Influence of arrangement field on magnetostrictive and
mechanical properties of magnetostrictive composites
DONG Xu-feng(董旭峰)1, GUAN Xin-chun(关新春)2, QI Min(齐 民)1, OU Jin-ping(欧进萍)2, 3
1. School of Materials Science and Engineering, Dalian University of Technology, Dalian 116024, China;
2. School of Civil Engineering, Harbin Institute of Technology, Harbin 150090, China;
3. School of Civil and Hydraulic Engineering, Dalian University of Technology, Dalian 116024, China
Received 10 August 2009; accepted 15 September 2009
Abstract: Non-aligned and aligned polymer-bonded Tb0.3Dy0.7Fe2 composites with 20% particle volume fraction were prepared under different arrangement fields (i.e. 0, 10 kA/m, 20 kA/m, 30 kA/m, 60 kA/m and 100 kA/m) during their gel process. Static magnetostriction, dynamic magnetostriction, elastic modulus and compressive strength of all specimens were tested and compared. Experimental results indicate that all the parameters are positively dependent on the arrangement field. The dependence is significant at low field levels, the critical value of which is 30 kA/m for the composites fabricated. No obvious improvement of the properties can be observed for a larger field. Such critical values are defined as the optimal arrangement field to manufacture magnetostrictive composites.
Key words: magnetostrictive composites; Terfenol-D; arrangement field; magnetostriction
1 Introduction
Magnetostrictive effect, the phenomenon of the change in elastic state exhibited by certain ferromagnetic materials subjected to changes in the magnetization state, was first observed by JOULE more than 150 years ago[1].
Early researchers studied magnetostriction of iron, nickel, and cobalt. With the discovery of the significant magnetostriction found in the rare earth materials in 1960s, a new era in magnetostrictive materials was begun. An excellent magnetostrictive material discovered by CLARK in 1970s was called Terfenol-D, a specially formulated compound of terbium, dysprosium, and iron, which has saturation magnetostriction more than 1×10-2 at room temperature and relatively small applied field. Therefore, it has been a commercially available magnetostrictive material for application in many fields[2]. However, the brittleness in tension and the development of eddy currents have limited its useful frequency range[3]. In response to these shortcomings, SANDLUND et al[4] combined Terfenol-D particles with a passive polymer matrix to form magnetostrictive particulate composites. The insulating layer created by the matrix between the particles eliminates eddy current losses at high frequencies. Moreover, the polymer matrix produces a relatively tough material that can better accommodate tensile and shear loading states[5-7].
Composites based on 1-3 architectures were produced by aligning Terfenol-D particles in the matrix with an applied magnetic field during the thermal cure process[8-10]. Previous studies proved that the 1-3 type magnetostrictive composites exhibit larger magnetostriction than those based on 0-3 architectures, which are fabricated by dispersing Terfenol-D particles randomly in a polymer matrix[11-13]. But surprisingly, studies have not agreed on the degree of influence of the arrangement field on the properties of the composites. For instance, in SANDLUND’s experiment, the saturation magnetostriction of 1-3 composites was 40% higher than that of 0-3 ones, while in LIM’s experiment, the promotion was 10%[12-13]. We conjecture the reason to the different arrangement fields applied, which needs to be verified by experiments. In addition, the effects of the arrangement fields on magnetomechanical coupling coefficient, elastic modulus and compressive strength need to be studied.
In this study, 0-3 type and 1-3 type magnetostrictive composites were prepared by dispersing Terfenol-D particles with 20% particle volume fraction in an unsaturated polyester resin matrix. Different arrangement magnetic fields (i.e. 0, 10 kA/m, 20 kA/m, 30 kA/m, 60 kA/m and 100 kA/m) were applied during the gel process to align the particles along the longitudinal direction of the mold. The static and dynamic magnetostrictive properties, magnetomechanical coupling coefficient, elastic modulus and compressive strength of the samples were tested.
2 Experimental
2.1 Materials
Terfenol-D powder was supplied by the Gansu Tianxing Rare Earth Functional Materials Co. Ltd., China. The particle shape is irregular (Fig.1) and the particles size is normally distributed in the range of 30-500 μm (Fig.2). A three-part unsaturated polyester resin with a low viscosity of 0.2 Pa?s(25 ℃) and a high elastic modulus of 3 GPa was used as polymer matrix. The gel time of the resin system was 30 min at room temperature.
Fig.1 Irregular shape of Terfenol-D particles
Fig.2 Particle size distribution
2.2 Specimen fabrication
All the ingredients were mixed by using a stirrer bar for about 5 min at room temperature. After all ingredients were evenly mixed, the mixture was put into a duralium mold with a cavity of d10 mm×25 mm. Then the mold was placed in a vacuum for degassing. Three minutes later, the mold was sealed and placed in an arrangement field (i.e. 0, 10 kA/m, 20 kA/m, 30 kA/m, 60 kA/m and 100 kA/m) which was generated by a PEM-1022LS magnetic field system. Twenty-two minutes later, the resin became gel and the mixture system was supposed to be stable. Then, the mold was placed in an oven and cured at 80 ℃ for 2 h. Finally, specimens were obtained by removing from the mold. To minimize the end effects, very small bi-directional strain gauges (1 mm×1 mm) located at the center of the length direction of the composite samples were used to measure magnetostriction, as shown in Fig.3.
Fig.3 Photo of specimen
2.3 Apparatus and procedures
The magnetic field was generated by a PEM-1022LS magnetic field system, the intensity of which was adjusted in the range of 0-796 kA/m by tuning the electric current intensity. A Hall probe connected with a CST-Ⅱmagnetometer was used to measure the magnetic field. A YE2537 strain measuring instrument was used to measure the magnetostriction.
Under constant external preload, dynamic magnetostriction d33 is obtained from
(1)
where λ is the magnetostriction and H is the applied field. To obtain d33 behavior with the applied field, a three- parameter sigmoidal function was fitted to λ vs H plots. The pointwise slopes of the fitting curves correspond to d33, which can be achieved by differentiating the curves.
In order to measure the dynamic magnetomechanical properties of the composites in the longitudinal direction at room temperature with zero stress bias, specimens were placed inside two solenoids: an inner solenoid for generating an AC magnetic drive field, and an outer solenoid for providing a DC magnetic bias field. The drive solenoid was made by winding a layer of 80-turn magnetic wire, connected electrically parallel to each other, on a 60 mm-long, 12 mm-inner diameter Teflon bobbin. The bias solenoid had a length of 60 mm, an inner diameter of 18 mm, with a multi-layer of 830 turns. A pair of steel rods, situated at both ends of the sample, homogenized the magnetic flux distribution at or near the ends[14]. An impedance analyzer (SI 1260) was used to generate a swept sinusoidal voltage of constant amplitude at a prescribed frequency range. The output impedance frequency spectra at different bias field levels were recorded. The magnetomechanical coupling coefficient (k33) was calculated as
(2)
where fr and fa are the resonance and anti-resonance frequencies, respectively.
To test the mechanical properties including the elastic modulus and the compressive strength under zero magnetic field bias, the samples were mechanically loaded using an MTS system. All the compression tests reported were conducted in load control mode. The YE2537 strain measuring instrument was used to measure the strain.
3 Results and discussion
3.1 Effects of arrangement field on static magnetostriction
Fig.4 shows the relationship between the static magnetostriction and the applied field of the specimens. The magnetostriction is a function of the applied field. Under the same applied field, the strain of the specimens prepared under larger arrangement field is larger than that under smaller field. Fig.5 presents the relationship between saturation magnetostriction and arrangement field. It is indicated that the saturation magnetostriction increases fast with increasing the arrangement field at low field levels (i.e. 0-30 kA/m), while it becomes slower at high levels (30-100 kA/m). The specimen prepared under 10 kA/m arrangement magnetic field presents a saturation magnetostriction 32% larger than that of non-aligned one, while the specimen prepared under a 30 kA/m magnetic field presents a 61% larger saturation magnetostriction. However, the promotion of static magnetostrictive property becomes insignificant by applying an arrangement field larger than 30 kA/m.
Fig.4 Relationship between static magnetostriction and applied field for specimen prepared under different arrangement fields
Fig.5 Relationship between saturation magnetostriction and arrangement field
The reason for this phenomenon is possibly that more particles are rotated in their easiest magnetic direction along the field direction with increasing the arrangement field. When the field reaches a certain value, nearly all the particles have been fully rotated, and no obvious increase of magnetostriction can be observed with a larger field.
3.2 Effects of arrangement field on dynamic magnetostriction
Fig.6 presents plots of d33 as a function of the applied field for all specimens gelled under different arrangement fields. It is shown that, for all the composites, d33 increases initially up to a certain peak and then decreases with increasing the applied field. As the arrangement field increases, the maximum dynamic magnetostriction increases, as shown in Fig.7. With the same reason of variation of the static magnetostriction, the tendency becomes insignificant when the arrangement field is larger than 30 kA/m.
Fig.6 Relationship between dynamic magnetostriction and applied field
Fig.7 Relationship between maximum dynamic magneto- striction and arrangement field
3.3 Effects of arrangement field on magnetomechanical coupling coefficient
The magnetomechanical coupling coefficients under different bias fields (i.e. 14 kA/m, 28 kA/m, 42 kA/m and 50 kA/m) for all specimens are obtained by substituting their resonance and anti-resonance frequencies into Eq.(2). The dependence of k33 on magnetic bias field for all specimens is similar, as shown in Fig.8. k33 reaches its maximum value when the bias field approximates 40 kA/m, then decreases with increasing the bias field, although the data at high magnetic bias field levels are not given due to the limit of the DC current source. In the same magnetic bias field level, specimens prepared at high arrangement field levels present a larger magnetomechanical coupling coefficient than those prepared at low levels. In Fig.9 the relationship between k33 and the arrangement field at a 42 kA/m bias field is plotted. The improvement of k33 with increasing the arrangement field is significant at low field levels (i.e. 0-30 kA/m), while it becomes less obvious at high levels (i.e. 30-100 kA/m).
Fig.8 Relationship between magnetomechanical coupling coefficient and magnetic bias field
Fig.9 Relationship between magnetomechanical coupling coefficient and arrangement field at 42 kA/m bias field
3.4 Effects of arrangement field on mechanical properties
Fig.10 shows the relationship between the elastic modulus at zero bias field and the arrangement field. It is indicated that the elastic modulus dramatically increases with increasing the arrangement field at low field levels. By applying an arrangement field larger than 30 kA/m, the promotion of the elastic modulus becomes slight. The specimens prepared under the arrangement field of 10 kA/m and 30 kA/m present elastic modulus 8.6% and 15.6% larger than non-aligned specimen, respectively. The phenomenon is consistent with the indication of the rule-of-mixtures approach. With the increase of arrangement field, the modulus values are closer to the theoretical upper bound, suggesting that the composites are closer to a 1-3 configuration rather than 0-3 type [15].
Fig.10 Relationship between elastic modulus and arrangement field
The dependence of compressive strength on arrangement field is similar to that of the elastic modulus, as shown in Fig.11. The specimens prepared under the arrangement field of 10 kA/m and 30 kA/m present 13.9% and 15.1% larger compressive strength than non-aligned specimen, respectively.
Fig.11 Relationship between compressive strength and arrangement field
4 Conclusions
1) Magnetostrictive properties of aligned magnetostrictive composites are larger than those of non-aligned ones.
2) The saturation magnetostriction and peak dynamic magnetostriction increase with increasing the arrangement field.
3) At the same bias field level, the magneto- mechanical coupling coefficient increases with increasing the arrangement field.
4) The elastic modulus and the compressive strength at zero bias fields positively depend on arrangement field.
5) An optimal arrangement field exists at 30 kA/m for no obvious improvement of the properties could be expected for a field larger than the optimal value.
References
[1] Joule J P. On a new class of magnetic forces [J]. Annals of Electricity, Magnetism, and Chemistry, 1842, 8: 219-224.
[2] Clark A E. Magnetostrictive rare earth-Fe2 compounds [C]// WOLFARTH E P. Ferromagnetic Materials. Amsterdam: North-Holland Publishing Company, 1980: 531-589.
[3] Anjanappa M, WU Y. Magnetostrictive particulate actuators: Configuration, modeling and characterization [J]. Smart Mater Struct, 1997, 6: 393-402.
[4] Sandlund L, Fahlander M, Cedell T, CLARK A E, Restorff J B, Wun-Fogle M. Magnetostriction, elastic moduli, and coupling factors of composite Terfenol-D [J]. J Appl Phys, 1994, 75(10): 5656-5658.
[5] de Angulo L R, Abell J S, Harris I R. Magnetostrictive properties of polymer bonded Terfenol-D [J]. J Magn Magn Mater, 1996, 157/158: 508-509.
[6] Duenas T A, Carman G P. Experimental results for magnetostrictive composites [J]. American Society of Mechanical Engineers, Aerospace Division, 1998, 57: 63-71.
[7] Duenas T A, Carman G P. Large magnetostrictive response of Terfenol-D resin composites [J]. J Appl Phys, 2000, 87: 4696-4700.
[8] OR S W, Yung C S, Lo C Y. A 64-kHz sandwich transducer fabricated using pseudo 1-3 magnetostrictive composite [J]. IEEE Transactions on Magnetics, 2006, 42(1): 47-50.
[9] Altin G, HO K K, Henry C P, Carman G P. Static properties of crystallographically aligned Terfenol-D/polymer composites [J]. J Appl Phys, 2007, 101: 033537.
[10] L? Xiang-ke, OR S W, Liu Wei, Liu Xiong-hua, Zhang Zhi-dong. Magnetomechanical properties of epoxy-bonded (Tb0.3Dy0.7)1-xPrxFe1.55 (0≤x≤0.4) pseudo-1–3magnetostrictive composites [J]. J Phys D: Appl Phys, 2009, 42: 035002.
[11] Rodríguez C, Barrio A, Orue I, Vilas J L, León L M, Barandiarán J M, Fdez-GubiedaRuiz M L. High magnetostriction polymer-bonded Terfenol-D composites [J]. Sensors and Actuators A, 2008, 142: 538-541.
[12] Sandlund L, Cedell T. Terfenol-D powder composite with high frequency performance [C]// MCCOLLOM M D, HAMONIC B F, WILSON O B. Transducers for Sonic and Ultrasonics. Technomic Publishing Co., INC, 1993: 113-118.
[13] Lim S H, Kim S R, Kang S Y, Park J K, Nam J T, Son D. Magnetostrictive properties of polymer-bonded Terfenol-D composites [J]. J Magn Magn Mater, 1999, 191: 113-121.
[14] OR S W, Nersessian N, Carman G P. Dynamic magnetomechanical behavior of Terfenol-D/epoxy 1–3 particulate composites [J]. IEEE Transaction on Magnetics, 2004, 40(1): 71-77.
[15] Nersessian N, OR S W, Carman G P. Magneto-thermo- mechanical characterization of 1–3 type polymer-bonded Terfenol-D composites [J]. J Magn Mag Mater, 2003, 263: 101-112.
Foundation item: Projects(50478024, 50738002) supported by the National Natural Science Foundation of China; Project (2007CB714204) supported by the National Basic Research Program of China
Corresponding author: DONG Xu-feng; Tel: +86-411-84708441; E-mail: dongxf@dlut.edu.cn
DOI: 10.1016/S1003-6326(09)60050-9
(Edited by YANG Bing)
