cond-mat0701077/main.tex
1: \documentclass[prb,  onecolumn, square,showpacs, aps] {revtex4}
2: %\documentclass[a4,prl,showpacs,showkeys, sort&compress, floatfix, square]{revtex4}
3: 
4: %\usepackage[pdftex]{graphicx}
5: \usepackage[dvips]{graphicx}
6: 
7: 
8: %\usepackage{color}
9: 
10: %\usepackage{times, mathptm}
11: 
12: % Forcing printing figure labels in Roman numerals
13: %\renewcommand{\thefigure}{\Roman{figure}}
14: 
15: % Forcing printing tables labels in Roman numerals
16: %\renewcommand{\thetable}{\Roman{table}}
17: 
18: % In order to process the file in A4: dvi to ps
19: % dvips -t a4 -o ratio.ps ratio.dvi
20: % This is important to show within xdvi and ghostview the page numbering
21: % and helps format correctly the document.
22: 
23: % In order to process the file in A4: ps to pdf
24: % ps2pdf -sPAPERSIZE=a4  ratio.ps
25: 
26: \begin{document}
27: 
28: \title{Effect of the metal-to-wire ratio on the high-frequency magnetoimpedance 
29: of glass-coated CoFeBSi amorphous microwires}
30: 
31: \author{R. Valenzuela\P, A. Fessant, J. Gieraltowski and C. Tannous} 
32:  \affiliation{Laboratoire de Magn\'{e}tisme de Bretagne, CNRS-FRE 2697, 
33: Universit\'{e} de Bretagne Occidentale, BP 809, Brest CEDEX, 29285, France \\
34: \P Departamento de Materiales Met\'{a}licos y Cer\'{a}micos,
35: Instituto de Investigaciones en Materiales, Universidad Nacional 
36: Aut\'{o}noma de M\'{e}xico, P.O. Box 70-360, Coyoac\'{a}n, Mexico D.F., 04510, Mexico}
37: 
38: \begin{abstract}
39: High frequency [1-500 MHz] measurements of the magnetoimpedance (MI) of 
40: glass-coated Co$_{69.4}$Fe$_{3.7}$B$_{15.9}$Si$_{11}$ microwires are carried 
41: out with various metal-to-wire diameter ratios. A twin-peak, anhysteretic 
42: behaviour is observed as a function of magnetic field. A maximum in the 
43: normalized impedance, $\Delta Z$/$Z$, appears at different values of the 
44: frequency $f$, 125, 140 and 85 MHz with the corresponding diameter ratio $p$ = 
45: 0.80, 0.55 and 0.32. We describe the measurement technique and interpret our 
46: results with a thermodynamic model that leads to a clearer view of the 
47: effects of $p$ on the maximum value of MI and the anisotropy field. The 
48: behavior of the real and imaginary components of impedance is also 
49: investigated; they display a resonance that becomes a function of the DC 
50: field $H_{DC}$ for values larger or equal to $H_{K}$ the circumferential 
51: anisotropy field for each $p$ value. These results are interpreted in terms
52:  of a rotation model of the outer shell magnetization.
53: \end{abstract}
54: 
55: \pacs {75.50.Kj;   72.15.Gd;   75.30.Gw;  75.80.+q}
56: 
57: \maketitle
58: 
59: \textit{Index Terms}---Amorphous magnetic wires. Giant magnetoimpedance. Magnetic anisotropy. 
60: Magnetostriction.
61: 
62: \section{INTRODUCTION}
63: 
64: When a ferromagnetic conductor is traversed by a current of low amplitude 
65: and high frequency, its impedance, or rather its Magnetoimpedance (MI) can 
66: be altered by applying a DC magnetic field. This phenomenon, first described 
67: [1] in the 1930's, has been receiving special attention over the last 15 
68: years [2,3] due to its potential technological applications [4,5] in 
69: sensors, devices and instruments. Its fundamental physics is also being 
70: deeply examined [6]. MI has been observed in a wide variety of materials, 
71: geometries and structures, particularly in amorphous wires having 
72: diameters of a few hundred microns. Wires with smaller diameters (a few 
73: microns) coated with a glass sheath show an increase of the working 
74: frequency, and introduce an additional structural feature that alter the 
75: physical parameters [7,8]. 
76: 
77: Since glass exerts some mechanical stress on the 
78: metallic wire, a change in the magnetic response is expected; other works 
79: have rather investigated the effect of external stress [9,10], as well as 
80: that of various annealing methods [11,12]. Consequently, it is of interest 
81: to finely tune the physical properties through the control of the thickness 
82: and nature of the glass sheath. In this paper, we report on MI measurements 
83: of Co-rich amorphous microwires with various ratios of the metal-to-wire 
84: diameter, in the [1-500 MHz] frequency range, carried out with a novel [13] 
85: broadband technique. This technique allows a complete determination of MI as 
86: a function of both frequency and magnetic field. 
87: 
88: In contrast with most 
89: published works, where MI is measured at a single frequency, or at a 
90: discrete set of frequencies, this technique provides a quasi-continuous 
91: ensemble of results over the 1-500 MHz frequency range. The effects of the 
92: thickness of the glass sheath are clearly illustrated and the variation of 
93: the anisotropy field $H_{K}$ is evaluated directly as a function of stress. 
94: 
95: Additionally, we carried out an analysis of the real and imaginary 
96: components of impedance. The resonance character observed can be attributed 
97: to the rotation of the outer shell magnetization as described further below. 
98: Possible sensor applications are discussed in the conclusion.
99: 
100: 
101: 
102: \section{MEASUREMENT TECHNIQUES}
103: 
104: 
105: 
106: Glass-coated amorphous microwires of nominal composition 
107: Co$_{69.4}$Fe$_{3.7}$B$_{15.9}$Si$_{11}$ were prepared by fast cooling with 
108: the Taylor-Ulitovski technique \cite{Torcunov}. Several metal-to-wire ratio values, 
109: $p =\phi_{m}/\phi_{w}$, with $\phi_{m}$ the metallic core diameter and $\phi_{w}$ the total 
110: wire diameter, were produced and characterized. For values of metal core 
111: diameters of 24, 12 and 7 $\mu $m, with corresponding total diameters of 30, 
112: 21.8 and 21.9 $\mu $m, we get the ratio $p$ = 0.80, 0.55 and 0.32, 
113: respectively. In order to make electrical contacts, the glass sheath was 
114: etched away over a few mm on both microwire ends, with a solution of 
115: hydrofluoric acid. Silver paste contacts were then made in order to proceed 
116: with the electrical measurements. 
117: 
118: MI measurements were carried out in the 
119: [1-500 MHz] range, on pieces of microwires $\sim $12 mm in length, with an 
120: HP 8753C Network Analyzer using a novel broadband measurement technique 
121: described in [13]. Helmholtz coils served as source of axial DC magnetic 
122: fields ranging from -80 Oe to +80 Oe. Ferromagnetic resonance 
123: (FMR) measurements were carried out 
124: using 3 mm long samples, in a JEOL JES-RES 3X spectrometer operating at 9.4 
125: GHz (X-band), at room temperature. Non-resonant low-field absorption (LFA) 
126: measurements were taken using a JEOL ES-ZCS2 zero-cross sweep unit, which 
127: compensates for any magnetic remanence, allowing precise determination of 
128: low magnetic fields around zero.
129: 
130: 
131: 
132: \section{EXPERIMENTAL RESULTS AND MODELING}
133: 
134: The results obtained are plotted in a continuous 3D representation 
135: of $\Delta Z/Z$, with:
136: \begin{equation} 
137: \Delta Z/Z=\frac{[Z(H_{DC})-Z(H_{DC}=80 \mbox{ Oe})]}{Z(H_{DC}=80 \mbox{ Oe})}, 
138: \end{equation}
139: where $Z(H_{DC})$ is the total impedance modulus
140: $Z =\sqrt{ (Z'^{2} + Z''^{2})}$, with $Z$' the real part and $Z$'' the imaginary part of impedance. 
141: $Z(H_{DC})$ is a function of the DC field, $H_{DC}$ (in Oersteds), and 
142: frequency, $f$. The results for $p$ = 0.8 are shown in Fig. 1. A symmetrical 
143: double peak MI plot is obtained as a function of $H_{DC}$; the peaks are 
144: associated with $\pm  H_{K}$, the circumferential anisotropy field. We 
145: obtain $H_{K}  \sim $ 3.5 Oe and no hysteresis by cycling the DC field 
146: $H_{DC}$.
147: 
148: Regarding frequency $f$, the MI shows a maximum of $\sim $ 250 {\%} at 
149: about 100 MHz. Similar plots were obtained with the other $p$ ratios, albeit 
150: with significant differences in the values of the anisotropy field and peak 
151: frequency values. Instead of making a comparison at a single frequency, as 
152: typically done, we choose the frequency at which the maximum in $\Delta Z/Z$ appears
153:  and compare results as a function of $H_{DC}$, as shown in Fig. 2. 
154: Note that the sample with $p$ = 0.32 exhibits, as a function of field, several 
155: peaks that can be associated with a distribution of the anisotropy axis 
156: orientation. This introduces a large uncertainty in the numerical value of 
157: the anisotropy field $H_{K}$. 
158: 
159: Figure 2 shows clear trends in the results: MI response increases as $p$ 
160: increases, while the anisotropy field decreases. $p$ indicates the importance 
161: of the metal core with respect to the total diameter of the wire and the 
162: stress increases as the thickness of the glass sheath increases. During 
163: fabrication, glass-coated microwires are subjected to strong stresses, 
164: generally proportional to the thickness of the glass coating that varies 
165: inversely proportional to $p$. The origin of such stresses can be understood, 
166: since glass possesses a smaller thermal contraction coefficient than metals. 
167: In the cooling process, the metallic core tends to contract faster and more 
168: substantially than the surrounding glass sheath; however glass hampers such 
169: contraction. 
170: 
171: Torcunov [14] modeled the thermoelastic and quenching stresses that occur in 
172: glass-coated wires and evaluated with a thermodynamic model the stress 
173: components in terms of their axial $\sigma_{zz}$, radial $\sigma_{rr}$ and azimuthal 
174: $\sigma_{\phi \phi }$ components (in a cylindrical system of coordinates $(r, \phi, z)$ 
175: with the $z$ direction along the wire axis). The following expressions 
176: (providing the Poisson's coefficients of the glass and metal are equivalent 
177: $\nu _{g}  \sim   \nu _{m}  \sim $ 1/3 ) are obtained and adapted to 
178: our case: 
179: 
180: \begin{eqnarray}
181: \sigma_{rr} & = & \frac{\epsilon E_{g} (1-p^2)}{(\frac{k}{3} +1) (1-p^2) +\frac{4 p^2}{3} } \\
182:  \sigma_{\phi \phi} & = &  \sigma_{rr}  \\
183:  \sigma_{z z} & = &  \sigma_{rr}  \frac{(k+1)(1-p^2) + 2 p^2}{k (1-p^2) + p^2} 
184: \end{eqnarray}
185: 
186: where $E_{g}$ is the glass Young modulus, $k=E_{g}/E_{m}$ and $E_{m}$ is the metallic wire
187: Young modulus. \\
188: 
189: The term $\epsilon$ is given by the difference of the glass
190: and metal expansion coefficients $\alpha_g, \alpha_m$ (respectively) times the difference
191: of the minimum glass solidification  temperature $T^{*}$ and room temperature $T$,
192: $\epsilon=(\alpha_m-\alpha_g)(T^{*}-T)$. \\
193: 
194: We apply this variation to the anisotropy field $H_K=2K_\sigma/\mu_0 M_s$ with
195: $K_\sigma$ the anisotropy constant of the wire 
196: under stress $\sigma$. We consider that the latter induces a change in the 
197: anisotropy constant according to 
198: $K_\sigma= K_{(\sigma=0)} -\frac{3}{2}\lambda_s (\sigma_{zz} - \sigma_{\phi \phi}) $
199: with the additional assumption of no extra
200: applied stress ($\mu_0$, $M_s$ and $\lambda_s$ are vacuum permeability and saturation
201: magnetization and magnetostriction respectively). Using physical 
202: parameters of wires [15] with a composition 
203: (Co$_{0.94}$Fe$_{0.06})_{72.5}$B$_{15}$Si$_{12.5}$ similar to ours, we get 
204: in Fig. 3, a reasonable agreement with the experimental behaviour, despite a 
205: faster tapering off of $H_{K}$ at low values of $p$ where we observe 
206: experimentally a large uncertainty in the value of $H_{K}$, due to a broad 
207: distribution of anisotropy axis orientation. 
208: 
209: Since our measurement technique provides also the real and imaginary 
210: components of impedance, we now investigate the frequency behavior of the 
211: imaginary component of impedance, $X$, for selected values of the DC field, 
212: Fig. 4, for $p$ = 0.80. For high frequencies ($f >$ 100 MHz), $X$ goes through the 
213: axis and take negative values; it changes from an inductive to a capacitive 
214: character. This is a common feature of resonance phenomena. We observe that 
215: for low DC fields, the plots join in a common point ($f \sim $ 141 MHz, 
216: $X   \sim  7 \Omega $); for fields larger than the anisotropy field ($H_{K}   
217: \sim $ 4 Oe), the resonance frequency becomes a function of the field. A 
218: plot of the imaginary part of impedance as a function of the real component, 
219: i.e., a Cole-Cole plot, is a more direct evidence of resonance, when the 
220: locus of the points forms a full circle, as shown in Fig. 5, for the $p$ = 0.8 
221: sample. A comparison of Cole-Cole plots for the other $p$ values appear in Fig. 
222: 6, for DC field values corresponding to the anisotropy field of each sample. 
223: The $p$ = 0.32 plot exhibits a deformed circle, probably because the high level 
224: of mechanical stresses leads to a complex distribution of the anisotropy 
225: axis. 
226: 
227: High-frequency MI and FMR have led to some 
228: confusion in the past [16,17]; however, we feel that the differences between 
229: FMR and MI are now well established. Recent results show [18] that as 
230: frequency increases (in the 200 MHz -- 6 GHz range for Co-rich amorphous 
231: ribbons), a divergence in the MI response appears with two maxima in the 
232: impedance response, corresponding to $\pm H_K$ (the value of which remains
233: virtually constant at all frequencies), and the FMR response, which becomes 
234: field dependent, with a Larmor relation that depends 
235: on the geometry of the sample. FMR experiments in Co-rich amorphous ribbons 
236: at even higher frequencies (X-band at 9.4 GHz) have exhibited both signals 
237: clearly resolved [19], the non-resonant low-field absorption (LFA) similar 
238: to MI at fields lower than 50 Oe, and the expected FMR absorption at 1,682 Oe. 
239: Therefore our present results show the beginning of this separation. The 
240: response associated with FMR shows effectively an increase in resonance 
241: frequency as the field increases as observed in the case $p$ = 0.80.
242: 
243: The relationship between the resonance frequency, $f_{RES}$, and the resonance 
244: field $H_{RES}$, however, shows a $f ^{3}_{RES}  \sim   H_{RES}$ dependence that does 
245: not fit the Larmor equation for a cylindrical geometry [20]; 
246: an FMR experiment at 9.4 GHz in these wires leads to a resonance field of 1,132 Oe,
247: about two orders of magnitude 
248: smaller than an extrapolation of the relationship exhibited in Fig. 7. 
249: In order to understand the behaviour of 
250: the resonance frequency as a function of the DC field, we follow the work of 
251: Panina et al. [21] describing the wire as containing an axial core 
252: magnetization and a circumferential magnetization in an outer shell region 
253: transverse to the wire axis. From the rotational permeability of the outer 
254: shell magnetization, one may define a resonance frequency assuming 
255: negligible Landau-Gilbert damping coefficient:
256: 
257: 
258: \begin{equation}
259: \label{eq1}
260: f_{RES} = \frac{\gamma }{2\pi }\sqrt {[H_{DC} \sin (\theta + \theta _K ) + H_K 
261: \cos ^2(\theta ) + 4\pi M_S ][H_{DC} \sin (\theta + \theta _K ) + H_K \cos 
262: (2\theta )]} 
263: \end{equation}
264: 
265: 
266: where $\gamma $ is the gyromagnetic ratio, $\theta $ is the angle the 
267: magnetization makes with the circumferential anisotropy axis (CAA) and 
268: $\theta _{K }$ the angle the CAA makes with a direction perpendicular to 
269: the wire axis [21]. In principle, $\theta $ is determined from the 
270: equilibrium condition, however for simplicity we consider the magnetization 
271: along the CAA (that means we take $\theta $=0) and therefore the simplified 
272: resonance formula becomes:
273: 
274: 
275: \begin{equation}
276: \label{eq2}
277: f_{RES} = \frac{\gamma }{2\pi }\sqrt {[H_{DC} \sin (\theta _K ) + H_K + 4\pi M_S 
278: ][H_{DC} \sin (\theta _K ) + H_K ]} 
279: \end{equation}
280: 
281: 
282: Replacing the sin($\theta _{K})$ term by its average value (since the CAA 
283: fluctuates randomly with respect to the direction perpendicular to the wire 
284: axis) in the above formula, we are able to make a direct comparison with the 
285: experimental values we obtain for the resonance frequency versus the DC 
286: field as displayed in Fig.7. The agreement we obtain is very good given our 
287: simplifying assumptions and the fact we have no free parameter other than 
288: the average value of sin($\theta _{K})$. The low field region is a 
289: crossover region from the domain relaxation to the magnetization rotation in 
290: the shell whereas in the high field region we ought to observe another 
291: transition from the magnetization rotation mode to the FMR mode of the axial 
292: magnetization precession (Fig. 8).
293: 
294: \section{CONCLUSION}
295: 
296: In conclusion, the measurement of the MI response of microwires with a novel 
297: broadband technique provides a satisfactory view of the interplay between 
298: different physical phenomena operating in the glass or metal side. In 
299: addition to the increase of anisotropy field as $p$ decreases, a larger 
300: distribution of $H_{K}$ is observed for small metal cores. Besides, an 
301: analysis of the real and imaginary components of impedance has been carried 
302: out, leading to the observation of a crossover region from the domain 
303: relaxation to the magnetization rotation in the outer shell. A second 
304: crossover region to the axial magnetization precession FMR at higher fields 
305: is observed (see Fig. 8) making the MI a valuable tool to observe and 
306: identify various modal transitions in these glass-covered microwires. 
307: 
308: Several applications of the present results are possible. One of them is the 
309: ability to select or tune the physical properties such as a better microwire 
310: might be produced and suited for a specific application. In particular, the 
311: presence of a resonant absorption peak at very low fields, that depends on 
312: $H_{DC}$, allows the possibility of engineering devices that can be designed for 
313: tunable stopband filtering applications at microwave frequencies [22].
314: 
315: 
316: 
317: \textbf{Acknowledgments}
318: 
319: The authors acknowledge Prof. M. Vazquez (Spain) for providing the 
320: microwire samples; R.V. thanks DGAPA-UNAM, Mexico, for partial support 
321: through grant PAPIIT IN119603-3.
322: 
323: 
324: \begin{thebibliography}{00}
325: 
326: \bibitem{Harrison} E.P. Harrison, G.L. Turney, H. Rowe, "An impedance magnetometer" 
327: Nature \textbf{135},  961 (1935). 
328: 
329: \bibitem{Makhotkin} V.E. Makhotkin, B.P. Shurukhin, V.A. Lopatin, P.Y. Marchukov, Y. K. 
330: Levin, "Magnetic field sensors based on amorphous ribbons"
331: Sens. Act. A \textbf{25-27}, 759-762 (1991). 
332: 
333: \bibitem{Mandal} K. Mandal, S.K. Ghatak, "Large magnetoresistance in an amorphous 
334: Co$_{68.1}$Fe$_{4.4}$Si$_{12.5}$B$_{15}$ ferromagnetic wire"
335: Phys. Rev. B \textbf{47},  14233-14236 (1993). 
336: 
337: \bibitem{Mohri} K. Mohri, T. Uchiyama and L. V. Panina, "Recent advances of micro magnetic sensor and sensing
338: applications"  Sens. Act. A \textbf{59}, 1-8  (1997). 
339: 
340: \bibitem{Vazquez} M. Vazquez, M. Knobel, M.L. Sanchez, R. Valenzuela and A.P. Zhukov, 
341: "Giant Magnetoimpedance Effect in Soft Magnetic Wires for Sensor Applications"
342: Sens. Act. A \textbf{59},  20-29 (1997). 
343: 
344: \bibitem{Knobel} M. Knobel, M. Vazquez, L. Kraus,
345: "Giant Magnetoimpedance" in  Handbook of Magnetic Materials, Vol.\textbf{15},
346: Ed. K.H.J. Buschow, ISBN: 0-444-51459-7 (Elsevier Science, Holland, 2003). 
347: 
348: \bibitem{Chiriac} H. Chiriac and T.A. Ovari. Prog. Mater. Sci. \textbf{40}, 233 (1996)
349: 
350: \bibitem{Hernando} M. V\'{a}zquez and A. Hernando. J. Phys. D. \textbf{29}, 939 (1996)
351: 
352: \bibitem{Carara} M. Carara, K.D. Sossmeier, A.D.C. Viegas, J. Geshev, H. Chiriac and R.L. 
353: Sommer. J. Appl. Phys. \textbf{98}, 033902 (2005)
354: 
355: \bibitem{Makh} L.V. Panina, S.I. Sandacci and D.P. Makhnovskiy. J. Appl. Phys. \textbf{97}, 
356: 013701 (2005)
357: 
358: \bibitem{Zhukova}  V. Zhukova, A.F. Cobe\~{n}o, A. Zhukov, J.M. Blanco, S. Puerta, J. 
359: Gonzalez and M. Vazquez. J. Non-Cryst. Solids \textbf{287}, 31 (2001).
360: 
361: \bibitem{Kraus} L. Kraus, M. Knobel, S.N. Kane and H. Chiriac. J. Appl. Phys. \textbf{85}, 5435 
362: (1999)
363: 
364: \bibitem{Fessant} A. Fessant, J. Gieraltowski, C. Tannous and R. Valenzuela, "A novel
365: broadband measurement method for the magnetoimpedance of ribbons and thin films"
366: J. Mag. Mag.  Mat. \textbf{272-276}, 1871-1872 (2004). 
367: 
368: \bibitem{Torcunov} A. V. Torcunov, S.A. Baranov and V. S. Larin, "The internal stresses dependence
369: of the magnetic properties of cast amorphous microwires covered with glass insulation" 
370: J. Mag. Mag. Mat. \textbf{196-197}, 835-836 (1999). 
371: 
372: \bibitem{Humphrey} K. Mohri, F.B. Humphrey, K. Kawashima, K. Kimura, M. Mizutani,
373: "Large Barkhausen and Matteucci effects in FeCoSiB, FeCrSiB and FeNiSiB amorphous
374: wires" \textbf{IEEE} Trans. Mag. \textbf{26}, 1789-1791 (1990). 
375: 
376: \bibitem{Duque} J.G.S. Duque, C. G\'{o}mez-Polo, A. Yelon, P. Ciureanu, A.E.P. de 
377: Araujo, and M. Knobel. J. Magn. Magn. Mater. \textbf{271}, 390 (2004).
378: 
379: \bibitem{Yelon} A. Yelon, D. M\'enard, M. Britel, and P. Ciureanu. Appl. Phys. Lett. \textbf{69}, 
380: 3084 (1996).
381: 
382: \bibitem{Baran} J.M. Barandiar\'{a}n, A. Garc\'{\i}a-Arribas, and D. de Cos. J. Appl. 
383: Phys. \textbf{99}, 103904 (2006).
384: 
385: \bibitem{Montiel} H. Montiel, G. Alvarez, I. Betancourt, R. Zamorano, and R. Valenzuela. 
386: Appl. Phys. Lett. \textbf{86}, 072503 (2005).
387: 
388: \bibitem{Kittel} C. Kittel. Phys. Rev. \textbf{73}, 155 (1948).
389: 
390: \bibitem{Panina} L.V. Panina, K. Mohri, T. Uchiyama, M. Noda and K Bushida \textbf{IEEE} Trans. Mag. 
391: \textbf{31}, 1249 (1995).
392: 
393: \bibitem{Huynen} I. Huynen, G. Goglio, D. Vanhoenacker, and A. Vander Vorst. \textbf{IEEE} Microw. Guid. Wave Lett. \textbf{9},  401 (1999).
394: 
395: \bibitem{Zhukov} A. Zhukov,  J. Magn. Magn. Mater. \textbf{242-245}, 216 (2002).
396: 
397: 
398: \end{thebibliography}
399: 
400: \newpage
401: 
402: \section{FIGURES}
403: 
404: \begin{figure}[!ht]
405: \begin{center}
406: %\scalebox{0.3}{\includegraphics[angle=0]{fig1.pdf}}
407: \scalebox{0.3}{\includegraphics[angle=0]{fig1.eps}}
408: \end{center}
409: \caption{Magnetoimpedance plot for $p$ = 0.8 as a function of axial DC field and 
410: frequency.}
411: \label{fig1}
412: \end{figure}
413: 
414: 
415: \begin{figure}[!ht]
416: \begin{center}
417: %\scalebox{0.3}{\includegraphics[angle=0]{fig2.pdf}}
418: \scalebox{0.3}{\includegraphics[angle=0]{fig2.eps}}
419: \end{center}
420: \caption{$\Delta Z/Z$ plot as a function of DC field at the selected frequencies 
421: of the MI maximum for each diameter ratio $p$}
422: \end{figure}
423: 
424: \begin{figure}[!ht]
425: \begin{center}
426: %\scalebox{0.3}{\includegraphics[angle=0]{fig3.pdf}}
427: \scalebox{0.8}{\includegraphics[angle=0]{fig3.eps}}
428: \end{center}
429: \caption{Effect of the diameter ratio $p$ on the anisotropy field $H_{K}$.
430: Our measured data is for $p=$0.32, 0.55 and 0.8. The $H_{K}$ values corresponding to
431: $p=$ 0.35, 0.789, 0.816 and  0.98 are adapted from the literature [23].
432: The values of $\mu _{0}M_{s}$ = 0.8 T, zero stress anisotropy $K_{(\sigma=0)}$~=~40~J/m$^3$, 
433: and $\lambda_s = -0.1 \times 10^{-6}$ are taken from Ref. [15].
434: Agreement with the theory improves and uncertainty decreases for larger values of $p$.}
435: \label{fig3}
436: \end{figure}
437: 
438: \begin{figure}[!ht]
439: \begin{center}
440: %\scalebox{0.8}{\includegraphics[angle=0]{fig4.pdf}}
441: \scalebox{0.8}{\includegraphics[angle=0,clip=]{fig4.ps}}
442: \end{center}
443: \caption{Imaginary part of impedance as a function of frequency, at selected 
444: values of applied field, for the $p$ = 0.8 sample. The arrow indicates the 
445: point where plots merge for low values of field.}
446: \label{fig4}
447: \end{figure}
448: 
449: 
450: \begin{figure}[!ht]
451: \begin{center}
452: %\scalebox{0.8}{\includegraphics[angle=0]{fig5.pdf}}
453: \scalebox{0.8}{\includegraphics[angle=0]{fig5.eps}}
454: \end{center}
455: \caption{Cole-Cole plot for the $p$ = 0.8 sample, for selected values of the 
456: applied field.}
457: \label{fig5}
458: \end{figure}
459: 
460: 
461: \begin{figure}[!ht]
462: \begin{center}
463: %\scalebox{0.8}{\includegraphics[angle=0]{fig6.pdf}}
464: \scalebox{0.8}{\includegraphics[angle=0,clip=]{fig6.ps}}
465: \end{center}
466: \caption{Cole-Cole plots for the three $p$ values, taken at $H_{DC}$ 
467: corresponding to the anisotropy field value $H_K$.
468: Note that the $H_K$ are slightly different from those indicated in fig.3.
469: $H_K$ is taken as 4, 12 and 24 Oe for  $p=$0.8, 0.55 
470: and 0.32 instead of 3.5, 11.8 and 25 Oe respectively}. 
471: \label{fig6}
472: \end{figure}
473: 
474: 
475: \begin{figure}[!ht]
476: \begin{center}
477: %\scalebox{0.6}{\includegraphics[angle=0]{fig7.pdf}}
478: \scalebox{0.6}{\includegraphics[angle=0]{fig7.eps}}
479: \end{center}
480: \caption{Comparison between the experimentally determined resonance 
481: frequencies (crosses with errorbars) and the theoretical model (continuous 
482: line) based on Panina et al's [21] involving rotational motion of the outer 
483: shell magnetization as shown in the inset. The low field region [0 Oe -10 Oe]
484: is a crossover region from domain relaxation to magnetization rotation in the shell.}
485: \label{fig7}
486: \end{figure}
487: 
488: \begin{figure}[!ht]
489: \begin{center}
490: %\scalebox{0.6}{\includegraphics[angle=0]{fig8.pdf}}
491: \scalebox{0.4}{\includegraphics[angle=-90]{fig8.eps}}
492: \end{center}
493: \caption{Microwave absorption of $p$ = 0.8 microwire at 9.4 GHz. The FMR 
494: resonance field is 1,132 Oe. A low-field absorption (LFA) double peak 
495: appears at $H_{DC}   \sim   H_{K}$ =~4 Oe, where MI was observed at low 
496: frequencies. The full separation between MI and FMR has therefore taken 
497: place. }
498: \label{fig8}
499: \end{figure}
500: 
501: 
502: 
503: \end{document}
504: