Dy substitution effect on the temperature dependences of magnetostriction in Pr1–xDyxFe1.9 alloys
Tang Yan-Mei1, 2, †, , Huang Hai-Fu2, Tang Shao-Long2, ‡, , Du You-Wei2
College of Physics and Technology, Guangxi Normal University, Guilin 541004, China
Institute of Materials Engineering, National Laboratory of Microstructures and Department of Physics, Nanjing University, Nanjing 210093, China

 

† Corresponding author. E-mail: tangym0707@163.com

‡ Corresponding author. E-mail: tangsl@nju.edu.cn

Project supported by the National Natural Science Foundation of China (Grant No. U1232210), the Science Foundation of Guangxi Zhuang Autonomous Region, China (Grant No. 2015GXNSFBA139020), and the Enhancement of the Basic Ability of Teachers of Guangxi Zhuang Autonomous Region, China (Grant No. KY2016YB068).

Abstract
Abstract

The temperature dependences of magnetostriction in Pr1−xDyxFe1.9 (0 ≤ x ≤ 1.0) alloys between 5 K and 300 K were investigated. An unusual decrease of magnetostriction with temperature decreasing was found in Pr-rich alloys (0 ≤ x ≤ 0.2), due to the change of the easy magnetization direction (EMD). Dy substitution reduces the magnetostriction in high-magnetic field (10 kOe ≤ H ≤ 90 kOe) at 5 K, while a small amount of Dy substitution (x = 0.05) is beneficial to increasing the magnetostriction in low-magnetic field between 10 K and 50 K. This makes the alloys a potential candidate for low temperature applications.

1. Introduction

The polycrystalline PrFe2 alloy is of significant technological interest due to its giant magnetostriction (5600 ppm) at 0 K and the cheaper price of Pr than heavy rare-earth elements.[1,2] Recently, we reported that the magnetostriction λ111 of PrFe1.9 alloy is as large as 6700 ppm at 70 K.[3] However, its application is somewhat limited, because PrFe1.9 possesses large magnetocrystalline anisotropy below 100 K, which makes it difficult for the material to get saturated at low temperature.[3] Therefore, much attention should be paid to the preparation of RR′Fe2 alloys or the substitution of Fe with the other metal element in R′Fe2 alloys, for the purpose of developing new magnetostrictive materials in low temperature applications.[410] Among these elements, Dy is regarded as being ideal. Ren et al. reported that the anisotropies of PrFe2 and DyFe2 can compensate each other in the Dy1−xPrxFe2 system, as proved phenomenologically by a single-ion approach.[4,5] They found that the anisotropy of Dy1−xPrx(Fe0.9B0.1)1.93 alloys decreases with the increase of x when 0 ≤ x ≤ 0.3,[4] and the saturation magnetization Ms at 5 K or 295 K for Dy1−xPrx(Fe0.35Co0.55B0.1)2 alloys decreases to a minimum, then increases with the increase of the Pr content.[5] This meaningful work indicates that Dy substitution can help to decrease the magnetocrystalline anisotropy of PrFe1.9 alloy, and might help to increase the low-field magnetostrictive properties at low temperature. Furthermore, Clark et al. found that the magnetostriction of DyFe2 shows some abnormal features at low temperature (0 K ≤ x ≤ 300 K).[1,11] And the investigation on step-scanned {440} and {222} x-ray diffraction (XRD) reflections of PrFe1.9 alloy indicated that the easy magnetization direction (EMD) of the alloy changes from [111] to [100] with temperature decreasing.[3] The change of the EMD leads to the change of the magnetostriction in PrFe1.9 alloy at some temperatures. Therefore, some particular features might be expected in the magnetostrictive property of Pr1−xDyxFe1.9 alloys, due to the particular magnetostrictive properties in PrFe1.9 and DyFe2 alloys.

In this paper, the temperature dependences of magnetostriction in the polycrystalline Pr1−xDyxFe1.9 (0 ≤ x ≤ 1.0) alloys between 5 K and 300 K were investigated in detail.

2. Experimental

The samples were prepared by the arc-melting and subsequent high pressure annealing.[12,13] Conventional XRD powders analysis was carried out using Cu K α radiation with a Rigaku D/Max-gA diffractometer. The powder XRD patterns for Pr1−xDyxFe1.9 alloys with different Dy concentrations at room temperature are shown Fig. 1. A single MgCu2-type cubic Laves phase in all the samples investigated is confirmed, which could be ascribed to the high-pressure annealing method. The magnetostriction (λ||) was measured by standard strain-gauge technique in the parallel direction of magnetic fields for each sample (a rectangular block with a size of about 6 mm×4 mm×1 mm for each sample), the magnetic field was supplied by a physical property measurement system (PPMS). A superconducting quantum interference device magnetometer (SQUID) was used to measure the magnetization curves of the samples.

Fig. 1. XRD patterns for Pr1−xDyxFe1.9.
3. Results and discussion

Figure 2 shows the magnetostrictions (λ||) as a function of the applied field at 5 K. It is found that the magnetostrictions of all the alloys with different x are far from saturation with the maximum magnetic field of 90 kOe. This indicates that the magnetocrystalline anisotropy of the alloys is quite large at 5 K. The slopes of the magnetostrictions curves in the alloys with 0 ≤ x ≤ 0.2 are larger than those of the alloys with 0.2 ≤ x ≤ 1.0. This indicates that the magnetocrystalline anisotropy of the alloys with 0 ≤ x ≤ 0.2 is larger than that of the alloys with 0.2 ≤ x ≤ 1.0 at 5 K. This tendency is similar to the results of the Dy1−xPrx(Fe0.9B0.1)1.93 system and Dy1−xPrxFe1.9 system at room temperature.[4,14] In high-magnetic field (10 ≤ H ≤ 90 kOe), Dy substitution reduces the magnetostriction. The magnetostriction decreases sharply from 5000 ppm (x = 0.0) to 320 ppm (x = 1.0) in a magnetic field of 90 kOe. This might be due to that the magnetostriction of PrFe1.9 is larger than DyFe1.9.[1,2] However, in low magnetic field (0 ≤ H < 10 kOe), Dy substitution yields an increased magnetostriction, which is shown in the inset of Fig. 2. It shows that the alloy with x = 0.05 has larger magnetostriction than the other alloys in the magnetic field of 0 ≤ H < 10 kOe, which indicates that a small amount of Dy substitution is beneficial to increase the magnetostriction at low temperature and in low magnetic field. This result is similar to the result reported by Ren.[5] They found in Dy1−xPrx(Fe0.35Co0.55B0.1)2 system that, at room temperature, λ|| increases with the increase of Dy content when 0.8 ≤ x ≤ 1.0, and Dy0.2Pr0.82(Fe0.35Co0.55B0.1)2 possesses the largest magnetostriction.

Fig. 2. The magnetostriction (λ||) of Pr1−xDyxFe1.9 as a function of the applied field at 5 K. The inset shows the magnetostriction in the magnetic field of 0 ≤ H ≤ 1.2 kOe.

The temperature dependences of magnetostriction (λ||) during cooling from 300 K to 10 K in a magnetic field of 50 kOe were measured, which are shown in Fig. 3(a). Striking dissimilarity is found among these alloys. For Dy-rich alloys (0.4 ≤ x ≤ 1.0), the magnetostriction shows some peculiar features similar to that of DyFe2, which are shown in Fig. 3(b). As seen in Fig. 3(b), as temperature decreases, the magnetostriction first decreases with the decrease of temperature to some temperature (at near 80 K for the sample x = 1.0), and then even changes sign with the further temperature decreasing. The temperature, at which the magnetostriction changes sign is near 80 K, 105 K, 185 K, and 150 K for x = 1.0, x = 0.8, x = 0.6, and x = 0.4, respectively. This particular feature was first reported by Clark in DyFe2 alloy, and cannot fit a simple single-ion expression.[1,11] Therefore, the peculiar feature in Dy-rich alloys can be attributed mainly to the small and complex nature of λ100 of DyFe2.[1,11,15] Recently, Bowden theoretically pointed out that this abnormality is due to the competition between n = 2, 4, and 6 crystal field distortions in DyFe2 alloy, of which temperature dependences can be accurately modeled using the Callen and Callen model.[15] While for the Pr-rich alloys (0 ≤ x ≤ 0.2), the magnetostriction of the alloys has another unusual kind of feature, as shown in Fig. 3(a). It was found from Fig. 3(a) that, as the temperature decreases, the magnetostriction first increases to some temperature (at near 50 K for the sample x = 1.0), and then unusually decreases with the further temperature increasing. This peculiar feature is not like the feature of TbFe2.[1] Recent study on the step-scanned {440} and {222} XRD reflections of PrFe1.9 indicated that the EMD of the alloy changes from [111] to [100] with temperature decreasing between 30 K and 70 K. Therefore, we consider the abnormal decrease of the magnetostriction in Pr-rich alloys is probably due to the change of the EMD in the alloys.[3] For PrFe1.9 alloy, the change of the EMD occurs between 70 K and 300 K, and for Pr0.95Dy0.05Fe1.9 alloy, the change of the EMD might occur between 35 K and 120 K. This indicates that the increase of Dy increases the spinreorientation temperature TSR of the alloys. This result is similar to the results of TbxDy1−x(Fe0.8Co0.2)2 and TbxDy0.1−xPr0.3(Fe0.9B0.1)1.93 system reported by Liu et al.[16,17]

Fig. 3. (a) The temperature dependences of magnetostriction (λ||) in Pr1−xDyxFe1.9 (0 ≤ x ≤ 1.0) alloys (H = 50 kOe). (b) The magnetostriction (λ||) of the alloys with 0.4 ≤ x ≤ 1.0 (H = 50 kOe). (c) The magnetostriction (λ||) of the alloys with 0.6 ≤ x ≤ 1.0 between 200 K and 300 K (H = 50 kOe).

Furthermore, it is found that Dy substitution for Pr with x = 0.6 yields a minimum λ|| between 250 K and 300 K. As shown in Fig. 3(c), the magnetostriction of the sample with x = 0.6 increases from 184 ppm at 250 K to 313 ppm at 300 K, which is even smaller than the 197 ppm at 250 K to 344 ppm at 300 K of the sample with x = 0.0. This result agrees well with the result reported by Shi et al.,[14] in which the magnetostriction of Dy0.6Pr0.4Fe1.9 also exhibits a minimum at room temperature. According to Clark, the magnetic moment of Pr1−xDyxFe1.9 can be described as: μs = xμDy − 1.9μFe − (1 − x)μPr.[1,5,14] And the decrease in Curie temperature with the increase in the atomic number is directly reflected in the room temperature value of the rare-earth sublattice magnetization.[1,4] Therefore, the magnetic moment compensation point at x = 0.6 between 250 K and 300 K can be well explained. The atomic number of Pr3+ is less than that of Dy3+, so the sublattice magnetization of Pr3+ decreases more rapidly than that of Dy3+ with temperature increasing,[4] which leads to the minimum magnetic moment μs of the alloys between 250 K and 300 K, and hence the minimum value of the magnetostriction in the alloy with x = 0.6.

For practical magnetostrictive materials, the magnetostriction in low field is more important than in high field. Therefore, the temperature dependences of the magnetostriction (λ||) and magnetization (M) of PrFe1.9 and Pr0.95Dy0.05Fe1.9 were measured during cooling from 200 K to 10 K at a magnetic field of 5 kOe, which are shown in Figs. 4(a) and 4(b), respectively. Two interesting phenomena could be observed in the figures. Firstly, a sharp decrease of the magnetostriction can be evidently observed. As shown in Fig. 4(a), the magnetostriction of PrFe1.9 decreases sharply between 30 K and 70 K and the magnetostriction of Pr0.95Dy0.05Fe1.9 decreases relatively gradually between 35 K and 120 K. Correspondingly, the sharp decrease of magnetization over the corresponding temperature interval is also evidently detected in Fig. 4(b). This illustrates that a sharp change of the anisotropy may occur in the corresponding temperature range,[3,18,19] which is attributed to the change of the EMD. Then, the abnormal decrease of magnetostriction in Pr-rich alloys shown in Fig. 3 can be well explained.[3] Take PrFe1.9 as an example, the EMD of the alloy lies along [100] between 10 K and 30 K. In this temperature range, the magnetostriction decreases monotonously to a minimum at 30 K due to the decrease of the magnetization. However, between 30 K and 70 K, a change of the EMD from [100] to [111] occurs, which leads to the abnormal decrease of the magnetocrystalline anisotropy to a minimum at 70 K, and hence the maximum of magnetostriction at 70 K at a given magnetic field. The changes of the EMD with temperature decreasing were also found in Mn substitute for Fe in Sm0.88Dy0.12(Fe1−xCox),[18,19] Tb0.3Dy0.7Fe2,[20] Tb0.27Dy0.73(Fe1−xCox),[21] and Ho0.85Tb0.15(Fe1−xCox).[22] Secondly, Dy substitution yields an increased magnetostriction over a long temperature interval. As seen in Fig. 4(a), the magnetostriction of Pr0.95Dy0.05Fe1.9 is larger than that of PrFe1.9 between 10 K to 50 K. This result indicates that a small amount of Dy substitution can help to enhance the low-field magnetostriction, not merely at 5 K (as discussed in the inset of Fig. 2), but in a long range of temperature. This particular feature makes the alloy a potential candidate for low-field applications at low temperature.

Fig. 4. The temperature dependences of magnetostriction (λ||) (a) and magnetization (M) (b) of the alloys with x = 0.0 and x = 0.05 in a magnetic field of 5 kOe.
4. Conclusion

The temperature dependences of the magnetostriction in polycrystalline Pr1−xDyxFe1.9 (0 ≤ x ≤ 1.0) alloys between 5 K and 300 K were investigated in detail. Distinct features have been found among the alloys with different Dy contents. For the Dy-rich alloys (0.4 ≤ x ≤ 1.0), the magnetostriction abnormally changes sign at some temperature due to the small and complex nature of λ100 of DyFe2. While for the Pr-rich alloys (0 ≤ x ≤ 0.2), the magnetostriction decreases unusually with temperature decreasing. A small amount of Dy substitution (x = 0.05) is beneficial to enhance the magnetostriction in low-magnetic field (H = 5 kOe) between 10 K and 50 K. These findings have significant impacts and applications in low-temperature and low-magnetic fields.

Reference
1Clark A E1980Ferromagnetic MaterialsNorth-HollandAmsterdam531589531–89
2Ren W JZhang Z D 2013 Chin. Phys. 22 077507
3Tang Y MXie RZhang LGao J LChen L YXia W BWang YTang S LDu Y W 2013 J. Appl. Phys. 113 233902
4Ren W JZhang Z DZhao X GYou C YGeng D Y 2003 J. Alloys. Compd. 359 119
5Ren W JLi DSui Y CLui WZhao X GLiu J JLi JZhang Z D 2006 J. Appl. Phys. 99 08M701
6Tang Y MChen L YWei JTang S LDu Y W 2014 Chin. Phys. 23 077503
7Wang YTang S LLi Y LXie RDu Y W 2013 Chin. Phys. 22 037503
8Wang B WWu C HTang S LJin X MZhuang Y ZZhong X PLi R QLi J Y 1996 Chin. Phys. Lett. 13 231
9Ren W JLi DLui WLi JZhang Z D 2008 J. Appl. Phys. 103 07B311
10Ren W JZhang Z DZhao X GLiu WGeng D Y 2004 Appl. Phys. Lett. 84 562
11Clark A EBelson H S 1972 Phys. Rev. 5 3642
12Shi Y GZhai LTang S LChen R J 2011 J. Appl. Phys. 109 07A915
13Zhao X GWu G HWang J HJia K CZhan W S 1996 J. Appl. Phys. 79 6225
14Shi Y GTang S LLv L YFan J Y 2010 J. Alloys. Compd. 506 533
15Bowden G Jde Groot P A JO’Neil J DRainford B DZhukov A A 2004 J. Phys.: Condens. Matter 16 2437
16Liu J JRen W JLi DSun N KZhao X GLi JZhang Z D 2007 Phy. Rev. 75 064429
17Liu J JRen W JZhang Z DLi DLi JZhao X GLiu WOr S W 2006 J. Appl. Phys. 100 023904
18Guo H QGong H YYang H YLi Y FYang L YShen B GLi R Q 1996 Phy. Rev. 54 4107
19Guo Z JZhang Z DWang B WZhao X G 2000 Phy. Rev. 61 3519
20Funayama TKobayashi TSakai ISahashi M 1992 Appl. Phys. Lett. 61 114
21Dhilsha K RRama R K V S 1993 J. Appl. Phys. 73 1380
22Dhilsha K RRama R K V S 1990 J. Appl. Phys. 68 259