The field sensitivity of both the longitudinal permeability ratio and the magnetoimpedance ratio for the an-nealed Co 70 Fe 5 Si 15 Nb 2.2 Cu 0.8 B 7 ribbon increase exponentially as the
Trang 1Enhanced GMI effect in a Co70Fe5Si15B10 ribbon due to Cu and Nb substitution for B
M H Phan*, 1
, H X Peng1
, M R Wisnom1
, S C Yu2
, and N Chau 3
1 Department of Aerospace Engineering, Bristol University, Queen's Building, University Walk, Bristol, BS8 1TR, United Kingdom
2
Department of Physics, Chungbuk National University, Cheongju, 361-763, South Korea
3
Center for Materials Science, National University of Hanoi, 334 Nguyen Trai, Hanoi, Vietnam
Received 24 November 2003, revised 28 January 2004, accepted 5 February 2004
Published online 25 March 2004
PACS 75.50.Kj, 75.60.Ch, 75.75.+a
We present here, the results of an investigation on giant magnetoimpedance (GMI) effect in both annealed and as-quenched Co 70 Fe 5 Si 15 B 10 and Co 70 Fe 5 Si 15 Nb 2.2 Cu 0.8 B 7 ribbons Substitution of Cu and Nb for B in
an initial Co 70 Fe 5 Si 15 B 10 composition forming the Co 70 Fe 5 Si 15 Nb 2.2 Cu 0.8 B 7 composition improves both GMI effect and its field sensitivity The GMI effect was more pronounced in the annealed samples The field sensitivity of both the longitudinal permeability ratio and the magnetoimpedance ratio for the an-nealed Co 70 Fe 5 Si 15 Nb 2.2 Cu 0.8 B 7 ribbon increase exponentially as the testing temperature is increased, indi-cating that the magnetic permeability is very sensitive to the temperature The results obtained are of sig-nificant importance in developing quick-response magnetic sensors
© 2003 WILEY-VCH Verlag GmbH & Co KGaA, Weinheim
1 Introduction
Recently, a giant magnetoimpedance (GMI) effect, which was discovered in soft magnetic amorphous alloys, has generated growing interest owing to its high potential for magnetic sensors application [1, 2] The GMI phenomenon can be understood as the conjunction of a skin effect and a strong field depend-ence of the transverse magnetic permeability associated with transverse domain wall motions [1] At low frequencies, the GMI effect is demonstrated to originate from the contribution of the induced magneto-inductive voltage to magnetoimpedance (MI) As frequency increases, in the high-frequency regime, the GMI effect can be interpreted in terms of the applied dc magnetic field dependence of impedance as a result of the transverse magnetization with respect to the ac current direction flowing through the sample and the skin effect due to this ac current In such a magnetic material, the transverse permeability µ af-fects the magnetic penetration depth (δm) through δm = (ρ/πfµ)1/2
, where f is the frequency and ρ is the electrical resistivity It is worth noting that as the skin effect becomes dominant (a/δmⰇ 1, a is the
thick-ness of the ribbon), the impedance Z is proportional to (fµ)1/2
[1] Hence, µ decreases rapidly upon the dc applied magnetic field, causing a significant change in MI, i.e the GMI effect
In general, Co- and Fe-based amorphous alloys with nearly zero magnetostriction exhibit very large change of magnetoimpedance (MI) caused by the application of an external magnetic field [2–4]
Addi-tionally, Knobel et al [5] suggested that the time relaxation of impedance plays an important role in
identifying what soft magnetic amorphous material is suitable for a quick-response magnetic sensor application This observed relaxation of the impedance is related to the transverse magnetic permeability, also known as magnetic permeability aftereffect (MAE) or simply disaccomodation [6] There the
maxi-*
Corresponding author: e-mail: M.H.Phan@bristol.ac.uk, Phone: +44 (0)117 928 7697, Fax: +44 (0)117 927 2771
Trang 2mum GMI was found corresponding to a maximum transverse permeability It is therefore desirable to seek more suitable GMI candidate material for novel magnetic sensors
In this paper, we report the results of an investigation on the GMI effect in annealed and as-quenched
Co70Fe5Si15B10 and Co70Fe5Si15Nb2.2Cu0.8B7 amorphous alloys Interestingly, substitution of Cu and Nb for B in an initial Co70Fe5Si15B10 composition forming the Co70Fe5Si15Nb2.2Cu0.8B7 composition favors both GMI and its field sensitivity The GMI effect in both the amorphous samples can be significantly improved by appropriate heat treatment These results are very beneficial for magnetic sensor applica-tion
2 Experimental
The Co70Fe5Si15B10 and Co70Fe5Si15Nb2.2Cu0.8B7 ribbons with a width of 4 mm and a thickness of 20 µm were prepared by the rapid quenching method The as-quenched ribbons were annealed in vacuum at
550 K for 1 hour X-ray diffraction analysis confirmed the quality of the samples The hysteresis loops,
M-H, were measured using a vibrating sample magnetometer The resistivity measurements for the two
kinds of samples were carried out using the four-probe method The obtained resistivity values for the
Co70Fe5Si15B10 and the Co70Fe5Si15Nb2.2Cu0.8B7 compositions are ρ = 1.1 × 10–5Ωm and 0.75 × 10–5Ωm, respectively Magnetoimpedance measurements were carried out along the ribbon axis with the longitu-dinal applied magnetic field The samples with a length of about 15 mm were used for all MI measure-ments A schematic diagram of the magnetic impedance measurement system is depicted in Fig 1 A computer-controlled RF signal generator with its power amplifier was connected to the sample with a series of resistors for monitoring the driving ac current We measured the ac current and voltage across the sample using a digital multimeters (DMM) with RF/V probes to compute the impedance The exter-nal dc field, applied by a solenoid, can be swept through the entire cycle equally divided by 800 intervals from –150 to 150 Oe The frequency of the ac current was varied from 1 to 10 MHz, while its magnitude was kept constant at 10 mA
3 Results and discussion
The magnetoimpedance ratio ∆Z/Z can be defined as ∆Z/Z(%) = Z(H)/Z(Hmax) - 1, where Hmax is the external magnetic field sufficient to saturate the impedance and equals to 150 Oe in the present work Similarly, the longitudinal permeability ratio ∆µ/µ can also be defined as ∆µ/µ(%) = µ(H)/µ(Hmax) – 1
In several works [2–4], it has been shown that the change of MI is closely related to that of the longitudinal
Cryogenic
system
Solenoid
RF power Amp.
R
R’
Fig 1 (online colour at: www.interscience.wiley.com) Schematic diagram of the magnetoimpedance
measurement system
Trang 3permeability Hence, the evaluation for the GMI effect in such a soft magnetic amorphous alloy can be realized either by ∆Z/Z measurements or
by ∆µ/µ measurements
As shown in Fig 2, the ∆Z/Z curves for both
the as-quenched samples, measured at
frequen-cies up to f = 4.1 MHz, have a single peak at
zero field At the frequency of 3.1 MHz, it
is noted that the maximum ∆Z/Z is ~55%
for Co70Fe5Si15B10 while it is ~89% for
Co70Fe5Si15Nb2.2Cu0.8B7 The higher ∆Z/Z value for the latter compound at f = 3.1 MHz is likely
due to the presence of its special domain struc-ture as transverse domains formed by a magne-tomechanical coupling between internal stress and magnetostriction [1–3] Substitution of Cu and Nb for B in the initial Co70Fe5Si15B10 com-position seems to have promoted the formation
of a transverse domains structure, because of the presence of Cu and Nb allowing the forma-tion of well-differentiated microstructures [7]
In other words, a higher transverse permeability
of the latter compound could result in the larger ∆Z/Z Additionally, the maximum ∆Z/Z for
Co70Fe5Si15Nb2.2Cu0.8B7 composition is larger than that for Co70Fe5Si15B10 composition at all the meas-ured frequencies, indicating a favorably formed transverse domain structure in Co70Fe5Si15Nb2.2Cu0.8B7 Another reason causing the difference in ∆Z/Z for the two samples to be noted is the difference in their
electrical resistivities As reported earlier in Ref [8], the higher the electrical resistivity of amorphous alloy, the lower the obtained ∆Z/Z is In the present case, the resistivity is ρ = 1.1 × 10–5
Ωm for
Co70Fe5Si15B10 composition, and this is higher than ρ = 0.75 × 10-5Ωm for Co70Fe5Si15Nb2.2Cu0.8B7 com-position Hence, the higher ∆Z/Z for Co70Fe5Si15Nb2.2Cu0.8B7 composition was obtained Furthermore, one can see from Fig 2(b) that there is a smaller full width at haft maximum (FWHM) in ∆Z/Z curves for
Co70Fe5Si15Nb2.2Cu0.8B7 composition This indicates a high field sensitivity of ∆Z/Z (or the so-called magnetic response), which is ~17–18%/Oe at a current driving frequency of f = 3.1 MHz for
Co70Fe5Si15Nb2.2Cu0.8B7 composition More interestingly, the high magnetic response of this sample remains unchanged at high frequencies, which is ideal for magnetic sensors application in the high-frequency regime As compared to the Co70Fe5Si15Nb2.2Cu0.8B7 composition, Co70Fe5Si15B10 shows the much lower magnetic response (~1%/Oe) at the same frequency of 3.1 MHz This difference can be understood in terms of the local magnetic anisotropy in Co70Fe5Si15B10 which was much larger than that
in Co70Fe5Si15Nb2.2Cu0.8B7 (as estimated by the hysteresis loops) It is the local anisotropy that consid-erably reduces the transverse magnetization associated with the transverse permeability, thereby leading
to the broadening in ∆Z/Z curves and the smaller ∆Z/Z for Co70Fe5Si15B10 composition [9] It is also in-teresting to note that, due to the internal stress relief [7–11], the GMI effect was significantly improved
in both annealed samples relative to their as-quenched counterparts (see Fig 3)
Fig 2 ∆Z/Z vs the external magnetic field
at various frequencies for the as-quenched amorphous samples: (a) Co 70 Fe 5 Si 15 B 10 composition and (b)
Co70Fe5Si15Nb2.2Cu0.8B7 composition
0
10
20
30
40
50
60
0
15
30
45
60
75
90
2.1 MHz 3.1 MHz 4.1 MHz
(b)
H (Oe)
1.1 MHz 2.1 MHz 3.1 MHz 4.1 MHz
Trang 41 2 3 4 5 6 7 8 9
0
20
40
60
80
100
120
140
Frequency (MHz)
No 1 (as-quenched)
No 1 (annealed)
No 2 (as-quenched)
No 2 (annealed)
As shown in Fig 3, the maximum ∆Z/Z is plotted against frequency for all annealed and as-quenched
samples At the measured frequencies, ∆Z/Z is larger for the annealed samples than that for the
as-quenched samples ∆Z/Z first increases with increasing frequency up to f = 3.1 MHz and then decreases at higher frequencies This feature can be interpreted as follows: at frequencies below 1.1 MHz (a < δm), the maximum ∆Z/Z was not very large because of the contribution of the induced magneto-inductive voltage
to magnetoimpedance When 1.1 MHz ≤ f ≤ 3.1 MHz (a ≈ δ m), where the skin effect is dominant, the higher ∆Z/Z is reported Beyond f = 3.1 MHz, the maximum ∆Z/Z decreases with increasing frequency The reason here is that, in this frequency region (f ≥ 3.1 MHz), domain wall displacements were strongly damped owing to eddy currents thus contributing less to the transverse permeability, i.e., a small ∆Z/Z
Since the temperature dependence of GMI profile plays an important role in identifying what amor-phous soft magnetic alloy is useful for a quick-response magnetic sensor In the present study, we inves-tigated the influences of the temperature on the longitudinal permeability and GMI profile in the an-nealed Co70Fe5Si15Nb2.2Cu0.8B7 alloy The results show that the magnitude of ∆µ/µ decreases with
0
100
200
300
400
500
600
H (Oe)
T = 10 K
T = 300 K
Fig 4 Permeability change ∆µ/µ as a function of
magnetic field at f = 0.1 MHz and at T = 10, 300 K
Fig 3 [∆Z/Z]max vs frequency for the an-nealed and as-quenched Co 70 Fe 5 Si 15 B 10 (No 1) and
Co 70 Fe 5 Si 15 Nb 2.2 Cu 0.8 B 7 (No 2) samples
0 50 100 150 200 250 300 40
50 60 70 80 90 100
5 10 15 20 25 30 35
40
LPR
max (T) = LPR
max (0)exp(cT 1/2 ) with
c = const.
f = 1 MHz
] ma
] ma
T (K)
MIRmax(T) = MIRmax(0)exp(cT 1/2
) with
c = const.
f = 1MHz
Fig 5 (online colour at: www.interscience.wiley.com)
Temperature dependence of [∆µ/µ]max /∆H and
[∆Z/Z]max /∆H at a frequency of 1 MHz for the
an-nealed Co 70 Fe 5 Si 15 Nb 2.2 Cu 0.8 B 7 amorphous alloy The solid line indicates the fit
Trang 5increasing frequency from 100 kHz to 5 MHz, but it increases with the testing temperature between 10 and 300 K (for example, see in Fig 4) Note that there was a remarkable change in the shape of ∆µ/µ
curves with increase of the measured frequency and temperature It was found that the ∆µ/µ curves
be-came narrower with further increase of the frequency and temperature In contrast, the ∆Z/Z curves
be-came broader with further increases in the measured frequency and temperature This difference is caused by an increase of impedance and a reduction in the longitudinal permeability with increasing frequency [3] Also, the increase in ∆Z/Z and ∆µ/µ with further increase in the temperature results
mainly from enhanced effective magnetic permeability of the sample [12, 13], because, at low tempera-tures (~10 K), the exchange energy between magnetic moments is larger than that at high temperatempera-tures (~300 K) Thereby, the circular motion of magnetic moments at low temperature might be frozen thus leading to the lower permeability and the smaller ∆Z/Z As shown in Fig 5, for the temperature
depend-ence of ∆µ/µ and ∆Z/Z at f = 1 MHz, the field sensitivity of ∆µ/µ and ∆Z/Z, defined as [∆µ/µ]max/∆H
and [∆Z/Z]max/∆H respectively, increase exponentially as the temperature is increased This reflects the
fact that the magnetic permeability is very sensitive to the temperature, which is very useful in develop-ing so-called quick-response magnetic sensors [12–15] The decrease of anisotropy field and the increase
of GMI profile with the testing temperature in the present study are believed to be attributed to the in-creased magnetic softness [16]
4 Conclusions
We investigated the GMI effect in annealed and as-quenched Co-based amorphous ribbons Substitution
of Cu and Nb for B in the initial Co70Fe5Si15B10 composition forming the Co70Fe5Si15Nb2.2Cu0.8B7 com-position favors both GMI and its field sensitivity The GMI effect was significantly improved in the samples annealed at 550 K The results obtained are beneficial for quick-response magnetic sensor appli-cation Besides, the field sensitivity of the longitudinal permeability ratio and the magnetoimpedance ratio for the annealed Co70Fe5Si15Nb2.2Cu0.8B7 ribbon increases exponentially as the temperature is in-creased, reflecting that the magnetic permeability is very sensitive to the temperature The decrease of anisotropy field and the increase of GMI profile with the testing temperature could be attributed to the increased magnetic softness
Acknowledgements The authors are grateful to the support from the Korean Science and Engineering Foundation
through the Research Center for Advanced Magnetic Materials at Chungnam National University and to the support from the Vietnam National Program for Fundamental Research Grant No 420110
References
[1] L V Panina, K Mohri, T Uchiyama, and M Noda, IEEE Trans Magn 31, 1249 (1995)
[2] Y K Kim, W S Cho, T K Kim, C O Kim, and H B Lee, J Appl Phys 83, 6575 (1998)
[3] H B Lee, K J Lee, Y K Kim, T K Kim, C O Kim, and S C Yu, J Appl Phys 87, 5269 (2000)
[4] M H Phan, S C Yu, J S Chung, J S Kim, and Y M Kim, J Appl Phys 93, 9913 (2003)
[5] M Knobel, M L Sartorelli, and J P Sinnecker, Phys Rev B 57, R3362 (1997)
[6] H Kronmuller and N Moser, in: Amorphous Metallic Alloys, edited by F E Luborsky (Butterworths, Lodon/Washington, 1983), p 341
[7] P Marin, M Vázquez, A O Olofinjana, and H A Davies, Nanostruct Mater 10, 299 (1998)
[8] H Chiriac, T A Ovari, and C S Marinescu, Nanostruct Mater 12, 775 (1999)
[9] Y Yoshizawa, S Oguma, and K Yamauchi, J Appl Phys 64, 6044 (1988)
[10] K S Byon, S C Yu, J S Kim, and C G Kim, IEEE Trans Magn 36, 3439 (2000)
[11] H Chiriac, T A Ovari, and C S Marinescu, J Appl Phys 83, 6584 (1998)
[12] G Chen, X L Yang, L Zeng, J X Yang, F F Gong, D P Yang, and Z C Wang, J Appl Phys 87, 5263 (2000)
[13] C G Kim, Y W Rheem, C O Kim, S S Yoon, E A Ganshina, M Yu Kochneva, and D A Zaichenko, J
Magn Magn Mater 258–259, 170 (2003)
[14] D P Makhnovskly, L V Panina, and D J Mapps, Appl Phys Lett 77, 121 (2000)
[15] M H Phan, S C Yu, C G Kim, and M Vázquez, Appl Phys Lett 83, 2871 (2003)
[16] S S Yoon, S C Yu, G H Ryu, and C G Kim, J Appl Phys 85, 5432 (1999)