Cite this Article
Cheng Yong-Guang, Mao Yan-Chao, Liu Xain-Sheng, Yuan Bao-He, Chao Ming-Ju, Liang Er-Jun. Near-zero thermal expansion of In2(1−x)(HfMg)xMo3O12 with tailored phase transition. Chinese Physics B, 2016, 25(8): 086501  
Permissions
Near-zero thermal expansion of In2(1−x)(HfMg)xMo3O12 with tailored phase transition
Cheng Yong-Guang1, 2, Mao Yan-Chao1, Liu Xain-Sheng3, Yuan Bao-He1, Chao Ming-Ju1, Liang Er-Jun1, †,
School of Physical Science & Engineering and Key Laboratory of Materials Physics of Ministry of Education of China, Zhengzhou University, Zhengzhou 450051, China
College of Science, Center of Analysis and Testing of Henan Institute of Engineering, Zhengzhou 451191, China
Key Laboratory of Photovoltaic Materials of Henan Province and School of Physics and Electronics, Henan University, Kaifeng 475004, China

 

† Corresponding author. E-mail: ejliang@zzu.edu.cn

Project supported by the National Natural Science Foundation of China (Grant Nos. 11574276, 51302249, and 51503185) and the Doctoral Fund of the Ministry of Education of China (Grant No. 20114101110003).

Abstract
Abstract

Solid solutions of In2(1−x)(HfMg)xMo3O12 are synthesized by solid state reaction with the aim to reduce the phase transition temperature of In2Mo3O12 and improve its thermal expansion property. The effects of (HfMg)6+ incorporation on the phase transition and thermal expansion are investigated. It is shown that the monoclinic-to-orthorhombic phase transition temperature obviously decreases and the coefficient of thermal expansion (CTE) of the orthorhombic becomes less negative and approaches to zero with increasing the content of (HfMg)6+. A near zero thermal expansion covering the case at room temperature (RT) is achieved for the solid solutions with x ≥ 0.85, implying potential applications of this material in many fields.

PACS: 61.50.Ks;65.40.De;64.60.–i;78.30.–j
Keyword:phase transition;thermal expansion;thermal analysis;Raman spectrum
1. Introduction

The negative thermal expansion (NET) phenomena in some materials have received considerable attention in the past few years.[1–7] NTE materials have the promising potential applications in tailoring the coefficient of thermal expansion (CTE) of materials to reduce thermal stress.[3,8–10] However, a series of NTE materials such as BiNiO3, LaCu3Fe4O12, and Mn3NX needs extreme synthesis conditions, like high pressure/temperature or exclusion of oxygen,[5,11–13] which restricts their applications in industry. Recently, some ceramics consisting of an open-framework with low thermal expansion or NTE[14–21] and low synthesis requirements, such as A2O(A = Ag/Cu),[18] ABO3 (A = Bi/Pb, B = Ni/Ti),[5] AMO5 (A = Ta/Nb, M = V/P),[15,16] AM2O7 (A = Zr/Hf, M = V/P),[17] AM2O8 (A = Zr/Hf, M = Mo/W)[1] and A2M3O12 (A is transition metal or rare earth, M is Mo or W) have been extensively investigated. The A2M3O12 family is extremely attractive due to its chemical flexibility within the orthorhombic Pbcn (No. 60) space group, and it possesses a variation of linear CTE from low positive to large negative values.[22–32] This orthorhombic feature of A2M3O12 makes it possible to be synthesized in solid solution state with controllable CTEs. Nevertheless, its heavy hygroscopicity[25–28] and abrupt change in thermal expansion property caused by phase transition from monoclinic to orthorhombic with increasing temperature[29–32] are still serious obstacles for practical applications of the A2M3O12 family.

In2Mo3O12 is a typical material of the A2M3O12 family with the monoclinic-to-orthorhombic phase transition temperature of about 608 K to 610 K.[31] Miller et al.[30] reported an In(HfMg)0.5Mo3O12 material and its monoclinic-to-orthorhombic structure transition occurring around 430 K. It means that the thermal expansion properties of In2Mo3O12 and In(HfMg)0.5Mo3O12 will be abruptly changed from giant positive to negative or near zero around their corresponding phase transition temperatures. The abrupt changes in thermal expansion deteriorate their mechanical properties and induce cracking, thus seriously hindering the applications of these materials. In order to reduce the phase transition temperature of In2Mo3O12, Cr3+ was reported to be introduced into In2Mo3O12,[32] but the phase transition temperature is still in a range of 523 K–673 K, which is much higher than room temperature (RT). It is because the phase transition (monoclinic-to-orthorhombic) temperature region of Cr2Mo3O12 is as high as 623 K–673 K.[29] Reducing the phase transition temperature of In2Mo3O12 to below room temperature (RT) and improving its thermal expansion property are still very attractive challenges.

In order to reduce the phase transition temperature of In2Mo3O12, the most effective method is to partially replace In3+ with a larger trivalent ion such as Y3+ or Yb3+. However, this may introduce crystal water at the same time such as in the cases of Yb2−xAlxMo3O12[33] and Fe2−xYxMo3O12.[27] Crystal water deteriorates the thermal expansion properties of materials because the releasing and admission of crystal water lead to serious expanding and contracting of the lattice during temperature circling. Besides, HfMgMo3O12 was shown to adopt an orthorhombic structure down to very low temperature.[34] It is expected that simultaneous doping with Hf4+ and Mg2+ into In2Mo3O12 can not only reduce the phase transition temperature but also avoid introducing crystal water. At the same time, charge balance in the formula is maintained. In this paper, solid solutions of In2(1−x)(HfMg)xMo3O12 are synthesized by solid state reaction in order to reduce the phase transition temperature of In2Mo3O12 and improve its thermal expansion property. It is found that the phase transition temperature decreases as the amount of (HfMg)6+ is increased and it is lower than RT when x ≥ 0.85. With further increasing the amount of (HfMg)6+, the near zero thermal expansion covering the case at RT is achieved so that the application scope of NTE materials can be further expanded.

2. Experiment

The starting materials of In2O3, MgO, HfO2, and MoO3 were mixed according to the mole ratios of the destination of materials of In2(1−x)(HfMg)xMo3O12 with x = 0.0, 0.1, 0.3, 0.5, 0.7, 0.75, 0.8, and 1.0. The mixtures were ground for 2 h in an agate mortar followed by sintering at 1073 K for 8 h in a furnace.

XRD measurements were carried out with a Bruker D8 Advance x-ray diffractometer. Raman spectra were recorded with a Renishaw inVia Raman spectrometer with 532-nm laser wavelength excitation. The differential scanning calorimeter (DSC) curves were obtained by the simultaneous thermal analyzer of Netzsch STA 449F3 in a temperature range of 298 K–673 K, under the 50 mL/min nitrogen (N2) purge, with a heating rate of 10 K/min. The linear CTE was measured with a dilatometer of Linseis DIL L76 in an atmosphere environment and with a heating rate of 5 K/min.

3. Results and discussion
3.1. The phase transition through the introduction of (HfMg)6+ component

Figure 1 shows the x-ray diffraction patterns of the In2(1−x)(HfMg)xMo3O12 solid solutions when x = 0, 0.1, 0.3, 0.5, 0.7, 0.9, and 1.0. With increasing the amount of (HfMg)6+ component, the diffraction peaks increase but shift first to higher degrees and then to lower degrees. The crystal structures can be identified as being monoclinic for x ≤ 0.7 and orthorhombic for x ≥ 0.9 from the XRD patterns.[22,31]

According to previous reports, the Raman spectra are sensitive to the monoclinic-to-orthorhombic phase transition.[27,35,36] In order to confirm the phase transition after the introduction of (HfMg)6+, Raman spectra of the solid solutions of In2(1−x)(HfMg)xMo3O12 are measured and shown in Fig. 2. The peak located at about 995 cm−1 originates from the monoclinic structure.[27,35,36] This Raman band decays gradually with increasing the content of (HfMg)6+ and disappears finally at about x = 0.85, suggesting a transition from the monoclinic to orthorhombic phase occurring at this composition ratio.

Fig. 1. The x-ray diffracion patterns of the solid solutions of In2(1−x)(HfMg)xMo3O12.
Fig. 2. The Raman spectra of the solid solutions of In2(1−x)(HfMg)xMo3O12.
3.2. Phase transition through increasing temperature

Figure 3(a) shows the DSC curves of In2(1−x)(HfMg)x Mo3O12. An endothermic peak appears at 616 K when x = 0, indicating the monoclinic-to-orthorhombic phase transition occurring at this temperature. The endothermic peak shifts to the lower temperature with increasing the amount of (HfMg)6+. When x = 0.5, the endothermic peak appears at 425 K, which is close to the reported value of 430 K. With increasing the content of (HfMg)6+, the endothermic peak becomes progressively weak so that the endothermic peak of the phase transition for x ≥ 0.7 cannot be observed clearly as indicated by the arrow in Fig. 3(a). Through the fitting data for x = 0, 0.1, 0.3, and 0.5 as shown in Fig. 3(b), it can be deduced that a room-temperature orthorhombic phase requires x ≈ 0.84. This accords well with the XRD and Raman results.

Fig. 3. (a) The DSC of the solid solutions of In2(1−x)(HfMg)xMo3O12 and (b) linear fit and interpolation of the phase transition temperature vis (HfMg)6+ content.

The progressively weakening and shifting to lower temperature of the endothermic peak indicate that the monoclinic-to-orthorhombic phase transition requires less energy with increasing the content of (HfMg)6+, the physics behind which could be understood by the difference in electronegativity between (HfMg)6+ and In3+. The electronegativity values of Hf3+ (1.30) and Mg2+ (1.31) are smaller than that of In3+ (1.78). With increasing the content of (HfMg)6+, the bridging oxygen connected with Hf4+ or Mg2+ gains more negative partial charge, which enhances the oxygen-oxygen repulsion force and makes the monoclinic-to-orthorhombic structure transition easier. When the sample with a higher content of (HfMg)6+ is heated, its phase transition requires a lower temperature and less heat, resulting in the weakening and shifting of the endothermic peak.

3.3. Phase transition and near-zero thermal expansion

Figure 4 shows the variations of relative length of the In2(1−x)(HfMg)xMo3O12 solid solutions with temperature. For In2Mo3O12 (x = 0.0), its relative length increases slowly with temperature increasing from RT to 590 K, and then expands abruptly from 590 K to 623 K. The result indicates that the monoclinic-to-orthorhombic phase transition starts at about 590 K and ends at about 623 K, which is consistent with the phase transition temperature of the DSC result. The monoclinic-to-orthorhombic phase transition temperature decreases obviously with the increase of the amount of (HfMg)6+, and the phase transition occurs at about 326 K for x = 0.75. The phase transition temperature even decreases to lower than RT when x ≥ 0.85. It is worth noting that the CTEs of the orthorhombic phase become less negative and approach to zero with increasing the amount of (HfMg)6+. A near zero thermal expansion for each of the solid solutions with x ≥ 0.8 is achieved from RT to about 773 K.

Fig. 4. Variations of relative length of the solid solutions of In2(1−x)(HfMg)xMo3O12 with temperature.

The mechanisms for the near-zero CTE can be understood by the cation size effect of the six-coordinated In3+, Hf4+, and Mg2+ in the center of octahedron. As is well known, the NTE increases with a larger A3+ cation size for the A2M3O12 family in the orthorhombic structure. The cation radius of In3+ is 0.080 nm while the average size of (HfMg)6+ is 0.0715 nm. The In3+ cation size plays a decisive role for the NTE in In2Mo3O12 above 623 K. However, with increasing the substitution amount of (HfMg)6+ for In3+, the NTE is reduced and even down to near-zero CTE due to the smaller average cation size. On the other hand, the thermal expansion of the In(HfMg)0.5M3O12 is anisotropic,[30] the a- and c-axis contract while the b-axis expands on heating. The two factors result in the average thermal expansion of the cell volume approaching to zero.

4. Conclusions

Solid solutions of In2(1−x)(HfMg)xMo3O12 are synthesized and the effects of (HfMg)6+ incorporation on the phase transition and thermal expansion are investigated. It is shown that the monoclinic-to-orthorhombic phase transition temperature obviously decreases with increasing the content of (HfMg)6+. The CTE of the orthorhombic structure becomes less negative and approaches to zero with increasing the content of (HfMg)6+. A near zero thermal expansion over the range from RT to 773 K is achieved for the solid solutions with x ≥ 0.8. The In2(1−x)(HfMg)xMo3O12 solid solutions with near-zero thermal expansion property covering the case at RT hold great potential applications in many fields.

Reference
1Mary T AEvans J S OVogt TSleight A W 1996 Science 272 90
2Grigoriadis CHaase NButt H JMullen KFloudas G 2010 Adv. Mater. 22 1403
3Evans J S O 1999 J. Chem. Soc. Dalton Trans. 19 3317
4Suzuki TOmote A 2006 J. Am. Ceram. Soc. 89 691
5Azuma MChen WSeki HCzapski MOlga SOka KMizumaki MWatanuki TIshimatsu NKawamura NIshiwata STucker M GShimakawa YAttfield J P 2011 Nat. Commun. 2 347
6Mohn P 1999 Nature 400 18
7Chen JHu LDeng J XXing X R 2015 Chem. Soc. Rev. 44 3522
8Atkinson ABarnett SGorte R JIrvine J T SMcEvoy A JMogensen MSinghal S CVohs J 2004 Nat. Mater. 3 17
9Haertling G H 1999 J. Am. Ceram. Soc. 82 797
10Ravi VFirdosy SCaillat TBrandon Evan der Walde KMaricic LSayir A 2009 J. Electron. Mater. 38 1433
11Long Y WHayashi NSaito TAzuma MMuranaka SShimakawa Y 2009 Nature 458 60
12Ding LWang CWen Y CChu L HNa Y YYan J 2010 Acta Phys. Sin. 60 097507 (in Chinese)
13Tong PWang B SSun Y P 2013 Chin. Phys. 22 067501
14Marinkovic B AJardim P MSaavedra ALau L YBaehtz Cde Avillez R RRizzo F 2004 Microporous Mesoporous Mater. 71 117
15Amos T GYokochi ASleight A W 1998 J. Solid State Chem. 141 303
16Wang X WHuang Q ZDeng J XYu R BChen JXing X R 2011 Inorg. Chem. 50 2685
17Yuan B HYuan H LSong W BLiu X SCheng Y GChao M JLiang E J 2014 Chin. Phys. Lett. 31 076501
18Kennedy B JKubota YKato K 2005 Solid State Commun. 136 177
19Liang YXing H ZChao M JLiang E J 2014 Acta Phys. Sin. 63 248106 (in Chinese)
20Evans J S OMary T A 2000 Int. J. Inorg. Mater. 2 143
21Gates S DColin J ALind C J 2006 J. Mater. Chem. 16 4214
22Marinkovic B AJardim P MAri Mde Avillez R RRizzo FFerreira F F 2008 Phys. Stat. Sol. 245 2514
23Song W BLiang E JLiu X SLi Z YYuan B HWang J Q 2013 Chin. Phys. Lett. 30 126502
24Wu M MPeng JCheng Y ZWang HYu Z XChen D FHu Z B 2006 Solid State Sci. 8 665
25Sumithra SUmarji A M 2004 Solid State Sci. 6 1313
26Liang E JHuo H LWang J PChao M J 2008 J. Phys. Chem. 112 6577
27Li Z YSong W BLiang E J 2011 Phys. Chem. 115 17806
28Sumithra SUmarji A M 2005 Mater. Res. Bull. 40 167
29Tyagi A KAchary S NMathews M D 2002 J. Alloys Compd. 339 207
30Miller K JRomao C PBieringer MMarinkovic B APrisco LWhite M A 2013 J. Am. Ceram. Soc. 96 561
31Marinkovic B AAri MJardim P Mde Avillez R RRizzo FFerreira F F 2010 Thermochim. Acta 499 48
32Wu M MZu YPeng JLiu R. D.Hu Z BLiu Y TChen D F 2012 Cryst. Res. Technol. 47 793
33Li Q JYuan B HSong W BLiang E JYuan B 2012 Chin. Phys. 21 046501
34Miller K JJohnson M BWhite M AMarinkovic B A 2012 Solid State Commun. 152 1748
35Song W BWang J QLi Z YLiu X SYuan B HLiang E J 2014 Chin. Phys. 23 066501
36Song W BYuan B HLiu X SLi Z YWang J QLiang E J 2014 J. Mater. Res. 29 849