- Email: [email protected]
Study on the microstructure of cement-based piezoelectric ceramic composites
Construction and Building Materials 72 (2014) 133–138
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Study on the microstructure of cement-based piezoelectric ceramic composites Biqin Dong a, Yuqing Liu a, Ningxu Han a, Hongfang Sun a, Feng Xing a, Daoding Qin b,⇑ a b
School of Civil Engineering, Guangdong Province Key Laboratory of Durability for Marine Civil Engineering, Shenzhen University, Shenzhen 518060, PR China Department of Mechanical and Aerospace Engineering, Hong Kong University of Science & Technology, Clear Water Bay, Hong Kong, China
h i g h l i g h t s The interaction of Ti–O and Si–O in cement-based piezoelectric ceramic composites occurs. SIMS images provide a direct indication of element spatial distribution in the composite. The chemical reaction results in the diffusion of Ti into the cement paste.
a r t i c l e
i n f o
Article history: Received 29 May 2014 Received in revised form 12 August 2014 Accepted 23 August 2014
Keywords: Piezoelectric ceramic Cement Composite Microstructure
a b s t r a c t In this paper, a study is presented on the microstructure of a cement-based PZT (Lead Zirconate Titanate) piezoelectric ceramic composite. A mechanism is proposed for chemical bonding between piezoelectric ceramic particles and cement material in the composite. The microstructure of the composite, as well as chemical bonding at the ceramic–cement interface is investigated using XRD, IR and XPS. Experimental results indicate that Ti–O in the ceramic particles links to Si–O in the cement paste via bridging oxygen, forming a Ti–O Si–O bonding. And the change of chemical environment takes place for all of Ti, Si, Ca and Zr for cement-based piezoelectric ceramic composites. One prominent result pertaining to chemical reactions at the interface between ceramic particles and cement paste shows that Ti diffuses into the cement paste via Ti–O Si–O bonding. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction With the development and application of the ‘‘Performanceand Reliability-based Service Life Design’’ (PRSLD) principle in civil engineering, the importance of health monitoring processes within the PRSLD framework has become clear [1–5]. To facilitate monitoring of targeted functionality, a variety of different sensor types have been developed for use in civil engineering projects. Among the approaches used in the design of sensors and actuators adopted in smart structures, piezoelectricity has proven to be one of the most efficient [6–9], and piezoelectric materials have accordingly received much attention for potential applications in the field of smart materials [10–16]. One such smart material – suitable for structures in civil engineering – is 0–3 type [17,18] cement-based piezoelectric ceramic composite, incorporating lead zirconate ⇑ Corresponding author. E-mail addresses: [email protected] (B. Dong), [email protected] (Y. Liu), [email protected] (N. Han), [email protected] (H. Sun), [email protected] (F. Xing), [email protected] (D. Qin). http://dx.doi.org/10.1016/j.conbuildmat.2014.08.058 0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
titanate (PZT) ceramic powder into the cement paste. Its properties and potential applications in civil engineering have been previously explored by Li et al. [19] However, the microstructure of this new composite, and the relationship between the microstructure and the material’s macroscopic properties, are still not well understood. The present paper focuses on the microstructure of PZT ceramic composite. In particular, attention is given to the interaction of the solid phases (ceramic and cement) [20]. In general, the material properties of a composite, including its piezoelectric, dielectric, and mechanical properties, depend strongly on the microstructure. Therefore, a full understanding of the microstructure of composites is crucial in order to predict the relevant material properties. In addition to the separate material properties of cement paste and piezoelectric ceramic particles, attention should be paid to the interface between the two phases and especially to the bonding behavior between them. The investigation of microstructure can provide a fundamental basis for understanding and predicting the performance of cement-based piezoelectric ceramic composites.
134
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
Table 1 Properties of PZT powder, hardened Portland cement paste, and normal concrete.
2000
P-85 Cement paste Concrete 390 23 1700 58 2000 – 7.5 3.40 –
– – 56 – – 72 2.0 2.64 5.3
– – – – – 30 2.4 3.73 9.0
* * *
1500
Intensity
Piezoelectric strain factor d33 (1012 C/N) Piezoelectric voltage factor g33 (103 Vm/N) Dielectric constant er (at 1KHz) Electromechanical coupling coefficient KP (%) Mechanical quality Qm Elastic compliance s33 (1012 m2/N) Density q (103 kg/m3) Acoustic velocity V(103 m/s) Acoustic impedance q V (106 kg/m2 s)
1000
Piezoelectric Ceramic
500 Cement-based Piezoelectric Ceramic composite #
0 20
25
30
35
1423
Adsorption 400
1006
1126
45
50
55
60
65
70
75
80
microscope. The chemical environments of each element were measured by X-ray Photoelectron Spectroscopy (XPS, Surface analysis PHI5600, PHI 5600) and Secondary Ion Mass Spectroscopy (SIMS, Model PHI 7200 (Physical Electronics)).
1
980 1111
2
980
3 4 5 6
1108 980 1100 980
800
40
Fig. 2. X-ray diffraction analysis of cement paste, piezoelectric ceramic and cement-based piezoelectric ceramic composite.
713
600
#
2 Theta
874
529 597
Cement Paste
1000
1200
1400
1600
3. Results and analysis
1800
2000
-1
Wave number (cm ) Fig. 1. IR spectra of piezoelectric ceramic, cement paste and cement-based piezoelectric composite with different ceramic/cement volume ratios. 1: White cement, 2: cement-based piezoelectric composite with ceramic/cement ratio = 20:80, 3: cement-based piezoelectric composite with ceramic/cement ratio = 35:65, 4: cement-based piezoelectric composite with ceramic/cement ratio = 50:50, 5: cement-based piezoelectric composite with ceramic/cement ratio = 75:25, 6: piezoelectric ceramic.
2. Experiment Lead zirconate titanate (PZT) piezoelectric ceramic powder (Hong Kong Functional Ceramic Co. Ltd.) and cement (H.S.L. Enterprises Co. Ltd.) were used to make cement-based piezoelectric composites. The properties of the piezoelectric ceramic and cement paste are listed in Table 1. Piezoelectric ceramic powder and white cement with various volume ceramic/ cement ratios were mixed together to make a 0–3 type cement-based piezoelectric ceramic composite. In order to improving the fluidity of the fresh mixture, a superplasticizer (W19, W. R. Grace) was added. To achieve a uniform mixture, the cement and ceramic particles were first dry-mixed for 2 min. Then water and superplasticizers were added, and mixing continued until a uniform mixture was obtained. The resulting mixture was cast in a 13 mm 13 mm 3 mm mould. After casting, the specimens were covered with a plastic plate and rested at room temperature for about 24 h, after which they were removed from the mould. Then the specimens were placed in a curing room at a temperature of 65 °C and relative humidity of 98% for a further 24 h. The composition of each sample was measured with XRD (High Resolution X-ray Diffraction System, Model PW1825 (Philips)). The bonding structures were analyzed by IR (FT-IR System, FTS 6000). The morphology was observed by optical
Table 2 The peak assignment of vibration in IR spectra. Wave number (cm1)
Assignment
597 713, 874 1423 1006, 1126
TiO6 octahedron Si–O–H Ca2+ Si–O–Si
It is well known that Si–O chemical bonding occurs during the hardening of Portland cement, leading to the formation of Si–O– Si (siloxane). According to the pioneering work of Damidot et al. [21], silicic acids are unstable after liberation from a silicate. Studies with 1H NMR shows that Si–OH group become detectable immediately after mixing [22]. This indicates the formation of hydroxylated species at the C3S surface. Ca–OH groups in combination with Si–OH groups, corresponding to C–S–H formation, become detectable during the induction period. Later on, Ca–OH groups belonging to Ca(OH)2 are also formed. At the nanometer scale the C–S–H phase formed by the hydration of C3S at room temperature seems to be structurally related to the crystalline phase, i.e. 1.4 nm tobermorite and jennite, and also to the poorly crystalline phases C–S–H (I) and C–S–H (II) [23]. Both comprise [SiO4] square planar units, condensed into linear chains, which are linked so that they repeat at intervals of three [SiO4] units. In contrast, PZT ceramic has a perovskite structure with a threedimensional network of [BO6] octahedra; it may be regarded also as a cubic close-packed arrangement of Pb and O ions with Ti (or Zr) ions filling the octahedral interstitial positions (spontaneous polarization will occur when each of the Ti4+ ions moves to one side of its octahedron) [24]. Moreover, the experimental analysis was to obtain both qualitative and quantitative clues to elucidate the microstructure and chemical bonding situation in cement-based piezoelectric ceramic composites. Results from IR vibrational spectra were found to be fully consistent with our theoretical predictions. Fig. 1 shows the IR spectra in the wave number range 1300–400 cm1 for cement paste, piezoelectric ceramic, and composite. Interpretations of each peak are summarized in Table 2. IR spectra for silicate compounds exhibit a large absorption between 1200 and 900 cm1, which corresponds to the asymmetrical stretching vibration (v3), whereas absorptions at 900 cm1 or below correspond to out-of-plane (v4) and in-plane (v2) [24–28] skeletal vibrations. There is a tendency for v3 to shift to higher wave numbers (‘‘blue shift’’) as the degree of polymerization or condensation increases. Conversely, for v3 to shift to lower wave numbers (‘‘red shift’’) means that the degree of polymerization or condensation decreases. Since v ¼ 1=2pcðf =mÞ1=2 , as long as the atoms involved are the same,
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
135
Fig. 3. XPS results for white cement and cement-based piezoelectric ceramic composite with different ceramic/cement volume ratios. (a) Ca2p; (b) Si2p; (c) Ti2p; (d) Zr3d. 1: cement paste, 2: cement-based piezoelectric composite with ceramic/cement volume ratio = 20:80, 3: cement-based piezoelectric composite with ceramic/cement volume ratio = 35:65, 4: cement-based piezoelectric composite with ceramic/cement volume ratio = 50:50, 5: cement-based piezoelectric composite with ceramic/cement volume ratio = 75:25, 6: piezoelectric ceramic.
the observed shift to low wave number can be interpreted as a reduction in Si–O–Si bonding [29,30]. Fig. 2 also shows the influence of the ceramic/cement volume ratio on Si–O–Si bonding. As the relative proportion of ceramic is increased, the Si–O–Si bonding peaks shift to lower wave numbers (from 1126 cm1 to 1063 cm1, and from 1006 cm1 to 980 cm1). This suggests that PZT ceramic
particles attack Si–O–Si bonds, breaking down polymeric chains of [SiO4]. In contrast with Si–O–Si bonding, the Si–O–H bonding peaks (874 cm1 and 713 cm1) show no influence from the presence of PZT particles in the mixture. The same is true for the Ca2+ vibration peak (1423 cm1). This indicates that Ti–O in the ceramic
136
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
Fig. 3 (continued)
particles links to Si–O in the cement paste via a bridging oxygen, forming a Ti–O Si–O bonding structure. Further evidence for this mechanism comes from examining the Ti–O bond ([TiO6] octahedron), which in ceramic particles shows only one peak at 597 cm1 [31]. As the ceramic particles are added to the cement paste, the peak splits into two subsidiary bands, one corresponding to the original Ti–O bond at 597 cm1 and the other to the new Ti–O Si–O bond at 529 cm1. This shows that only a
small fraction of the PZT ceramic takes part in chemical reactions with the cement paste, whereas the majority remains in the original ceramic form. X-ray diffraction patterns directly show the differences in microstructure between the composite and cement paste. In Fig. 2, the abundant small broad peaks for hydrated cement paste indicate the presence of all kinds of amorphous microstructure, and a few Bragg peaks (marked by (⁄)) refer to the crystalline
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
137
Fig. 4. SIMS images with element mapping mode of cement-based piezoelectric ceramic composite.
parameters of tobermorite, which are in agreement with the results obtained by Hamid [32,33]. X-ray diffraction patterns of pure piezoelectric ceramic show the same positions as that of the cement–ceramic composite, which demonstrates that the PZT ceramic particles retain their crystalline structure in the composite. However, the tobermorite pattern seen in hydrated cement paste disappears in the composite, and new patterns (marked by (#) in Fig. 2) appear, implying that the addition of ceramic powder restricts the growth of the long [SiO4] chains that make up tobermorite, as well as reducing the size of cement paste particles. These results are in accord with the IR spectroscopy results. The chemical composition of the cement–ceramic composite was further studied with XPS. Fig. 3a–d shows the XPS spectra of Ca, Si, Ti, and Zr in the composite, covering the binding energy range of 0–500 eV. The binding energy of various peaks is calibrated by using the binding energy of C1s (285.0 eV). As shown in Fig. 3, chemical bonding of all these elements (Ca, Si, Ti, Zr) occurs in the composite. The binding energy of Ca2p (Fig. 3a) is calibrated to 347 eV and it matches well with Ca in cement paste. The results suggest that Ca exists mainly in a form similar to that in cement paste. In contrast, the binding energy peak of Si2p (Fig. 3b) shifts from 458.1 eV (in cement paste) to 458.6 eV (in the composite). This demonstrates that the chemical environment of Si differs significantly between cement paste and composite. Similar results can be found for Ti2p (Fig. 3c) and Zr3d (Fig. 3d) [34]. It can be concluded that a change of chemical environment takes place for all of Ti, Si, Ca and Zr during their incorporation into a cement-based piezoelectric ceramic composite. One prominent result pertaining to chemical reactions at the interface between ceramic particles and cement paste shows that Ti diffuses into the cement paste via Ti–O Si–O bonding. In order to check for such a process in the composite, the spatial distributions of the elements Ca, Si, Ti, Zr and Pb are all measured using the SIMS method with element mapping mode. As PZT ceramic is added to the cement paste, fine cubic particles are formed, and they are uniformly dispersed throughout the cement paste. This phenomenon can be clearly observed in SEM images [35]. In the
element-mapping SIMS images shown as Fig. 4, each point is shaded according to the concentration of the element at that point. The concentration of Pb remains confined to the location of PZT ceramic particles; in contrast, Ti – which, along with Pb, originates in the PZT ceramic – is distributed throughout the composite, not just within the ceramic particles. It is clear that Ti diffuses from the PZT ceramic particles into the cement paste once they come into contact. The SIMS images provide a direct measure of the elemental distribution in the cement-based piezoelectric composite, and these are found to be consistent with the occurrence of Ti– O Si–O bonding at the interface between the PZT ceramic particles and cement paste. This reaction results in the diffusion of Ti element into the cement matrix. As consider the chemical reaction between PZT ceramic particles and cement grains, it will affect influence of the component of composite on the material properties, such as dielectric, piezoelectric and elastic properties, of the composite. And it has been confirmed by our previous study [36]. 4. Conclusion In this study, the chemical reaction mechanisms and microstructure of a cement-based piezoelectric ceramic composite were investigated. The following conclusions can be drawn: 1. From IR spectra, there is no change of Ca2+ chemical environment as increasing of volume ratio of piezoelectric ceramic particle to cement paste, whereas [SiO4] takes an obvious ‘‘red shifting’’ in adsorption bands from 1106 cm1 to 980 cm1 and 1126 cm1 to 1100 cm1, respectively. Additionally, the [TiO6] in 597 cm1 splits into two subsidiary bands at 597 cm1 and 529 cm1. The variety of adsorption bands in IR spectra due to the contribution of the distortion of Si–O band which causes local change in the crystal field potential and hence induce the splitting of Ti–O (or Zr–O) adsorption bands. Thus the interaction of Ti–O and Si–O in cement-based piezoelectric ceramic composites occurs.
138
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
2. XRD results reveal that the patterns of tobermorite observed in hydrated cement paste disappear in that of composite. Some new patterns are found in the composite, indicating that the addition of ceramic powder both restrict the growing long [SiO4] chains of tobermorite and make cement paste particles smaller. These results are consistent with those obtained by means of IR spectrum. 3. Details of the chemical composition of cement-based piezoelectric ceramic composites were further studied with XPS. The spectra suggest that a significant change in chemical environment occurs for every element type, including Ti, Si, Ca and Zr, present in the composite. SIMS images provide a direct indication of element spatial distribution in the composite, which can be understood as the result of Ti–O Si–O bonding via chemical reactions at the interface between the PZT ceramic particles and the cement paste. The chemical reaction results in the diffusion of Ti into the cement paste.
Acknowledgements The authors would like to acknowledge financial support provided by National Key Basic Research Program funded by MOST(Project No. 2011CB013600; Issue No. 2011CB013604) and National Natural Science Foundation of China (No. 51120185002/ 51272160/U1301241); Foundation for Technology Innovation Project in Higher Education of Guangdong, China (No. 2012KJCX0091) and Collaborative Innovation Center for Advanced Civil Engineering Materials, Nanjing, PR China. References [1] Han N, Xing F. Service life consideration of concrete structures in a broad perspective, keynote speech. In: Proceedings, the 11th international symposium on structural engineering, December 18-20, 2010, Guangzhou, China, Sponsored by National Natural Science Foundation of China. [2] Han N. Role of NDE in quality control during construction of concrete infrastructures on the basis of service life design. Constr Build Mater, Elsevier Science 2004;18:163–78. [3] Chang Fu-Kuo, editor. Structural health monitoring 2000. Lancester, Pennsylvania: Technomic Publishing Co., Inc.; 1999. [4] Tzou HS, Guran A, editors. Structronics systems: smart structures, devices and systems (Parts I & II). Singapore: World Scientific; 1998. [5] Aizawa S, Kakizawa T, Higasino M. Case studies of smart materials for civil structures. Smart Mater Struct 1998;7(5):617–26. [6] Hilczer B, Mailecki J. Electrets and piezopolymers. Warsaw: PWN; 1992 (in polish). [7] Banks HT, Smith RC, Wang Y. Smart material structures: modeling, estimation, and control. New York: John Wiley & Sons Inc.; 1996. [8] George EP, Gotthardt R, Otsuka K, Trolier-McKinstry S, Wun-Fogle M, editors. Materials for smart systems II, Symposium proceedings of Materials Research Society, vol. 459. [9] Tao B. Smart/intelligent materials and structures. Beijing: Defense Industry Press; 1997 (In Chinese).
[10] Wetherhold RC, Panthalingal Nandakumar. Piezoelectric PZT/expoxy composites for sensing and actuating torsional motion. Smart Struct Mater 1993; SPIE: 266–74. [11] Yuhuan Xu. Ferroelectric materials and their applications. Amsterdam: NorthHolland; 1991. [12] Furukawa T. Piezoelectricity and pyroelectricity in polymers. IEEE Trans Electr Insul 1989;24(3):375–94. [13] Okazaki K. Developments in fabrication of piezoelectric ceramics. In: Taylor GW, Gagnepain JJ, Meeker TR, Nakamura T, Shuvalov LA, editors. Piezoelectricity. Switzerland: Gordon and Breach Science Publishers; 1985. p. 131–50. [14] Banno H. Recent developments of piezoelectric composites in Japan. In: Saito S, editor. Advanced ceramics. Oxford University Press; 1988. p. 8–26. [15] Safari A, Sa-gong G, Giniewicz J, Newnham RE. Composites piezoelectric sensors. In: Rosen CZ, Hiremath BV, Newnham R, editors. Piezoelectricity. College Park, MD: American Institute of Physics; 1992. p. 195–204. [16] Latour M, Jolivet S, Rahmoune M, Lagarrigue O, Roure A. Piezopolymer transducers in the active control of vibrations. In: Proceedings of 8th international symposium on electrets, 1994. p. 985–90. [17] Safari A. Novel piezoelectric ceramics and composites for sensor and actuator applications. Mater Res Innovat 1999;2:263–9. [18] Mazur K. Polymer-ferroelectric ceramic composites. In: Nalwa HS, editor. Ferroelectric polymers: chemistry, physics, and applications. New York: M. Dekker Inc.; 1995. p. 539–610. [19] Li Zongjin, Dong Biqin, Zhang Dong. Influence of polarization on properties of 0–3 Cement-based PZT composite. Cem Concr Compos 2005;27(1): 27–32. [20] Gleize PJP, Muller A, Roman HR. Microstructural investigation of a silica fumecement-lime mortar. Cem Concr Compos 2003;25:171–5. [21] Damidot D, Nonat A. In: Nonat A, Mutin C, editors. Hydration and setting of cement. E & FN Spon; 1992. p. 23. [22] Hewlett PC. LEA’S Chemistry of cement and concrete. ARNOLD; 1998. [23] Struble Leslie, Skalny Jan, Mindess Sidney. A review of the cement–aggregate bond. Cem Concr Res 1980;10(2):277–86. [24] Cotton Albert. Advanced inorganic chemistry. John Wiley & Sons; 1988. [25] Okada Yoshihiko, Masuda Tsuguya, Takada Minori, Xu Lingling, Mitsuda Takeshi. Relationship between NMR 29Si chemical shifts and FT-IR wave numbers in calcium silicates. In: Colombet P, Grimmer AR, Zanni H, Sozzani P, editors. Nuclear magnetic resonance spectroscopy of cement-based materials. Springer; 1998. p. 69–78. [26] Henning O. In: Farmer VC, editor. The infrared spectra of minerals. London: Mineralogical Society; 1974. p. 445. [27] Bensted J, Prakash S. Cem Technology Sep/Oct.; 1974. pp. 440. [28] Bensted J, Prakash S. Cem Technology July/Aug.; 1974. pp. 378. [29] Bell G, Bensted J, Glasser FP. Adv Cem Res 1998; (2):61. [30] Gong Hongyu, Zhang Yujun, Quan Jing, Che Songwei. Preparation and properties of cement based piezoelectric composites modified by CNTs. Curr Appl Phys 2011(11):653–6. [31] Rao CNR. Chemical application of Infrared Spectroscopy. New York: Academic Press; 1963. p. 8. [32] Mazen SA, Metawe F, Mansour SF. IR absorption and dielectric properties of Li– Ti ferrite. J Phys D: Appl Phys 1997;30:1799–808. [33] Hamid SA. Z Krist; 1981. p. 54. [34] Klur I, Pollet B, Virlet J, Nonat A. C–S–H structure evolution with Calcium content by multinuclear NMR. In: Colombet P, Grimmer AR, Zanni H, Sozzani P, editors. Nuclear magnetic resonance spectroscopy of cement-based materials. Springer; 1998. p. 119–41. [35] Bhaskar S, Majumder SB, Jain M, Dobal PS, Katiyar RS. Studies on the structural, microstructural and optical properties of sol–gel derived lead lanthanum titanate thin films. Mater Sci Eng, B 2001;87:178–90. [36] Xing Feng, Dong Biqin, Li Zongjin. Dielectric, piezoelectric, and elastic properties of cement-based piezoelectric ceramic composites. J Am Ceram Soc 2008;91(9):2886–91.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Study on the microstructure of cement-based piezoelectric ceramic composites Biqin Dong a, Yuqing Liu a, Ningxu Han a, Hongfang Sun a, Feng Xing a, Daoding Qin b,⇑ a b
School of Civil Engineering, Guangdong Province Key Laboratory of Durability for Marine Civil Engineering, Shenzhen University, Shenzhen 518060, PR China Department of Mechanical and Aerospace Engineering, Hong Kong University of Science & Technology, Clear Water Bay, Hong Kong, China
h i g h l i g h t s The interaction of Ti–O and Si–O in cement-based piezoelectric ceramic composites occurs. SIMS images provide a direct indication of element spatial distribution in the composite. The chemical reaction results in the diffusion of Ti into the cement paste.
a r t i c l e
i n f o
Article history: Received 29 May 2014 Received in revised form 12 August 2014 Accepted 23 August 2014
Keywords: Piezoelectric ceramic Cement Composite Microstructure
a b s t r a c t In this paper, a study is presented on the microstructure of a cement-based PZT (Lead Zirconate Titanate) piezoelectric ceramic composite. A mechanism is proposed for chemical bonding between piezoelectric ceramic particles and cement material in the composite. The microstructure of the composite, as well as chemical bonding at the ceramic–cement interface is investigated using XRD, IR and XPS. Experimental results indicate that Ti–O in the ceramic particles links to Si–O in the cement paste via bridging oxygen, forming a Ti–O Si–O bonding. And the change of chemical environment takes place for all of Ti, Si, Ca and Zr for cement-based piezoelectric ceramic composites. One prominent result pertaining to chemical reactions at the interface between ceramic particles and cement paste shows that Ti diffuses into the cement paste via Ti–O Si–O bonding. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction With the development and application of the ‘‘Performanceand Reliability-based Service Life Design’’ (PRSLD) principle in civil engineering, the importance of health monitoring processes within the PRSLD framework has become clear [1–5]. To facilitate monitoring of targeted functionality, a variety of different sensor types have been developed for use in civil engineering projects. Among the approaches used in the design of sensors and actuators adopted in smart structures, piezoelectricity has proven to be one of the most efficient [6–9], and piezoelectric materials have accordingly received much attention for potential applications in the field of smart materials [10–16]. One such smart material – suitable for structures in civil engineering – is 0–3 type [17,18] cement-based piezoelectric ceramic composite, incorporating lead zirconate ⇑ Corresponding author. E-mail addresses: [email protected] (B. Dong), [email protected] (Y. Liu), [email protected] (N. Han), [email protected] (H. Sun), [email protected] (F. Xing), [email protected] (D. Qin). http://dx.doi.org/10.1016/j.conbuildmat.2014.08.058 0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
titanate (PZT) ceramic powder into the cement paste. Its properties and potential applications in civil engineering have been previously explored by Li et al. [19] However, the microstructure of this new composite, and the relationship between the microstructure and the material’s macroscopic properties, are still not well understood. The present paper focuses on the microstructure of PZT ceramic composite. In particular, attention is given to the interaction of the solid phases (ceramic and cement) [20]. In general, the material properties of a composite, including its piezoelectric, dielectric, and mechanical properties, depend strongly on the microstructure. Therefore, a full understanding of the microstructure of composites is crucial in order to predict the relevant material properties. In addition to the separate material properties of cement paste and piezoelectric ceramic particles, attention should be paid to the interface between the two phases and especially to the bonding behavior between them. The investigation of microstructure can provide a fundamental basis for understanding and predicting the performance of cement-based piezoelectric ceramic composites.
134
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
Table 1 Properties of PZT powder, hardened Portland cement paste, and normal concrete.
2000
P-85 Cement paste Concrete 390 23 1700 58 2000 – 7.5 3.40 –
– – 56 – – 72 2.0 2.64 5.3
– – – – – 30 2.4 3.73 9.0
* * *
1500
Intensity
Piezoelectric strain factor d33 (1012 C/N) Piezoelectric voltage factor g33 (103 Vm/N) Dielectric constant er (at 1KHz) Electromechanical coupling coefficient KP (%) Mechanical quality Qm Elastic compliance s33 (1012 m2/N) Density q (103 kg/m3) Acoustic velocity V(103 m/s) Acoustic impedance q V (106 kg/m2 s)
1000
Piezoelectric Ceramic
500 Cement-based Piezoelectric Ceramic composite #
0 20
25
30
35
1423
Adsorption 400
1006
1126
45
50
55
60
65
70
75
80
microscope. The chemical environments of each element were measured by X-ray Photoelectron Spectroscopy (XPS, Surface analysis PHI5600, PHI 5600) and Secondary Ion Mass Spectroscopy (SIMS, Model PHI 7200 (Physical Electronics)).
1
980 1111
2
980
3 4 5 6
1108 980 1100 980
800
40
Fig. 2. X-ray diffraction analysis of cement paste, piezoelectric ceramic and cement-based piezoelectric ceramic composite.
713
600
#
2 Theta
874
529 597
Cement Paste
1000
1200
1400
1600
3. Results and analysis
1800
2000
-1
Wave number (cm ) Fig. 1. IR spectra of piezoelectric ceramic, cement paste and cement-based piezoelectric composite with different ceramic/cement volume ratios. 1: White cement, 2: cement-based piezoelectric composite with ceramic/cement ratio = 20:80, 3: cement-based piezoelectric composite with ceramic/cement ratio = 35:65, 4: cement-based piezoelectric composite with ceramic/cement ratio = 50:50, 5: cement-based piezoelectric composite with ceramic/cement ratio = 75:25, 6: piezoelectric ceramic.
2. Experiment Lead zirconate titanate (PZT) piezoelectric ceramic powder (Hong Kong Functional Ceramic Co. Ltd.) and cement (H.S.L. Enterprises Co. Ltd.) were used to make cement-based piezoelectric composites. The properties of the piezoelectric ceramic and cement paste are listed in Table 1. Piezoelectric ceramic powder and white cement with various volume ceramic/ cement ratios were mixed together to make a 0–3 type cement-based piezoelectric ceramic composite. In order to improving the fluidity of the fresh mixture, a superplasticizer (W19, W. R. Grace) was added. To achieve a uniform mixture, the cement and ceramic particles were first dry-mixed for 2 min. Then water and superplasticizers were added, and mixing continued until a uniform mixture was obtained. The resulting mixture was cast in a 13 mm 13 mm 3 mm mould. After casting, the specimens were covered with a plastic plate and rested at room temperature for about 24 h, after which they were removed from the mould. Then the specimens were placed in a curing room at a temperature of 65 °C and relative humidity of 98% for a further 24 h. The composition of each sample was measured with XRD (High Resolution X-ray Diffraction System, Model PW1825 (Philips)). The bonding structures were analyzed by IR (FT-IR System, FTS 6000). The morphology was observed by optical
Table 2 The peak assignment of vibration in IR spectra. Wave number (cm1)
Assignment
597 713, 874 1423 1006, 1126
TiO6 octahedron Si–O–H Ca2+ Si–O–Si
It is well known that Si–O chemical bonding occurs during the hardening of Portland cement, leading to the formation of Si–O– Si (siloxane). According to the pioneering work of Damidot et al. [21], silicic acids are unstable after liberation from a silicate. Studies with 1H NMR shows that Si–OH group become detectable immediately after mixing [22]. This indicates the formation of hydroxylated species at the C3S surface. Ca–OH groups in combination with Si–OH groups, corresponding to C–S–H formation, become detectable during the induction period. Later on, Ca–OH groups belonging to Ca(OH)2 are also formed. At the nanometer scale the C–S–H phase formed by the hydration of C3S at room temperature seems to be structurally related to the crystalline phase, i.e. 1.4 nm tobermorite and jennite, and also to the poorly crystalline phases C–S–H (I) and C–S–H (II) [23]. Both comprise [SiO4] square planar units, condensed into linear chains, which are linked so that they repeat at intervals of three [SiO4] units. In contrast, PZT ceramic has a perovskite structure with a threedimensional network of [BO6] octahedra; it may be regarded also as a cubic close-packed arrangement of Pb and O ions with Ti (or Zr) ions filling the octahedral interstitial positions (spontaneous polarization will occur when each of the Ti4+ ions moves to one side of its octahedron) [24]. Moreover, the experimental analysis was to obtain both qualitative and quantitative clues to elucidate the microstructure and chemical bonding situation in cement-based piezoelectric ceramic composites. Results from IR vibrational spectra were found to be fully consistent with our theoretical predictions. Fig. 1 shows the IR spectra in the wave number range 1300–400 cm1 for cement paste, piezoelectric ceramic, and composite. Interpretations of each peak are summarized in Table 2. IR spectra for silicate compounds exhibit a large absorption between 1200 and 900 cm1, which corresponds to the asymmetrical stretching vibration (v3), whereas absorptions at 900 cm1 or below correspond to out-of-plane (v4) and in-plane (v2) [24–28] skeletal vibrations. There is a tendency for v3 to shift to higher wave numbers (‘‘blue shift’’) as the degree of polymerization or condensation increases. Conversely, for v3 to shift to lower wave numbers (‘‘red shift’’) means that the degree of polymerization or condensation decreases. Since v ¼ 1=2pcðf =mÞ1=2 , as long as the atoms involved are the same,
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
135
Fig. 3. XPS results for white cement and cement-based piezoelectric ceramic composite with different ceramic/cement volume ratios. (a) Ca2p; (b) Si2p; (c) Ti2p; (d) Zr3d. 1: cement paste, 2: cement-based piezoelectric composite with ceramic/cement volume ratio = 20:80, 3: cement-based piezoelectric composite with ceramic/cement volume ratio = 35:65, 4: cement-based piezoelectric composite with ceramic/cement volume ratio = 50:50, 5: cement-based piezoelectric composite with ceramic/cement volume ratio = 75:25, 6: piezoelectric ceramic.
the observed shift to low wave number can be interpreted as a reduction in Si–O–Si bonding [29,30]. Fig. 2 also shows the influence of the ceramic/cement volume ratio on Si–O–Si bonding. As the relative proportion of ceramic is increased, the Si–O–Si bonding peaks shift to lower wave numbers (from 1126 cm1 to 1063 cm1, and from 1006 cm1 to 980 cm1). This suggests that PZT ceramic
particles attack Si–O–Si bonds, breaking down polymeric chains of [SiO4]. In contrast with Si–O–Si bonding, the Si–O–H bonding peaks (874 cm1 and 713 cm1) show no influence from the presence of PZT particles in the mixture. The same is true for the Ca2+ vibration peak (1423 cm1). This indicates that Ti–O in the ceramic
136
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
Fig. 3 (continued)
particles links to Si–O in the cement paste via a bridging oxygen, forming a Ti–O Si–O bonding structure. Further evidence for this mechanism comes from examining the Ti–O bond ([TiO6] octahedron), which in ceramic particles shows only one peak at 597 cm1 [31]. As the ceramic particles are added to the cement paste, the peak splits into two subsidiary bands, one corresponding to the original Ti–O bond at 597 cm1 and the other to the new Ti–O Si–O bond at 529 cm1. This shows that only a
small fraction of the PZT ceramic takes part in chemical reactions with the cement paste, whereas the majority remains in the original ceramic form. X-ray diffraction patterns directly show the differences in microstructure between the composite and cement paste. In Fig. 2, the abundant small broad peaks for hydrated cement paste indicate the presence of all kinds of amorphous microstructure, and a few Bragg peaks (marked by (⁄)) refer to the crystalline
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
137
Fig. 4. SIMS images with element mapping mode of cement-based piezoelectric ceramic composite.
parameters of tobermorite, which are in agreement with the results obtained by Hamid [32,33]. X-ray diffraction patterns of pure piezoelectric ceramic show the same positions as that of the cement–ceramic composite, which demonstrates that the PZT ceramic particles retain their crystalline structure in the composite. However, the tobermorite pattern seen in hydrated cement paste disappears in the composite, and new patterns (marked by (#) in Fig. 2) appear, implying that the addition of ceramic powder restricts the growth of the long [SiO4] chains that make up tobermorite, as well as reducing the size of cement paste particles. These results are in accord with the IR spectroscopy results. The chemical composition of the cement–ceramic composite was further studied with XPS. Fig. 3a–d shows the XPS spectra of Ca, Si, Ti, and Zr in the composite, covering the binding energy range of 0–500 eV. The binding energy of various peaks is calibrated by using the binding energy of C1s (285.0 eV). As shown in Fig. 3, chemical bonding of all these elements (Ca, Si, Ti, Zr) occurs in the composite. The binding energy of Ca2p (Fig. 3a) is calibrated to 347 eV and it matches well with Ca in cement paste. The results suggest that Ca exists mainly in a form similar to that in cement paste. In contrast, the binding energy peak of Si2p (Fig. 3b) shifts from 458.1 eV (in cement paste) to 458.6 eV (in the composite). This demonstrates that the chemical environment of Si differs significantly between cement paste and composite. Similar results can be found for Ti2p (Fig. 3c) and Zr3d (Fig. 3d) [34]. It can be concluded that a change of chemical environment takes place for all of Ti, Si, Ca and Zr during their incorporation into a cement-based piezoelectric ceramic composite. One prominent result pertaining to chemical reactions at the interface between ceramic particles and cement paste shows that Ti diffuses into the cement paste via Ti–O Si–O bonding. In order to check for such a process in the composite, the spatial distributions of the elements Ca, Si, Ti, Zr and Pb are all measured using the SIMS method with element mapping mode. As PZT ceramic is added to the cement paste, fine cubic particles are formed, and they are uniformly dispersed throughout the cement paste. This phenomenon can be clearly observed in SEM images [35]. In the
element-mapping SIMS images shown as Fig. 4, each point is shaded according to the concentration of the element at that point. The concentration of Pb remains confined to the location of PZT ceramic particles; in contrast, Ti – which, along with Pb, originates in the PZT ceramic – is distributed throughout the composite, not just within the ceramic particles. It is clear that Ti diffuses from the PZT ceramic particles into the cement paste once they come into contact. The SIMS images provide a direct measure of the elemental distribution in the cement-based piezoelectric composite, and these are found to be consistent with the occurrence of Ti– O Si–O bonding at the interface between the PZT ceramic particles and cement paste. This reaction results in the diffusion of Ti element into the cement matrix. As consider the chemical reaction between PZT ceramic particles and cement grains, it will affect influence of the component of composite on the material properties, such as dielectric, piezoelectric and elastic properties, of the composite. And it has been confirmed by our previous study [36]. 4. Conclusion In this study, the chemical reaction mechanisms and microstructure of a cement-based piezoelectric ceramic composite were investigated. The following conclusions can be drawn: 1. From IR spectra, there is no change of Ca2+ chemical environment as increasing of volume ratio of piezoelectric ceramic particle to cement paste, whereas [SiO4] takes an obvious ‘‘red shifting’’ in adsorption bands from 1106 cm1 to 980 cm1 and 1126 cm1 to 1100 cm1, respectively. Additionally, the [TiO6] in 597 cm1 splits into two subsidiary bands at 597 cm1 and 529 cm1. The variety of adsorption bands in IR spectra due to the contribution of the distortion of Si–O band which causes local change in the crystal field potential and hence induce the splitting of Ti–O (or Zr–O) adsorption bands. Thus the interaction of Ti–O and Si–O in cement-based piezoelectric ceramic composites occurs.
138
B. Dong et al. / Construction and Building Materials 72 (2014) 133–138
2. XRD results reveal that the patterns of tobermorite observed in hydrated cement paste disappear in that of composite. Some new patterns are found in the composite, indicating that the addition of ceramic powder both restrict the growing long [SiO4] chains of tobermorite and make cement paste particles smaller. These results are consistent with those obtained by means of IR spectrum. 3. Details of the chemical composition of cement-based piezoelectric ceramic composites were further studied with XPS. The spectra suggest that a significant change in chemical environment occurs for every element type, including Ti, Si, Ca and Zr, present in the composite. SIMS images provide a direct indication of element spatial distribution in the composite, which can be understood as the result of Ti–O Si–O bonding via chemical reactions at the interface between the PZT ceramic particles and the cement paste. The chemical reaction results in the diffusion of Ti into the cement paste.
Acknowledgements The authors would like to acknowledge financial support provided by National Key Basic Research Program funded by MOST(Project No. 2011CB013600; Issue No. 2011CB013604) and National Natural Science Foundation of China (No. 51120185002/ 51272160/U1301241); Foundation for Technology Innovation Project in Higher Education of Guangdong, China (No. 2012KJCX0091) and Collaborative Innovation Center for Advanced Civil Engineering Materials, Nanjing, PR China. References [1] Han N, Xing F. Service life consideration of concrete structures in a broad perspective, keynote speech. In: Proceedings, the 11th international symposium on structural engineering, December 18-20, 2010, Guangzhou, China, Sponsored by National Natural Science Foundation of China. [2] Han N. Role of NDE in quality control during construction of concrete infrastructures on the basis of service life design. Constr Build Mater, Elsevier Science 2004;18:163–78. [3] Chang Fu-Kuo, editor. Structural health monitoring 2000. Lancester, Pennsylvania: Technomic Publishing Co., Inc.; 1999. [4] Tzou HS, Guran A, editors. Structronics systems: smart structures, devices and systems (Parts I & II). Singapore: World Scientific; 1998. [5] Aizawa S, Kakizawa T, Higasino M. Case studies of smart materials for civil structures. Smart Mater Struct 1998;7(5):617–26. [6] Hilczer B, Mailecki J. Electrets and piezopolymers. Warsaw: PWN; 1992 (in polish). [7] Banks HT, Smith RC, Wang Y. Smart material structures: modeling, estimation, and control. New York: John Wiley & Sons Inc.; 1996. [8] George EP, Gotthardt R, Otsuka K, Trolier-McKinstry S, Wun-Fogle M, editors. Materials for smart systems II, Symposium proceedings of Materials Research Society, vol. 459. [9] Tao B. Smart/intelligent materials and structures. Beijing: Defense Industry Press; 1997 (In Chinese).
[10] Wetherhold RC, Panthalingal Nandakumar. Piezoelectric PZT/expoxy composites for sensing and actuating torsional motion. Smart Struct Mater 1993; SPIE: 266–74. [11] Yuhuan Xu. Ferroelectric materials and their applications. Amsterdam: NorthHolland; 1991. [12] Furukawa T. Piezoelectricity and pyroelectricity in polymers. IEEE Trans Electr Insul 1989;24(3):375–94. [13] Okazaki K. Developments in fabrication of piezoelectric ceramics. In: Taylor GW, Gagnepain JJ, Meeker TR, Nakamura T, Shuvalov LA, editors. Piezoelectricity. Switzerland: Gordon and Breach Science Publishers; 1985. p. 131–50. [14] Banno H. Recent developments of piezoelectric composites in Japan. In: Saito S, editor. Advanced ceramics. Oxford University Press; 1988. p. 8–26. [15] Safari A, Sa-gong G, Giniewicz J, Newnham RE. Composites piezoelectric sensors. In: Rosen CZ, Hiremath BV, Newnham R, editors. Piezoelectricity. College Park, MD: American Institute of Physics; 1992. p. 195–204. [16] Latour M, Jolivet S, Rahmoune M, Lagarrigue O, Roure A. Piezopolymer transducers in the active control of vibrations. In: Proceedings of 8th international symposium on electrets, 1994. p. 985–90. [17] Safari A. Novel piezoelectric ceramics and composites for sensor and actuator applications. Mater Res Innovat 1999;2:263–9. [18] Mazur K. Polymer-ferroelectric ceramic composites. In: Nalwa HS, editor. Ferroelectric polymers: chemistry, physics, and applications. New York: M. Dekker Inc.; 1995. p. 539–610. [19] Li Zongjin, Dong Biqin, Zhang Dong. Influence of polarization on properties of 0–3 Cement-based PZT composite. Cem Concr Compos 2005;27(1): 27–32. [20] Gleize PJP, Muller A, Roman HR. Microstructural investigation of a silica fumecement-lime mortar. Cem Concr Compos 2003;25:171–5. [21] Damidot D, Nonat A. In: Nonat A, Mutin C, editors. Hydration and setting of cement. E & FN Spon; 1992. p. 23. [22] Hewlett PC. LEA’S Chemistry of cement and concrete. ARNOLD; 1998. [23] Struble Leslie, Skalny Jan, Mindess Sidney. A review of the cement–aggregate bond. Cem Concr Res 1980;10(2):277–86. [24] Cotton Albert. Advanced inorganic chemistry. John Wiley & Sons; 1988. [25] Okada Yoshihiko, Masuda Tsuguya, Takada Minori, Xu Lingling, Mitsuda Takeshi. Relationship between NMR 29Si chemical shifts and FT-IR wave numbers in calcium silicates. In: Colombet P, Grimmer AR, Zanni H, Sozzani P, editors. Nuclear magnetic resonance spectroscopy of cement-based materials. Springer; 1998. p. 69–78. [26] Henning O. In: Farmer VC, editor. The infrared spectra of minerals. London: Mineralogical Society; 1974. p. 445. [27] Bensted J, Prakash S. Cem Technology Sep/Oct.; 1974. pp. 440. [28] Bensted J, Prakash S. Cem Technology July/Aug.; 1974. pp. 378. [29] Bell G, Bensted J, Glasser FP. Adv Cem Res 1998; (2):61. [30] Gong Hongyu, Zhang Yujun, Quan Jing, Che Songwei. Preparation and properties of cement based piezoelectric composites modified by CNTs. Curr Appl Phys 2011(11):653–6. [31] Rao CNR. Chemical application of Infrared Spectroscopy. New York: Academic Press; 1963. p. 8. [32] Mazen SA, Metawe F, Mansour SF. IR absorption and dielectric properties of Li– Ti ferrite. J Phys D: Appl Phys 1997;30:1799–808. [33] Hamid SA. Z Krist; 1981. p. 54. [34] Klur I, Pollet B, Virlet J, Nonat A. C–S–H structure evolution with Calcium content by multinuclear NMR. In: Colombet P, Grimmer AR, Zanni H, Sozzani P, editors. Nuclear magnetic resonance spectroscopy of cement-based materials. Springer; 1998. p. 119–41. [35] Bhaskar S, Majumder SB, Jain M, Dobal PS, Katiyar RS. Studies on the structural, microstructural and optical properties of sol–gel derived lead lanthanum titanate thin films. Mater Sci Eng, B 2001;87:178–90. [36] Xing Feng, Dong Biqin, Li Zongjin. Dielectric, piezoelectric, and elastic properties of cement-based piezoelectric ceramic composites. J Am Ceram Soc 2008;91(9):2886–91.