separate isolated topics, but rather that the field is veritably bursting with activity and crossdisciplinary innovation. Bottom-up and top-down fabrication approaches were illustrated. Advances in instrumentation were essential for the characterization of these materials, ranging from major facility synchrotron X-ray spectroscopy to scanning probes, magneto-optics, magnetometry, and electron microscopies. Geometric confinement, physical proximity, and chemical self-organization were enunciated as the main principles of nanoscience. These principles were illustrated with examples involving creation of differential magnetic domains of FePt nanoclusters on a topographic pattern of self-assembled colloidal crystal templates, stable magnetic recording media signified by a narrow switching field distribution, and novel class of interfacial magnetic 1D organic-inorganic nanosystems. The magnetic properties were accounted for via surface chemistry, morphology, spin-orbit coupling and magnetostatic interactions. Magnetism in bulk solids is a well-developed research area which has provided a fertile testing ground for quantum mechanics, theories of many-body interactions and collective phenomena, and critical phenomena. Surfaces, interfaces, and thin films represent new magnetic systems and, furthermore, are the, building blocks for more 162 complex systems such as multilayers. Size effects and lower dimensionality add interesting new facets to the study of magnetic properties of surfaces, interfaces, and thin films. From an experimental perspective, the understanding of how interface anisotropy can be controlled will determine whether future devices can be developed. For example, what is the effect of the semiconductor materials on the perpendicular or in-plane anisotropy at the interface of the nanostructure? Can surface anisotropy be used to compensate bulk anisotropy? Can the dispersion in anisotropy be reduced by growing more aligned and/or well-defined nanostructured arrays? These very practical questions require a more thorough understanding, both theoretical and experimental, of many materials. A new class of 1D interfacial magnetic nanostructures has been developed by the structural and crystalline modulation of 1D organic-inorganic hybrid nanostructures. The onset of interfacial magnetism in thiol-capped semiconductors would imply that other functional groups capping the semiconductor nanostructures with variable charge transfer properties can be a strategy to tune the magnetic property of the organic-inorganic hybrid nanostructures. Thus there are two approaches for the further study of this class of interfacial inorganic-organic magnets: (1) the continued developments in growth and characterization of materials at the nanoscale, coupled with (2) the choice of organic capping. Tuning these two aspects provides a detailed investigation of effect of charge transfer on the magnetic moment of the inorganic-organic hybrid structure, which would 163 100 nm FePt S (iv) Ar Aggre state Plasma (ii (i) )i)M gation Metal RIE FePt Therm al (i) (ii) (iii) (iv) Vortex Twisted Diamond Triangle µm Remanent tLower Low High H (ii) rrP anne gas nanoclusters nan un(He) aling derlayer ocluster states P lower SFD SFDat dhigher electrons deposition different H -3 cm ) 11 X-rays1000 S d set the foundation for the proliferation of the interfacial phenomena for the creation of a new class of ferromagnetic and semiconducting materials. The future of nanomagnetism is multidisciplinary and irrevocably linked to advances in chemistry and chemical processing with their historical bottom-up strategy. Creating a new materials system and understanding its magnetic properties will potentially make a significant contribution to technology. We have new materials exhibiting new properties and presenting many fascinating fundamental questions to be answered when one combines magnetism and transport at the nanoscale. With a better control of magnetism at the nanoscale, new technologies will emerge. It is clear that magnetic nanostructures will play a major role in data storage, memory, and sensors applications. 164 [...]... coordination number of the nearest Ag atom due to chemical bonding with S, which results in the weakening of the Ag-Ag d-d interaction 24 These observations support that the strong Ag-S interactions make the 4d electrons of the core Ag NCs behave as Normalized absorption (a.u.) partially localized holes that have lost itinerancy 1.0 Ag L3 edge AgNC-ZnO NW 0.5 0.0 3. 32 AgNC 3. 36 3. 40 3. 44 3. 36 3. 40 3. 44 Energy... synthesis methods of dilute ferromagnetic semiconductors involve the doping of magnetic (nonmagnetic) elements into semiconductor nanoparticles to induce 3d-host hybridization and strong Coulomb interactions between 3d-3d electrons.4 However, the nature and origin of the observed magnetic behavior of dilute magnetic semiconductors is still unclear This is complicated by the presence of intrinsic point... alignment of the orbital moments via spin-orbit coupling 22 Table 5.1 Coercivity and saturation magnetization values for C9dithiol-capped, Adamantanethiol (AT) and 1H,1H,2H,2H-perfluoroalkanethiol (FT) Ag NC-ZnO NW Sample Hc,5K (Oe) Hc ,35 0K (Oe) Ms,5K (μemucm2 ) Ms ,35 0K (μemucm2 ) C9-(13nm) 1 73 78 23 22 C9-(4nm) 124 48 59 96 AT-(13nm) 1 73 83 14 13 AT-(4nm) 85 48 126 158 FT-(13nm) 107 59 19 16 FT-(4nm) 2 23. .. development of a new class of magnetic semiconductor nanostructures for various applications Interfacial magnetic behavior has been observed for organically passivated semiconductor nanostructures including in ZnO, 13 CdSe,14 PbS nanoparticles assembly on GaAs.15 Of these, ZnO is particularly attractive because of its ability to retain ferromagnetism at room temperature and possibility of producing... out -of- plane direction Low temperature hysteresis loops were measured with a superconducting quantum interference device (SQUID, QUANTUM DESIGN) magnetometer The sample sizes range from 3 3mm to 5×5mm The as-measured hysteresis loop consists of a ferromagnetic signal superimposed on a diamagnetic background The diamagnetic background was substracted to display the ferromagnetic signal only 134 6 .3 Results... the NTs of outer diameter ~30 0 nm The 45° tilted SEM views for the NW and NT in parts d and g of Figure 6.1 show the vertical alignment of the NWs and NTs The TEM image of a single NT in part h of Figure 6.1 clearly shows the hollow structure of the NTs with well-defined traversal contrast and the inner/outer wall surfaces are quite smooth The 30 0 nm NTs have a wall thickness (dwall) of ∼ 30 nm Similar... 6 .3 Results and Discussion 6 .3. 1 Characterizations of bare and thiol-capped ZnO NWs/NTs Figure 6.1 SEM and TEM of single-crystalline NWs and NTs (a) 45° tilted SEM of vertically aligned VPT-grown NWs array (b) TEM image of a several VPT-grown NWs (c) SAED and HRTEM of VPT-grown NW (d) 45° tilted SEM of hydrothermally grown NWs array (e) TEM image of a single NW (f) HRTEM of a single NW Inset shows the... ferromagnetic NTs In addition, due to the higher specific surface and porosity, ZnO NTs are an attractive alternative to ZnO NPs and NWs for the understanding of the effect of 129 morphology on the physical properties of nanomaterials,19 and in the fabrication of new nanodevices.20,21 Scheme 6.1 Ordered alignment of magnetic moments in thiol-capped single crystalline ZnO NW/NT vs disordered alignment of. .. diameter of 80 nm with wall thickness of ~15-20 nm for 100 ALD cycles Part b of Figure 6.2 is a TEM image of a single NT which shows that the wall of NT is uniform in thickness, confirming that successive ALD preserves the porous structure of the AAO template The polycrystalline nature of the walls was confirmed by high resolution TEM and SAED as shown in part b and c of Figure 6.2 The bottom of the... Since the Ag-ZnO before thiol capping was shown to be diamagnetic, the occurrence of ferromagnetism cannot be due to intrinsic defects in ZnO or ferromagnetic impurities There is a concern that the ferromagnetism might be due to the presence of magnetic impurities since magnetic impurities can induce polarization of the conduction electrons of Ag, giving rise to paramagnetism which rapidly decreases . charge rectification 3. 32 3. 36 3. 40 3. 44 0.0 0.5 1.0 3. 36 3. 40 3. 44 AgNC Normalized absorption (a.u.) Energy (keV) Ag L 3 edge AgNC-ZnO NW 121 properties (i.e., the flow of electrons and holes. H c,5K (Oe) H c ,35 0K (Oe) M s,5K (μemucm - 2 ) M s ,35 0K (μemucm - 2 ) C9-(13nm) 1 73 78 23 22 C9-(4nm) 124 48 59 96 AT-(13nm) 1 73 83 14 13 AT-(4nm) 85 48 126 158 FT-(13nm) 107 59 19. thiolate-bound monolayers (figure 5 .3) . 5 .3. 2 Effect of Ag NC size on Magnetic Properties of Thiol-capped Ag NC-ZnO NWs Figure 5.4 (a) M-H plot for 13 nm Ag NC-ZnO substrate at 35 0K (i) before and (ii)