Accepted Manuscript
Static magnetic susceptibility of radiopaque NiTiPt and NiTiEr
Drahomír Chovan, Abbasi Gandhi, Jim Butler, Syed A.M. Tofail
PII: S0304-8853(17)32981-5
DOI: https://doi.org/10.1016/j.jmmm.2017.12.090
Reference: MAGMA 63559
To appear in: Journal of Magnetism and Magnetic Materials
Please cite this article as: D. Chovan, A. Gandhi, J. Butler, S.A.M. Tofail, Static magnetic susceptibility of
radiopaque NiTiPt and NiTiEr, Journal of Magnetism and Magnetic Materials (2017), doi: https://doi.org/10.1016/
j.jmmm.2017.12.090
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.
Static magnetic susceptibility of radiopaque NiTiPt and
NiTiEr
Drahomı́r Chovan∗, Abbasi Gandhi, Jim Butler, Syed A. M. Tofail∗
Department of Physics, and Materials and Surface Science Institute,
University of Limerick, Ireland
Abstract
Magnetic properties of metallic alloys used in biomedical industry are im-
portant for the magnetic resonance imaging (MRI). If the alloys were to be
used for long term implants or as guiding devices, safety of the patient as
well as the medical staff has to be ensured. Strong response to the external
magnetic field can cause mechanical damage to the patients body. In this
paper we present magnetic susceptibility of nickel rich, ternary NiTiPt and
NiTiEr to static magnetic field. We show that the magnetic susceptibility of
these radiopaque alloys has values in low paramagnetic region comparable to
the binary nickel-titanium. Furthermore, we studied the effect of the ther-
mal and mechanical treatments on magnetic properties. Despite deviation
from linear M (H) treated samples spanning small region around H = 0, the
linearity of the M (H) and χ = dM/dH values suggest that these ternary
alloys are safe to use under MRI conditions.
Keywords: static magnetic suscpeptibility, nitinol, rare-earth metal, MRI
∗
Corresponding authors
Email addresses: Drahomir.Chovan@ul.ie (Drahomı́r Chovan ),
Tofail.Syed@ul.ie (Syed A. M. Tofail)
Preprint submitted to Elsevier December 27, 2017
safety
1 1. Introduction
2 Medical diagnostic imaging is minimally invasive medical procedure that
3 require the patient to be exposed to different types of electromagnetic radi-
4 ation. Compatibility of a medical device to imaging techniques restricts the
5 use of particular material in such procedure. Both, high frequency ionizing
6 radiation and non-ionizing radiation in kHz to MHz range are used during
7 procedures, e.g. in X-ray fluoroscopy or magnetic resonance imaging (MRI)
8 [1, 2]. Electromagnetic properties and the geometry of the medical device
9 impacts the image quality, but more importantly, the energy transferred from
10 diagnostic radiation posses thermal risks and the mechanical response to the
11 fields used can cause damage to the surrounding tissues.
12 A high static magnetic field (up to 3 Tesla) is typically used during MRI.
13 Radiofrequency pulses in the MHz range are superimposed to a static mag-
14 nitic field to obtain such magnetic resonance images. A combination of these
15 two fields acting on a metallic object can result in serious damage if the object
16 is highly susceptible to the external magnetic field. Exposing human body
17 tissue to a static field B0 , as well as superimposed gradient fields dB/dx has
18 no side effects and static fields as high as 4 T are determined to be safe even
19 for infants [3]. RF pulses can however cause tissue heating or burning when
20 exposed for a substantial time. The increase of temperature depends on the
21 specific absorption rate (SAR) in which energy is absorbed in tissue. SAR for
22 a human body, up to 4 W/kg for body and up to 12 W/kg [3] for extremities
23 is not considered to be cause significant clinical hazard if exposure times are
2
24 controlled [4].
25 Binary, near equiatomic nickel-titanium (NiTi) alloy is the most com-
26 monly used alloy in medical industry due to its biocompatibility, elastic
27 properties and high compatibility with medical imaging procedures includ-
28 ing X-ray fluoroscopy or magnetic resonance imaging (MRI). NiTi belongs
29 to a family of shape memory alloys (SMA), which posses the unique prop-
30 erty of reverting back to their original shape upon heating after having been
31 deformed in a colder state. Properties such as pseudoelasticity, high fatigue
32 resistance, high corrosion resistance, biocompatibility of SMAs make these
33 alloys suitable for a wide range of applications. A broad class of applications
34 ranges from micro-actuators to aircraft engines and wing parts [5–7], and
35 from small stents to bone implants in medical applications [8, 9]. These al-
36 loys undergo temperature or stress induced phase transitions between a high
37 symmetry austenitic phase to a lower symmetry martensitic structure that
38 is capable of accommodating large strains [10]. Transition temperatures or
39 the level of stress needed for phase transitions can be modified by changing
40 the stoichiometry or composition of the alloy as well as the alloy’s thermo-
41 mechanical processing history [11–14]. Other physical properties such as
42 contrast enhancement under imaging scans used in medical examination can
43 also be modified.
44 Radiopacity of NiTi is low due to very low X-ray absorption of Ti. This
45 low X-ray absorption has been improved by adding high atomic density ele-
46 ments such as platinum, palladium, tungsten or gold into binary NiTi [15].
47 Recently, a range of ternary NiTi alloys has been developed by alloying with
48 rare-earth elements e.g. erbium, galolinium or neodymium [15]. Despite a
3
49 very high amount of ferromagnetic nickel (over 50 w.t %) in NiTi, the alloy
50 is paramagnetic and therefore suitable to use under MRI.
51 Here we investigate the impact of addition of such radiopaque elements
52 Pt and Er on magnetic properties of nickel rich ternary NiTiPt and NiTiEr
53 alloys. These alloys were found to be highly radiopaque in X-ray fluoroscopy
54 imaging [15, 16]. We pay particular attention to the NiTiEr alloy, since
55 binary and ternary rare-earth transition metal alloys are known strong fer-
56 romagnets.
57 2. Experimental Methods
58 Specimen were prepared by Vacuum Induction Melting from high purity
59 metals. Subsequently, a series of treatments - homogenisation, hot rolling,
60 extrusion were applied to the samples to improve its microstructural, and
61 thermal and mechanical properties. Fabrication and processing routes are
62 detailed in reference [23].
63 Vibrating sample magnetometry (VSM) was used to measure response
64 of the binary near-equiatomic NiTi, and ternary nickel-rich NiTiPt and Ni-
65 TiEr to static magnetic field using a Lakeshore 7400 vibrating magnetometer
66 (Lakeshore, USA [17]). The electromagnet was capable of achieving fields as
67 high as 2 Tesla (T ), but the strongest field used during our experiments
68 was set to be Bmax = 1.5 T , which is typical field used in standard MRI
69 procedures. Sample dimensions and mass were chosen to be sufficient to
70 produce measurable signal while allowing for using a point magnetic-dipole
71 approximation. The mass of the samples varied between 0.85 and 1 gram.
72 The sample geometry was approx. 7 × 7 × 1.5 mm prism. Calibration of
4
73 the VSM was performed prior to every sample measurement to ensure the
74 probe reads zero response in the field H = 0 Am−1 . Field step-size was
75 300 × 103 /4π Am−1 with a ramp-rate of 50 × 103 /4π Am−1 s−1 . We em-
76 ployed full H0 → Hmax → −Hmax → H0 loop with H0 = 0 and |Hmax | = 15
77 ×103 /4π Am−1 to measure the magnetic response.
78 3. Results and Discussions
79 3.1. Binary NiTi alloy
80 As near equiatomic NiTi is paramagnetic with a low magnetic susceptibil-
81 ity, it is one of the most suitable materials for MRI imaging. The magnitude
82 of the magnetic susceptibility of binary NiTi varies between 2.4 × 10−6 and
83 3.7×10−6 103 /4π m3 · kg−1 for martensite and austenite, respectively [18–22].
84 Literature data, however, does not clarify if the samples were single-crystals
85 or polycrystalline material. The data-sheets of binary NiTi can be found in
86 the product specifications, [18, 19], [20] for martensitic and [21] for austenite
87 phase of N i55 T i45 , [22] where an average value 3 × 10−6 103 /4π m3 · kg−1 is
88 given.
89 Our measurements of polycrystalline samples show that the susceptibility
90 is between that of purely martensite and bare austenite phase as reported in
91 the literature [21, 22], indicating the presence of both, austenite and marten-
92 site phases in our samples. Response of hot-rolled and extruded NiTi alloy
93 to the static magnetic field was evaluated in order to establish the effect of
94 thermo-mechanical treatment on magnetic properties of NiTi. Figure 1 de-
95 picts magnetization loops of as-cast NiTi and hot-rolled and extruded NiTi.
96 Magnetization-field dependence was calculated as a mean value from three
5
97 M(H) measurements. Black squares denote as-cast alloy and red squares
98 represent hot-rolled and extruded NiTi. It is clear from the present M (H)
99 dependence that the thermal and mechanical treatments do not affect the
100 static magnetic susceptibility of the NiTi alloy.
101 3.2. Ternary NiTiPt alloy
102 The static field magnetization of nickel-rich N i50 T i42.5 P t7.5 (in atomic
103 percent) was measured on both, as-cast alloy and the alloy homogenized at
104 925 ◦ C for 72 hours. Current research of NiTiPt alloy is directed towards
105 Ti-rich composition, which is thermodynamically stable and easier to process
106 than the Ni-rich alloy [23]. Nickel-rich NiTiPt with the same chemistry and
107 thermal post-processing route has been reported by O’Donoghue and co-
108 workers in [24]. Their research reveals the evolution of the microstructure,
109 hardness and chemical composition of multiphase structure as a result of
110 processing of the alloy.
111 As-cast N i50 T i42.5 P t7.5 alloy studied here is composed of a matrix that
112 is rich in nickel N i61.4 T i34 P t4.6 and a phase with the higher content of plat-
113 inum N i46.1 T i42.9 P t11 [24]. As-cast alloy shows nearly linear M (H) depen-
114 dence wit the mass magnetic susceptibility χρ = 1.78×10−6 ×4π/103 m3 kg −1 .
115 The non-linear region spans only small field symmetric around the zero field
116 (switching field direction). The initial susceptibility calculated from M/H
117 dependence is 2.5 × 10−6 ×4π/103 m3 kg −1 with a standard deviation 8.12%.
118 The small increase in susceptibility around H = 0 is caused due to the mi-
119 crostructure evolution of the material as a result of thermal treatment at 925
◦
120 C. In polycrystalline materials grain boundaries act as the defects creating
121 energy barriers that have to be overcome by the applied field. This results
6
122 in deviation from linear, paramagnetic M (H) in low magnetic fields.
123 The ternary NiTiPt alloy can be modelled as composed of binary phases
124 of N ix T iy , N ix P ty and T ix P ty . Binary N ix T iy systems of near equiatomic
125 compositions do not posses any magnetic moment [25]. Nickel-platinum sys-
126 tems in near equiatomic compositions were experimentally studied by Kumar
127 et. al. [26] and the temperature dependence of susceptibility showed very
128 low values above 100 Kelvin. Furthermore, as documented in [27] and [28],
129 magnetism is strongly dependent on ordering of the structure. T ix P ty alloys
130 exhibit the shape-memory effect for near-equiatomic concentrations and there
131 is no evidence present that the alloys possess high magnetic susceptibility at
132 ambient temperatures. It is therefore reasonable to expect that if a ternary
133 phase is present, this will also be non-magnetic, since local environment of
134 the atoms in the ternary alloy is almost the same as in binary alloys. This
135 holds true for high and low temperature phases, as T iP t undergoes marten-
136 sitic transformation as well and N iP t is essentially non-magnetic down to
137 liquid nitrogen temperatures.
138 Homogenization of the as-cast alloy at 925◦ C for 72 hours results in struc-
139 tural changes [24] as the matrix breaks and precipitates form due to the heat
140 treatment. Homogenization also affects magnetic properties as the shape of
141 the magnetization loop becomes more sigmoidal within a small range around
142 zero field (Fig. 2) and a small hysteresis loop is developed as shown in (Fig.
143 3). This behaviour may have originated due to the separation of phases and
144 precipitates that are consequently blocking spin reorientation perhaps due
145 to formation of grain boundaries. Similar behaviour has been observed in
146 N ix P t1−x [29]. Vasumathi et. all have found that the deposition tempera-
7
147 ture and thermal treatment (annealing) alters the shape of the magnetiza-
148 tion loop, although the values of magnetization in higher fields are almost
149 the same for all specimen tested.
150 The extent of hysteresis in homogenized NiTiPt is very small and spans
151 approximately over ±750×103 /4π Am−1 range. For fields higher than ±750×
152 103 /4π Am−1 , magnetization linearly increases with increasing external field.
153 The following fields were identified from the data: coercive field at M = 0,
154 Hc = (Hc+ +Hc− )/2 = ±103.35 ×103 /4π Am−1 and remanent magnetization
155 at H = 0 also calculated as an average of Mr+ and Mr− , Mr = ±8 × 10−4
156 Am2 kg −1 . The energy density (area of a hysteresis loop) is estimated to be
157 2.89 × 10−4 Jkg −1 .
158 3.3. Ternary NiTiEr alloy
159 When alloying rare-earth (RE ) elements with transition metals (TM ), a
160 wide range of magnetic behaviour has been reported in the literature. The
161 interaction of strongly localized magnetic moments on RE site and the itin-
162 erant magnetism observed in TM give rise to different types of magnetic
163 ordering and transitions at low temperatures [30]. Strong permanent mag-
164 netic moments can be obtained by alloying RE, TM with elements like boron
165 or carbon [31, 32], which enhance magnetic moments and also anisotropy of
166 such compounds due to directional bonding and crystal field [31]. An exten-
167 sive review of structure and magnetic properties of RE - TM intermetallics
168 was published by Shytula and Leciejewicz in the reference [32].
169 Here we first present the magnetization loop of metallic erbium which
170 is known as paramagnetic at room temperature. A small hysteresis loop
171 is developed in region around H = 0, with the remanent magnetization
8
172 Mr = ±2.7 × 10−2 Am2 kg −1 and coercive fields Hc = ±64.59 ×103 /4π Am−1
173 and the energy density 213 × 10−4 Jkg −1 (Fig. 4).
174 For external fields above 1000 × 103 /4π Am−1 we observed a linear M/H
175 with a suceptibility of χρ = 2.87 × 10−4 4π × 10−3 m3 kg −1 and with standard
176 deviation less than 10−8 4π × 10−3 m3 kg −1 , and is in close agreement with the
177 value of susceptibility for metallic erbium recalculated from the molar sus-
178 ceptibility in [33]. Initial susceptibility calculated from initial magnetization
179 curves is 3.19 × 10−4 4π × 10−3 m3 kg −1 with a standard deviation 1.8 × 10−6
180 4π × 10−3 m3 kg −1 . The paramagnetism is typical for all lanthanide metals at
181 higher (room) temperatures [32].
182 Magnetic properties of ternary N i50 T i50−x Erx with x = 6, 7.5, 9, 10,
183 12.5 and 15 atomic percent erbium content were measured. As found for
184 the other Er containing system, addition of the erbium to multicomponent
185 compounds changes mechanical properties due to the formation of binary
186 precipitates and the segregation of the excess Er on the grain boundaries for
187 > 0.2 weight addition of Er into Al − M g − M n − Zr alloy [34]. Ternary
188 systems are often found to be composed of pseudobinaries [35, 36], which also
189 may affect magnetic behaviours as different RE − metal binary systems can
190 contribute to the magnetic response differently. The formation of a mixture
191 of different ternary phases in a multicomponent nickel-rich alloy is confirmed
192 for NiErP and NiErAs [37]. The magnetic phase diagram of T bHoCuSn
193 [38] has shown that the magnetic moments on lanthanide site at an ordered
194 state deviates from the calculated RE +3 values, whereas at paramagnetic
195 state the magnetic moment is close to that of the free-ion value.
196 For ternary NiTiEr alloys we first investigate magnetization loops as a
9
197 function of erbium content in as-cast specimens. A linear M/H dependence
198 is observed through the range of erbium content in ternary alloys. Calculated
199 mass magnetic susceptibilities of N i50 T i50−x Erx are collected in Table 1.
200 We then elaborate the effect of thermal and mechanical treatments on
201 N i50 T i42.5 Er7.5 alloy. A set of extruded and homogenized (850 ◦ C for 24
202 hours) specimens were measured to establish effects of such treatments on
203 the shape of magnetization curve and magnitude of mass susceptibility. We
204 present our results in Fig. (5).
205 Several authors studied the effects of thermo-mechanical treatment of
206 alloys containing RE elements [39–45] and have found changes in structure
207 and texture, and improved corrosion resistance. Investigation of the effects
208 of thermo-mechanical treatment on magnetic properties of alloys containing
209 RE-elements are not available to the best of our knowledge.
210 As clearly seen from the graph presented in Fig.(5), the extrusion process
211 (red squares on the graph), does not change the linearity of the M/H loop,
212 but there is slight increase in the susceptibility (slope). This is the effect
213 of structural changes during extrusion process, which primarily breaks down
214 the as-cast structure and causes a small change of chemical composition of
215 respective phases present in the specimen [23]. The effect of extrusion on
216 magnetic susceptibility can be seen from the Fig. 6, where the values of
217 mass susceptibility of the extruded and the as-cast specimens lie close to
218 each other. Calculated susceptibility of extruded sample is χρ = 4.95 × 10−5 ,
219 standard deviation is 5.69 × 10−8 4π/103 m3 kg −1 . Initial susceptibility from
220 H = 0 to H = Hmax curve is 5.87 × 10−5 , std. deviation 1.21 × 10−6 in
221 4π/103 m3 kg −1 . The values of χρ of as-cast and extruded alloy differ only by
10
222 1 × 10−6 .
223 Homogenization process is different from mechanical extrusion as during
224 homogenization and eventual cooling, diffusion of elements takes place to
225 disperse the solutes evenly in the alloy. In such process, the composition as
226 well as the size of grains change and compositional segregation occurs.
227 In Table 2 we present summary of the susceptibility of binary NiTi and
228 ternary NiTiPt and NiTiEr to the static magnetic field for all measured
229 specimen.
230 4. Conclusion
231 Magnetization in static field of the NiTi, NiTiPt, different compositions
232 a ternary N i50 T i50−x Erx (x < 15 at. %) alloys were studied here. We also
233 investigated effect of different thermal and mechanical treatments on selected
234 alloys. The magnetic response of a NiTiPt to static field is as good as the
235 response of a binary NiTi with very low values of paramagnetic susceptibility
236 in the order of ∼ 10−9 m3 kg −1 . For alloys containing erbium, values of
237 χρ are an order of magnitude higher than for binary alloy. Low values of
238 susceptibility to static magnetic field make ternary, radiopaque NiTiPt and
239 NiTiEr alloys save to use under MRI.
240 5. Acknowledgements
241 Authors wish to acknowledge Enterprise Ireland Innovation Partnership
242 for providing research funding.
11
243 References
244 [1] I. N. Bankman (Ed.), Handbook of Medical Imaging: Processing and
245 Analysis Management, Academic Press, 2000.
246 URL http://www.sciencedirect.com/science/book/
247 9780120777907
248 [2] Diagnostic imaging, Online (2013).
249 URL http://www.nlm.nih.gov/medlineplus/diagnosticimaging.
250 html
251 [3] Guidelines for ethics approval of research protocols involving human
252 exposure to magnetic resonance imaging, Online (September 2009).
253 URL http://www.uwo.ca/research/_docs/ethics/MRI_
254 guidelines_2-G-004_Sept-revision.pdf
255 [4] N. Jagannathan, Magnetic resonance imaging (MRI): effects of electro-
256 magnetic radiation and safety aspects, in: Electromagnetic Interference
257 and Compatibility ’99. Proceedings of the International Conference on,
258 1999, pp. 469 – 470.
259 [5] C. Megnin, J. Barth, M. Kohl, A bistable SMA microvalve for 3/2-way
260 control, Sensors and Actuators A: Physical 188 (2012) 285 – 291.
261 [6] M. Bergamasco, P. Dario, F. Salsedo, Shape memory alloy microactua-
262 tors, Sensors and Actuators A: Physical 21 (13) (1990) 253 – 257.
263 [7] D. J. Hartl, D. C. Lagoudas, Aerospace applications of shape memory
264 alloys, Journal of Aerospace Engineering 221 (4) (2007) 535 – 552.
12
265 [8] A. Biesiekierski, J. Wang, M. A. H. Gepreel, C. Wen, A new look at
266 biomedical Ti-based shape memory alloys, Acta Biomaterialia 8 (5)
267 (2012) 1661 – 1669.
268 [9] M. Niinomi, M. Nakai, J. Hieda, Development of new metallic alloys for
269 biomedical applications, Acta Biomaterialia 8 (11) (2012) 3888 – 3903.
270 [10] D. Mantovani, Shape memory alloys: Properties and biomedical appli-
271 cations, JOM 52 (10) (2000) 36 – 44.
272 [11] K. Otsuka, X. Ren, Physical metallurgy of Ti-Ni-based shape memory
273 alloys, Progress in Materials Science 50 (5) (2005) 511 – 678.
274 [12] C. Li, H. Luo, Q. Hu, R. Yang, B. Johansson, L. Vitos, First-
275 principles investigation of the composition dependent properties of
276 N i2+x M n1−x Ga shape-memory alloys, Phys. Rev. B 82 (2010) 024201.
277 [13] V. V. Khovailo, T. Abe, V. V. Koledov, M. Matsumoto, H. Nakamura,
278 R. Note, M. Ohtsuka, V. G. Shavrov, T. Takagi, Influence of Fe and
279 Co on Phase Transitions in Ni-Mn-Ga Alloys, Materials Transactions
280 44 (12) (2003) 2509 – 2525.
281 [14] J. Lu, Q. Hu, R. Yang, Composition-dependent elastic properties and
282 electronic structures of off-stoichiometric TiNi from first-principles cal-
283 culations, Acta Materialia 56 (17) (2008) 4913 – 4920.
284 [15] S. Carr, The development of a radiopaque nitinol alloy: Radiopacity and
285 thermodynamic stability, Ph.D. thesis, University of Limerick (2011).
13
286 [16] J. Carlson, S. Carr, P. Devereaux, D. Haverty, S. Lavelle, T. Mcglough-
287 lin, S. A. M. Tofail, Nickel-titanium alloy including a rare earth element,
288 cA 2658580 A1 (March 2008).
289 [17] L. S. Cryotronics, 7400 Series VSM, Online (2014).
290 URL http://www.lakeshore.com/products/
291 Vibrating-Sample-Magnetometer/Pages/Model-Landing.aspx
292 [18] J. Electronics, Jameco part number 141874 (July 2013).
293 URL http://www.jameco.com/Jameco/Products/ProdDS/141874.
294 pdf
295 [19] D. G. Heart, Nitinol: A new metal that can drive a stirling cycle heat
296 engine? (Dec. 2012).
297 URL http://agreenroad.blogspot.sk/2012/12/
298 nitinol-new-metal-that-can-drive-free.html
299 [20] MatWeb.com, Nitinol - NiTi shape memory alloy; Low-temperature
300 phase (July 2013).
301 URL http://www.matweb.com/search/datasheet.aspx?matguid=
302 44afc7d3c6eb4829bc2df27884fd2d6c&ckck=1
303 [21] MatWeb.com, Nitinol - NiTi shape memory alloy; High-temperature
304 phase (July 2013).
305 URL http://www.matweb.com/search/DataSheet.aspx?MatGUID=
306 de9dd08433714f698d513766dccea437&ckck=1
307 [22] J. M. M. Components, Nitinol technical properties (July 2013).
308 URL http://jmmedical.com/resources/
14
309 221/Nitinol-Technical-Properties.html#
310 nitinol-electrical-and-magnetic-properties
311 [23] J. Butler, Development of a workable radiopaque rare earth doped niti-
312 nol, Ph.D. thesis, University of Limerick (2011).
313 [24] L. O’Donoghue, A. Gandhi, J. Butler, W. Redington, P. Tiernan, T. Mc-
314 Gloughlin, J. Carlson, S. Lavelle, S. Tofail, X-ray and microstructural
315 investigation of NiTiPt alloys homogenised at intermediate to high tem-
316 peratures, Nuclear Instruments and Methods in Physics Research Sec-
317 tion B: Beam Interactions with Materials and Atoms 268 (3-4) (2010)
318 287 – 290.
319 [25] S. Wang, Z. Wang, W. Setyawan, N. Mingo, S. Curtarolo, Assessing the
320 thermoelectric properties of sintered compounds via high-throughput
321 ab-initio calculations, Phys. Rev. X 1 (2011) 021012.
322 [26] U. Kumar, K. Padmalekha, P. Mukhopadhyay, D. Paudyal, A. Mooker-
323 jee, Magnetic transition in NiPt alloy systems: experiment and theory,
324 Journal of Magnetism and Magnetic Materials 292 (0) (2005) 234 – 240.
325 [27] C. E. Dahmani, M. C. Cadeville, J. M. Sanchez, J. L. Morán-López, Ni-
326 Pt phase diagram: Experiment and theory, Phys. Rev. Lett. 55 (1985)
327 1208 – 1211.
328 [28] R. E. Parra, J. W. Cable, Neutron study of magnetic-moment distribu-
329 tion in Ni-Pt alloys, Phys. Rev. B 21 (1980) 5494 – 5504.
330 [29] D. Vasumathi, A. Shapiro, B. Maranville, F. Hellman, Growth-induced
15
331 perpendicular magnetic anisotropy and clustering in N ix P t1−x alloys,
332 Journal of Magnetism and Magnetic Materials 223 (3) (2001) 221 – 232.
333 [30] D. Gignoux, D. Schmitt, Metamagnetism and complex magnetic phase
334 diagrams of rare earth intermetallics, Journal of Alloys and Compounds
335 225 (1-2) (1995) 423 – 431.
336 [31] L. Wei, Z. Zhi-Dong, X. Sun, Y. Chuang, F.-M. Y, F. de Boer, Room
337 temperature magnetic anisotropy of rare-earth transition-metal inter-
338 metallics R2 F e14 C, (R = Lu, Gd, Nd) and R2 F e14 B, (R = Gd, Nd),
339 Solid State Communications 76 (12) (1990) 1375 – 1378.
340 [32] A. Szytula, J. Leciejewicz, Handbook of Crystal Structures and Mag-
341 netic Properties of Rare Earth Intermetallics, 1st Edition, CRC Press,
342 1994.
343 [33] D. L. Huffman, Magnetic susceptibility of the elements and inorganic
344 compounds (July 2013).
345 URL http://mri-q.com/uploads/3/4/5/7/34572113/
346 susceptibility_of_inorganic_compounds.pdf
347 [34] S. Wen, Z. Xing, H. Huang, B. Li, W. Wang, Z. Nie, The effect of erbium
348 on the microstructure and mechanical properties of Al-Mg-Mn-Zr alloy,
349 Materials Science and Engineering: A 516 (1-2) (2009) 42 – 49.
350 [35] B. Chevalier, F. Fourgeot, L. Fournes, P. Gravereau, G. L. Caer,
351 J. Etourneau, Complex magnetic phase diagram of the ternary stan-
352 nide N d2 N i2 Sn, Physica B: Condensed Matter 226 (4) (1996) 283 –
353 292.
16
354 [36] P. von Ranke, N. de Oliveira, S. Gama, Understanding the influence of
355 the first-order magnetic phase transition on the magnetocaloric effect:
356 application to Gd5 (Six Ge1−x )4, Journal of Magnetism and Magnetic
357 Materials 277 (1-2) (2004) 78 – 83.
358 [37] M. Zelinskaa, O. Zhaka, S. Oryshchyna, T. Polianskaa, J. Y. Pivanb,
359 Solid state phase equilibria in the Er-Ni-P and Er-Ni-As systems at 800
◦
360 C, ChemInform 38 (45).
361 [38] A. Szytula, T. Jaworska-Golab, K. Nenkov, B. Penc, A. Zarzycki, Mag-
362 netic phase diagram of T b3−x Hox Cu4 Sn4 system, Journal of Magnetism
363 and Magnetic Materials 324 (6) (2012) 1017 – 1020.
364 [39] B. Wu, G. Wan, X. Du, Y. Zhang, F. Wagner, C. Esling, The quasi-
365 static mechanical properties of extruded binary Mg-Er alloys, Materials
366 Science and Engineering: A 573 (2013) 205 – 214.
367 [40] F. Mirza, D. Chen, D. Li, X. Zeng, Effect of rare earth elements on
368 deformation behavior of an extruded Mg-10Gd-3Y-0.5Zr alloy during
369 compression, Materials and Design 46 (2013) 411 – 418.
370 [41] X. Zhang, G. Yuan, L. Mao, J. Niu, P. Fu, W. Ding, Effects of extru-
371 sion and heat treatment on the mechanical properties and biocorrosion
372 behaviors of a Mg-Nd-Zn-Zr alloy, Journal of the Mechanical Behavior
373 of Biomedical Materials 7 (2012) 77 – 86.
374 [42] T. Zhang, G. Meng, Y. Shao, Z. Cui, F. Wang, Corrosion of hot ex-
375 trusion AZ91 magnesium alloy. Part II: Effect of rare earth element
17
376 neodymium (Nd) on the corrosion behavior of extruded alloy, Corrosion
377 Science 53 (9) (2011) 2934 – 2942.
378 [43] H. Song, Z. Wang, X. He, Improving in plasticity of orthorhombic
379 T i2 AlN b-based alloys sheet by high density electropulsing, Transactions
380 of Nonferrous Metals Society of China 23 (1) (2013) 32 – 37.
381 [44] S. Wang, L. Meng, S. Yang, C. Fang, H., S. Dai, X. Zhang, Microstruc-
382 ture of Al-Zn-Mg-Cu-Zr-0.5Er alloy under as-cast and homogenization
383 conditions, Transactions of Nonferrous Metals Society of China 21 (7)
384 (2011) 1449 – 1454.
385 [45] X. Chen, Z. Liu, S. Bai, Y. Li, L. Lin, Alloying behavior of erbium in an
386 Al-Cu-Mg alloy, Journal of Alloys and Compounds 505 (1) (2010) 201 –
387 205.
18
388 Tables and Figures
Table 1:
Erbium χρ SD(±)
at. % 4π × 10−3 m3 kg − 1
0 2.83 ×10−6 1.49 ×10−8
6 3.8 ×10−5 6.3 ×10−7
7.5 4.85 ×10−5 3.4 ×10−7
9 5.37 ×10−5 2.7 ×10−7
10 6.02 ×10−5 1.3 ×10−7
12.5 7.1 ×10−5 2.6 ×10−7
15 7.89 ×10−5 7.3 ×10−8
100 2.87 ×10−4 2.71 ×10−9
389
19
Table 2:
χρ initial SD(±) χρ linear SD(±)
×4π × 10−3 m3 kg − 1
N iT i
as-cast 2.83 × 10−6 1.49 × 10−8
hot-rolled, extruded 2.83 × 10−6 2.95 × 10−8
N i50 T i42.5 P t7.5
as-cast 2.5 × 10−6 2.03 × 10−7 1.78 × 10−6 2 × 10−8
homogenized 4.99 × 10−6 2.47 × 10−7 1.72 × 10−6 3.34 × 10−8
N i50 T i50−x Erx
6 3.8 × 10−5 6.3 × 10−7
7.5 4.85 × 10−5 3.4 × 10−7
9 5.31 × 10−5 2.7 × 10−7
10 6.02 × 10−5 1.3 × 10−7
12.5 7.1 × 10−5 2.6 × 10−7
15 7.89 × 10−5 7.3 × 10−8
Er 3.19 × 10−4 1.8 × 10−6 2.87 × 10−4 2.71 × 10−9
N i50 T i42.5 Er7.5
as-cast 4.85 × 10−5 3.4 × 10−7
extruded 5.87 × 10−5 1.21 × 10−6 4.95 × 10−5 5.69 × 10−8
homogenized 1.61 × 10−4 8.81 × 10−8 4.59 × 10−5 1.23 × 10−8
S.Steel 316L 4.59 × 10−3 2.65 × 10−7 1.38 × 0−5 1.39 × 10−7
390
20
Figure 1:
391
21
Figure 2:
392
22
Figure 3:
393
23
Figure 4:
394
24
Figure 5:
395
25
Figure 6:
396
26
397 Captions
Table 1: Mean value mass magnetic susceptibility of as-cast N i50 T i50−x Erx . Alloy con-
tained 6 − 15 at. % erbium keeping the alloy rich in nickel. First data row is for binary
NiTi in order of magnitude 10−6 in 4π × 10−3 m3 kg − 1 units. Alloying erbium into binary
alloy results in an increased magnetic susceptibility by an order of magnitude compared
to that of binary NiTi.
Table 2: Mass magnetic susceptibility χρ ×4π/103 m3 kg −1 of binary NiTi, ternary NiTiPt
and NiTiEr. Susceptibility at initial stage of magnetization is noted where appropriate
(χρ initial ) and for all alloys we list χρ in linear region of magnetization loop (χρ lin-
ear ). Standard deviation (SD) from mean value is tabulated to the right of susceptibility.
Stainless steel 316L is noted for comparison.
Figure 1: As-cast (black squares) and hot-rolled, extruded (red squares) equiatomic NiTi
magnetization curve (line). Note that values of magnetization are almost same for both
paramagnetic samples
Figure 2: Magnetization loop of as-cast (black squares) and homogenized (red squares)
N i50 T i42.5 P t7.5 alloy. Nearly linear M/H of as-cast alloy changes shape to sigmoidal after
the homogenization. This effect spans only over a small range of field around H = 0 and
continues as a linear M/H with the slope (susceptibility) similar to that of the as-cast
alloy.
Figure 3: Magnetization loop of NiTiPt homogenized at 925 ◦ C for 72 hours. Homogeniza-
tion process causes deviation from linear M/H dependence. The effect is most profound
near the field direction switching region. Inset: a small hysteresis loop developed as re-
sult of homogenization. Scale of magnetization (y axis) in top left inset is ±7.5 × 10−3
Am2 kg −1 and field is ±1500 × 103 /4π Am−1 . Right bottom inset field scale spans the
whole range and the susceptibility scale is up to 1 × 10−5 ×4π/103 m3 kg −1 . The energy
density (specific energy) enclosed by hysteresis loop is 2.89 × 10−4 Jkg −1 .
27
Figure 4: Detail of magnetization loop of polycrystalline erbium. Coercive Hc and rema-
nent Mr fields were identified from M(H) graph. The energy density 213 × 10−4 Jkg −1
has been estimated from hysteresis loop. Top left inset is the whole scale magnetization
loop of metallic Er. right bottom inset is calculated susceptibility χρ = dM/dH.
Figure 5: Magnetization loop of thermo-mechanically treated N i50 T i42.5 Er7.5 . As evi-
dent, extrusion process does not change the shape of M (H) loop, whereas non-linearity is
developed as a result of homogenization.
Figure 6: Differential mass susceptibility χρ of thermo-mechanically treated
N i50 T i42.5 Er7.5 alloy. In linear part, homogenized and as-cast alloys have similar sus-
ceptibility. Non-linear M/H relationship has developed in the alloy as a result of homog-
enization (blue peak near H = 0). Susceptibility of extruded alloy differs from that of as
cast by 1 × 10−6 4π/103 m3 kg −1 .
28
