Journal of Alloys and Compounds 475 (2009) 817–821
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jallcom
Nanosized lead lanthanum titanate (PLT) ceramic powders synthesized by the oxidant peroxo method Emerson R. Camargo a,∗ , Cristiano M. Barrado a , Caue Ribeiro b , Elson Longo c , Edson R. Leite a a
LIEC-Laboratório Interdisciplinar de Eletroquímica e Cerâmica, Department of Chemistry, UFSCar-Federal University of São Carlos, Rod.Washingtin Luis km 235, CP 676, São Carlos SP 13565-9905, Brazil b EMBRAPA Instrumentac¸ão Agropecuária, Rua XV de Novembro 1452, São Carlos SP 13560-970, Brazil c Department of Biochemistry, Chemistry Institute of Araraquara, UNESP-São Paulo State University, Rua Francisco Degni, CP 355, Araraquara SP 14801-907, Brazil
a r t i c l e
i n f o
Article history: Received 6 May 2008 Received in revised form 1 August 2008 Accepted 7 August 2008 Available online 14 October 2008 Keywords: Oxide materials Chemical synthesis X-ray diffraction Crystal structure Inelastic light scattering
a b s t r a c t For the first time it is reported the synthesis of lead titanate modified with rare earth by the oxidantperoxo method (OPM). Lanthanum was added up to 20% in mol through the dissolution of lanthanum oxide in nitric acid, followed by the addition of a solution of lead and lanthanum nitrate into an aqueous solution of titanium peroxo complexes. The amorphous precipitate formed was heat-treated at different temperatures in the range from 400 to 900 ◦ C for crystallization. Powders were characterized by Raman spectroscopy and X-ray diffraction. Tetragonal perovskite structure was observed for the samples up to 15% of lanthanum substitution and cubic perovskite for sample with 20% of lanthanum. Crystallographic domains calculated by Scherrer equation showing a probable suppression of the crystallite growth in function of lanthanum content. It was observed shifting to lower frequencies of Raman modes in the range between 100 and 400 cm−1 and the vanishing of the A1(2TO) and E(1LO) modes could be attributed to transition phase from tetragonal to cubic. Electronic microscopy image revealed that the powders annealed at height temperature are spherical with sharp size distribution. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Isomorphic substitution of lead by lanthanum atoms induces some interesting changes in the physical properties of lead titanate (PT), for instance the linear decrease in the Curie temperature (Tc ) that follows the modification in the phase transition [1]. As consequence, structural characteristics of the Pb1−x Lax TiO3 solid solutions, thereafter referred to as PLT, depend strongly on the lanthanum concentration. However, because of nanoparticles generally display properties that differ from those of bulk material, alternative synthetic routes to the solid-state reaction have been developed to obtain a wider number of compounds at this scale [2–6]. Particularly, PLT and PT have been synthesized by several wet-chemical routes [7–9], but recently one of us developed a new synthetic route called the “oxidant peroxo method”, sometimes referred simply by the acronym OPM [10–16]. This wet-chemical method of synthesis is characterized by the fundamental oxyreduction reaction between lead(II) ions and some water soluble peroxo complexes that leads to the formation of an amorphous
and highly reactive precipitate. This precipitate is free of any common contaminants usually found in materials synthesized by others chemical routes, such as halides or graphitic carbon formed during the decomposition of organic material. Since this precipitate is formed by a molecular-level mechanism, its composition can be efficiently controlled. It was demonstrated that the crystallization temperature is below than those reported for this and similar systems synthesized by solid-state reaction or different sol–gel routes even. Moreover, the OPM technique uses water as solvent and a relatively simple experimental apparatus, without the necessity of dry atmosphere or toxic compounds In this paper, we are reporting by the first time the synthesis of lead-lanthanum titanate by the oxidant-peroxo method (OPM) by means of the heat treatment of the OPM-amorphous precipitates with different contents of lanthanum, demonstrating that the OPM route can be efficiently used to prepare rare earth doped lead titanate compositions. 2. Experimental Procedure 2.1. Synthesis
∗ Corresponding author. Tel.: +55 16 33518090; fax: +55 16 33518350. E-mail addresses:
[email protected] (E.R. Camargo),
[email protected] (C. Ribeiro),
[email protected] (E. Longo),
[email protected] (E.R. Leite). 0925-8388/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2008.08.035
Five compositions of Pb1−x Lax TiO3 , with nominal composition of “x” from 0.00 to 0.20 were prepared, and they are abbreviated thereafter to as “LX”. For instance, the composition Pb0.9 La0.1 TiO3 where “x = 0.10” is referred to as L10. Samples were pre-
818
E.R. Camargo et al. / Journal of Alloys and Compounds 475 (2009) 817–821
pared by the OPM route [10–16] through the addition of 1 g (0.02 mol) of titanium metal powder (Aldrich, USA, 99.99%) to an aqueous solution consisting of 80 mL of hydrogen peroxide (Synth, Brazil) and 20 mL ammonia aqueous solution (Synth, Brazil). This solution was kept in rest using an ice-water bath for approximately 10 h, resulting in a yellow transparent aqueous solution of soluble peroxytitanate [Ti(OH)3 O2 ]− ion with concentration of 0.14 mol L−1 . Amounts of lanthanum oxide (Merck, Germany, 99.5%) previously heat treated at 900 ◦ C for 1 h was weighted and added to a diluted aqueous solution of nitric acid (Synth, Brazil) at pH 2. After this dissolution, the volume was corrected to 50 mL with distilled water, and lead nitrate (Merck, Germany, 99.5%) was added to complete the stoichiometric ratio for each sample composition. The solution of lanthanum and lead nitrates was slowly dropped into the peroxytitanate solution under ice-water bath and stirring, resulting in a vigorous evolution of gas. An orange precipitate was immediately formed and the solution lost its yellow colour. This precipitate was filtered and washed with acetone (Synth, Brazil) to eliminate the adsorbed water and the nitrate ions. The washed amorphous precipitates were dried at 50 ◦ C for 5 h and ground using a mortar. Amounts of 0.30 g of this powder, thereafter referred to as “precursor”, were calcined between 400 and 900 ◦ C for 1 h using closed alumina boats under a heating rate of 10 ◦ C min−1 . 2.2. Characterization The dried-precipitate and the calcined powders were characterized by elemental (carbon, hydrogen and nitrogen) and ICP (Pb, La, and Ti) analyses, Raman spectroscopy with Fourier transform (FT-Raman) and X-ray diffraction (XRD). Raman spectra were collected at room temperature between 100 and 950 cm−1 using a FT-Raman Bruker RFS 100/S spectrometer using the 1064 nm line of a air cooled Nd:YAG laser. The X-ray powder patterns were also collected at room temperature in the 2 range from 10◦ to 80◦ with step scans of 0.02◦ , using a Rigaku D/MAX 200 diffractometer with a rotary anode (Cu K␣ radiation) operating at 150 kV and 40 mA. Crystallographic coherence lengths were calculated according Scherrer’s equation [17] deconvoluting each peak using a Lorentzian approximation to determine the full width at half maximum (FWHM) using as reference FWHM the (1 0 1) reflexion of a monocrystalline silicon wafer. An average of three peaks was considered as an estimative of the crystallite size.
3. Results and discussion Lanthanum oxide (La2 O3 ) can be easily dissolved in aqueous acid solution, for example using diluted nitric acid, what results in a solution of hydrated lanthanum and nitrate ions that can be referred as dissolved lanthanum nitrate. Therefore, it is possible to prepare a stable solution of lanthanum and lead nitrates to be dropped on the peroxititanic solution, previously prepared from the dissolution of titanium metal in H2 O2 /NH3 solution at high pH, to obtain a precise and stoichiometric precipitate with the Pb:La:Ti mole ratio desired. This precipitate can be described, at least, as a complex mixture of amorphous and hydrated oxides PbO2 , TiO2 and La2 O3 that will result in the crystallized PLT after a firing step. Pb(OH)4 2− + La(H2 O)6 3+ + [Ti(OH)3 O2 ]− → {PbO2 · TiO2 · La2 O3 }nH2 O → Pb1−x Lax TiO3
(1)
Usually, samples of PLT are prepared by the conventional solidstate reaction. In this method, PbO, La2 O3 and TiO2 are weighed according to the stoichiometry and mixed by ball milling. On the other hand, the OPM approach to obtain PLT is based on the molecular level reaction between soluble compounds, following a “bottom-up” strategy of synthesis. First, it means that less energy is necessary to assembly nanosized crystalline particles, and second, final materials are more homogeneous and uniform than those obtained by traditional milling and firing process. Table 1 shows ICP results obtained from the dissolution of four PLT compositions in diluted nitric acid. Considering the substitution of lead by lanthanum, all of the results were calculated in function of the molar amount of titanium. The experimental amount of lanthanum is quite similar to the theoretical calculated value, demonstrating that the experimental procedure is reliable concerning the rare earth addition. However, the amount of lead is relatively higher than those calculated. This excess is a good evidence that lead is not lost during the amorphous precursor formation. Of course, the com-
Table 1 Mole composition of different amorphous precipitate obtained from ICP analysis Sample
L5 L10 L15 L20
Ti (mol)
Pb (mol)
Exp
Exp
Theo
Exp
La (mol) Theo
1 1 1 1
0.9808 0.9495 0.8954 0.8175
0.95 0.90 0.85 0.80
0.0448 0.1005 0.1551 0.2064
0.05 0.10 0.15 0.20
Standard deviation 0.40 0.11 0.14 0.08
Exp are the experimental results while Theo are those calculated from the Pb1−x Lax TiO3 formula.
position of crystalline powder obtained after the heat-treatment can be slightly different from the values of Table 1, since to keep the electroneutrality some vacancies in the lead site is expected, as discussed below. What is also necessary to note is that the peroxo chemical environment is quite oxidizing, and it is not expected to find oxygen vacancies to compensate the aliovalent substitution of Pb2+ by La3+ . One of most important characteristic of the OPM route is the absence of impurities when compared to other wet-chemical based methods, such as graphitic carbon, chlorine ion, sulphur and nitrogen [2–6]. The amorphous precipitates were subjected to elemental analysis to determine the amounts of C, N, H and S, and the results are shown in Table 2. These numbers confirm the purity of OPM precursors, where ND means not detected or that hydrogen and sulphur can be present but at amounts below than the sensibility of the detector. 3.1. Formation of the PLT In La modified PbTiO3 , La3+ replaces Pb2+ rather than Ti4+ . To keep the electrical neutrality, site vacancies should be created. On the other hand, it is possible that some Ti4+ are reduced to Ti3+ . Therefore, the properties of PLT should be influenced not only by the vacancy concentration, but in some extension also by the Ti3+ concentration. The local defects introduced in the structure by Bsite vacancies (occupied by the titanium cation) have much more inhibiting effects on the macroscopic cubic-to-tetragonal transition than the A-site vacancies (occupied by the lead cation), in spite of the fact that the electrical neutrality can be compensated in both A and B sites. But what interests us now is that this isomorphic substitution of lead by lanthanum atoms induces some changes in the structural properties of PbTiO3 host material. For instance, when lanthanum content is higher than 25% in mol, a diffuse character of the ferroelectric-to-paraelectric phase transition (DPT), which is a tetragonal-to-cubic transition, is observed. It has been reported that the lanthanum-induced modification concerning PbTiO3 results in structural changes that can be directly related to the nature of the phase transition. From this point of view, in the OPM approach to synthesize PLT, lanthanum and lead ions are added together at molecular level. This particular characteristic of the OPM method should result in Table 2 Mass percentage of nitrogen, carbon, hydrogen and sulphur in different amorphous precipitates Sample
L5 L10 L15 L20
Elements in percentage of mass N
C
H
S
0.1295 0.0984 0.1120 0.1136
0.1741 0.1643 0.1777 0.2045
ND ND ND ND
ND ND ND ND
ND means that the presence of the element can be below than the detector sensibility.
E.R. Camargo et al. / Journal of Alloys and Compounds 475 (2009) 817–821
819
Fig. 2. X-ray patterns of PLT with 20% (L20) of substitution. Samples were annealed for 1 h at different temperatures in the range form 400 to 800 ◦ C. Sample referred to as L20 means that there is 20% in mol of lanthanum. Miller indexes on the 800 ◦ C pattern refer to cubic phase.
Fig. 1. X-ray patterns of the lead titanate modified with different amounts of lanthanum collected at room temperature. Sample referred to as L20 means that there is 20% in mol of lanthanum. Miller indexes on the L0 pattern refer to tetragonal lead titanate phase, whereas on the L20 pattern refer to cubic phase.
a homogeneous distribution of the La in substitution of Pb in the A-site, and because of the oxidizing environment, the electrical neutrality results mainly because of vacancies than the presence of Ti3+ or oxygen vacancies. Fig. 1 shows XRD patterns of five PLT samples calcined at 800 ◦ C for 1 h with different amounts of lanthanum, from pure lead titanate (L0) up to 20% of lead substituted (L20). It is easy to observe, as indicated by the Miller indexes, a modification regarding the splitting of the peaks as a consequence of the transition from tetragonal to cubic phase [18]. On the other hand, the transition from the amorphous precursor to the crystalline PLT phase sometimes occurs through metastable or intermediate undesirable phases. To check it, the L20 sample was heat treated at different temperatures for 1 h and subjected to XRD analysis. It is possible to observe in Fig. 2 the presence of diffraction peaks in the powder treated at low temperature as 450 ◦ C. Of course, this temperature cannot be used as the lowest temperature for the crystallization of the PLT phase, since powders calcined at lower temperature but for several hours can also result in crystalline PLT, or powders calcined at this temperature for just few minutes can be unaffected. However, it is evident in Fig. 2 that intermediate phases or metastable structures are not present, but the PLT phase crystallizes directly from the amorphous OPM precursor, even when 20% of the lead is substituted by lanthanum. The absence of intermediate phases during the crystallization of the PLT structure with 20% of lanthanum is a strong evidence of the atomic level homogeneity of the amorphous precursor. Previous works regarding the synthesis of PLT and correlated samples by wet chemical routes [7,9,19] reported single PLT phase after heat treatments at temperatures near to 550 ◦ C, but sometimes it was possible to find diffraction peaks of secondary phases when the samples were treated at lower temperatures. It was also possible
to find direct transition from the amorphous precursor for samples with low lanthanum content. What is remarkable in Fig. 2 is the direct transition at low temperature for samples with 20% of lanthanum. An important note from XRD data is the influence of lanthanum substitution in crystallographic coherence lengths (crystallite size). The estimated crystallite sizes calculated by Scherrer’s equation (Table 3) show a slight reduction with the lanthanum content, indicating a probable suppression of the growth of the crystallite with the substitution. This observation is remarkable when compared with the behaviour of L20 samples with temperature, where the crystallite growth is clearly identified, however, with few differences in the range 500–700 ◦ C. These results show that the OPM synthesis is a good alternative to retain well-crystallized small nanoparticles, even in intermediary temperature conditions. These samples were also analysed by Raman spectroscopy to characterize the tetragonal distortion [9] and the presence of small secondary phases. The spectrum of pure lead titanate (L0) in Fig. 3(left) shows its usual profile, with well-defined characteristic peaks. On the other hand, the substitution of only 5% of lead by lanthanum ions resulted in broadness of the scattering peaks of L5 spectrum. Because of Raman scattering spectroscopy is a quite sensitive tool for local distortion, much more than XRD [20], it can be used to valuate how the rare earth ion affects the crystal structure of the host material. The Silent mode at 290 cm−1 remains almost unchanged, as observed in others studies [21], but it is visible that Table 3 Estimated crystallite sizes from Scherrer’s equation obtained from data in Figs. 1 and 2 Different samples annealed at 800 ◦ C D (nm) L0 L5 L10 L15 L20
57.5 43.47 41.51 34.86 39.62
± ± ± ± ±
4.32 8.01 9.69 13.87 6.81
Sample L20 annealed at different temperatures D (nm) 400 ◦ C 500 ◦ C 600 ◦ C 700 ◦ C 800 ◦ C
– 20.53 24.01 25.00 39.62
± ± ± ±
1.5 3.42 4.06 6.81
820
E.R. Camargo et al. / Journal of Alloys and Compounds 475 (2009) 817–821
Fig. 3. Left: Raman of the lead titanate modified with different amounts of lanthanum collected at room temperature and annealed for 1 h at 900 ◦ C. Right: Dependence of some Raman modes in function of lanthanum composition.
the modes in the range between 100 and 400 cm−1 shift to lower frequencies, what can be better visualized in Fig. 3(right). The vanishing of A1 (2TO) and E(1LO) modes resulted from the tetragonal to cubic phase transition. From this point of view, the downshift can be attributed to the decrease of tetragonality (c/a) with increasing lanthanum concentration [22]. Some authors affirm that there is dependence between the particle size and the Raman shift [9], however, spectra of Fig. 3 were collected from powders heat treated at 900 ◦ C for 1 h (shown in Fig. 4). When OPM-synthesized powders are treated at so high temperature, all of the particles usually growth uniformly, showing average size around 0.1 m, what eliminates the particle size influence on the Raman shift. Therefore, we
can ascribe the Raman modes shifting only to the influence of the lanthanum on the tetragonality. Fig. 4 shows some images of the heat-treated powders at high temperature, where can be observed a sharp size distribution, and particles with uniform spherical shape. The average diameter was estimated to be near to 80 nm. It is also evident that these powers are highly reactive, as observed by the presence of necks and some degree of sinterization between the particles. It is important to observe that when the precursor is annealed at so high temperature, reactive single crystalline particles tend to joint to form a bigger particle, decreasing the surface area. In this case, particles can show smaller crystallographic coherence lengths (crystallite
Fig. 4. Scanning electronic images from of PLT powders annealed at 900 ◦ C for 1 h with different amounts of lanthanum.
E.R. Camargo et al. / Journal of Alloys and Compounds 475 (2009) 817–821
size) than the diameter observed by electronic microscopy. Scherrer’s equation estimates the size of one monocrystal, but if single crystalline particles are sufficiently reactive, it is possible that the final particle observed resulted from the union of several single crystalline particles that retain their crystallographic coherence even after their consolidation. Therefore, the difference between the observed particle size and the crystallite size indicates that each powder particle observed is better described as a polycrystalline particle than a single crystal. 4. Conclusions It was demonstrated that it is possible to prepare nanosized powders of lanthanum modified lead titanate Pb1−x Lax TiO3 , with x in the range from 0 to 20, by means of the oxidant peroxo method (OPM). Transition tetragonal to cubic phase was observed by X-ray diffraction at compositions of 15% of lanthanum, what was confirmed by Raman spectroscopy. SEM images of the powders are formed of polycrystalline particle with a sharp size distribution around 100 nm, with uniform spherical shape. Acknowledgements This work was supported by the Brazilian agencies FAPESP through of the CMDMC/Cepid (Project 98/14324-0), CNPq (Project 555644/2006-5) and CAPES.
821
References [1] P.P. Neves, A.C. Doriguetto, V.R. Mastelaro, L.P. Lopes, Y.P. Mascarenhas, A. Michalowicz, J.A. Eiras, J. Phys. Chem. B. 108 (2004) 14840–14849. [2] A.P. Alivisatos, J. Phys. Chem. 100 (1996) 13226–13239. [3] N.L. Wu, S.Y. Wang, I.A. Rusakova, Science 285 (1999) 1375–1377. [4] E.R. Leite, in: H.S. Nalwa (Ed.), Encyclopedia of Nanoscience and Nanotechnology, vol. VI, American Scientific Publishers, 2004, pp. 537–550. [5] R.C.R. Franco, E.R. Camargo, M.A.L. Nobre, E.R. Leite, E. Longo, J.A. Varela, Ceram. Int. 25 (1999) 455–460. [6] E.R. Camargo, M. Popa, M. Kakihana, Chem. Mater. 14 (2002) 2365–2368. [7] J. Chen, X.R. Xing, R.B. Yu, J.X. Deng, G.R. Liu, J. Alloys Compd. 388 (2005) 308–313. [8] H.P. Soon, J.M. Xue, J. Wang, J. Appl. Phys. 95 (2004) 4981–4988. [9] Q.F. Zhou, H.L.W. Chan, Q.Q. Zhang, C.L. Choy, J. Appl. Phys. 89 (2001) 8121–8126. [10] E.R. Camargo, M. Kakihana, Chem. Mater. 13 (2001) 1181–1184. [11] E.R. Camargo, M. Popa, J. Frantti, M. Kakihana, Chem. Mater. 13 (2001) 3943–3948. [12] E.R. Camargo, J. Frantti, M. Kakihana, J. Mater. Chem. 11 (2001) 1875–1879. [13] E.R. Camargo, M. Kakihana, J. Am. Ceram. Soc. 85 (2002) 2107–2109. [14] E.R. Camargo, E. Longo, E.R. Leite, V.R. Mastelaro, J. Solid State Chem. 177 (2004) 1994–2001. [15] E.R. Camargo, F.L. Souza, E.R. Leite, M. Kakihana, J. Appl. Phys. 96 (2004) 2169–2172. [16] L.P.S. Santos, E. Longo, E.R. Leite, E.R. Camargo, J. Alloys Compd. 372 (2004) 111–115. [17] B.D. Cullity, S.R. Stock, Elements of X-Ray Diffraction, third ed., Prentice-Hall, Englewood Cliffs, NJ, 2001. [18] H. Wang, L. Wang, J.B. Liu, B. Wang, H. Yan, Mater. Sci. Eng. B 99 (2003) 495– 498. [19] L.H. Xue, Q. Li, Q.Y. Guo, R. Liu, Y.L. Zhang, J. Mater. Sci. 40 (2005) 2697–2699. [20] M. Kakihana, Trends Appl. Spectrosc. 1 (1993) 261. [21] Y. Deng, M.S. Zhang, J.M. Hong, Z. Yin, J. Matter. Sci. Lett. 21 (2002) 1371–1373. [22] Y. Deng, Z. Yin, Q. Chen, M.S. Zhang, W.F. Zhang, Mater. Sci. Eng. B 84 (2001) 248–251.
