ISSN 00201685, Inorganic Materials, 2014, Vol. 50, No. 8, pp. 793–802. © Pleiades Publishing, Ltd., 2014. Original Russian Text © M.V. Kuznetsov, O.V. Belousova, D. Ortega, Yu.G. Morozov, 2014, published in Neorganicheskie Materialy, 2014, Vol. 50, No. 8, pp. 856–866.
Ferromagnetic Sn–O Nanoparticles M. V. Kuznetsova, O. V. Belousovab, D. Ortegac, and Yu. G. Morozovb a
Department of Chemistry, Materials Chemistry Research Centre, University College London, 20 Gordon St., London WC1H 0AJ, United Kingdom b Institute of Structural Macrokinetics and Materials Science, Russian Academy of Sciences, ul. Akademika Osip’yana 8, Chernogolovka, Noginskii raion, Moscow oblast, 142432 Russia c Instituto Madrile@n[tilde]@o de Estudios Avanzados en Nanociencia (IMDEANanociencia), Madrid, Spain email:
[email protected] Received February 3, 2014
Abstract—Tin oxide nanoparticles ranging in average size from 12 to 315 nm have been prepared by levita tionjet aerosol synthesis through condensation of tin vapor in a flow of inert gases and oxygen (air). The nanoparticles have been characterized by transmission electron microscopy, Xray diffraction, BET measure ments, vibratingsample magnetometry, and Raman scattering spectroscopy. The results indicate that the nanoparticles may exhibit roomtemperature ferromagnetism, with their magnetization having a maximum at O : Sn = 1. The ferromagnetic order is tentatively attributed to the presence of localized states on the Sn/SnO and SnO/SnO2 interfaces. DOI: 10.1134/S0020168514080111
INTRODUCTION The discovery of roomtemperature ferromag netism (RTFM) in nanoparticles of undoped main group metal oxides (ZnO, TiO2, HfO2, and others) has opened up new possibilities for creating spinbased electronic devices [1–4]. The origin of this unex pected RTFM has been one of the most extensively discussed topics among researchers all over the world [5–8]. The systematic dependence on the methods used to prepare the nanomaterials and the similar magnetization values in all reports are convincing evi dence that RTFM is a characteristic feature of such oxides. It is well known that tin dioxide, SnO2, is a material of technological importance, which is widely used in catalysts, solar cells, optoelectronic devices, and transparent semiconductor coatings [9–11]. This material has great potential for use in gas sensors, especially in the context of emerging fuel cell technol ogies [12]. It is known also that tin oxide exists in the form of two phases, SnO2 and SnO, in which tin is in the oxi dation states 4+ and 2+, respectively. Tin(IV) oxide is an ntype semiconductor containing native oxygen vacancies, and its properties have been well studied. On the other hand, SnO offers high рtype conductiv ity due to native Sn vacancies. Analysis of reports con cerned with the crystal structure and properties of the tin oxides shows that SnO and its properties have been the subject of very few detailed experimental studies, primarily because it has reduced stability and readily converts to SnO2.
In addition, except for several results reported pre viously [8], the magnetic properties of nanoparticulate tin oxides have not yet been investigated in sufficient detail, in contrast to those of, e.g., ZnO and TiO2. In particular, in experimental studies of undoped SnO2 nanoparticles, magnetic measurements yielded maxi mum roomtemperature saturation magnetizations from 8 × 10 –4 to 1.5 × 10 –3 A m2/kg [13–17], which considerably exceeded uncertainty arising from vari ous experimental conditions [18]. Even though the maximum saturation magnetization in undoped films reached 12 A m2/kg [14], this value appears unlikely because the weight of such films is difficult to accu rately determine. Studies of RTFM in nanoparticles of pure ZnO in combination with organic ligands [19, 20] or in rela tively large zinc nanoparticles coated with small ZnO nanoparticles (see Kuznetsov et al. [21], where this issue was addressed in detail) highlight the importance of understanding the fine surface structure of nano particles. In most studies concerned with the proper ties of undoped tin dioxide, RTFM is assigned to low dimensionality and various defects. Even though Raman et al. [22] tentatively attributed RTFM to tin vacancies in the crystal lattice of tin dioxide, the VSn configuration is not regarded as thermodynamically favorable [11]. For this reason, it is thought that the effect is primarily due to oxygen vacancies, which are located in the surface layer because of the large surface area to volume ratio in nanomaterials. At the same time, the high density of oxygen defects in tin dioxide essentially shifts the local phase composition of tin dioxide toward that of tin monoxide and metallic tin.
793
794
KUZNETSOV et al.
This issue has not been addressed in the literature, so our purpose was to prepare Sn–O nanoparticles with controlled phase composition and investigate the prin cipal physicochemical and magnetic characteristics of the materials obtained. Such research is of vital impor tance to technological applications, as has recently been demonstrated by the example of studies of the enhanced resistive hydrogen sensing response of nanocomposites based on SnO/SnO2 heterojunctions [23]. The possibility of magnetically controlling this characteristic of RTFM materials offers the promise of creating radically new sensing elements. Special interest in the study of Sn–O nanoparticles is also aroused by the fact that compact samples of slightly oxidized tin nanoparticles were found previ ously to have a higher superconducting transition tem perature in comparison with bulk tin [24, 25]. The observed behavior was probably due to the formation of current contact arrays, which were very similar in appearance to “grainboundary foam” [26] in systems exhibiting RTFM. Such an approach is consistent with the recently discovered relationship between the origins of roomtemperature superconductivity [27] and RTFM in such systems [21], which will of course be the subject of further detailed investigation.
WIN ver. 2.02, release 1999) and the Crystallographica SearchMatch ver. 3.102 program. Rietveld profile analysis of Xray diffraction patterns with PowderCell 2.0 was used to evaluate the relative concentrations of crystalline phases in the nanomaterials obtained. Pow der morphology was examined by transmission elec tron microscopy (TEM) on a JEM1200EX II (JEOL, Japan) operated at an accelerating voltage of 120 kV. In a number of cases, highresolution TEM (HRTEM) images were obtained on a JEM2100F microscope (JEOL, Japan). The specific surface area of the nanoparticles was determined by fourpoint nitrogen physisorption BET measurements using a META SORBIM instrument. Magnetic characteris tics of the powders were measured in magnetic fields of up to 0.8 MA/m at room temperature using an M4500 vibratingsample magnetometer (EG&G PARC, USA). Diamagnetic signals from a nylon ampule was subtracted from the total magnetization of the nano particles in the form of loose powder. The nanoparti cles were characterized by Raman scattering spectros copy on an InVia Raman spectrometer (Renishaw, United Kingdom) and a DMLM confocal microscope (Leica, Germany) with He–Cd and Ar lasers (excita tion wavelengths of 325 and 514 nm, respectively).
EXPERIMENTAL Sn–O aerosol nanoparticles were prepared using a Gen levitation jet generator described in detail else where [28, 29]. To this end, a 3.5mmdiameter OVCh0000 (99.9999%) tin wire situated in a quartz tube 14 mm in inner diameter was introduced into an rf inductor encasing the tube. The end of the wire was heated by the electromagnetic field of a countercur rent inductor until the formation of a levitating liquid droplet and the onset of metal vaporization. The drop let was blown by helium or argon at a constant gas flow rate and fed with the source wire at regular time inter vals, as the metal vaporized. In the series of experi ments under consideration, tin nanoparticles were oxidized by adding a gaseous oxidant (oxygen or air) to the inert gas flow upstream of the levitating droplet [30] or by running the process only in flowing air. At high oxidant flow rates and gas pressures in the system under 40 kPa, we observed the deposition of a large amount of material on the inner wall of the quartz tube. The metallic deposit absorbed most of the power of the rf inductor, distorting gas mixture flow laminar ity and strongly reducing the duration of the process (to 10 min). After the process reached completion, the resultant gray to dark gray powder was collected on fil ter cloth and/or from the surface of a cooler. Measures were taken to preclude the presence of foreign mag netic contaminants [21]. The crystal structure of the nanoparticles was determined by Xray diffraction on an ADP2 diffrac tometer (CuKα radiation). Their phase composition was determined using JCPDS PDF data (PCPDF
RESULTS AND DISCUSSION Morphological characterization and BET results. Figure 1 shows micrographs of some typical samples of the nanoparticles obtained. As seen in Fig. 1a, the large particles obtained in flowing argon (table, S1) are nearly spherical in shape. The particles less than 40 nm in average size (S13) obtained in flowing air when the reaction zone was pumped had the form of polyhedra formed during the oxidation of the tin nanoparticles. In Figs. 1c and 1d, it is well seen at high magnification that samples S8 (lower pressure in comparison with the previous sample) and S10 (considerably reduced gas flow rate and lower pressure) contain highly agglomerated, irregularly shaped small nanoparticles. Since the preparation of these nanoparticles involved cluster growth from vaporized atoms and concurrent agglomeration of the atoms, particle growth and nucleation processes control both the size and mor phology of the aerosol nanoparticles. Nuclei of small crystalline blocks have a tendency to coalesce into the most thermodynamically stable nanoparticles [31]. Next, reducing the surface energy of the oxide in the oxidation process, van der Waals forces cause such nanostructures to selfassemble [23]. Figure 2a is an HRTEM image of the nanoparticles in sample S8, which characterizes the crystal structure of the material, and Fig. 2b shows the intensity profiles along lines A and B in the image. Analysis of the inten sity profiles leads us to conclude that they correspond to the {110} planes of the SnO2 and SnO phases, respectively, which seem to be concentrated in indi vidual nanoparticles. INORGANIC MATERIALS
Vol. 50
No. 8
2014
FERROMAGNETIC Sn–O NANOPARTICLES
200 nm
(а)
50 nm
(c)
(b)
(d)
795
50 nm
50 nm
Fig. 1. Micrographs of the (a) S1, (b) S13, (c) S8, and (d) S10 nanoparticles.
The specific surface areas S of the various nanopar ticles are indicated in the table. From these data, we evaluated the average particle size dBET (nm). It was found to be rather close to the average nanoparticle size estimated from micrographs and was subsequently used as one of the principal characteristics of the material. Xray diffraction characterization. Figure 3 shows typical Xray diffraction patterns of the nanoparticles whose micrographs are presented in Fig. 1. It is well seen that the large particles (S1) are essentially unoxi dized and that their diffraction pattern corresponds to the tetragonal lattice of metallic tin. The nanoparticles obtained in the presence of a gaseous oxidant contain tin oxides of different stoichiometries. Sample S13 consists of pure tin dioxide. Samples S8 and S10 con INORGANIC MATERIALS
Vol. 50
No. 8
2014
sist of a mixture of the SnO and SnO2 oxides, with considerably broadened diffraction peaks in the case of the smallest particles in S10. In the Xray diffraction pattern of sample S8, the peaks of SnO are broader than those of SnO2, in full accord with the conclusions above. Analysis of all the diffraction patterns obtained showed that there were only reflections from the tet ragonal lattice of Sn (JCPDS card no. 040673) with lattice parameters a = 0.5831 nm and c = 0.3182 nm; from tetragonal tin monoxide, SnO (JCPDS card no. 721012), with a = 0.3803 nm and c = 0.4838 nm; and from tetragonal tin dioxide, SnO2, (JCPDS card no. 721147) with a = 0.4737 nm and c = 0.3185 nm. It should be emphasized that, in all of the samples stud ied, the lattice parameters of the nanoparticles varied little with particle size (in contrast to those of zinc
796
KUZNETSOV et al.
B
6. 06
nm
A
6.01 nm
10 nm
(а)
(b) 130 120
Intensity, arb. units
110 100
A
90 80 70 60 50 40 30 20
B
10 0
0.5
1.0
1.5 2.0 2.5 3.0 3.5 4.0 Distance, nm
4.5 5.0 5.5 6.0
Fig. 2. (a) HRTEM image of nanoparticles in sample S8; (b) intensity profiles along lines А and B in the image. INORGANIC MATERIALS
Vol. 50
No. 8
2014
FERROMAGNETIC Sn–O NANOPARTICLES σ, A m2/kg
80
SnO SnO2
70 Intensity, %
797
0.10
60
0.05
50 40
0 S8
30
–0.05
S13
10 0 20
S8
S10
20
40
50 2θ, deg
60
70
S6
–0.10 S5
S1 30
S13
S10
80
–0.8 –0.6 –0.4 –0.2
0.2 0 H, МА/m
0.4
0.6
0.8
Fig. 3. Xray diffraction patterns of the nanoparticles (the curves are numbered as the samples in the table).
Fig. 4. Ferromagnetic hysteresis loops of the nanoparticles (the curves are numbered as the samples in the table).
nanoparticles [21]). The percentages of the observed phases, evaluated by profile analysis of Xray diffrac tion patterns, are presented in the table together with the principal controlled parameters of the nanoparti cle preparation process: the metallic tin feed rate, the inert and reactant gas flow rates, and the gas pressure p in the system.
considerably exceeds that of bulk SnO2 (–5 × 10–11 m3/kg [32]). At the same time, the magnetization curve obtained by us for SnO powder with a particle size near 1 µm (Aldrich, CAS no. 21 651194, 99.99 wt %) showed diamagnetic behavior with an even higher magnetic susceptibility: –5.2 × 10–10 m3/kg (data on diamagnetism of bulk SnO are not available in the lit erature). Given this, it cannot be ruled out that the excess diamagnetism may originate from the existence of some superconducting regions on the surface of the material [27]. Another approach takes into account that the electronegativity of a Sn atom is higher than that of the oxygen atoms bonded to it in the SnO2 compound and that calculations predict a very large chemical shift in a nanoparticulate state [33]. There fore, the diamagnetic shielding in SnO2 also increases, and this may lead to changes in the magnetic proper ties of the oxide [33].
It can be seen from the data in the table that the most suitable way of varying the phase composition of the nanoparticles is by pumping the reaction zone, through which the air stream flows [29], but this is accompanied by a marked decrease in average nano particle size. It is worth pointing out that the composition of the phasepure SnO2 nanoparticles did not vary during storage in air. However, like in the case of zinc nano particles [21], the storage of the multiphase nanopar ticles was accompanied by a gradual variation in their phase composition (toward the formation of pure tin dioxide as mentioned above). Because of this, we per formed repeated magnetic measurements on particu lar samples, almost concurrently with the determina tion of their crystal structure and phase composition. The table presents the corresponding measurement results for our samples after storage for about three years under normal conditions, except for the aspre pared samples S3 and S4. Magnetic measurements. Roomtemperature mag netic measurements showed that most of the nanopar ticles obtained had a ferromagnetic hysteresis loop. Figure 4 shows the magnetization of the nanoparti cles, σ, versus applied magnetic field, H, for some samples of the nanoparticles. Samples S10 and S13 exhibit weak ferromagnetism, accompanied by dia magnetic behavior in high fields, with a roomtempera ture magnetic susceptibility of –3 × 10–10 m3/kg, which INORGANIC MATERIALS
Vol. 50
No. 8
2014
All of the other curves are similar in shape, indica tive of softmagnetic behavior. Since the coercive force of all the nanoparticles was within 8 kA/m, the table gives only their specific saturation magnetization σs. It is well seen that the saturation magnetization of the nanoparticles is independent of their average size. The maximum roomtemperature saturation magne tization obtained by us is 0.1 A m2/kg, which a is a rather high value compared to data in the literature. It should be kept in mind that the weight of the entire sample was used in calculation [18], even though RTFM exists most likely not throughout the sam ple [21]. Figure 5 demonstrates that RTFM in the Sn–O system depends on stoichiometry: there is a well defined positive correlation between the saturation magnetization and the O : Sn ratio in the different nanoparticles. In addition, the magnetization of the
INORGANIC MATERIALS
Vol. 50
No. 8
* Feed or flow rate.
87
89
S14
S15
42
S8
S13
23
S7
73
20
S6
S12
58
S5
140
19
S4
S11
19
S3
13
31
S2
S10
12
S1
17
310
Sample
S9
dBET, nm
0
0
0
1.0
2.5
0
0
0
0
13.3
6.5
4.8
20
46.7
100
Sn
0
0
0
0
0
22.9
37.2
39.7
58.7
39.3
74.5
86.5
70
53.3
0
SnO
99.0
97.5
77.1
62.8
60.3
41.3
47.4
19.0
8.7
0
0
0
SnO2
100
100
100
at % (XRD data)
Principal parameters of the nanoparticles
0.001
0.008
0.008
0.004
0.003
0.003
0.01
0.034
0.043
0.074
0.1
0.069
0.064
0.0165
0.0038
σs, A m2/kg
2.62 ± 0.17
6.32 ± 0.10
2
2
2
9.87 ± 0.13
9.62 ± 0.13
20.24 ± 0.16
1.941 11.81 ± 0.01
1.857
1.692 66.12 ± 0.39
1.531 54.35 ± 0.86
1.503 38.71 ± 1.36
1.319 44.81 ± 0.97
1.217 15.23 ± 0.52
0.997 49.41 ± 2.97
0.963 41.97 ± 2.45
0.636 27.69 ± 0.35
0.363 72.00 ± 0.67
0
O : Sn
S × 10–3, m2/kg
8.3
3.3
6.9
9.8
9.8
4.2
5.6
5.6
5.6
4.2
4.2
4.2
5.6
4.2
2.5
–
–
–
14
–
3.2
–
–
–
–
–
–
–
8.6
–
8.3
–
–
–
2.3
–
–
–
–
–
–
–
–
–
2.3
Sn* × 106, He* × 105, Ar* × 105, kg/s m3/s m3/s
–
12
12
–
–
6.4
12
12
3.3
2.7
2.7
2.7
2.7
3.6
–
Air* × 105, m3/s
Preparation conditions
1.4
–
–
1.4
1.4
–
–
–
–
–
–
–
–
–
–
O2* × 105, m3/s
–
–
40
–
–
1
3
10
8
10
6
8
5
0.5
–
P, kPa
798 KUZNETSOV et al.
2014
FERROMAGNETIC Sn–O NANOPARTICLES
nanoparticles is clearly seen to rise sharply at O : Sn = 1. Most likely, local magnetic moments appear in a thin layer at the Sn/SnO (to a higher degree) and SnO/SnO2 interfaces as a result of the formation of a highly imperfect structure [26], in which longrange magnetic interactions are possible. It is reasonable to assume that, like in zinc nanoparticles [21], the con centration of all vacancies has a maximum in the interfacial region, so this structure is similar to a ferro magnetic “grainboundary foam” [27] (Fig. 1c, 1d). Unfortunately, in contrast to that of zinc nanopar ticles [21], the hightemperature behavior of RTFM in nanoparticles of tin oxides proved rather difficult to investigate because of the low magnitude of the ferro magnetic response compared to the appreciable dia magnetic contribution from the hightemperature boron nitride ampule used. It should also be kept in mind that, given the above observation as to the possi bility of the coexistence of weakly temperaturedepen dence effects—excess diamagnetism (superconduc tivity) and ferromagnetism (like in the NiO–NiO sys tem [27])—the overall magnetic response of the nanoparticles to an applied magnetic field may have a nonmonotonic compensation character, depending on temperature. Near critical temperatures, magnetic signals from different physical subsystems of one sys tem are rather difficult to separate out if these temper atures differ only slightly. Nevertheless, RTFM in the system under consideration was observed to at least 750 K, in full accord with earlier findings for Zn/ZnO nanoparticles [21]. As to the fine structure of defect subsystems, the following can be proposed: It is known that the types and density of possible defects depend primarily on nanoparticle preparation conditions [34]. During syn thesis, the oxidant and metal form a complex which is at very high temperatures in a near condensation zone (>1500 K). This high temperature is, however, main tained for only a fraction of a second, and then the material is quenched to room temperature [35, 36]. This leads to the formation of high vacancy concentra tions within nanocrystallites [4]. The oxygen vacancies generated at high tempera tures persist at room temperature [4] and may exist in a neutral or filled state because in metal oxides elec tron hopping from the conduction band is possible. The oxygen vacancies may exist as neutral (VO), indi vidual (VO− ), or doubly filled (VO2 − ) vacancies, among which VO2 − acts as a shallow donor, whereas the VO− vacancies act as deep donors. The individual filled vacancies (that is, VO− ) become bound and give rise to longrange ferromagnetic ordering, in contrast to the doubly filled and neutral oxygen vacancies [4]. At the same time, understanding the relationship between complex defect subsystems and magnetism requires more indepth studies [21]. For this purpose, we tried to use Raman scattering spectroscopy. INORGANIC MATERIALS
Vol. 50
No. 8
2014
799
σ, A m2/kg 0.10
S5
0.08 S3
0.06
S6
S4
S7
0.04
S8
0.02
S2 S11 S12
S9
S1
S10
0 0
0.5
1.0
1.5
S13
2.0 O : Sn
Fig. 5. Maximum magnetization as a function of the O : Sn ratio (the data points are numbered as the samples in the table).
Raman spectroscopy characterization. Raman scattering spectroscopy is one of the most informative methods for analysis of the crystal structure, crystal line phases, and oxygen vacancyrelated defects in nanostructured materials [4, 23, 31]. In Ramanactive 1 modes, oxygen atoms experience vibrations, whereas Sn atoms are at rest [37]. Figures 6 and 7 show roomtemperature Raman spectra of the nanoparticles. For convenience, the spectra are divided into two frequency ranges. In the lowfrequency part of the spectrum (Fig. 6), the peak at 208 cm–1 may be due to the active vibrational modeA1g of SnO [38], and its position varies little with nanoparticle size. The peak at 167 cm–1 may arise from SnOx transition mixed oxide states with 1 < x < 2 (170 cm–1 [38–40]), in particular from Sn3O4 (171 cm–1 [41]). This peak is extremely weak in the spectra of theS8, S9, and S12 nanoparticles and has the highest intensity in the spectrum of the S6 nano particles, in full accord with Xray diffraction data that indicate the presence of trace levels of metallic tin in these samples. Even in the case of the S1 nanoparticles (pure tin according to Xray diffraction data), Raman spectra indicate the presence of a nonstoichiometric oxide. In addition, the spectrum of the S6 nanoparti cles contains un unidentified peak at 237 cm–1, attrib utable to interfacial disorder. In the highfrequency part of the spectrum (Fig. 7), the peaks well seen at 473, 633, and 772 cm–1 in the case of the S14 nanoparticles may arise from the active vibrational modes Eg (double degenerate), A1g, and B2g of SnO2, respectively [23]. The Eg peak corresponds to inplane vibrations of the oxygens [42, 43]. The A1g 2 and B2g peaks are attributable to the expansion and contraction of the Sn–O stretching vibrational mode, respectively [44]. In addition, the spectrum of this
800
KUZNETSOV et al. Sn3O4
100
A1g SnO
Intensity, arb. units
80
S1 S12 S8 S6
60
40
20
0 200
150
250
300
350 400 Raman shift, cm–1
Fig. 6. Raman spectra of the nanoparticles in the lowfrequency region.
A1g SnO2 160 S14 S6 S13 S15
140
Intensity, arb. units
120 100 80 Eg S2
B2g
S1
60
S3
40 20 400
500
600
700
800 900 Raman shift, cm–1
Fig. 7. Raman spectra of the nanoparticles in the highfrequency region.
sample contains a broad feature at 530–570 cm–1. This feature may be due to surface defects of SnO2, which are observed when the particle size decreases considerably (< 20 nm), and to a surface phonon mode that prevails in a nanoparticulate state [45]. The other peaks are attributable to modes inactive in the case of bulk SnO2. The peaks at 549, 495, and 694 cm–1 are assignable to similar modes, S1, S2, and S3, respec
tively, and emerge as a consequence of disorder activa tion on the nanocrystal surface [44, 46]. On the whole, in analyzing the Raman spectra it is worth pointing out that the S6 nanoparticles, which possess one of the largest ferromagnetic moments, are characterized by broad peaks of all the active modes, which are shifted to lower frequencies. Characteristi cally, these particles have a prominent peak of nonsto INORGANIC MATERIALS
Vol. 50
No. 8
2014
FERROMAGNETIC Sn–O NANOPARTICLES
ichiometric oxides, which points to an essential con nection with the development of RTFM. Most likely, such oxides form Sn/SnO and SnO/SnO2 heterojunc tions, which may also play a marked role in this pro cess [23]. CONCLUSIONS Tin oxide aerosol nanoparticles differing in stoichi ometry and average size (12–315 nm) can be prepared by levitation jet synthesis through condensation of tin metal vapor in an inertgas flow. The material may be a ferromagnet, with a saturation magnetization of up to 0.1 A m2/kg (at Sn : O = 1) and a coercive force of up to 8 A/m. Such behavior can be interpreted in terms of the defect structure of the Sn/SnO and SnO/SnO2 interfacial layers, containing oxygen vacancies, whose concentration and interactions can be controlled by varying nanoparticle preparation conditions. ACKNOWLEDGMENTS This research was supported by the Russian Foun dation for Basic Research, project no. 130312407. REFERENCES 1. Wolf, S.A., Awschalom, D.D., Buhrman, R.A., et al., Spintronics: a spinbased electronics vision for the future, Science, 2001, vol. 294, no. 5546, pp. 1488– 1495. 2. Prellier, W., Fouchet, A., and Mercey, B., Oxide diluted magnetic semiconductors: a review of the experimental status, J. Phys.: Condens. Matter, 2003, vol. 15, no. 37, pp. R1583–R1601. 3. Pearton, S.J., Heo, W.H., Ivill, M., et al., Dilute mag netic semiconducting oxides, Semicond. Sci. Technol., 2004, vol. 19, no. 10, pp. R59–R74. 4. Kamble, V.D., Bhat, S.V., and Umarji, A.M., Investi gating thermal stability of structural defects and its effect on d0 ferromagnetism in undoped SnO2, J. Appl. Phys., 2013, vol. 113, no. 24, paper 244 307. 5. Venkatesan, M., Fitzgerald, C., and Coey, J.M.D., Unexpected ferromagnetism in a dielectric oxide, Nature, 2004, vol. 430, p. 630. 6. Seshadri, R., Zinc oxidebased diluted magnetic semi conductors, Curr. Opin. Solid State Mater. Sci., 2005, vol. 9, nos. 1–2, pp. 1–7. 7. Coey, J.M.D. and Chambers, S.A., Oxide dilute mag netic semiconductors—fact or fiction?, MRS Bull., 2008, vol. 33, no. 11, pp. 1053–1058. 8. Ogale, S.B., Dilute doping, defects, and ferromag netism in metal oxide systems, Adv. Mater., 2010, vol. 22, no. 29, pp. 3125–3155. 9. Patil, K.C., Hegde, M.S., Rattan, T., and Aruna, S.T., Chemistry of Nanocrystalline Oxide Materials, Combus tion Synthesis, Properties and Applications, Singapore: World Scientific, 2008. INORGANIC MATERIALS
Vol. 50
No. 8
2014
801
10. Yang, F. and Forrest, S.R., Organic solar cells using transparent SnO2–F anodes, Adv. Mater., 2006, vol. 18, no. 15, pp. 2018–2022. 11. Kiliç, Ç. and Zunger, A., Origins of coexistence of con ductivity and transparency in SnO2, Phys. Rev. Lett., 2002, vol. 88, no. 9, paper 095 501. 12. Hubert, T., BoonBrett, L., Black, G., and Banach, U., Hydrogen sensors—a review, Sens. Actuators, B, 2011, vol. 157, pp. 329–352. 13. Sundaresan, A., Bhargavi, R., Rangarajan, N., et al., Ferromagnetism as a universal feature of nanoparticles of the otherwise nonmagnetic oxides, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, vol. 74, no. 16, paper 161 306. 14. Hong, N.H., Poirot, N., and Sakai, J., Ferromag netism observed in pristine SnO2 thin films, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, vol. 77, no. 3, paper 033 205. 15. Aragon, F.H., Coaquira, J.A.H., Hidalgo, P., et al., Structural and magnetic properties of pure and nickel doped SnO2 nanoparticles, J. Phys.: Condens. Matter, 2010, vol. 22, no. 49, paper 496 003. 16. Wang, C., Ge, M., and Jiang, J.Z., Magnetic behavior of nanosheets at room temperature, Appl. Phys. Lett., 2010, vol. 97, no. 4, paper 042 510. 17. Alanko, G.A., Thurber, A., Hanna, C.B., and Pun noose, A., Size, surface structure, and doping effects on ferromagnetism in SnO2, J. Appl. Phys., 2012, vol. 111, no. 7, paper 07C321. 18. Garcia, M.A., Fernandez Pinel, E., de la Venta, J., et al., Sources of experimental errors in the observation of nanoscale magnetism, J. Appl. Phys., 2009, vol. 105, no. 1, paper 013 925. 19. Garcia, M.A., Merino, J.M., Fernández Pinel, E., et al., Magnetic properties of ZnO nanoparticles, Nano Lett., 2007, vol. 7, no. 6, pp. 1489–1494. 20. Ortega, D., Chen, S.J., Suzuki, K., and Garitaonandia, J.S., Room temperature spontaneous magnetization in calcined trioctylphosphine–ZnO nanoparticles, J. Appl. Phys., 2012, vol. 111, no. 7, paper 07C314. 21. Kuznetsov, M.V., Morozov, Yu.G., Belousova, O.V., and Ortega, D., Ferromagnetic Zn/ZnO nanoparticles, Inorg. Mater., 2014, vol. 50, no. 4, pp. 369–378. 22. Rahman, G., GarciaSuarez, V.M., and Hong, S.C., Vacancyinduced magnetism in SnO2: a density func tional study, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, vol. 78, no. 18, paper 184 404. 23. Shanmugasundaram, A., Basak, P., Satyanarayana, L., and Manorama, S.V., Hierarchical SnO/SnO2 nano composites: formation of in situ p–n junctions and enhanced H2 sensing, Sens. Actuators, B, 2013, vol. 185, pp. 265–273. 24. Morozov, Yu.G., Naumenko, I.G., and Petinov, V.I., Superconducting transition temperature of small tin particles, Fiz. Tverd. Tela (Leningrad), 1973, vol. 15, pp. 3077–3078. 25. Morozov, Y.G. and Petinov, V.I., Superconductivity in the ensembles of small metallic particles, Solid State Commun., 1981, vol. 40, no. 11, pp. 991–994. 26. Straumal, B.B., Mazilkin, A.A., Protasova, S.G., et al., Magnetization study of nanograined pure and Mn doped ZnO films: formation of a ferromagnetic grain
802
27.
28. 29.
30.
31.
32. 33.
34.
35. 36.
KUZNETSOV et al. boundary foam, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, vol. 79, no. 20, paper 205 206. Morozov, Yu.G., Belousova, O.V., Ortega, D., et al., Some peculiarities in the magnetic behavior of aerosol generated NiO nanoparticles, J. Alloys Compd., 2013, vol. 572, pp. 150–157. Gen, M.Ya. and Miller, A.V., Levitation synthesis of ultrafine metal powders, Poverkhnost, 1983, no. 2, pp. 150–154. Morozov, Yu.G., Belousova, O.V., and Kuznetsov, M.V., Preparation of nickel nanoparticles for cata lytic applications, Inorg. Mater., 2011, vol. 47, no. 1, pp. 36–40. Morozov, Yu.G., Belousova, O.V., Kuznetsov, M.V., et al., Electric fieldassisted levitationjet aerosol synthe sis of Ni/NiO nanoparticles, J. Mater. Chem., 2012, vol. 22, no. 22, pp. 11 214–11 223. Varshney, D. and Verma, K., Effect of stirring time on size and dielectric properties of SnO2 nanoparticles prepared by coprecipitation method, J. Mol. Struct., 2013, vol. 1034, pp. 216–222. National Research Council, International Critical Tables, New York: McGrawHill, 1926. Ayeshamariam, A., Ramalingam, S., Bououdina, M., and Jayachandran, M., Preparation and characteriza tions of SnO2 nanopowder and spectroscopic (FTIR, FTRaman, UV–visible and NMR) analysis using HF and DFT calculations, Spectrochim. Acta, Part A, 2014, vol. 118, pp. 1135–1143. Hu, J., Bando, Y., Liu, Q., and Golberg, D., Laser ablation growth and optical properties of wide and long singlecrystal SnO2 ribbons, Adv. Funct. Mater., 2003, vol. 13, no. 6, pp. 493–496. Patil, K.C., Advanced ceramics: combustion synthesis and properties, Bull. Mater. Sci., 1993, vol. 16, no. 6, pp. 533–554. Barsan, N., Koziej, D., and Weimar, U., Metal oxide based gas sensor research: how to?, Sens. Actuators, B, 2007, vol. 121, no. 1, pp. 18–35.
37. Batzill, M. and Diebold, U., The surface and materials science of tin oxide, Prog. Surf. Sci., 2005, vol. 79, nos. 2–4, pp. 47–154. 38. Sangaletti, L., Depero, L.E., Allieri, B., et al., Oxida tion of Sn thin films to SnO2. MicroRaman mapping and Xray diffraction studies, J. Mater. Res., 1998, vol. 13, no. 9, pp. 2457–2460. 39. Geurts, J., Rau, S., Richter, W., and Schmitte, F.J., SnO films and their oxidation to SnO2: Raman scatter ing, IR reflectivity and Xray diffraction studies, Thin Solid Films, 1984, vol. 121, no. 3, pp. 217–225. 40. Gole, J.L., Iretskii, A.V., White, M.G., et al., Suggested oxidation state dependence for the activity of submi cron structures prepared from tin/tin oxide mixtures, Chem. Mater., 2004, vol. 16, no. 25, pp. 5473–5481. 41. Mcguire, K., Pan, Z.W., Wang, Z.L., et al., Raman studies of semiconducting oxide nanobelts, J. Nanosci. Nanotechnol., 2002, vol. 2, no. 5, pp. 499–502. 42. Lee, J.S., Sim, S.K., Min, B., et al., Structural and optoelectronic properties of SnO2 nanowires synthe sized from ballmilled SnO2 powders, J. Cryst. Growth, 2004, vol. 267, nos. 1–2, pp. 145–149. 43. Ghosh, S., Das, K., Chakrabarti, K., and De, S.K., Template free synthesis of SnO2 nanoflower arrays on Sn foil, CrystEngComm, 2012, vol. 14, no. 3, pp. 929–935. 44. Diéguez, A., RomanoRodríguez, A., Vilà, A., and Morante, J.R., The complete Raman spectrum of nanometric SnO2 particles, J. Appl. Phys., 2001, vol. 90, no. 3, pp. 1550–1557. 45. Baranauskas, V., Fontana, M., Guo, Z.J., et al., Field emission properties of nanocrystalline tin oxide films, Sens. Actuators, B, 2005, vol. 107, no. 1, pp. 474–478. 46. Kaur, J., Shah, J., Kotnala, R.K., and Verma, K.C., Raman spectra, photoluminescence and ferromag netism of pure, Co and Fe doped SnO2 nanoparticles, Ceram. Int., 2012, vol. 38, no. 7, pp. 5563–5570.
Translated by O. Tsarev
SPELL: 1. nanostructured, 2. oxygens INORGANIC MATERIALS
Vol. 50
No. 8
2014
