Journal of Electroceramics

, Volume 18, Issue 3, pp 283–288

Structure and microwave dielectric properties of Sm(2−x)/3LixTiO3

Authors

    • Department of Inorganic MaterialsShanghai University
  • K. Yan
    • Department of Inorganic MaterialsShanghai University
  • R. Ubic
    • Department of Materials Science & EngineeringBoise State University
Article

DOI: 10.1007/s10832-007-9126-2

Cite this article as:
Bian, J.J., Yan, K. & Ubic, R. J Electroceram (2007) 18: 283. doi:10.1007/s10832-007-9126-2
  • 77 Views

Abstract

The structural evolution, and microwave dielectric properties of \( {\text{Sm}}_{{{{\left( {2 - x} \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - x} \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{x} {\text{TiO}}_{3} \) ceramics (x = 0.0 ≤ x ≤ 0.5) were investigated in this work. X-ray diffraction (XRD) results show that samples with x > 0.3 exhibit a single perovskite phase. Impurity phases of Sm2Ti2O7 and TiO2 appear and their amount increases with the decrease of x when x ≤ 0.3. TEM observation indicates that the A-site is ordered in x = 0.5, but not in x = 0.3). The dielectric constant decreases with the increase of x for 0.1 ≤ x ≤ 0.4 and then increases with further increase in x up to x = 0.5. The Q×f value decreases with the decrease of x due to the increased occurrence of Sm2T2O7 secondary phase, defects and twinning boundaries. The temperature coefficient of resonant frequency is negative and its absolute value decreases greatly with the decrease of x value.

Keywords

A-site deficient perovskiteOrder-disorderMicrowave dielectric properties

Introduction

Complex perovskites with general formula \( {\text{A}}^{{2 + }} {\left( {{\text{B}}^{{2 + }}_{{1 \mathord{\left/ {\vphantom {1 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{B}}^{{{\text{5 + }}}}_{{2 \mathord{\left/ {\vphantom {2 3}} \right. \kern-\nulldelimiterspace} 3}} } \right)}{\text{O}}_{3} \) have received widespread interest in the wireless microwave communications community. Compared to the numerous examples of B-site ordered systems, A-site ordered perovskites are relatively rare. A-site order-disorder reaction can play a critical role in affecting the properties of perovskites. Several studies have focused on the ionically conducting perovskites, \( {\left( {{\text{La}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } } \right)}{\text{TiO}}_{3} \) [13]. The ionic conductivities of the ordered A-site samples were approximately an order of magnitude lower than their disordered counterparts [1]. The A-site ordering degree of \( {\left( {{\text{La}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } } \right)}{\text{TiO}}_{3} \) was governed by the Li content and thermal treatment [3, 4]. In Li poor compounds the A-site cations adopt usual (001) ordered structure, and in Li rich compounds (x ≥ 0.25) or quenched samples a disordered A-site structure was adopted. Authors have studied the structure and microwave dielectric properties of \( {\text{La}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Na}}_{ \times } {\text{TiO}}_{3} {\left( {0.02 \leqslant x \leqslant 0.5} \right)} \) and \( {\text{Nd}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} {\left( {0.0 \leqslant \times \leqslant 0.5} \right)} \) [5, 6]. In the case of \( {\text{La}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Na}}_{ \times } {\text{TiO}}_{3} {\left( {0.02 \leqslant x \leqslant 0.5} \right)} \), the A-site ordering degree decreases with the increase of sodium content and becomes zero when x > 0.2. The A-site vacancy concentration and the ordering degree have opposite effect on the Q×f value of \( {\text{La}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Na}}_{ \times } {\text{TiO}}_{3} \). The τf decreases almost linearly with the decrease of sodium content due to the decrease of tilting angle of TiO6-octahedra. The same arguments hold true for \( {\text{Nd}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} {\left( {0.0 \leqslant x \leqslant 0.5} \right)} \) except that second phases of Nd2Ti2O7 and Nd2Ti4O11 appeared in addition to Nd2/3TiO3 when x ≤ 0.1. The purpose of this paper is to systematically investigate the structural evolution and microwave dielectric properties of \( {\text{Sm}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} \) ceramics(x = 0.0 ≤ x ≤ 0.5).

Experiment

\( {\text{Sm}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} {\left( {0.0 \leqslant x \leqslant 0.5} \right)} \) ceramic samples were prepared by a conventional solid-state reaction process from the starting materials including TiO2 (99.8%), Li2CO3(99.6%) and Sm2O3(99.9%). The \( {\text{Sm}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} {\left( {0.0 \leqslant x \leqslant 0.5} \right)} \) compounds were weighed according to the stoichmetric composition and milled with ZrO2 media in ethanol for 24 h, dried and calcined at 1100 °C for 2 h in an alumina crucible. The calcined powders were remilled for 24 h, dried and mixed with 7 wt% PVA. The mixtures were pressed into pellets. The compacts were sintered between 1300–1350 °C for 2 h. In order to prevent the evaporation loss of Li2O, all samples were muffled with powder of the same composition during sintering.

The phase constituents of the sintered samples were identified by X-ray powder diffraction (XRD) with Ni-filtered CuKα radiation (Model Dmax-RC, Japan). Bulk densities of the sintered specimens were identified by the Archimedes’ method. The microstructure of the sintered samples were characterized by scanning electron microscopy (SEM; Model XL20, Philips Instruments, Netherlands). All samples were polished and thermally etched at a temperature which was 100 °C lower than the sintering temperature. Some samples underwent thinning by conventional ceramographic techniques followed by ion milling (PIPS 691, Gatan, USA) for observation in the transmission electron microscope (JEM 2010, JEOL, Japan). Microwave dielectric properties of the sintered samples were measured between 4 and 6 GHz using a network analyzer (Hewlett Packard, Model HP8720C, USA). The quality factor was measured by the transmission cavity method. The relative dielectric constant (ɛr) was measured according to the Hakki-Coleman method using the TE011 resonant mode, and the temperature coefficient of the resonant frequency (τf) was measured using invar cavity in the temperature range from −20 to 80 °C.

Results and disussions

Figure 1 shows the XRD patterns of as-sintered surface of \( {\text{Sm}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} {\left( {0.0 \leqslant x \leqslant 0.5} \right)} \) samples. It indicates that the samples with x ≥ 0.4 are composed of single perovskite phase. Impurity phases of Sm2Ti2O7 and TiO2 appear and their amount increases with the decrease of x value when x ≤ 0.3. Compared with the analogous \( {\text{La}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Na}}_{ \times } {\text{TiO}}_{3} \) and \( {\text{Nd}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} \) series [5, 6], the perovskite solubility range for \( {\text{Sm}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{x} {\text{TiO}}_{3} \) is narrower. Sm(2 – ×) / 3 LiTiO3 seems to exhibit a different type of perovskite structural distortion. The trend in such distortions is obviously related to the decrease of tolerance factor t with the decrease of the A-site ionic radii from La3+ to Nd3+ to Sm3+. Based on the assumption that the perovskite type structural framework is dominated by the larger rare earth ions orther than small alkali ions, tolerance factors for La0.5Na0.5TiO3, Nd0.5Li0.5TiO3 and Sm0.5Li0.5TiO3 are calculated as 0.97, 0.94 and 0.93, respectively. Only antiphase tilt of octahedra occurs when 0.965 <t < 0.985, and both in-phase and anti-phase tilt occur simultaneously for t < 0.965 [7]. The variation of tolerance factor as function of x for \( {\text{Sm}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} \) is shown in Fig. 2. The tolerance factor t decreases linearly with the decrease of x. A tolerance factor lower than 0.95 suggests a heavily tilted structure, and the tolerance factors for compositions corresponding to x ≤ 0.3 are below the typical stability limit of even distorted perovskites
https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig1_HTML.gif
Fig. 1

XRD patterns of sintered \(Sm_{{{{\left( {2 - x} \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - x} \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} Li_{x} TiO_{3} samples{\left( {x = 0.0 \leqslant \times \leqslant 0.5} \right)}\)(https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_IEqa_HTML.gif: TiO2;https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_IEqb_HTML.gif: Sm2Ti2O7,https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_IEqc_HTML.gif: (Sm1/2Li1/2)TiO3)

https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig2_HTML.gif
Fig. 2

Variation of tolerance factor t as function of x value

Figure 3 shows the SEM micrographs for the samples sintered at 1325 °C/2 h for x = 0.1, and 0.5. The sample with x = 0.5 exhibits single phase while the sample with x = 0.1is multiphase, in agreement with the results from the XRD analysis.
https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig3_HTML.gif
Fig. 3

SEM micrographs of the sample with x = 0.1 and x = 0.5

The selected area diffraction patterns (SADPs) in Fig. 4, obtained from several grains in x = 0.3 and x = 0.5 samples, have been indexed self-consistently according to the simple pseudocubic perovskite unit cell (a = 3.8002 Å). Evidence of in-phase octahedral tilting can be found by examining the <100> pseudocubic SADPs (Fig. 4(a–c)). Each pattern can reveal such tilting about a single axis by allowing, according to the Weiss zone law, a unique set of 1/2{odd, even, odd}-type γ reflections to appear. As can clearly be seen, only Fig. 4(b) contains γ reflections, indicating the presence of in-phase tilting about a single pseudocubic axis. This conclusion is further supported by the <111> pseudocubic pattern (Fig. 4(d)), which would simultaneously allow every type of γ reflection to appear and so show all in-phase tilt axes present, and yet still shows only a single set of γ reflections. As indexed, this axis would be b (b+ in Glazer’s notation [9]). Antiphase tilting can be revealed by studying, instead, the <110> pseudocubic SADPs, each one allowing a single set of 1/2{odd, odd, odd}-type α reflections corresponding to antiphase tilting about two possible axes. Such patterns are shown in Fig. 4(e–g), two of which show the presence of superlattice α reflections. The lack of α reflections in the [110] pattern indicates that octahedra are not tilted in antiphase about a or b. As there is also no in-phase tilting about a (Fig. 4(a,d)), octahedra must remain untilted about this axis (a0 in Glazer’s notation [9]). The α reflections in the [101] correspond to antiphase tilting about either a or c; however, as it has already been established that octahdra are not tilted about a, these reflections must indicate antiphase tilts about c instead (c in Glazer’s notation [9]). The [011] pattern only confirms this conclusion. Its α reflections correspond to antiphase tilts about either b or c. As it has already been established that octahedra are tilted in-phase about b, this pattern must be further evidence of antiphase tilting about c. Thus, from the collection of superlattice reflections observed, the tilt system of \( {\text{Sm}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{x} {\text{TiO}}_{3} \) may be described as aob+c; however, the extreme similarity between [100] and [001] patterns and [101] and [011] patterns means that a strict interpretation of all the patterns can only conclude that octahedra are tilted in-phase about a single axis and in antiphase about at least one other axis. The presence of 1/2{odd, even, even}-type β reflection in Fig. 4(a,c,d) could be indicative of antiparallel A-site cation displacements, but are in fact almost certainly an artefact caused by double diffraction.
https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig4_HTML.gif
Fig. 4

SADPs of \( Sm_{{{{\left( {2 - x} \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - x} \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} Li_{x} TiO_{3} \) indexed according to the simple pseudocubic perovskite unit cell (a = 3.8002 Å) corresponding to (a) [100], (b) [010], (c) [001], (d) [111], (e) [110], (f) [101], and (g) [011]

For distorted perovskites, when the two A-site cations become very different, such as alkali metal Li and rare earth Sm, the tilt system achieved will be directly dependent upon the ratio of large cations, AL, to small ones, As. When the AL/As ratio is 1:1, such as in the case of Sm0.5Li0.5TiO3, tilt system a0b+c should be favored [8]. The in-phase tilt transition can coincide with the appearance of tilting of the BO6 octahedra with an antiparallel shifts of the A-site cations [9]. Determination of the exact space group is then complicated by the presence of twinning, antiparallel A-site displacements, and possible A-site ordering. Glazer’s work, which only examined disordered perovskites, would suggest space groups Pmmn, P21/m, Pmnb, or Bmmb as possible symmetries [9], but none of these space groups can satisfactorily explain all the SADPs observed. Even the \( {\text{R}}\overline{{\text{3}}} {\text{c}} \) model which Alonso et al. [10] suggested for ordered \( La_{{1 \mathord{\left/ {\vphantom {1 2}} \right. \kern-\nulldelimiterspace} 2}} Li_{{1 \mathord{\left/ {\vphantom {1 2}} \right. \kern-\nulldelimiterspace} 2}} TiO_{3} \) is unable to interpret the SADPs of \( {\text{Sm}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} \). Howard et al. [11], who examined perovskites with 1:1 ordering on the B-site, would suggest either P42/n, C2/c, or P21/c; but P21/c cannot account for A-site ordering, C2/c cannot account for antiparallel displacements of A-site cations, and there are no known examples of a perovskite with space group P42/n. Indeed, none of these space groups can be used consistently to explain all the observed SADPs either.

Figure 5 shows a dark-field image and SADP of a twin variants in the x = 0.5 sample imaged with g = 1/2(120). The twinning is the common {110} perovskite twinning, with variants rotated by 90° about<001>. In this example, the twin plane is (011) and the two variants, oriented along [100] and \( {\left[ {\overline{1} 00} \right]} \), are related by a 90° rotation about [100]. Striations apparent within the bright variant are parallel to <100> and are shown more clearly in Fig. 6. The defects visible have traces of [100] and [001] on the (010) plane and, given the frequent streaking in <100> directions in SADPs of this composition, probably lie on {100} planes. They may be evidence of the usually illusive 180° twins. Such twins are rotated by 180° about pseudocubic <001>; and due to the almost exact coincidence of reflections in electron diffraction patterns, it is not usually possible to detect them. The slight spot splitting frequently observed in such grains both corroborates this interpretation and implies a crystalline symmetry lower than orthorhombic (otherwise there would be an exact coincidence of reflections). The streaking, a small amount of which is visible in the inset SADP in Fig. 6, would then correspond to some one-dimensional disorder in this direction caused by the spacing of these twin boundaries. Observable in both Figs. 5 and 6 are ribbons of antiphase boundaries (APBs) caused by the ordering of Sm3+ and Li1+ cations on the A-site. From this observation alone it is possible to conclude that Sm0.5Li0.5TiO3 is ordered on the A-site, although the ordered regions are very small (the density of APBs is large).
https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig5_HTML.jpg
Fig. 5

Typical dark-field image of a grain of Sm0.5Li0.5TiO3, g = ½(120)

https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig6_HTML.jpg
Fig. 6

Sm0.5Li0.5TiO3 grain with defects on {100} and APBs

APBs are not observed in the x = 0.3 composition, although the 180° twin boundaries are still present. A dark-field image is shown in Fig. 7. In addition, another type of defect is also observed which is reminiscent of the modulated {111} nanotwins in tetragonal BaTiO3 observed by Wu et al. [12]. Both types of defect are visible in Fig. 8. The boundaries between 180° variants are still present (there are three of them visible in the figure), but the variants themselves are modulated by a ripple contrast. These ripples have traces of \( {\left[ {11\overline{1} } \right]} \) and \( {\left[ {1\overline{1} 1} \right]} \) on the (011) plane. Streaking in SADPs of these samples along <110> is common, suggesting that these ripples lie on {110} planes and further suggesting nanotwinning as a possible explanation for this contrast. Indeed, spot splitting common to most twinning processes is visible in many SADPs of such grains. The <110> streaking can be attributed to disorder in the spacing of these twins and is probably exacerbated by the dimensions of the twins themselves, which are only about 10 nm thick in this direction.
https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig7_HTML.jpg
Fig. 7

Dark-field (g = 010) image of a twinned grain in Sm0.567Li0.3TiO3 with variants oriented along [100] and [010]. The twin plane is (110)

https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig8_HTML.jpg
Fig. 8

Ripple contrast in Sm0.567Li0.3TiO3 showing 90° {110} nanotwins between 180° {100} twin planes

Figure 9 shows a TEM image for the sample with x = 0. The sample contains an estimated 20–30 vol% free TiO2. The matrix phase is Sm2Ti2O7, an SADP of which is shown in the inset. These results are in good agreement with the XRD data.
https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig9_HTML.jpg
Fig. 9

TEM micrograhp of the sample with x = 0. The small white grains were identified as TiO2; the matrix phase is Sm2Ti2O7. The inset SADP is of the Sm2Ti2O7 phase (zone axis is<110>)

Figure 10 shows the variation of dielectric constant as a function of x value. The dielectric constant decreases slightly up to x = 0.4 and then increases with the further decrease of x. The increase of dielectric constant with the decrease of x when x ≤ 0.4 is mainly related to the increased amount of high dielectric constant phase of TiO2 (ɛr = 100) [13]. The variation of Q×f value with x value is shown in Fig. 11. The Q×f value increases with the increase of x value, which is contrary to the \( {\text{La}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Na}}_{ \times } {\text{TiO}}_{3} \) and \( {\text{Nd}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} \) series but in agreement with the TEM analysis above, from which it can be inferred that A-site ordering increases with increasing x, although the ordered domains are tiny. It is well known that the losses in microwave dielectrics are made up of both intrinsic and extrinsic losses. The formers are related to the crystal structure such as order–disorder, while the latters are related to the microstructure (e.g. pores, boundaries and second phases). The x = 0.3 sample was more heavily twinned and defected than the x = 0.5 one, which suggests that Q×f should increase with increasing x, therefore contributing to the increase in Q×f with the increase in x is the decreasing contribution from extrinsic losses originating from Sm2Ti2O7 and other defects. Figure 12 shows the variation of temperature coefficient of resonant frequency τf as function of x. All samples exhibit negative τf values and the absolute value of τf decreases almost linearly with decreasing x. A minimum |τf| value of −5 ppm/°C was obtained when x = 0.1. The decrease of absolute τf value with the decrease of x is mainly related to the increase of oxygen octahedral tilt and the increase amount of TiO2 phase which has a high positive τf value (τf = 450 ppm/°C) [13]. Microwave dielectric properties of the x = 0 composition are not shown in Figs. 10, 11, 12 as its dielectric loss was too high to be measured by the resonant method.
https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig10_HTML.gif
Fig. 10

Variation of dielectric constant as a function of x value

https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig11_HTML.gif
Fig. 11

Variation of Q×f value as function of x value

https://static-content.springer.com/image/art%3A10.1007%2Fs10832-007-9126-2/MediaObjects/10832_2007_9126_Fig12_HTML.gif
Fig. 12

Variation of temperature coefficient of resonant frequency τf as function of x

Conclusions

The structural evolution, and microwave dielectric properties of \( {\text{Sm}}_{{{{\left( {2 - \times } \right)}} \mathord{\left/ {\vphantom {{{\left( {2 - \times } \right)}} 3}} \right. \kern-\nulldelimiterspace} 3}} {\text{Li}}_{ \times } {\text{TiO}}_{3} \) ceramics (x = 0.0 ≤ x ≤ 0.5) were investigated in this work. In conclusion, samples with x > 0.3 exhibit a single perovskite phase. Impurity phases of Sm2Ti2O7 and TiO2 appear and their amount increases with the decrease of x when x ≤ 0.3. The A-site ordering degree increases with the increase of lithium content, and the sample was more heavily twinned and defected with the decrease of lithium content. Dielectric constant decreases slightly up to x = 0.4 and then increases with the further increase of x. The Q×f value decreases with the decrease of x due to the increased occurrence of second phase (Sm2T2O7), defects and twinning boundaries. All samples exhibit negative τf values and the absolute value of τf decreases almost linearly with decreasing x. A minimum ∣τf∣ value of −5 ppm/°C was obtained when x = 0.1.2

Acknowledgement

This work was supported by the National Science Foundation of China (NSFC), (project number: 50572060), partially by Key founding for basic research from he Shanghai Sience and Technology Committee (06JC14070) and by the Shanghai Pujiang Program(D).”

Copyright information

© Springer Science+Business Media, LLC 2007