CeO2 nanoparticles growth by chemical bath and ... - ScienceDirect.com

2 downloads 0 Views 2MB Size Report
Aug 28, 2017 - Contents lists available at ScienceDirect .... The FT-IR spectra were recorded using a Perkin Elmer spectrophotometer in the ..... [21] C. Ho, J.C. Yu, T. Kwong, A.C. Mak, S. Lai, Morphology-controllable synthesis of mesoporous CeO2 ... Chen, C.H. Tsai, M.Z. Huang, D.C. Yan, T.W. Huang, A. Gloter, C.-L.
Optik 148 (2017) 142–150

Contents lists available at ScienceDirect

Optik journal homepage: www.elsevier.de/ijleo

Full length article

CeO2 nanoparticles growth by chemical bath and its thermal annealing treatment in air atmosphere O. Portillo Moreno a,∗ , R. Gutiérrez Pérez a , R. Palomino Merino b , M. Chávez Portillo a , G. Hernandez Tellez a , E. Rubio Rosas c , M. Zamora Tototzintle a a Materials Science Laboratory, Facultad de Ciencias Químicas, Universidad Autónoma de Puebla. P.O. Box 1067, Puebla, Pue, 72001, Mexico b Facultad de Ciencias Físico Matemáticas, Posgrado en Física Aplicada, Universidad Autónoma de Puebla, P.O. Box 1067, CP 7200, Mexico c Centro Universitario de Vinculación y Transferencia de Tecnología, Universidad Autónoma de Puebla, Ciudad Universitaria. P.O. Box 1067, CP 72001, Mexico

a r t i c l e

i n f o

Article history: Received 11 May 2017 Accepted 28 August 2017 Keywords: Growth kinetics Nanocrystals Quantum size effect Band gap energy

a b s t r a c t A novel synthesis of CeO2 powder by using chemical bath (CB) as green method is presented. The process was completed by thermal annealing (TA) at 1000 ◦ C during 2 h. CeO2 growth was confirmed by: X-ray diffraction (XRD), UV–vis spectra, Energy dispersive Xray spectroscopy (EDS), Fourier transform infrared spectroscopy (FTIR) and Raman spectra. By FTIR, bands associated with CO32− and OH− ions were identified, disappearing with the TA process. The X-ray diffraction (XRD) patterns of the as grown and CeO2 TA samples showed the reflection characteristic of CeO2 structure with a cubic symmetric pattern. By XRD diffractograms the grain size calculated were ∼2.7 nm and ∼28.5 nm for as-grown and CeO2 TA samples, respectively. The grain size of ∼2.7 nm is the smallest among the reported values for CeO2 . Absorbance of as-grown spectra showed three bands located at ∼382, ∼357 and ∼271 nm while the CeO2 TA sample displayed only two bands at ∼271 and ∼320 nm. Direct optical energy gap were Eg = 4.4 eV and Eg = 4.3 eV for as-grown and CeO2 TA samples, respectively, showing quantum size effect of CeO2 nanoparticles. Raman spectra indicated that the as-grown and CeO2 samples have the fluorite structure. © 2017 Elsevier GmbH. All rights reserved.

1. Introduction Cerium oxide (CeO2 ) presents a fcc crystal fluorite structure and is a compound that has been widely investigated for their chemical and physical properties of interest for technological applications and scientific research, for example as an electrolyte for solid oxygen fuel cells [1], as reactive oxygen species to protect human breast-line cells from radiation damage during cancer treatment and for mimicking superoxide dismutase activity [2], etc. On the other hand, for nanoparticles, the oxidation state observed is influenced by growth conditions and environment associated with key parameters in the synthesis of these nanoparticles and Cerium is a rare earth element that easily changes its Ce3+ → Ce4+ oxidation state [3]. These facts combine to influence physical properties of CeO2 nanoparticles of interest due to its morphological, structural and optical properties and the difference between CeO2 nanoparticles and micron-size particles [4]. A reported blue-shift in

∗ Corresponding author. E-mail address: [email protected] (O.P. Moreno). http://dx.doi.org/10.1016/j.ijleo.2017.08.133 0030-4026/© 2017 Elsevier GmbH. All rights reserved.

O.P. Moreno et al. / Optik 148 (2017) 142–150

143

optical absorption with decreasing grain size (GS), consistent with the exciton confinement (quantum size effect) that leads to blue-shift in optical spectra widely observed in semiconductors, and the CeO2 nanoparticles has presented this behavior. The literature on such quantum confinement effects in CeO2 nanoparticles has been inconsistent and has stimulated various debates on this line of research [5]. This effect is of interest in association with the Eg shifts of greater wavelengths caused by the decrease of GS, as reported [6]. There are different techniques for obtaining CeO2 nanoparticles such as the microemulsion method [7], by microwaves [8], spray pyrolysis [9], Polyvinyl pyrrolidone (PVP) −assisted hydrothermal method [10], etc. Such techniques require long reaction times and consumption of energy, and are not environmentally friendly because of multistep syntheses or aging processes. In addition, when nanocrystals are synthesized, the cost for manufacture will increase greatly. Hence, simple and cost-effective routes to synthesize CeO2 nanoparticles by utilization of cheap, nontoxic, and environmentally benign precursors are still key issues. The objective of this report was to develop an economically feasible precipitation method suitable for a large-scale production of CeO2 powders. To the best of our knowledge, this is the first report of CeO2 growth by chemical bath (CB) at room temperature to prepare CeO2 nanoparticles and its post thermal annealing (TA). The conditions to deposit and grow CeO2 powders by CB technique were modified and the proposed mechanism for the formation of CeO2 nanostructures was studied. Morphology of the powder was investigated by scanning electron microscopy (SEM) using a Jeol JSM-64-60 LV operating at 10 kV. Energy Dispersive X-ray spectroscopy (EDX) was performed to determine elemental composition of the samples. X-ray Diffraction (XRD) patterns were registered in a Siemens D500 diffractometer, using the Cu K␣ line, to carry out crystalline structure characterization. Morphological images were obtained by scanning electronic microscopy (SEM) utilizing a Voyager II X-ray quantitative microanalysis in an 1100/1110 SEM system from Noran Instruments. The FT-IR spectra were recorded using a Perkin Elmer spectrophotometer in the 500–4000 cm−1 wavelength region. The optical absorption spectra, measured employing a Unicam 8700 Spectrometer, allow to calculate the band gap energy by using the (␣h)2 vs. h plot, where ␣ is the optical absorption coefficient and h the photon energy. Thermal annealing treatments (TA) were carried out in a quartz tubular furnace, at normal pressure at 1000◦ C temperature gradient. The powder was placed in appropriate position in this temperature gradient during 2 h. 2. Experimental and chemical reactions The growth kinetics of CeO2 powder can be presented as: hydrolysis of thiourea SC(NH2 )2 with the formation of CO32− and S2− ions, according to previous reports [11,12]: SC(NH2 )2 + 3OH − ⇔ CO32− + HS − + 2NH3

(1)

Using the CB technique, the coordination of the [Ce(NH3 )6 ]4+ complex ion is proposed, and under alkaline medium this complex generates Ce(OH)4 : T = 23◦ C [Ce(NH3 )6 ] CO32−

4+

+ 4OH−

stirrer ⇒ Ce(OH)4 + 6NH3

(2)

ion generated by the hydrolysis thiourea and Ce(OH)4 are favored according to the following reactions: T = 23◦ C

Ce(NO3 )4 + 3KOH + NH4 NO3 + SC(NH2 )2

stirrer 4+ ⇒ [Ce(NH3 )6 ] + 2NO3− + 3OH − + 3K + + CO32− + S 2−

(3)

T = 25◦ C Ce(OH)4

stirrer ⇒ CeO2 + 2H2 O

(4)

For our surprise, in our experiments the reaction without thiourea cannot afford CeO2 precipitate, and from the above equations, it can be seen that thiourea plays a significant role in the formation of CeO2 nanocrystals and the reason for this comportment will be investigated in detail in future work as it has not been reported so far. CB is based on the formation of a solid phase from solution, which involves two steps in nucleation and particle growth. In the nucleation process, the clusters of molecules formed undergo rapid decomposition and particles combine to growth up to powder. Because of its convenient advantages as cost-effective and green reagent, thiourea has been increasingly used to synthesize novel phases and nanocrystals, and as far as we are aware, there is no report about the synthesis of CeO2 nanocrystals by CB. In this paper, thiourea was employed as a homogeneous precipitator to synthesize the CeO2 nanocrystals without any surfactant. With the hydrolysis of thiourea, the Ce(OH)4 (and probably Ce(OH)3 which is thermodynamically less stable) monomer is formed through formation Ce-OH coordination bonds. Polymerization of the hydroxide monomers is a crucial step in the next reaction process and such phenomenon has been discussed in some reports [13]. As the polymerization reaction increases slowly, the Ce4+ ions form backbones with serial Ce–OH covalent bonds. These backbones are connected together by Ce-O coordination bonds, and this result in the formation of the network structures, thus creating CeO2 nanoparticles [14]. It is known that the basic medium is favourable for the formation of Ce4+ ions and therefore, the Ce(OH)4 . In the other hand, the

144

O.P. Moreno et al. / Optik 148 (2017) 142–150

Fig. 1. EDX analysis and SEM surface micrographs: (a) as-grown (b) CeO2 TA powder.

oxidation of Ce3+ ions could occur during the hydrolysis and precipitation reaction of Ce4+ ions. However, the formation of Ce3+ is thermodynamically probable according to the above-mentioned working conditions. The powders were prepared using the following reagents (all precursor reagents are of 99.9% purity): Ce(NO3 )2 (0.2 M), KOH (0.1 M), NH4 NO3 (1.2 M), SC(NH2 )2 (0.1 M) and the solutions were sequentially mixed. The resulting solution was kept at 23±2◦ C and with constant stirring by 4 h. The experimental details have been previously reported by us [11,15]. After that, a precipitate-gel was obtained, which was filtered and washed several times with deionized water to remove ionic remnants. Finally, the powder was submitted to TA at 1000◦ C, during 2 h in air atmosphere in a cylindrical quartz tube. 3. Results and discussion Electronic dispersion spectroscopy (EDS) spectrum and micrographs of scanning electron microscopy (SEM) are showed in Fig. 1(a) as-grown and (b) CeO2 TA. In Table 1 the atomic concentration of Ce, O, and C obtained by EDS measurements are compiled. The surface roughness effect as well as the presence of some intrinsic defects within the powder (pores, interstices, stacking faults, grain boundaries, etc.) have some effects during the chemical analysis [16]. EDS is a semi-quantitative technique to calculate the concentration of elements, nevertheless the values listed in Table 1 are not so far from the true atomic concentrations. Images by SEM can be seen in Fig. 1(c) and (d). In these images, a spongy appearance can be appreciated for the powders. The same spongy soft-rounded morphology is observed in CeO2 TA sample, indicating an improvement of homogeneity. Table 1 C, O, and Ce concentrations in as-grown and CeO2 TA samples measured by means of electronic dispersion spectroscopy. Element

C Ce O

% Atomic As- grown

TA 1000 ◦ C

22.06 13.53 64.41

10.75 27.10 62.15

O.P. Moreno et al. / Optik 148 (2017) 142–150

145

Fig. 2. FTIR spectra of the as-grown and CeO2 TA powder.

FTIR spectra of as grown and CeO2 powders are shown in Fig. 2. For the as-grown sample, a broad absorption band at ∼3503 cm−1 can be attributed to stretching of the OH groups ( OH stretch, hydrogen bonded at 3500–3200 cm−1 ) of the defective sites and the physically adsorbed water molecules; this band disappears completely for CeO2 TA sample, and the same spectrum shows a sharp stretching mode that can be seen at ∼1300 cm−1 , attributed to stretching vibration mode of CO32− ion [17] and bands located in the ∼1260–1050 cm−1 range can be attributed to C O stretch, see equations (1) and (3). The band located at ∼1619 cm−1 can be assigned to various vibrational modes of amides groups, mainly due to the C O stretch absorptions. A comparison of these spectra clearly shows that many carboxylate groups remain on the surface of nanocrystals [18]. The sharp band centered at ∼1063 cm−1 are consistent with the stretching vibration of Ce O Ce bond [19]. The band due to stretching frequency of Ce O bond can be seen below 700 cm−1 [20,21] and these bands are not possible to locate in our experimental results. The peak at ∼1062 cm−1 are similar to those of commercial CeO2 powder [22]. The FTIR and EDS analysis confirm the bands associated with the CO32− ions and are in the form of pollutant and post-TA is removed together with the OH− groups. Such behavior can be considered according to the following equation: CO32−

T = 1000◦ C 1 ⇒ CO2(g) ↑ + O2(g) + 2e 2

The XRD patterns of the as grown and CeO2 TA powders are shown in Fig. 3(a). The as-grown powder showed four diffraction peaks located at: 2␪ = 28.58◦ , 33.65◦ , 47.31◦ , 56.66◦ , and CeO2 TA sample showed peaks at: 2␪ = 28.58◦ , 33.65◦ , 47.31◦ , 56.66◦ , 59.08◦ , 69.44◦ , 76.83◦ and 79.34◦ ; all peaks corresponding to the face-centered cubic phase structure of CeO2 according to PDF 04-013-4361 data. The diffractograms for CeO2 TA sample displayed that the reflection peaks become sharper and narrower owing to TA, indicating that the crystal size increases and the crystallinity of CeO2 become better defined. In the as-grown sample, very broad crystalline peaks can be seen and these are associated with nanocrystals [23]. The presence of some other different compound to CeO2 was not identified in these powders. In the inset of Fig. 3, deconvolution of the peak (111) crystal plane is performed, exhibiting the peak (200), and such plane is not clearly seen by an overlapping with the plane (111). The average crystallite sizes (GS) of the CeO2 samples were calculated from X-ray line broadening of the reflections of (111) using Scherer’s equation k␭/(dcos ), where  is the wavelength of the X-ray radiation, k is a constant taken as ∼0.89, ␪ is the diffraction angle, and d is the full width at half-maximum and (FWHM) for the as-grown and CeO2 TA samples. Table 2 shows the GS and FWHM for different crystal planes of as-grown and CeO2 TA samples. The most intense plane (111) GS for as-grown and CeO2 TA samples were ∼2.28 nm and ∼46.8 nm with FMWH of ∼3.48 and ∼0.369 Å, respectively. The effect of TA is significant in these structural properties of as-grown sample. Worthnoting is that the values of the lattice parameter a0 for the as-grown sample ∼5418 Å is slightly larger than that reported for CeO2 5.4113 Å (PDF 04-0134361), whereas the value for the CeO2 TA sample is ∼5.413 Å. According to these results, we can make a comparison with the XRD pattern of as-grown and CeO2 samples, observing that the width of the peaks is broadened for the as-grown as a feature of nanocrystals. Sayle et al. [24] and Conesa [25] have calculated the surface energies of some low (hkl) surfaces of CeO2 concluding that the (111) surface is the most stable surface of CeO2 owing to electrostatic (Coulombic) interaction energy, the effects of short-range interactions and the polarization energies and from their results, and such preferred orientation (111) plane is present in CeO2 TA. Due to the instability caused by polarization, the (200) surfaces are considered impossible in CeO2 [26]. According to the above-mentioned reports, this behavior occurs when the O2− ions are eliminated from the

146

O.P. Moreno et al. / Optik 148 (2017) 142–150

Fig. 3. XRD patterns of the CeO2 and TA powder. Table 2 Grain size (GS) and FWHM for crystal planes of as-grown and CeO2 TA samples. as-grown

CeO2 TA

2␪

FWHM

GS (nm)

2␪

FWHM

GS (nm)

28,53 32,84 47,29 56,57 77,71

3,48 2,6 3,05 3,53 4,8

2,28 3,05 2,6 2,25 1,65

28,56 33,09 47,5 56,37 76,75 79,06

0,169 0,185 0,216 0,262 0,369 0,158

46,8 42,8 36,6 30,2 21,5 50

lattice sites, and then the zero dipole energetic requirements are fulfilled. It has also been considered that although the surface energy of (200) plane is still higher than that of the (220) plane, the energy of the (200) plane is then in a reasonable energy level that renders the (200) plane appearance in CeO2 nanoparticles. In consequence, additional mechanisms for lowering further the (200) surface energy is suggested by their predominating appearance in the CeO2 nanoparticles. The absorbance spectra of the as-grown and CeO2 TA samples are shown in Fig. 4(a). The as-grown sample showed three absorption bands located at ∼382 nm (3.24 eV) and ∼357 nm (3.47 eV) and ∼271 nm (4.57 eV). These bands have been associated with defect states existing extensively between the Ce 4f state and O 2p valence band, originated from the charge transfer transitions from O 2p → Ce 4f [10,21]. The Ce 4f energy levels are localized at the band gap energy (Eg ) and lie at ∼3 eV above the valence band (O 2p) with 1.2 eV widths [27]. However, in our experimental results we identified for the CeO2 TA sample only two bands located at ∼271 nm (∼4.57 eV) and ∼320 nm (∼3.87 eV) that should be attributed to its better crystallinity compared to the as-grown samples and these bands are very close to those already reported [5]. A well-defined absorption band located at ∼320 nm can be also observed; this absorption band showing excitonic feature is in good agreement with the narrow size distribution of CeO2 nanocrystals [22]. Absorbance spectra in CeO2 nanocrystals have reported the following bands located at: (i) a broad UV emission band 393 nm (3.16 eV), (ii) a weak blue band at 419 nm (2.96 eV), (iii) a blue band at 443 nm (2.80 eV), (iv) a week blue-green band at 483 nm (2.57 eV), and (v) a green band at 529 nm (2.35 eV) [5]. We believe that these results are important because the absorption edge of CeO2 has shifted towards the visible region, which is of importance for applications in photovoltaic devices. In addition, the defect levels localized between the Ce 4f and O 2p band can also lead to wider emission bands (8 nm [10]. The Ce3+ ion is attributed to O2− vacancies in the crystal surface of the CeO2 [29]. However, the true mechanism of the red-shift and blue-shift of CeO2 nanostructures is still not clear so far. Two kinds of explanations have been proposed for the red-shift of the Eg of CeO2 nanoparticles; (a) it is known that the Eg will be blue-shifted with the reduction of GS due to the quantum confinement effect [18,37] and the quantum confinement effect may be ruled out when GS is