Journal of Alloys and Compounds 753 (2018) 167e175
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Structure and magnetic properties of synthesized fine cerium dioxide nanoparticles a, * a J. Lun cek a, O. Zivotský , P. Janos b, M. Dosek b, A. Chrobak c, M. Marysko d, J. Bursík e, skova e, f Y. Jira Department of Physics, V SB e Technical University of Ostrava, 17. listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic lova Výsina 7, 400 96 Ústí nad Labem, Czech Republic Faculty of the Environment, University of Jan Evangelista Purkyn e, Kra c Institute of Physics, University of Silesia in Katowice, Uniwersytecka 4, 40-007 Katowice, Poland d 10, 162 53 Prague 6, Czech Republic Institute of Physics, Academy of Sciences of the Czech Republic, Cukrovarnicka e zkova 22, 616 62 Brno, Czech Republic Institute of Physics of Materials, Academy of Sciences of the Czech Republic, Zi f zkova 22, 616 62 Brno, Czech Republic CEITEC IPM, Institute of Physics of Materials, Academy of Sciences of the Czech Republic, Zi a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 November 2017 Received in revised form 21 March 2018 Accepted 9 April 2018 Available online 22 April 2018
Cerium dioxide, which is diamagnetic in its bulk crystalline fluorite form, exhibits ferromagnetism when prepared as a nanopowder. This characteristic has recently been ascribed to localized electron spin moments formed at oxygen vacancies near the nanoparticle surfaces. In the present study, a number of structure-sensitive and magnetic methods were applied to CeO2 nanopowders prepared from the carbonate precursor by treatment at different calcination temperatures. The low-temperature magnetiza€ ssbauer spectroscopy revealed the presence of tion measurements subsequently performed via 57Fe Mo small amounts of iron impurities in the samples as well as in the chemicals used to prepare them. These experimental findings and our theoretical data analysis of iron impurities at the ppm level provide a plausible explanation for the ferromagnetic behaviour of cerium oxide. © 2018 Elsevier B.V. All rights reserved.
Keywords: CeO2 Microstructure €ssbauer spectroscopy Mo Magnetic properties Brillouin function
1. Introduction Cerium is a rare-earth element that can exist either as a free metal or in one of two oxide forms: cerous (cerium(III)) or ceric (cerium(IV)). Fig. 1 shows cerium dioxide, CeO2, in a fluorite-like structure in which each Ce site is surrounded by 8 O sites in a face-centred-cubic (fcc) arrangement and each O site has 4 neighbouring Ce sites. Typically, some Ce3þ ions are present in the structure of cerium (di)oxide; the number of these ions is related to the number of oxygen vacancies [1]. Cerium oxide is used in diverse industries as a glass-polishing agent [2], in protective coatings of metal alloys [3], in solar cells [4] and in gas sensors [5]. It plays a prominent role in heterogeneous catalysis (as a part of automobile exhaust systems [6]), and numerous studies have been devoted to its photocatalytic applications in environmental protection [7e9]. Recent studies have shown that some forms of cerium oxide can catalyse biologically
* Corresponding author. E-mail address:
[email protected] (O. Zivotský). https://doi.org/10.1016/j.jallcom.2018.04.115 0925-8388/© 2018 Elsevier B.V. All rights reserved.
relevant reactions in a manner similar to conventional enzymes. Hence, nanoparticulate cerium oxide (nanoceria) has been classified as a “nanozyme” and used, e.g., in the treatment of neurodegenerative diseases, for which its redox properties are exploited [10,11]. In addition, the phosphatase-mimetic ability of cerium oxide (its ability to break phosphoester bonds [12e14]) has been utilized to degrade dangerous organophosphate compounds, including some pesticides and chemical warfare agents [15,16]. Nanocrystalline metal oxides exhibit some unusual features not found in their bulk counterparts [17,18]. Most of these features are related to the dramatic changes in surface-to-volume ratio that occur when particle size is decreased to the nanometre level. Sizeinduced lattice expansions and relaxations [19,20], together with surface non-stoichiometry [21,22], generate active sites on the surface of cerium oxide, and these sites are responsible for its catalytic and enzyme-mimetic activities. Nanocrystalline cerium oxide also exhibits some purely physical phenomena that have attracted the attention of scientists, such as ferromagnetic behaviour [23,24], which has promising applications in spintronics [25,26]. The magnetic behaviour of transition-metalbased oxides (iron oxides, CrO2) originates from collective
168
cek et al. / Journal of Alloys and Compounds 753 (2018) 167e175 a J. Lun
changes in morphology, chemical composition and physical properties induced by calcination at various temperatures between 500 and 800 C. Detailed analysis of the obtained data should contribute to an understanding of the ferromagnetic behaviour of nanometre-sized cerium oxides. 2. Experimental 2.1. Materials and syntheses
Fig. 1. Cubic structure of cerium dioxide, CeO2, with labelling of atoms.
interactions of the magnetic moments of atoms or ions that constitute the material. The atoms or ions are arranged in a periodic crystalline lattice, and their moments interact through a molecular exchange field, resulting in long-range magnetic ordering. However, CeO2, in which cerium exists as Ce4þ (4f0 configuration), is a band insulator; its behaviour approaches diamagnetism. The hysteresis loop and the coercive field of approximately 8 kA/m with clear hysteresis dependence measured at 390 K in samples of Al2O3 nanoparticles smaller than 15 nm are surprisingly close to magnetic behaviour [23]. The origin of ferromagnetism in both of the aforementioned oxides has been ascribed to the presence of a large number of structural defects (oxygen vacancies) on the surfaces of the nanoparticles [24]. Although one can generalize that the magnetism is related to defects in the structure of the nanocrystals, its exact mechanism remains the subject of discussion; in particular, the role of dopants or impurities remains unclear [25,26]. As mentioned by Coey et al. [27], progress in this field is hampered by poor reproducibility of the published data, among other issues; this poor reproducibility may be related to the use of different cerium sources (their purity) and different procedures for cerium oxide preparation. Nanoparticulate cerium oxides can be prepared using various technologies that can be roughly divided into two categories. In the low-temperature synthetic routes, cerium oxide is created by ‘spontaneous’ conversion from cerium hydroxide in an aqueous medium at relatively low temperatures (500 C; such cerium oxides are typically prepared by thermal decomposition of suitable precursors (oxalates, carbonates) [2,16]. In the present work, a well-established precipitation/calcination procedure was used to prepare nanocrystalline cerium oxide. A relatively broad spectrum of experimental methods, some of which were applied at low temperatures, was used to examine the
A well-established precipitation/calcination synthetic route was used to prepare the cerium oxide samples [15,16]. Cerium(III) nitrate hexahydrate (p.a., >99%, Sigma-Aldrich) and ammonium bicarbonate (p.a., >99%, Sigma-Aldrich) were used as a cerium source and as a precipitant, respectively, in the cerium oxide synthesis. All solutions were prepared in deionized water obtained from a Demi Ultra 20 system (Goro, Prague, Czech Republic) in which reverse osmosis and mixed-bed ion-exchange were used for water purification. In a typical experimental arrangement, 500 ml of an aqueous solution of cerium nitrate (0.2 mol/l) was mixed with an excess of ammonium bicarbonate (0.5 mol/l) to obtain insoluble cerium carbonate; reaction with oxalic acid was used to check the completeness of the precipitation. The reaction mixture was agitated using a magnetic stirrer for one hour and was then left undisturbed overnight. The finely crystalline precipitate of cerium carbonate was separated by filtration, thoroughly washed with deionized water, dried at 110 C and stored in a closed PE vessel; this material is hereinafter called precursor E. Cerium oxides were prepared by annealing precursor E for 2 h at pre-determined calcination temperatures of 500, 600, 700, and 800 C in an open porcelain crucible in a muffle furnace. All the sample preparation steps were performed with extreme care to avoid iron contamination. The samples are denoted ET, where T is the calcination temperature. 2.2. Structure characterization Infrared (IR) spectra in the region 4000-400 cm1 were recorded on a Nicolet Impact 400D spectrometer equipped with accessories for diffuse reflectance measurements. Raman spectra were acquired on a DXR Raman microscope (Thermo Scientific). Thirtytwo two-second scans were accumulated with a laser (532 nm, 6 mW) under the 10 objective of an Olympus microscope. A Coulter SA3100 apparatus (Beckman, Krefeld, Germany) was used to determine the surface area of the particles and the porosity of the samples on the basis of physisorption via nitrogen adsorption/desorption. The Brunauer-Emmett-Teller and Barrett-JoynerHalenda methods were used to calculate the surface area and to determination of the pore size distribution, respectively. X-ray diffraction patterns were measured at room temperature (RT) using an X’PERT PRO diffractometer equipped with a Co Ka radiation source (l ¼ 0.17902 nm); the E500 sample was also measured at 4 K using a Cu Ka radiation source (l ¼ 0.1540598 nm). All measurements were performed in the 2q range from 25 to 135 in steps of 0.008 at a time per step of 3.5 s. The resultant powder patterns were evaluated by the Rietveld structure refinement method [30] in semiautomatic mode using the HighScore Plus program (Panalytical) and the ICSD database of inorganic and related structures [31]. In addition to the phase composition, the lattice parameters and the mean crystallite size, representing the size of coherently diffracting domains, were obtained from the pattern analysis. A TESCAN LYRA 3XMU FEG/SEM scanning electron microscope operated at an accelerating voltage of 20 kV and equipped with an XMax80 Oxford Instruments detector for energy-dispersive X-ray
cek et al. / Journal of Alloys and Compounds 753 (2018) 167e175 a J. Lun
analysis (EDX) was used to characterize the morphology and microstructure of the samples. Samples were prepared for SEM observation by spreading a small amount of powder onto a sticky conductive tape. 2.3. Magnetic measurements Room- and low-temperature hysteresis loops were obtained using a Quantum Design MPMS-XL superconducting quantum interference device magnetometer. Special care was devoted to the selection of the sample holder and to the sample handling procedure to prevent any contamination of the holder and/or sample. The CeO2 powders were loaded into plastic ampules for measurement, and their magnetic moment was subtracted from the hysteresis loop. The magnetic characteristics of coercive field and saturation magnetization were derived from the hysteresis loops at an accuracy of approximately ±1%. In addition to macroscopic magnetic measurements and tradi€ssbauer spectrometry was used to tional structural analysis, Mo detect and/or exclude prospective contamination of the samples with iron impurities. The measurements were performed at RT in standard transmission geometry using a57Co(Rh) source. Calibration of the velocity scale was performed with a-Fe, and the isomer €ssbauer spectrum of a-Fe. shifts are given with respect to the RT Mo All spectra were evaluated using the transmission integral €ssbauer spectra, approach in the program CONFIT [32]. In the Mo only double- and single-line components were detected. They are characterized by their isomer shift (d), quadrupole splitting (D), and their relative representations in the spectrum (A). 3. Results 3.1. Infrared and Raman spectroscopy Infrared spectra of the precursor E and the thermally treated samples are shown in Fig. 2a. The broadest band (at ~3400 cm1) observed in the spectra of all samples is assigned to the n(OH); the next bands visible in the spectra of the annealed samples are slightly above 1500 cm1 (d(OH)) and indicate adsorbed water and
169
OH groups. The shapes of the IR spectral curves and their band positions are in good agreement with the data reported by other authors, e.g., [33,34]. The intensities of bands at 3400 cm1 and 1500 cm1 decrease in the spectra of samples prepared at increasing temperatures, reflecting the decreasing abundance of free hydroxyl groups on the sample surface. Fig. 2b shows the Raman spectra of all the samples. The Raman spectra of the annealed samples arise from cerium dioxide nanoparticles, and the Raman peak values shift slightly from 461 to 463 cm1 as the calcination temperature is increased from 500 C to 800 C; simultaneously, the Raman line broadening (FWHM) decreases (from 23 to 16 to 11 to 10 cm1) due to increased particle size and crystal order. This Raman peak shift corresponds very well to the values reported by Meng et al. [35]. Also the intensity of the Raman band increased with increasing calcination temperature, which is consistent with recent findings of Babitha et al. [36] for cerium oxides prepared by calcination from carbonate precursors. We have not analysed this dependence in more details, as the theory behind this phenomenon (Raman signals of the nanosized materials) is rather complex and an interpretation difficult [37]. 3.2. Physisorption Physisorption (also called physical adsorption) was used to determine the surface area of the particles and the porosity of the samples. The obtained data are summarized in Table 1. After the first step of the calcination treatment, the total volume of pores of the calcined sample increased compared with that of precursor E; however, calcination at higher temperatures contributed to a decrease in the total volume of pores. The results in the table also document that only micropores (pores with widths not exceeding approximately 2 nm) are present. The volume of these pores in precursor E was very low, and it was below the limit of detection of the method in samples calcined at >500 C. 3.3. X-ray diffraction The diffraction patterns and the dependence of the analysed parameters on treatment temperature are shown in Fig. 3 and
Fig. 2. IR (a) and Raman (b) spectra of the precursor E and of the nanocrystalline CeO2 samples calcined at the indicated temperatures, ET.
cek et al. / Journal of Alloys and Compounds 753 (2018) 167e175 a J. Lun
170
Table 1 Porosity and surface area of particles in precursor E and in the nanocrystalline CeO2 samples calcined at the indicated temperatures, ET. Sample
Total pore volume (ml/g)
Micropore volume (ml/g)
Micropore surface area (m2/g)
Specific surface area (m2/g)
E E500 E600 E700 E800
0.0461 0.0667 0.0485 0.0314 0.0297
0.0008 0.0106 0.0000 0.0000 0.0000
1.0 24.3 0.0 0.0 0.0
13.2 65.5 25.9 1.5 6.0
Tables 2 and 3. The analysis was performed using data for CeO2 (ICSD 72155, the pattern shown in Fig. 3) [31]. The positions of the individual peaks correspond to fcc CeO2 and are only slightly shifted to higher 2q angles with increasing temperature, as is evident in Table 2. The pattern of the precursor E reflects peaks corresponding to cerium dioxide and a rather amorphous halo that is likely due to cerium carbonate and other compounds that could not be reliably identified entering the production process. The peaks of the annealed samples became sharper with increasing annealing temperature, indicating a fluorite-like crystalline structure with increasing tendency towards crystal order. No differences were observed in the XRD analysis of sample E500 conducted at 4 K. The recalculated data are depicted by a grey line in Fig. 3 for comparison with the measurement results obtained at RT using Co Ka radiation. The lattice parameter of the E sample (a z 0.5433 nm) decreases after annealing; however, it is still slightly higher than the value of 0.5411(1) nm observed for the bulk cerium dioxide [38]. This observation is consistent with the observations of other authors, see, e.g., [39]. The average crystallite size of the cerium oxides calculated from the X-ray diffraction patterns using the Scherrer equation [30] ranges from 9 nm to 70 nm. 3.4. Morphology and chemical analysis The morphology of the precursor E and of the samples treated at 500 C and 800 C, each for 2 h, are shown in Figs. 4 and 5, respectively. The EDX analysis did not reveal any important differences in the chemical compositions of the samples. No elements other than cerium and oxygen were detected. All the analysed samples were in the form of thin microplates approximately 1 mm thick, with dimensions ranging from 5 to 10 mm. Fine features with weak contrast (Fig. 5b) were observed only in sample E800 with the highest calcination temperature. Their size is of the order of the grain size measured by XRD. 3.5. Magnetic measurements
Fig. 3. Room-temperature XRD patterns of the precursor E and of the CeO2 samples calcined at the indicated temperatures, ET. The E500 sample was also measured at 4 K.
As previously stated, the magnetic measurements were performed very carefully to avoid contamination of the samples and the sample holders. Fig. 6 shows the hysteresis loops measured for the precursor E at 300 K and 2 K. The behaviour of the precursor at room temperature is paramagnetic (see Fig. 6, right); however, at low temperatures, saturation, which is also characteristic of lowtemperature paramagnetism, is observed. The coercive field is close to zero, approximately 2 kA/m, due to the small crystallite size of 9.6 nm, in agreement with the extraordinary D6 dependence of the coercive field in the region of small grain sizes [40]. The results obtained for the nanocrystalline CeO2 sample E500, which has a crystallite size of less than 9 nm, are presented in Figs. 7e9. The hysteresis loop measured at room temperature (Fig. 7) consists of three contributions: a diamagnetic, a paramagnetic and a ferromagnetic contribution.
cek et al. / Journal of Alloys and Compounds 753 (2018) 167e175 a J. Lun
171
Table 2 Peak positions of X-ray diffraction patterns obtained for precursor E and for the CeO2 samples calcined at the indicated temperatures, ET. hkl
E E500 E600 E700 E800
1
2
3
4
5
6
7
8
9
10
111
200
220
311
222
400
331
420
422
511
33.31 33.49 33.48 33.50 33.49
38.62 38.72 38.82 38.83 38.83
55.67 55.85 55.94 55.96 55.96
66.36 66.58 66.68 66.70 66.69
69.71 69.95 70.05 70.07 70.07
82.54 82.86 82.96 82.98 82.97
91.88 92.26 92.36 92.38 92.38
94.99 95.34 95.50 95.52 95.51
107.68 108.19 108.30 108.32 108.32
117.79 118.41 118.52 118.55 118.55
Table 3 Lattice parameter a and crystallite size d of the precursor E and of the CeO2 samples calcined at the indicated temperatures, ET. Sample
a (nm)
d (nm)
E E500 E600 E700 E800
0.5433(1) 0.5415(1) 0.5414(1) 0.5414(1) 0.5414(1)
9.60(4) 8.87(5) 13.00(0) 29.30(0) 70.35(0)
The diamagnetic contribution is responsible for the decrease in magnetization at higher magnetic fields. It is characterized by the (mass) susceptibility cD ¼ 6.98 1010 m3/kg. Nearly the same value of cD was also obtained from the low-temperature (2 K) hysteresis loop in Fig. 9 (top), confirming the weak temperature dependence of this contribution. By contrast, the paramagnetic contribution depends strongly on temperature and follows the Curie-Weiss law. The low-temperature dependence of magnetization depicted in Fig. 8 was characterized with the sample in an external field of 800 kA/m to saturate the ferromagnetic component. The results were used to estimate the Curie constant
c ¼ 4.11 109 (m3 K)/kg and the Weiss temperature (3 K). The paramagnetic susceptibility cP is clearly observed to be practically zero at room temperature, whereas it markedly increases at temperatures less than 10 K (Fig. 8). The hysteresis loop for the E500 sample measured at 2 K (Fig. 9, top) was subjected to detailed analysis. The diamagnetic correction due to the known susceptibility cD ¼ 6.98 1010 m3/kg resulted in the curve shown in Fig. 9 (middle). Because the paramagnetic response is important at low temperatures (Fig. 8), it must also be taken into account and separated from the low-temperature hysteresis loop. The corrected dependence of the magnetization M on the magnetic field H was further approximated by the Brillouin function BJ, which describes the dependence of the magnetization on the applied magnetic field and on the total angular momentum quantum number J of the microscopic magnetic moment of the material. The magnetization M is given by
M ¼ NgmB JBJ ðxÞ;
(1)
where N is the number of atoms per unit volume, g is the Lande's splitting factor, mB is the Bohr magneton, and x ¼ gmBH/kBT is the ratio of the Zeeman energy of the magnetic moment in the external
Fig. 4. The morphology of precursor E (a) and of a CeO2 sample calcined at 500 C (b).
Fig. 5. The morphology of the CeO2 sample calcined at 800 C (a) and a detail showing the presence of very small grains (b).
172
cek et al. / Journal of Alloys and Compounds 753 (2018) 167e175 a J. Lun
Fig. 6. Hysteresis loops measured for the precursor E at the indicated temperatures.
Fig. 7. A portion of the hysteresis loop of the E500 sample measured at room temperature.
magnetic field to the thermal energy, where kB is the Boltzmann constant. Cerium occurs as Ce4þ and Ce3þ ions. Whereas Ce4þ ions are diamagnetic (4f0), Ce3þ ions, because of their magnetic moment, can also be present in CeO2 and can influence its ferromagnetic behaviour, as suggested by Li et al. [41] and taken into account in
Fig. 8. Temperature dependence of magnetization of the E500 sample measured in a magnetic field of 800 kA/m.
the work of Chen et al. [42]. The electron configuration of the Ce3þ ion is 4f1, the total electronic angular momentum is J ¼ 5/2, and the Lande's g-factor is g ¼ 6/7. Using these parameters, we calculated the Brillouin function, which is depicted by the dotted line in Fig. 9 (bottom). From this dependence, the saturation paramagnetic moment Mspm is 0.0035 (A m2)/kg, which corresponds to 0.0002917 mB per formula unit for Ce3þ (0.0035/12). If the theoretical saturation magnetic moment of Ce3þ ions is 2.54 mB, then its content in the measured E500 sample is on the order of 104 and its contribution appears to be insignificant. In the next considerations, possible contamination of the samples with iron was taken into account. This possibility was suggested by a similar problem involving graphite contamination presented by Esquinazi et al. [43]. They measured the dependence of the magnetization at a magnetic field of 160 kA/m on the iron concentration in graphite. According to this dependence, our value of the saturation magnetization of 0.0014 (A m2)/kg due to the ferromagnetic contribution at low temperature (Fig. 9, bottom) should correspond to approximately 6 ppm Fe. The potential presence of such an amount of iron as an impurity prompted our €ssbauer spectrometry, the results of which are discussed use of Mo in the next section. Notably, all three magnetic contributions analysed for the E500 sample were detected also for CeO2 samples calcined at higher temperatures. Whereas the diamagnetic and paramagnetic parameters of other samples fluctuated around the values presented for the E500 sample, the ferromagnetic contribution decreased
Fig. 9. CeO2 sample annealed for 2 h at 500 C. Dependence of magnetization on the magnetic field measured at 2 K (top), recalculated (middle, bottom; see text) and supported by the Brillouin function (bottom, dotted line).
cek et al. / Journal of Alloys and Compounds 753 (2018) 167e175 a J. Lun
with increasing calcination temperature due to increasing particle size and possible oxidation. €ssbauer spectrometry 3.6. Mo Because the possible presence of a small amount of Fe-based impurity was considered on the basis of the analysis of magnetic € ssbauer spectrometry was used in this measurements, 57Fe Mo study as well. This technique can provide direct identification of all structurally different regions of resonant Fe atoms located inside grains, at their surfaces and/or at phase boundaries. The spectra of the precursor E and the E500 and E800 samples (Fig. 10) clearly confirm the presence of Fe atoms. A double-line component with the isomer shift d ¼ 0.190(19) mm/s and quadrupole splitting D ¼ 0.330(28) mm/s detected in the E sample is also present in the calcined samples. The slightly different hyperfine parameters and the appearance of an additional single-line component of d ¼ 0.302(34) mm/s, A ¼ 27% (E500) and d ¼ 0.246(20) mm/s,
173
A ¼ 25.5% (E800) indicate changes in atomic ordering in the regions surrounding resonating Fe atoms. At present, we can only speculate that some of the iron-resonant atoms are located in the cerium dioxide, making it visible by the €ssbauer effect, and/or that the effect arises from a contribution Mo of intergrain regions containing small clusters of the undesirable iron and/or iron oxide. Very similar values of d and D for the doubleline component were obtained in our previous investigations of cerium dioxide/iron oxide-reactive sorbents [44]. Additional €ssbauer measurements at low temperatures and in external Mo magnetic fields would likely be helpful in clarifying the nature of the iron-type impurities. Such experiments will be considered in future studies. Nevertheless, the finding that unexpected iron impurities are present in CeO2 samples has resulted in the initiation of €ssbauer measurements of selected chemicals that are Mo commonly used as starting materials for preparing cerium dioxide. The obtained results [45] clearly show that traces of iron may be present even in high-purity chemicals and that they likely influence the physical properties of the final products. 4. Discussion
€ssbauer spectra of the precursor E and of samples calcined at 500 C/2 h Fig. 10. Mo (E500) and 800 C/2 h (E800).
Our detailed analysis of the temperature dependence of magnetization on an applied magnetic field of a sample calcined at 500 C for 2 h indicates the presence of iron in the sample at a concentration of approximately 6 ppm. Subtraction of the effects of the parasitic iron impurity implies that the magnetic moment of the CeO2 nanoparticles consists primarily of a diamagnetic contribution of CeO2 (Ce4þ) because the paramagnetic contribution of the Ce3þ ion is on the order of 104. The assumed presence of a parasitic iron impurity in the E500 sample, which was subsequently confirmed by highly sensitive €ssbauer spectrometric analysis of precursor E and sample E800, Mo led us to perform measurements of selected input materials; the results confirmed the presence of iron impurities in these materials as well. The contamination of chemicals is typically described in certificates provided by the supplier; however, in some cases, such documentation may be missing. The present findings emphasize the influence of the purity of the chemicals used in the preparation of nanosized CeO2 and likely of the chemicals used to prepare other diamagnetic oxides, e.g., Al2O3. The detected iron impurities may be one of the factors contributing to the ferromagnetism of CeO2 and may affect its magnetic behaviour. Iron impurities present at the ppm level correspond to a magnetization equivalent to thousandths of (A$m2)/kg [43]. As indicated in the discussion below, such small values of magnetization have been reported by many authors who characterized CeO2 nanoparticles prepared by various methods. Sundaresan et al. [23,24] declared ferromagnetism (FM) to be a universal feature of inorganic nanoparticles without offering any information concerning the quality of chemicals or their processing. They reported that the saturation magnetization of CeO2 nanoparticles prepared by chemical methods was 0.001 (A m2)/kg for as-prepared 7-nm large particles and close to 0.002 (A m2)/kg for 15-nm large particles heated to 500 C. Liu et al. [46] studied size-dependent FM in pure CeO2 powders synthesized by a precipitation route. They observed FM only in samples with particle sizes of less than 20 nm; these samples displayed a saturation magnetization of 0.08 (A m2)/kg. The authors declared that none of the chemicals used in the preparation of their samples showed magnetic signals in the range from 10 to 300 K; however, the origin of FM was not reliably explained. Chen et al. [47] prepared CeO2 nanoparticles by a thermal decomposition procedure and reported magnetic behaviour in their samples; the obtained saturation magnetization of 0.12 (A m2)/kg is higher than the values reported
174
cek et al. / Journal of Alloys and Compounds 753 (2018) 167e175 a J. Lun
by others. No information concerning the purity of the input compounds was presented. Nevertheless, in other work [42], the authors proposed an explanation related to Ce3þ ions for the presence of FM in this compound. Ge et al. used commercial CeO2 nanoparticles (Aldrich Company, purity of 99.9%) and CeO2 nanocubes prepared by a chemical method for which no data on the input chemicals were presented [48]. These authors presented magnetic characteristics of 6.2 kA/m for a coercive field and 0.0007 (A m2)/kg for the saturation magnetization of commercial cerium oxide and slightly different values of 5.4 kA/m and 0.0057 (A m2)/ kg, respectively, for the self-produced samples. By contrast, Meng et al. [49] reported a value of 16 kA/m for the coercive field, which can be explained by structural sensitivity. Phokha et al. [50] observed RT-FM in monodisperse CeO2 nanospheres synthesized by hydrothermal treatment of cerium nitrate hexahydrate, Ce(NO3)3$6H2O (purity of 99.99%), cerium acetate hydrate (99.9%), cerium chloride heptahydrate (99.9%), cerium sulfate octahydrate (99.999%), and polyvinylpyrrolidone. The authors declared that Xray absorption near edge structure (XANES) results revealed the fraction of the Ce3þ ions contributing to a weak RT-FM with saturation magnetization ranging between 0.0026 and 0.016 (A m2)/kg. In another paper, Phokha et al. [51] also observed RT-FM in pure CeO2 nanoparticles prepared by simple direct thermal decomposition of acetate hydrate Ce(C2H3O2)3$H2O (purity of 99.99%, SigmaAldrich) in air at temperatures ranging from 300 C to 800 C for 3 h. The samples showed RT-FM after thermal decomposition above 500 C. The measured magnetic saturation values were between 0.007 and 0.023 (A m2)/kg. In almost all publications in which it is mentioned, ferromagnetic behaviour is ascribed to exchange interactions between localized electron spin moments resulting from oxygen vacancies at the nanoparticle surface. Nevertheless, the reasons for the very high dispersion of saturation magnetization values of three orders of magnitude (from 0.0007 to 0.12 (A m2)/kg) are at present unclear and are difficult to understand from the physical viewpoint. The origin of FM in pure cerium dioxide is also unclear and has not been confirmed by theoretical calculations [52]. 5. Conclusions This paper presents a detailed structural and physical characterization of cerium dioxide prepared by a well-known precipitation/calcination procedure. The chemical composition, morphology, and magnetic properties of the material were determined using a wide range of experimental methods, and almost all aspects of the sample preparation and the composition of input chemicals with respect to their purity were taken into account. The Raman spectroscopy and X-ray diffraction results confirmed the presence of a fluorite-like crystalline structure in the CeO2 precursor (in amounts of approximately 15%) and in the calcined samples. Increasing the calcination temperature was shown to contribute to a slight shift of the Raman peak corresponding to this phase in the nanoparticle form (from 461 to 463 cm1); the decrease in the lattice parameter was slightly higher than that of bulk CeO2, and an increase in crystallite size was accompanied by an increase in atomic ordering. The values of saturation magnetization obtained at room temperature and at a low temperature of 2 K fell well within the range of values reported by other authors. Detailed analysis of the magnetic data was first performed assuming that Ce3þ ions yield their own magnetic moment, unlike diamagnetic Ce4þ ions. Nevertheless, the influence of Ce3þ ions on the ferromagnetism of the prepared CeO2 samples was found to be insignificant. In contrast, the presence of iron impurities and their effect on the ferromagnetic behaviour of cerium dioxide obtained from the magnetic
€ssbauer spectrometry measurements and well confirmed by Mo was unambiguous. This result is the main finding of the present € ssbauer spectrometry, morestudy. The results obtained using Mo over, show that iron impurities are also detectable in the starting chemicals from which the nanosized CeO2 powders were prepared. The existence of iron impurities in oxides such as CeO2 and their possible effects on the magnetic properties of nanoparticles are not well known at present. This fact, together with other features of the particles, e.g., oxygen vacancies, leaves the subject open for further investigation of ferromagnetism. Acknowledgements The authors thank I. Turek (IPM) for his comments and helpful suggestions and the XRD laboratory of Charles University in Prague for performing the low-temperature measurements. This work was supported by projects SP2017/42, CEITEC 2020 - National Sustainability Programme II (No. LQ1601), and LM2015073 ‘the Research Infrastructure NanoEnviCz’, all funded by the Ministry of Education, Youth and Sports of the Czech Republic. Partial support from the National Science Centre of Poland via grant 2015/19/B/ST8/02636 and from ERDF/ESF (No. CZ.02.1.01/0.0/0.0/17_048/0007399) is acknowledged. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.jallcom.2018.04.115 References [1] C.T. Campbell, C.H.F. Peden, Chemistry. Oxygen vacancies and catalysis on ceria surfaces, Science 309 (2005) 713e714, https://doi.org/10.1126/ science.1113955. k, Preparation of ceria-based polishing powders from car[2] P. Janos, M. Petra bonates, J. Mater. Sci. 26 (1991) 4062e4066, https://doi.org/10.1007/ BF02402947. [3] H. Hassannejad, M. Moghaddasi, E. Saebnoori, A.R. Baboukani, Microstructure, deposition mechanism and corrosion behavior of nanostructured cerium oxide conversion coating modified with chitosan on AA2024 aluminum alloy, J. Alloys Compd. 725 (2017) 968e975, https://doi.org/10.1016/ J.JALLCOM.2017.07.253. [4] X. Wang, L.-L. Deng, L.-Y. Wang, S.-M. Dai, Z. Xing, X.-X. Zhan, X.-Z. Lu, S.Y. Xie, R.-B. Huang, L.-S. Zheng, Cerium oxide standing out as an electron transport layer for efficient and stable perovskite solar cells processed at low temperature, J. Mater. Chem. A 5 (2017) 1706e1712, https://doi.org/10.1039/ C6TA07541J. [5] D. Barreca, A. Gasparotto, C. Maccato, C. Maragno, E. Tondello, E. Comini, G. Sberveglieri, Columnar CeO2 nanostructures for sensor application, Nanotechnology 18 (2007), 125502, https://doi.org/10.1088/0957-4484/18/12/ 125502. [6] A. Trovarelli, Catalytic properties of ceria and CeO2 -Containing materials, Catal. Rev. 38 (1996) 439e520, https://doi.org/10.1080/01614949608006464. [7] S.K. Muduli, S. Wang, S. Chen, C.F. Ng, C.H.A. Huan, T.C. Sum, H.S. Soo, Mesoporous cerium oxide nanospheres for the visible-light driven photocatalytic degradation of dyes, Beilstein J. Nanotechnol. 5 (2014) 517e523, https:// doi.org/10.3762/bjnano.5.60. [8] C. Maria Magdalane, K. Kaviyarasu, J. Judith Vijaya, B. Siddhardha, B. Jeyaraj, Facile synthesis of heterostructured cerium oxide/yttrium oxide nanocomposite in UV light induced photocatalytic degradation and catalytic reduction: synergistic effect of antimicrobial studies, J. Photochem. Photobiol. B Biol. 173 (2017) 23e34, https://doi.org/10.1016/J.JPHOTOBIOL.2017.05.024. [9] S. Gnanam, V. Rajendran, Facile sol-gel preparation of Cd-doped cerium oxide (CeO2) nanoparticles and their photocatalytic activities, J. Alloys Compd. 735 (2018) 1854e1862, https://doi.org/10.1016/j.jallcom.2017.11.330. [10] H. Wei, E.I. Wang, Nanomaterials with enzyme-like characteristics (nanozymes): next-generation artificial enzymes, Chem. Soc. Rev. 42 (2013) 6060e6093, https://doi.org/10.1039/c3cs35486e. [11] S. Singh, Cerium oxide based nanozymes: redox phenomenon at biointerfaces, Biointerphases 11 (2016) 04B202, https://doi.org/10.1116/1.4966535. [12] M.H. Kuchma, C.B. Komanski, J. Colon, A. Teblum, A.E. Masunov, B. Alvarado, S. Babu, S. Seal, J. Summy, C.H. Baker, Phosphate ester hydrolysis of biologically relevant molecules by cerium oxide nanoparticles, Nanomedicine 6 (2010) 738e744, https://doi.org/10.1016/j.nano.2010.05.004. szov [13] P. Janos, I. Lova a, J. Pfeifer, J. Ederer, M. Dosek, T. Lou cka, J. Henych,
cek et al. / Journal of Alloys and Compounds 753 (2018) 167e175 a J. Lun
[14]
[15]
[16]
[17]
[18]
[19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28] [29]
[30] [31] [32] [33]
, D. Milde, T. Opletal, Accelerated dephosphorylation of adenosine Z. Kolska phosphates and related compounds in the presence of nanocrystalline cerium oxide, Environ. Sci.: Nano 3 (2016) 847e856, https://doi.org/10.1039/ C6EN00086J. , D. Milde, T. Opletal, P. Janos, J. Henych, J. Pfeifer, N. Zemanov a, V. Pilarova J. Tolasz, M. Malý, V. Stengl, Nanocrystalline cerium oxide prepared from carbonate precursor and its ability to breakdown biologically relevant organophosphates, Environ. Sci. Nano 4 (2017) 1283e1293. P. Janos, P. Kuran, M. Kormunda, V. Stengl, T.M. Grygar, M. Dosek, M. Stastny, J. Ederer, V. Pilarova, L. Vrtoch, Cerium dioxide as a new reactive sorbent for fast degradation of parathion methyl and some other organophosphates, J. Rare Earths 32 (2014) 360e370, https://doi.org/10.1016/S1002-0721(14) 60079-X. , L. Vrtoch, M. Kormunda, K. Mazanec, P. Janos, J. Henych, O. Pelant, V. Pilarova V. Stengl, Cerium oxide for the destruction of chemical warfare agents: a comparison of synthetic routes, J. Hazard Mater. 304 (2016) 259e268, https:// doi.org/10.1016/j.jhazmat.2015.10.069. E. Lucas, S. Decker, A. Khaleel, A. Seitz, S. Fultz, A. Ponce, W. Li, C. Carnes, K.J. Klabunde, Nanocrystalline metal oxides as unique chemical reagents/ sorbents, Chemistry 7 (2001) 2505e2510. S. Gangopadhyay, D.D. Frolov, A.E. Masunov, S. Seal, Structure and properties of cerium oxides in bulk and nanoparticulate forms, J. Alloys Compd. 584 (2014) 199e208, https://doi.org/10.1016/j.jallcom.2013.09.013. S. Tsunekawa, K. Ishikawa, Z.-Q. Li, Y. Kawazoe, A. Kasuya, Origin of anomalous lattice expansion in oxide nanoparticles, Phys. Rev. Lett. 85 (2000) 3440e3443, https://doi.org/10.1103/PhysRevLett.85.3440. X.-D. Zhou, W. Huebner, Size-induced lattice relaxation in CeO2 nanoparticles, Appl. Phys. Lett. 79 (2001) 3512e3514, https://doi.org/10.1063/1.1419235. Y.-M. Chiang, E.B. Lavik, I. Kosacki, H.L. Tuller, J.Y. Ying, Defect and transport properties of nanocrystalline CeO2x, Appl. Phys. Lett. 69 (1996) 185e187, https://doi.org/10.1063/1.117366. E.B. Lavik, I. Kosacki, H.L. Tuller, Y.-M. Chiang, J.Y. Ying, Nonstoichiometry and electrical conductivity of nanocrystalline CeO2, J. Electroceram. 1 (1997) 7e14, https://doi.org/10.1023/A:1009934829870. A. Sundaresan, R. Bhargavi, N. Rangarajan, U. Siddesh, C.N.R. Rao, Ferromagnetism as a universal feature of nanoparticles of the otherwise nonmagnetic oxides, Phys. Rev. B 74 (2006), 161306, https://doi.org/10.1103/ PhysRevB.74.161306. A. Sundaresan, C.N.R. Rao, Ferromagnetism as a universal feature of inorganic nanoparticles, Nano Today 4 (2009) 96e106, https://doi.org/10.1016/ J.NANTOD.2008.10.002. M.C. Dimri, H. Khanduri, H. Kooskora, J. Subbi, I. Heinmaa, A. Mere, J. Krustok, R. Stern, Ferromagnetism in rare earth doped cerium oxide bulk samples, Phys. Status Solidi A 209 (2012) 353e358, https://doi.org/10.1002/ pssa.201127403. H. Shi, T. Hussain, R. Ahuja, T.W. Kang, W. Luo, Role of vacancies, light elements and rare-earth metals doping in CeO2, Sci. Rep. 6 (2016) 31345, https:// doi.org/10.1038/srep31345. M. Coey, K. Ackland, M. Venkatesan, S. Sen, Collective magnetic response of CeO2 nanoparticles, Nat. Phys. 12 (2016) 694e699, https://doi.org/10.1038/ nphys3676. L.M. Dennis, W.H. Magee, Contributions to the chemistry of cerium, J. Am. Chem. Soc. 16 (1894) 649e664, https://doi.org/10.1021/ja02108a001. T. Naganuma, Shape design of cerium oxide nanoparticles for enhancement of enzyme mimetic activity in therapeutic applications, Nano Res. 10 (2017) 199e217, https://doi.org/10.1007/s12274-016-1278-4. R.A. Young (Ed.), The Rietveld Method, International Union of Crystallography, Oxford University Press, Oxford, 1995. ICSD Database, Version 1.9.4, Volume 1, NIST/FIZ, 2014. a k, Y. Jir , CONFIT: Mo € ssbauer spectra fitting program, Surf. Interface T. Z askova Anal. 38 (2006) 710e714. F. Caputo, M. Mameli, A. Sienkiewicz, S. Licoccia, F. Stellacci, L. Ghibelli, E. Traversa, A novel synthetic approach of cerium oxide nanoparticles with improved biomedical activity, Sci. Rep. 7 (2017) 4636, https://doi.org/ 10.1038/s41598-017-04098-6.
175
[34] S. Chaudhary, P. Sharma, R. Kumar, S.K. Mehta, Nanoscale surface designing of cerium oxide nanoparticles for controlling growth, stability, optical and thermal properties, Ceram. Int. 41 (2015) 10995e11003, https://doi.org/ 10.1016/j.ceramint.2015.05.044. [35] F. Meng, L. Wang, J. Cui, A. Li, H. Tang, Experimental and theoretical investigations of the origin of magnetism in undoped CeO2, Sci. Adv. Mater. 7 (2015) 663e669, https://doi.org/10.1166/sam.2015.2149. [36] K.K. Babitha, A. Sreedevi, K.P. Priyanka, B. Sabu, T. Varghese, Structural characterization and optical studies of CeO2 nanoparticles synthesized by chemical precipitation, Indian J. Pure Appl. Phys. 53 (2015) 596e603. [37] M.V. Pellow-Jarman, P.J. Hendra, R.J. Lehnert, The dependence of Raman signal intensity on particle size for crystal powders, Vib. Spectrosc. 12 (1996) 257e261, https://doi.org/10.1016/0924-2031(96)00023-9. [38] K.A. Gschneidner, L. Eyring, Handbook on the Physics and Chemistry of Rare Earths, North-Holland Publishing Co., Elsevier North Holland, Amsterdam, 1979. [39] R.K. Hailstone, A.G. DiFrancesco, J.G. Leong, T.D. Allston, K.J. Reed, A study of lattice expansion in CeO2 nanoparticles by transmission electron microscopy, J. Phys. Chem. C 113 (2009) 15155e15159, https://doi.org/10.1021/ jp903468m. [40] G. Herzer, in: K.H.J. Buschow (Ed.), Handbook of Magnetic Materials, Elsevier Science, Netherlands, 1997. [41] M. Li, S. Ge, W. Qiao, L. Zhang, Y. Zuo, S. Yan, Relationship between the surface chemical states and magnetic properties of CeO2 nanoparticles, Appl. Phys. Lett. 94 (2009), 152511, https://doi.org/10.1063/1.3079330). [42] S. Chen, C. Tsai, M. Huang, D. Yan, T. Huang, A. Gloter, C. Chen, H. Lin, C. Chen, C. Dong, Concentration dependence of oxygen vacancy on the magnetism of CeO2 nanoparticles, J. Phys. Chem. C 116 (2012) 8707e8713, https://doi.org/ 10.1021/jp2065634. € hne, C. Semmelhack, Y. Kopelevich, D. Spemann, [43] P. Esquinazi, A. Setzer, R. Ho €sche, Ferromagnetism in oriented graphite samT. Butz, B. Kohlstrunk, M. Lo ples, Phys. Rev. B 66 (2002), 024429, https://doi.org/10.1103/ PhysRevB.66.024429. a skov [44] J. Lun cek, O. Zivotský, Y. Jira a, J. Bursík, P. Janos, Thermally stimulated iron oxide transformations and magnetic behaviour of cerium dioxide/iron oxide reactive sorbents, Mater. Char. 120 (2016) 295e303, https://doi.org/ 10.1016/j.matchar.2016.09.009. €ssbauer Measurements of Iron Impurities in [45] See Supplemental Material for Mo Input Chemicals. [46] Y. Liu, Z. Lockman, A. Aziz, J. MacManus-Driscoll, Size dependent ferromagnetism in cerium oxide (CeO2) nanostructures independent of oxygen vacancies, J. Phys. Condens. Matter 20 (2008), 165201. [47] S.Y. Chen, Y.H. Lu, T.W. Huang, D. Ch. Yan, Ch. Dong, Oxygen vacancy dependent magnetism of CeO2 nanoparticles prepared by thermal decomposition method, J. Phys. Chem. C 114 (2010) 19576e19581, https://doi.org/ 10.1021/jp1045172. [48] M.Y. Ge, H. Wang, E.Z. Liu, J.F. Liu, J.Z. Jiang, Y.K. Li, Z.A. Xu, H.Y. Li, On the origin of ferromagnetism in CeO2 nanocubes, Appl. Phys. Lett. 93 (2008), 062505, https://doi.org/10.1063/1.2972118. [49] F. Meng, C. Zhang, Z. Fan, J. Gong, A. Li, Z. Ding, H. Tang, M. Zhang, G. Wu, Hydrothermal synthesis of hexagonal CeO2 nanosheets and their room temperature ferromagnetism, J. Alloys Compd. 647 (2015) 1013e1021, https:// doi.org/10.1016/j.jallcom.2015.06.186. [50] S. Phokha, S. Pinitsoontorn, P. Chirawatkul, Y. Poo-Arporn, S. Maensiri, Synthesis, characterization, and magnetic properties of monodisperse CeO2 nanospheres prepared by PVP-assisted hydrothermal method, Nanoscale Res. Lett. 7 (2012) 425, https://doi.org/10.1186/1556-276X-7-425. [51] S. Phokha, E. Swatsitang, S. Maensiri, Room-temperature ferromagnetism in pure CeO2 nanoparticles prepared by a simple direct thermal decomposition, Electron. Mater. Lett. 11 (2015) 1012e1020, https://doi.org/10.1007/s13391015-4164-4. [52] P.R.L. Keating, D.O. Scanlon, G.W. Watson, Intrinsic ferromagnetism in CeO2: dispelling the myth of vacancy site localization mediated superexchange, J. Phys. Condens. Matter 21 (2009), 405502, https://doi.org/10.1088/09538984/21/40/405502.
