- Email: [email protected]
Preparation for ultra-low loss dielectric ceramics of ZnZrNb2O8 by reaction-sintering process
Accepted Manuscript Preparation for utral-low loss dielectric ceramics of ZnZrNb2O8 by reaction-sintering process Yonggui Zhao, Ping Zhang PII:
S0925-8388(16)30440-6
DOI:
10.1016/j.jallcom.2016.02.171
Reference:
JALCOM 36789
To appear in:
Journal of Alloys and Compounds
Received Date: 5 January 2016 Revised Date:
17 February 2016
Accepted Date: 19 February 2016
Please cite this article as: Y. Zhao, P. Zhang, Preparation for utral-low loss dielectric ceramics of ZnZrNb2O8 by reaction-sintering process, Journal of Alloys and Compounds (2016), doi: 10.1016/ j.jallcom.2016.02.171. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Graphical Abstract:
ACCEPTED MANUSCRIPT Preparation for utral-low loss dielectric ceramics of ZnZrNb2O8 by reaction-sintering process Yonggui Zhao, Ping Zhang* School of Electronic and Information Engineering and Key Laboratory of Advanced Ceramics and
RI PT
Machining Technology of Ministry of Education, Tianjin University, Tianjin 300072, P. R. China Abstract
Microwave dielectric properties and microstructures of novel high-Q ceramics ZnZrNb2O8 prepared
SC
by new sintering method of reaction-sintering method (RS) and conventional solid-state reaction method
M AN U
(CS) have been investigated. The XRD patterns and SEM reveal that both of RS and CS methods present a single phase of ZnZrNb2O8 and uniform morphology at their optimal sintering temperature. In order to evaluate the effects of difference sintering methods on the microwave dielectric properties of ZnZrNb2O8 ceramics, the theoretical dielectric constant, packing fraction and bond valence were calculated based on
TE D
the Rietveld refinement method. The calculation results reveal that the RS method was more beneficial to the microwave dielectric properties of ZnZrNb2O8 ceramics in contrast with CS method. Excellent microwave dielectric properties for ZnZrNb2O8 ceramics with enhanced high Q×f value of 76, 400 GHz
EP
could be obtained using RS method sintered at 1175 oC for 4h.
AC C
Keywords: Microwave dielectric properties; Ceramics; High-Q; Packing fraction; Reaction-sintering 1. Introduction
The remands for microwave dielectric ceramics with high performance in the recent wireless
industry like microwave communication and intelligent transport systems is greatly increasing than before [1-3]. A vital factor for microwave dielectric ceramics was possessed a low dielectric loss to reduce the 1 ___________________________ *Corresponding author. Tel. : +86 13702194791 Email address: [email protected] (P. Zhang)
ACCEPTED MANUSCRIPT delays of signal transmissions delays. ZnZrNb2O8 is one of most popular microwave dielectric materials due to its fine properties with εr of 30, Q×f of 61, 000 GHz and τf of -52 ppm/oC [4]. Latter, Ramarao et al [5]. investigated the crystal
RI PT
structure refinement and microwave dielectric properties of similar family materials AZrNb2O8 (A: Mn, Mg, Zn and Co) and reported that the ZnZrNb2O8 ceramics possessed εr=16.5, Q×f=53, 400 GHz and τf= -49.8 ppm/oC. In order to meet the application for LTCC, Tang et al [6]. succeed in lowering the sintering
SC
temperature of ZnZrNb2O8 ceramic and an maximum Q׃ value of 56, 720 GHz could be obtained from
M AN U
ZnZrNb2O8 ceramics doped with 3 wt.% BaCu(B2O5) additions. Recently, Li et al [7-12]. investigated the effects of ions substitution on the microwave dielectric properties of ZnZrNb2O8 ceramics, and discovered that Zn0.95Ni0.05ZrNb2O8 ceramics exhibit a high Q׃ value of 83, 558 GHz. In our recent work [13], the effects of pentavalent ions substitution on the bond ionicity, lattice energy and bond energy for the
TE D
ZnZrNb2O8 were discussed, and with 0.08 mol Sb5+ substitution for Nb5+ could greatly enhance the Q׃ value of ZnZrNb2O8 ceramics. RS method has been attracted more attentions for its high effectively by avoiding calcination and second milling process. Several materials, such as MgNb2O6, ZnNb2O6, MgTiO3,
EP
Li2MgTi3O8 and ZnTiNb2O8 ceramics [14-17], have been successfully prepared through this method.
AC C
Although several works have been studied on the ZnZrNb2O8 ceramics, prepared low loss dielectric materials of ZnZrNb2O8 ceramics using RS method have not been reported so far. Therefore, the purpose of this paper is to prepare a much high-Q value material of ZnZrNb2O8
ceramics using a sample method. Besides, the effects of two different methods (RS and CS methods) on the microwave dielectric properties and microstructures are also discussed. Based on the Rietveld refinement method, the theoretical dielectric constant, packing fraction and bond valence were calculated
2
ACCEPTED MANUSCRIPT to evaluate the structural characteristics of ZnZrNb2O8 ceramics. In addition, the X-ray diffraction (XRD) pattern and the scanning electron microscopy (SEM) analysis were also used to analyze the crystal structure and the microstructures.
RI PT
2. Experimental procedure The ZnZrNb2O8 powders were prepared by mixing the raw materials according to desired stoichiometry. High-purity oxide powders of the starting materials are listed as follows: ZnO (99.9%),
SC
ZrO2 (99.0%) and Nb2O5 (99.9%). For the RS method, the mixed powders were milled for 10 h with
M AN U
distilled water in a nylon container with ZrO2 balls. After drying, all the slurries were crushed and sieved with a 40-mesh screen firstly, then with 6 wt% paraffin as a binder, granulated by sieved with an 80 mesh screen. Latter, the sieved powders were pressed into disk-type pellets with 10 mm diameter and 5 mm thickness at 100 MPa. While for the CS method [13], the mixed powders were milled 6 h firstly. Then
TE D
slurries were dried and pre-sintered at 1100 oC for 4 h and re-milled for 8 h before granulated and pressed. All these pellets were sintered at temperatures range 1125 oC -1300 oC for 4 h in air. The details process for RS method and CS method are given in Fig.1. Obvious that the RS method has great advantages over
EP
CS method in simply preparation process.
AC C
The crystalline phases of the sintered samples were identified by X-ray diffraction (XRD, Rigaku D/max 2550 PC, Tokyo, Japan) with Cu Kα radiation generated at 40 kV and 40 mA. The microstructure of the ceramic surfaces were performed and analyzed by a scanning electron microscopy (SEM, MERLIN Compact, Germany). The microwave dielectric properties were measured in the frequency range of 6-13 GHz by the TE01σ shielded cavity method [18]. The temperature coefficient of resonant frequency (τƒ) was measured in the temperature range from 25 oC to 85 oC and was calculated by the following formula:
3
ACCEPTED MANUSCRIPT f85 − f 25 ×106 (ppm/oC) f 25 (85 − 25)
τf =
(1)
where ƒ85 and f25 were the TE01σ resonant frequencies at 85oC and 25oC, respectively. The apparent densities of the sintered pellets were measured using Archimedes method (Mettler
ρt heory =
RI PT
ToledoXS64). To study the relative density of the sample, the theoretical density was defined as follows:
ZA VC N A
(2)
where VC, NA , Z, and A are volume of unit cell (cm3), avogadro number (mol-1), number of atoms in unit
ρ relative =
ρ ρ
bulk
SC
cell, and atomic weight (g/mol), respectively. The relative density was obtained by the Eq. (3): ×100%
(3)
M AN U
theory
3. Results and discussion
Fig.2 shows the XRD patterns of ZnZrNb2O8 ceramics prepared using RS method sintered at 1125 oC -1225 oC and CS method sintered at 1200 oC -1300 oC, respectively. Notice that Fig.2 (a) presents a
TE D
monoclinic-structured phase of ZnZrNb2O8 (PDF file no. 50-1725, space group C2/c no. 15) and litter second phase of Zn3Nb2O8 (PDF file no. 50-1725) using RS method. However, all the reflections in Fig.2 (b) are well matched with ZnZrNb2O8 phase, and no any secondary phases was observed. The formation
EP
of second phase for the RS method was due to the missing calcination in the sintering process. Therefore,
AC C
it is difficult to form pure ZnZrNb2O8 phase at lower sintering temperature. Moreover, the diffraction peaks of the sintered samples slightly shift to lower angle direction with the temperature increased both RS and CS method, which indicates the decreasing of the cell volume [20-21]. In order to investigate the effects of different sintering methods on the crystal structures and microwave dielectric properties for the ZnZrNb2O8 ceramics, the Rietveld refinement were carried out using the software of Full-prof based on the XRD data. Fig.3 shows the structural refinement patterns of the ZnZrNb2O8 ceramics using different
4
ACCEPTED MANUSCRIPT sintering methods. The details about the lattice parameters and bond lengths of all the samples are given in Table.1 and Table.2. Observed that the cell volume of RS method was smaller than CS method, which is the key factor to the microwave dielectric properties.
RI PT
The microstructures of the sintered samples ZnZrNb2O8 ceramics using RS and CS method at different sintering temperature are presented in Fig. 4. Notice that with the increasing of sintering temperature, grain size for both two different methods have an increasing tendency. Homogeneous grain
SC
and smooth surface were obtained at 1175 oC for RS method and at 1250 oC for CS method, respectively.
M AN U
In addition, all of the grain size was in the range of 5-7 µm, and well-densified SEM photos could be seen in Fig.4. With a further increase of the sintering temperature, no significant change was observed for RS method. However, abnormal grain growth and some pores was found sintered at 1300 oC for CS method, which indicates the sintering temperature for CS method have more influence on the microwave dielectric
TE D
than the RS method.
Fig.5 shows the relative densities and microwave dielectric properties of ZnZrNb2O8 ceramics using RS and CS method as a function of sintering temperature. Notice that both of the two different methods
EP
exhibit an increasing tendency with the sintering temperature increased. After reaching their maximum of
AC C
97.58 % at 1175 oC for RS specimens and 96.37 % at 1250 oC for CS specimens, respectively, the relative densities started to decrease. The high densification of specimens is due to the high apparent densities and well-grain growth, which would be demonstrated from Eq.(3) and Fig.4. In generally, the microwave dielectric properties at different sintering temperature have close relationship with the relative density, grain boundaries, pores and phases compositions [22-23]. In this paper, the εr and Q×f values of the two different method prepared specimens present a similar tendency with the relative densities. This is
5
ACCEPTED MANUSCRIPT because that with the relative densities increasing, the grain boundaries and pores (εr=1) would decrease, therefore, the dielectric constant increased and dielectric loss decreased. In this paper, the sintering temperature have no direct relationship with the τf value due to the single phase for the sintered specimens
RI PT
and no other extra additives. The τf value reached a maximum of -48.61 ppm/oC at 1175 oC for RS method and of -51.36 ppm/oC at 1250 oC for CS method, respectively. It suggests that RS method could help to enhance the microwave dielectric properties of ZnZrNb2O8 ceramics.
SC
In order to evaluate the intrinsic factors between the RS method and CS method, in the following
M AN U
study, the theoretical dielectric constant, packing fraction and bond valence was calculated. Based on the previous works [24-26], the theoretical dielectric constant εtr and ionic polarizabilities αD have the following relationship:
ε tr =
(Vm + 2bα D ) (Vm − bα D )
(4)
TE D
where Vm is the molar volume; b has the value of 4π/3. According to the additivity rule of molecular polarizabilities, αD for the ZnZrNb2O8 compounds can be broken up as follows: αD(ZnZrNb2O8)=αD(Zn2+)+αD(Zr4+)+αD(Nb5+)+8αD(O2-)
(5)
AC C
formula:
EP
The packing fraction, which is a vital factor to the Q×f values, is calculated using the following
Packing fraction ( % ) ==
3 3 4π / 3 × ( rZn + rZr3 + 2 × rNb + 8 × rO3 )
Vcell
(6)
Kityk et al.[27-28] reported the niobates principal role in ABO4 compounds play intrinsic cationic
defects, and the structural characteristics of B-site oxygen octahedral have closely relation with the τƒ value. The structural characteristics of B-site oxygen octahedral could be characterized by the B-site bond valence, which could be obtained as follows: 6
ACCEPTED MANUSCRIPT Vi = ∑ vij j
vij = exp(
(7)
Rij - dij ) b'
(8)
where Vi was defined as the sum of all of the valences from a given atom i, Rij was bond valence
RI PT
parameter and was equal to 1.911, dij was the distance between atom i and j, and b′ was 0.37.
Table.3 gives the calculated results of theoretical dielectric constant, packing fraction and Nb-site
SC
bond valence for the two different method sintered at their optimal sintering temperature. Notice that the εr was bigger than εtr, which is because that εtr was actually the real part of the complex dielectric constant
M AN U
in the Clausius-Mosotti equation. For the high-relative density specimens, the εr was mainly dependent on the ionic polarizability. Therefore, we just calculated the theoretical dielectric constant caused by the ionic polarizability part. In this paper, the εtr for the specimens prepared different method keep similar tendency. According to Table.3, a higher εtr value of 23.94 for RS method prepared sample was obtained compared
TE D
to εtr value 23.81 for CS method, therefore, the εr for RS method would be higher than the CS method. According to Kim et al [29]., when the relative density of the samples was over 95%, the Q×f values
EP
are mainly dependent on the intrinsic losses caused by packing fraction. With an increase in the packing fraction, the lattice vibrations would decrease. As Table.3 shows, higher Q×f values of 76,400 GHz
AC C
correlate with higher packing fraction of 64.65% and lower Q×f values of 65,100 GHz correspond to smaller packing fraction of 64.61%. Based on the results of SEM images for RS methods, the grain size showed more uniform morphology, which was another reason for the increase of the Q×f values. In combination with the analysis of grain growth and packing fraction, it is could be explain why the RS-prepared samples possess higher Q×f values. In general, the temperature coefficient of resonant frequency τf value was influenced by the 7
ACCEPTED MANUSCRIPT composition, additives and second phases [30-33]. In this paper, the above factors were not consideration in the τf values for the RS method and CS method, which was due to the single phase of specimens and no other extra additives. Therefore, the τf values could be characterized by the Nb-site bond valence VNb.
RI PT
With the increasing of VNb, the |τf| value would decrease, which was because that higher bond valence would led to a higher restoring force, the restoring force increased, the system would be more system. As Table.3 shows, the VNb of 5.0144 for the RS method is bigger than the 4.4644 for the CS method. The VNb
SC
increased, |τf| value would decrease.
M AN U
In summary, fine microwave dielectric properties of ZnZrNb2O8 ceramics with εr = 28.99, Q×f = 76,400 GHz and τf = -48.61 ppm/oC could be obtained at 1750 oC for 4 h prepared by RS method, and with a εr of 26.70, Q×f value of 65,100 GHz and τf value of-51.36 ppm/oC for the CS-prepared samples could be obtained at 1250 oC for 4 h, respectively.
TE D
4. Conclusions
ZnZrNb2O8 ceramics were successfully prepared with excellent microwave dielectric properties via reaction-sintering method for the first time. Single monoclinic structured phase of ZnZrNb2O8 was
EP
obtained from the RS and CS methods sintered at their optimal sintering temperature. Based on the
AC C
Rietveld refinement methods, the εr of RS sample is much higher than the CS sample due to the high εtr in RS sample. The enhance Q×f value for RS prepared specimen could be contribution to the higher packing fraction. For single phase samples, the Nb-site bond valence plays an important role in τf values. A fine microwave dielectric properties (εr ~ 28.99, Q×f value ~ 76,400 GHz and τf value ~ -48.61 ppm/oC) was obtained for RS-prepared sample sintered at 1175 oC for 4 h. It suggests that reaction-sintering method could be a simple and efficient method to produce ZnZrNb2O8 ceramics with enhanced microwave
8
ACCEPTED MANUSCRIPT dielectric properties. Acknowledgments The authors gratefully acknowledged supports from the Key Laboratory of Advanced Ceramics and
RI PT
Machining Technology, Ministry of Education (Tianjin University). References
[1] M.T. Sebastian, Dielectric materials for wireless communications, Elsevier Publishers, Oxford, 2008.
SC
[2] R. D. Richtmeyer, Dielectric resonators, J. Appl. Phys. 15 (1939) 391-398.
M AN U
[3] P. Zhang, Y. G. Zhao, J. Liu, Z. K. Song, X. Y. Wang, M. Xiao, J. Alloy. Compd. 650 (2015) 90-94. [4] Q.W. Liao, L.X. Li, X. Ren, X. X. Yu, D. Guo, M. J. Wang, J. Am. Ceram. Soc. 95 (2012) 3363-3365. [5] S.D. Ramarao, V. R. K. Murthy, Scripta. Mater. 69 (2013) 274-277.
[6] X. Tang, H. Yang, Q. L. Zhang, J. H. Zhou, Ceram. Int. 40 (2014) 12875-12881.
TE D
[7] L. X. Li, H. Sun, H. C. Cai, X. S. Lv, J. Alloy. Compd. 639 (2015) 516-519. [8] J. Ye, L. X. Li, H. Sun, X. S. Lv, J. Y. Yu, S. Li, J. Mater. Sci.: Mater. Eletron. 26 (2015) 8954-8959. [9] J. Ye, L. X. Li et al., J. Mater. Sci.: Mater. Eletron. 27 (2016) 97-102.
EP
[10] L. X. Li, J. Ye, et al., J. Mater. Sci.: Mater. Eletron. 27 (2016) 1232-1238.
AC C
[13] P. Zhang, Y. G. Zhao, H .T. Wu, Dalton. Trans. 44 (2015) 16684-16693. [14] Y. C. Liou, Y. L. Sung, Ceram. Int. 34 (2008) 371-377. [15] L. X. Li, X. Ding, Q. W. Liao, J. Alloy. Compd. 509 (2011) 7271-7276. [16] G. G. Yao, P. Liu, H. W. Zhang, J. Mater. Sci.: Mater. Electron. 24 (2013) 1128-1131. [17] H. B. Bafrooei, E. T. Nassaj, T. Ebadzadeh, et al., Ceram. Int. 42 (2016) 3296-3303. [18] W. E. Courtney. IEEE Trans Microwave Theory Tech. 18 (1970) 476-485.
9
ACCEPTED MANUSCRIPT [19] D. Zhou, L. X. Pang, J. Guo, et al., Inorg. Chem. 53 (2014) 1048-1055. [20] K. Nishiyama, T. Abe , T. Sakaguchi , N. Momozawa, J. Alloy. Compd. 355 (2003) 103-107. [21] P. Zhang, Y. G. Zhao, X. Y. Wang, J. Alloy. Compd. 644 (2015) 621-625.
RI PT
[22] D. Zhou, L. X. Pang, J. Guo, et al., Inorg. Chem. 50 (2011) 12733-12738. [23] D. Zhou, L. X. Pang, J. Guo, et al,, Inorg. Chem. 53 (2014) 53 1048-1055.
[25] Y. G. Zhao, P. Zhang, J. Alloy. Compd. 658 (2016) 744-748.
M AN U
[26] Y. G. Zhao, P. Zhang, J. Alloy. Compd. 662 (2016) 455-460.
SC
[24] Y. G. Zhao, P. Zhang, RSC. Adv. 5 (2015) 97746-97754.
[27] I. V. Kityk, M. M. Janusik, J. Phys. Chem B. 105 (2001) 12242-12248 [28] M. Piasecki, I. V. Kityk, J. Alloy. Compd. 639 (2015) 577-582.
[29] E. S. Kim, B. S. Chun, R. Freer, R. J. Cernik, J. Eur, Ceram. Soc. 30 (2010) 1731-1736.
TE D
[30] Q. J. Mei, C.Y. Li, J.D. Guo, H.T. Wu, J. Alloy. Compd., 626 (2015) 217-222. [31] H.T. Wu, Z.B. Feng, Q.J. Mei, J.D. Guo, J.X. Bi, J. Alloy. Compd. 643 (2015) 368-373. [32] H.T. Wu, Q.J. Mei, J. Alloy. Compd. 651 (2015) 393-398.
AC C
EP
[33] H. L. Pan, Z.B. Feng, J.X. Bi, H.T. Wu, J. Alloy. Compd. 651 (2015) 440-443.
10
ACCEPTED MANUSCRIPT Table.1 Refined lattice parameters (a, b, c, β and V) and discrepancy factors (Rp and Rwp) of ZnZrNb2O8 ceramics using different sintering process. Sintering Method
a (Å)
b( Å)
c( Å)
β(°)
1125 oC
4.8150
5.6769
5.0795
91.45
1150 oC
4.8153
5.6774
5.0791
91.45
1175 oC
4.8150
5.6778
5.0799
91.45
1200 oC
4.8154
5.6825
5.0806
1225 oC
4.8137
5.6828
5.0795
1200 oC
4.8189
5.6724
1225 oC
4.8160
1250 oC
4.8247
1275 oC
V(Å3)
Rp(%)
Rwp(%)
138.81
10.40
14.90
138.83
9.76
13.00
SC
11.40
9.18
11.70
91.41
138.91
9.66
12.30
5.0806
91.49
138.83
6.94
9.01
5.6768
5.0807
91.43
138.86
7.12
9.35
5.6654
5.0843
91.56
138.92
11.70
14.70
4.8163
5.6799
5.0823
91.42
138.99
8.53
10.60
4.8287
5.6670
5.0851
91.52
139.10
11.30
14.40
EP
1300 oC
8.61
138.98
M AN U
CS
138.80
91.43
TE D
RS
RI PT
Temperature
Table.2 The refined bond length (Å) of ZnZrNb2O8 ceramics using RS and CS methods. Zn/Zr-O(1)×2
Zn/Zr-O(2)1×2
Zn/Zr-O(2)2×2
Nb-O(1)1×2
Nb-O(1)2×2
Nb-O(2)×2
RS (1175 oC)
1.9772
2.2673
2.1707
2.0367
2.1538
1.8207
CS (1250 oC)
1.9348
2.2356
2.1647
2.0550
2.2755
1.8497
AC C Method
Table.3 The measured microwave dielectric properties, and calculated theoretical dielectric constant, packing fraction and Nb-site bond valence of ZnZrNb2O8 ceramics using RS and CS methods.
ACCEPTED MANUSCRIPT Measured
Calculated
Method Q×f(GHz)
τf(ppm/oC)
εtr.
P.F .%
VNb
RS (1175 oC)
28.99
76400
-48.61
23.94
64.65
5.0144
CS (1250 oC)
26.70
65100
-51.36
23.81
64.61
4.4644
AC C
EP
TE D
M AN U
SC
RI PT
εr
ACCEPTED MANUSCRIPT Figure captions Fig.1 Reaction-sintering and conventional solid-state reaction sintering process for ZnZrNb2O8 ceramics.
RI PT
Fig.2 The XRD patterns of ZnZrNb2O8 ceramics using (a): RS method sintered at 1125 oC-1225 oC; (b): CS method sintered at 1200 oC-1300 oC.
Fig.3 The structural refinement patterns of the ZnZrNb2O8 ceramics (a): RS method at 1175 oC; (b): CS
SC
method at 1250 oC.
M AN U
Fig.4 SEM graphs of ZnZrNb2O8 ceramics using RS and CS method.
Fig.5 The relative densities, εr values, Q×f values and τf values of ZnZrNb2O8 ceramics using RS and
AC C
EP
TE D
CS method as a function of sintering temperature.
Fig.1
Fig.2
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig.3
AC C
EP
TE D
M AN U
Fig.4
SC
RI PT
ACCEPTED MANUSCRIPT
Fig.5
S0925-8388(16)30440-6
DOI:
10.1016/j.jallcom.2016.02.171
Reference:
JALCOM 36789
To appear in:
Journal of Alloys and Compounds
Received Date: 5 January 2016 Revised Date:
17 February 2016
Accepted Date: 19 February 2016
Please cite this article as: Y. Zhao, P. Zhang, Preparation for utral-low loss dielectric ceramics of ZnZrNb2O8 by reaction-sintering process, Journal of Alloys and Compounds (2016), doi: 10.1016/ j.jallcom.2016.02.171. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Graphical Abstract:
ACCEPTED MANUSCRIPT Preparation for utral-low loss dielectric ceramics of ZnZrNb2O8 by reaction-sintering process Yonggui Zhao, Ping Zhang* School of Electronic and Information Engineering and Key Laboratory of Advanced Ceramics and
RI PT
Machining Technology of Ministry of Education, Tianjin University, Tianjin 300072, P. R. China Abstract
Microwave dielectric properties and microstructures of novel high-Q ceramics ZnZrNb2O8 prepared
SC
by new sintering method of reaction-sintering method (RS) and conventional solid-state reaction method
M AN U
(CS) have been investigated. The XRD patterns and SEM reveal that both of RS and CS methods present a single phase of ZnZrNb2O8 and uniform morphology at their optimal sintering temperature. In order to evaluate the effects of difference sintering methods on the microwave dielectric properties of ZnZrNb2O8 ceramics, the theoretical dielectric constant, packing fraction and bond valence were calculated based on
TE D
the Rietveld refinement method. The calculation results reveal that the RS method was more beneficial to the microwave dielectric properties of ZnZrNb2O8 ceramics in contrast with CS method. Excellent microwave dielectric properties for ZnZrNb2O8 ceramics with enhanced high Q×f value of 76, 400 GHz
EP
could be obtained using RS method sintered at 1175 oC for 4h.
AC C
Keywords: Microwave dielectric properties; Ceramics; High-Q; Packing fraction; Reaction-sintering 1. Introduction
The remands for microwave dielectric ceramics with high performance in the recent wireless
industry like microwave communication and intelligent transport systems is greatly increasing than before [1-3]. A vital factor for microwave dielectric ceramics was possessed a low dielectric loss to reduce the 1 ___________________________ *Corresponding author. Tel. : +86 13702194791 Email address: [email protected] (P. Zhang)
ACCEPTED MANUSCRIPT delays of signal transmissions delays. ZnZrNb2O8 is one of most popular microwave dielectric materials due to its fine properties with εr of 30, Q×f of 61, 000 GHz and τf of -52 ppm/oC [4]. Latter, Ramarao et al [5]. investigated the crystal
RI PT
structure refinement and microwave dielectric properties of similar family materials AZrNb2O8 (A: Mn, Mg, Zn and Co) and reported that the ZnZrNb2O8 ceramics possessed εr=16.5, Q×f=53, 400 GHz and τf= -49.8 ppm/oC. In order to meet the application for LTCC, Tang et al [6]. succeed in lowering the sintering
SC
temperature of ZnZrNb2O8 ceramic and an maximum Q׃ value of 56, 720 GHz could be obtained from
M AN U
ZnZrNb2O8 ceramics doped with 3 wt.% BaCu(B2O5) additions. Recently, Li et al [7-12]. investigated the effects of ions substitution on the microwave dielectric properties of ZnZrNb2O8 ceramics, and discovered that Zn0.95Ni0.05ZrNb2O8 ceramics exhibit a high Q׃ value of 83, 558 GHz. In our recent work [13], the effects of pentavalent ions substitution on the bond ionicity, lattice energy and bond energy for the
TE D
ZnZrNb2O8 were discussed, and with 0.08 mol Sb5+ substitution for Nb5+ could greatly enhance the Q׃ value of ZnZrNb2O8 ceramics. RS method has been attracted more attentions for its high effectively by avoiding calcination and second milling process. Several materials, such as MgNb2O6, ZnNb2O6, MgTiO3,
EP
Li2MgTi3O8 and ZnTiNb2O8 ceramics [14-17], have been successfully prepared through this method.
AC C
Although several works have been studied on the ZnZrNb2O8 ceramics, prepared low loss dielectric materials of ZnZrNb2O8 ceramics using RS method have not been reported so far. Therefore, the purpose of this paper is to prepare a much high-Q value material of ZnZrNb2O8
ceramics using a sample method. Besides, the effects of two different methods (RS and CS methods) on the microwave dielectric properties and microstructures are also discussed. Based on the Rietveld refinement method, the theoretical dielectric constant, packing fraction and bond valence were calculated
2
ACCEPTED MANUSCRIPT to evaluate the structural characteristics of ZnZrNb2O8 ceramics. In addition, the X-ray diffraction (XRD) pattern and the scanning electron microscopy (SEM) analysis were also used to analyze the crystal structure and the microstructures.
RI PT
2. Experimental procedure The ZnZrNb2O8 powders were prepared by mixing the raw materials according to desired stoichiometry. High-purity oxide powders of the starting materials are listed as follows: ZnO (99.9%),
SC
ZrO2 (99.0%) and Nb2O5 (99.9%). For the RS method, the mixed powders were milled for 10 h with
M AN U
distilled water in a nylon container with ZrO2 balls. After drying, all the slurries were crushed and sieved with a 40-mesh screen firstly, then with 6 wt% paraffin as a binder, granulated by sieved with an 80 mesh screen. Latter, the sieved powders were pressed into disk-type pellets with 10 mm diameter and 5 mm thickness at 100 MPa. While for the CS method [13], the mixed powders were milled 6 h firstly. Then
TE D
slurries were dried and pre-sintered at 1100 oC for 4 h and re-milled for 8 h before granulated and pressed. All these pellets were sintered at temperatures range 1125 oC -1300 oC for 4 h in air. The details process for RS method and CS method are given in Fig.1. Obvious that the RS method has great advantages over
EP
CS method in simply preparation process.
AC C
The crystalline phases of the sintered samples were identified by X-ray diffraction (XRD, Rigaku D/max 2550 PC, Tokyo, Japan) with Cu Kα radiation generated at 40 kV and 40 mA. The microstructure of the ceramic surfaces were performed and analyzed by a scanning electron microscopy (SEM, MERLIN Compact, Germany). The microwave dielectric properties were measured in the frequency range of 6-13 GHz by the TE01σ shielded cavity method [18]. The temperature coefficient of resonant frequency (τƒ) was measured in the temperature range from 25 oC to 85 oC and was calculated by the following formula:
3
ACCEPTED MANUSCRIPT f85 − f 25 ×106 (ppm/oC) f 25 (85 − 25)
τf =
(1)
where ƒ85 and f25 were the TE01σ resonant frequencies at 85oC and 25oC, respectively. The apparent densities of the sintered pellets were measured using Archimedes method (Mettler
ρt heory =
RI PT
ToledoXS64). To study the relative density of the sample, the theoretical density was defined as follows:
ZA VC N A
(2)
where VC, NA , Z, and A are volume of unit cell (cm3), avogadro number (mol-1), number of atoms in unit
ρ relative =
ρ ρ
bulk
SC
cell, and atomic weight (g/mol), respectively. The relative density was obtained by the Eq. (3): ×100%
(3)
M AN U
theory
3. Results and discussion
Fig.2 shows the XRD patterns of ZnZrNb2O8 ceramics prepared using RS method sintered at 1125 oC -1225 oC and CS method sintered at 1200 oC -1300 oC, respectively. Notice that Fig.2 (a) presents a
TE D
monoclinic-structured phase of ZnZrNb2O8 (PDF file no. 50-1725, space group C2/c no. 15) and litter second phase of Zn3Nb2O8 (PDF file no. 50-1725) using RS method. However, all the reflections in Fig.2 (b) are well matched with ZnZrNb2O8 phase, and no any secondary phases was observed. The formation
EP
of second phase for the RS method was due to the missing calcination in the sintering process. Therefore,
AC C
it is difficult to form pure ZnZrNb2O8 phase at lower sintering temperature. Moreover, the diffraction peaks of the sintered samples slightly shift to lower angle direction with the temperature increased both RS and CS method, which indicates the decreasing of the cell volume [20-21]. In order to investigate the effects of different sintering methods on the crystal structures and microwave dielectric properties for the ZnZrNb2O8 ceramics, the Rietveld refinement were carried out using the software of Full-prof based on the XRD data. Fig.3 shows the structural refinement patterns of the ZnZrNb2O8 ceramics using different
4
ACCEPTED MANUSCRIPT sintering methods. The details about the lattice parameters and bond lengths of all the samples are given in Table.1 and Table.2. Observed that the cell volume of RS method was smaller than CS method, which is the key factor to the microwave dielectric properties.
RI PT
The microstructures of the sintered samples ZnZrNb2O8 ceramics using RS and CS method at different sintering temperature are presented in Fig. 4. Notice that with the increasing of sintering temperature, grain size for both two different methods have an increasing tendency. Homogeneous grain
SC
and smooth surface were obtained at 1175 oC for RS method and at 1250 oC for CS method, respectively.
M AN U
In addition, all of the grain size was in the range of 5-7 µm, and well-densified SEM photos could be seen in Fig.4. With a further increase of the sintering temperature, no significant change was observed for RS method. However, abnormal grain growth and some pores was found sintered at 1300 oC for CS method, which indicates the sintering temperature for CS method have more influence on the microwave dielectric
TE D
than the RS method.
Fig.5 shows the relative densities and microwave dielectric properties of ZnZrNb2O8 ceramics using RS and CS method as a function of sintering temperature. Notice that both of the two different methods
EP
exhibit an increasing tendency with the sintering temperature increased. After reaching their maximum of
AC C
97.58 % at 1175 oC for RS specimens and 96.37 % at 1250 oC for CS specimens, respectively, the relative densities started to decrease. The high densification of specimens is due to the high apparent densities and well-grain growth, which would be demonstrated from Eq.(3) and Fig.4. In generally, the microwave dielectric properties at different sintering temperature have close relationship with the relative density, grain boundaries, pores and phases compositions [22-23]. In this paper, the εr and Q×f values of the two different method prepared specimens present a similar tendency with the relative densities. This is
5
ACCEPTED MANUSCRIPT because that with the relative densities increasing, the grain boundaries and pores (εr=1) would decrease, therefore, the dielectric constant increased and dielectric loss decreased. In this paper, the sintering temperature have no direct relationship with the τf value due to the single phase for the sintered specimens
RI PT
and no other extra additives. The τf value reached a maximum of -48.61 ppm/oC at 1175 oC for RS method and of -51.36 ppm/oC at 1250 oC for CS method, respectively. It suggests that RS method could help to enhance the microwave dielectric properties of ZnZrNb2O8 ceramics.
SC
In order to evaluate the intrinsic factors between the RS method and CS method, in the following
M AN U
study, the theoretical dielectric constant, packing fraction and bond valence was calculated. Based on the previous works [24-26], the theoretical dielectric constant εtr and ionic polarizabilities αD have the following relationship:
ε tr =
(Vm + 2bα D ) (Vm − bα D )
(4)
TE D
where Vm is the molar volume; b has the value of 4π/3. According to the additivity rule of molecular polarizabilities, αD for the ZnZrNb2O8 compounds can be broken up as follows: αD(ZnZrNb2O8)=αD(Zn2+)+αD(Zr4+)+αD(Nb5+)+8αD(O2-)
(5)
AC C
formula:
EP
The packing fraction, which is a vital factor to the Q×f values, is calculated using the following
Packing fraction ( % ) ==
3 3 4π / 3 × ( rZn + rZr3 + 2 × rNb + 8 × rO3 )
Vcell
(6)
Kityk et al.[27-28] reported the niobates principal role in ABO4 compounds play intrinsic cationic
defects, and the structural characteristics of B-site oxygen octahedral have closely relation with the τƒ value. The structural characteristics of B-site oxygen octahedral could be characterized by the B-site bond valence, which could be obtained as follows: 6
ACCEPTED MANUSCRIPT Vi = ∑ vij j
vij = exp(
(7)
Rij - dij ) b'
(8)
where Vi was defined as the sum of all of the valences from a given atom i, Rij was bond valence
RI PT
parameter and was equal to 1.911, dij was the distance between atom i and j, and b′ was 0.37.
Table.3 gives the calculated results of theoretical dielectric constant, packing fraction and Nb-site
SC
bond valence for the two different method sintered at their optimal sintering temperature. Notice that the εr was bigger than εtr, which is because that εtr was actually the real part of the complex dielectric constant
M AN U
in the Clausius-Mosotti equation. For the high-relative density specimens, the εr was mainly dependent on the ionic polarizability. Therefore, we just calculated the theoretical dielectric constant caused by the ionic polarizability part. In this paper, the εtr for the specimens prepared different method keep similar tendency. According to Table.3, a higher εtr value of 23.94 for RS method prepared sample was obtained compared
TE D
to εtr value 23.81 for CS method, therefore, the εr for RS method would be higher than the CS method. According to Kim et al [29]., when the relative density of the samples was over 95%, the Q×f values
EP
are mainly dependent on the intrinsic losses caused by packing fraction. With an increase in the packing fraction, the lattice vibrations would decrease. As Table.3 shows, higher Q×f values of 76,400 GHz
AC C
correlate with higher packing fraction of 64.65% and lower Q×f values of 65,100 GHz correspond to smaller packing fraction of 64.61%. Based on the results of SEM images for RS methods, the grain size showed more uniform morphology, which was another reason for the increase of the Q×f values. In combination with the analysis of grain growth and packing fraction, it is could be explain why the RS-prepared samples possess higher Q×f values. In general, the temperature coefficient of resonant frequency τf value was influenced by the 7
ACCEPTED MANUSCRIPT composition, additives and second phases [30-33]. In this paper, the above factors were not consideration in the τf values for the RS method and CS method, which was due to the single phase of specimens and no other extra additives. Therefore, the τf values could be characterized by the Nb-site bond valence VNb.
RI PT
With the increasing of VNb, the |τf| value would decrease, which was because that higher bond valence would led to a higher restoring force, the restoring force increased, the system would be more system. As Table.3 shows, the VNb of 5.0144 for the RS method is bigger than the 4.4644 for the CS method. The VNb
SC
increased, |τf| value would decrease.
M AN U
In summary, fine microwave dielectric properties of ZnZrNb2O8 ceramics with εr = 28.99, Q×f = 76,400 GHz and τf = -48.61 ppm/oC could be obtained at 1750 oC for 4 h prepared by RS method, and with a εr of 26.70, Q×f value of 65,100 GHz and τf value of-51.36 ppm/oC for the CS-prepared samples could be obtained at 1250 oC for 4 h, respectively.
TE D
4. Conclusions
ZnZrNb2O8 ceramics were successfully prepared with excellent microwave dielectric properties via reaction-sintering method for the first time. Single monoclinic structured phase of ZnZrNb2O8 was
EP
obtained from the RS and CS methods sintered at their optimal sintering temperature. Based on the
AC C
Rietveld refinement methods, the εr of RS sample is much higher than the CS sample due to the high εtr in RS sample. The enhance Q×f value for RS prepared specimen could be contribution to the higher packing fraction. For single phase samples, the Nb-site bond valence plays an important role in τf values. A fine microwave dielectric properties (εr ~ 28.99, Q×f value ~ 76,400 GHz and τf value ~ -48.61 ppm/oC) was obtained for RS-prepared sample sintered at 1175 oC for 4 h. It suggests that reaction-sintering method could be a simple and efficient method to produce ZnZrNb2O8 ceramics with enhanced microwave
8
ACCEPTED MANUSCRIPT dielectric properties. Acknowledgments The authors gratefully acknowledged supports from the Key Laboratory of Advanced Ceramics and
RI PT
Machining Technology, Ministry of Education (Tianjin University). References
[1] M.T. Sebastian, Dielectric materials for wireless communications, Elsevier Publishers, Oxford, 2008.
SC
[2] R. D. Richtmeyer, Dielectric resonators, J. Appl. Phys. 15 (1939) 391-398.
M AN U
[3] P. Zhang, Y. G. Zhao, J. Liu, Z. K. Song, X. Y. Wang, M. Xiao, J. Alloy. Compd. 650 (2015) 90-94. [4] Q.W. Liao, L.X. Li, X. Ren, X. X. Yu, D. Guo, M. J. Wang, J. Am. Ceram. Soc. 95 (2012) 3363-3365. [5] S.D. Ramarao, V. R. K. Murthy, Scripta. Mater. 69 (2013) 274-277.
[6] X. Tang, H. Yang, Q. L. Zhang, J. H. Zhou, Ceram. Int. 40 (2014) 12875-12881.
TE D
[7] L. X. Li, H. Sun, H. C. Cai, X. S. Lv, J. Alloy. Compd. 639 (2015) 516-519. [8] J. Ye, L. X. Li, H. Sun, X. S. Lv, J. Y. Yu, S. Li, J. Mater. Sci.: Mater. Eletron. 26 (2015) 8954-8959. [9] J. Ye, L. X. Li et al., J. Mater. Sci.: Mater. Eletron. 27 (2016) 97-102.
EP
[10] L. X. Li, J. Ye, et al., J. Mater. Sci.: Mater. Eletron. 27 (2016) 1232-1238.
AC C
[13] P. Zhang, Y. G. Zhao, H .T. Wu, Dalton. Trans. 44 (2015) 16684-16693. [14] Y. C. Liou, Y. L. Sung, Ceram. Int. 34 (2008) 371-377. [15] L. X. Li, X. Ding, Q. W. Liao, J. Alloy. Compd. 509 (2011) 7271-7276. [16] G. G. Yao, P. Liu, H. W. Zhang, J. Mater. Sci.: Mater. Electron. 24 (2013) 1128-1131. [17] H. B. Bafrooei, E. T. Nassaj, T. Ebadzadeh, et al., Ceram. Int. 42 (2016) 3296-3303. [18] W. E. Courtney. IEEE Trans Microwave Theory Tech. 18 (1970) 476-485.
9
ACCEPTED MANUSCRIPT [19] D. Zhou, L. X. Pang, J. Guo, et al., Inorg. Chem. 53 (2014) 1048-1055. [20] K. Nishiyama, T. Abe , T. Sakaguchi , N. Momozawa, J. Alloy. Compd. 355 (2003) 103-107. [21] P. Zhang, Y. G. Zhao, X. Y. Wang, J. Alloy. Compd. 644 (2015) 621-625.
RI PT
[22] D. Zhou, L. X. Pang, J. Guo, et al., Inorg. Chem. 50 (2011) 12733-12738. [23] D. Zhou, L. X. Pang, J. Guo, et al,, Inorg. Chem. 53 (2014) 53 1048-1055.
[25] Y. G. Zhao, P. Zhang, J. Alloy. Compd. 658 (2016) 744-748.
M AN U
[26] Y. G. Zhao, P. Zhang, J. Alloy. Compd. 662 (2016) 455-460.
SC
[24] Y. G. Zhao, P. Zhang, RSC. Adv. 5 (2015) 97746-97754.
[27] I. V. Kityk, M. M. Janusik, J. Phys. Chem B. 105 (2001) 12242-12248 [28] M. Piasecki, I. V. Kityk, J. Alloy. Compd. 639 (2015) 577-582.
[29] E. S. Kim, B. S. Chun, R. Freer, R. J. Cernik, J. Eur, Ceram. Soc. 30 (2010) 1731-1736.
TE D
[30] Q. J. Mei, C.Y. Li, J.D. Guo, H.T. Wu, J. Alloy. Compd., 626 (2015) 217-222. [31] H.T. Wu, Z.B. Feng, Q.J. Mei, J.D. Guo, J.X. Bi, J. Alloy. Compd. 643 (2015) 368-373. [32] H.T. Wu, Q.J. Mei, J. Alloy. Compd. 651 (2015) 393-398.
AC C
EP
[33] H. L. Pan, Z.B. Feng, J.X. Bi, H.T. Wu, J. Alloy. Compd. 651 (2015) 440-443.
10
ACCEPTED MANUSCRIPT Table.1 Refined lattice parameters (a, b, c, β and V) and discrepancy factors (Rp and Rwp) of ZnZrNb2O8 ceramics using different sintering process. Sintering Method
a (Å)
b( Å)
c( Å)
β(°)
1125 oC
4.8150
5.6769
5.0795
91.45
1150 oC
4.8153
5.6774
5.0791
91.45
1175 oC
4.8150
5.6778
5.0799
91.45
1200 oC
4.8154
5.6825
5.0806
1225 oC
4.8137
5.6828
5.0795
1200 oC
4.8189
5.6724
1225 oC
4.8160
1250 oC
4.8247
1275 oC
V(Å3)
Rp(%)
Rwp(%)
138.81
10.40
14.90
138.83
9.76
13.00
SC
11.40
9.18
11.70
91.41
138.91
9.66
12.30
5.0806
91.49
138.83
6.94
9.01
5.6768
5.0807
91.43
138.86
7.12
9.35
5.6654
5.0843
91.56
138.92
11.70
14.70
4.8163
5.6799
5.0823
91.42
138.99
8.53
10.60
4.8287
5.6670
5.0851
91.52
139.10
11.30
14.40
EP
1300 oC
8.61
138.98
M AN U
CS
138.80
91.43
TE D
RS
RI PT
Temperature
Table.2 The refined bond length (Å) of ZnZrNb2O8 ceramics using RS and CS methods. Zn/Zr-O(1)×2
Zn/Zr-O(2)1×2
Zn/Zr-O(2)2×2
Nb-O(1)1×2
Nb-O(1)2×2
Nb-O(2)×2
RS (1175 oC)
1.9772
2.2673
2.1707
2.0367
2.1538
1.8207
CS (1250 oC)
1.9348
2.2356
2.1647
2.0550
2.2755
1.8497
AC C Method
Table.3 The measured microwave dielectric properties, and calculated theoretical dielectric constant, packing fraction and Nb-site bond valence of ZnZrNb2O8 ceramics using RS and CS methods.
ACCEPTED MANUSCRIPT Measured
Calculated
Method Q×f(GHz)
τf(ppm/oC)
εtr.
P.F .%
VNb
RS (1175 oC)
28.99
76400
-48.61
23.94
64.65
5.0144
CS (1250 oC)
26.70
65100
-51.36
23.81
64.61
4.4644
AC C
EP
TE D
M AN U
SC
RI PT
εr
ACCEPTED MANUSCRIPT Figure captions Fig.1 Reaction-sintering and conventional solid-state reaction sintering process for ZnZrNb2O8 ceramics.
RI PT
Fig.2 The XRD patterns of ZnZrNb2O8 ceramics using (a): RS method sintered at 1125 oC-1225 oC; (b): CS method sintered at 1200 oC-1300 oC.
Fig.3 The structural refinement patterns of the ZnZrNb2O8 ceramics (a): RS method at 1175 oC; (b): CS
SC
method at 1250 oC.
M AN U
Fig.4 SEM graphs of ZnZrNb2O8 ceramics using RS and CS method.
Fig.5 The relative densities, εr values, Q×f values and τf values of ZnZrNb2O8 ceramics using RS and
AC C
EP
TE D
CS method as a function of sintering temperature.
Fig.1
Fig.2
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig.3
AC C
EP
TE D
M AN U
Fig.4
SC
RI PT
ACCEPTED MANUSCRIPT
Fig.5