• 沒有找到結果。

The effects of annealing temperature and sputtering power on the structure and magnetic properties of the Co-Fe-Zr-B thin films

N/A
N/A
Protected

Academic year: 2021

Share "The effects of annealing temperature and sputtering power on the structure and magnetic properties of the Co-Fe-Zr-B thin films"

Copied!
5
0
0

加載中.... (立即查看全文)

全文

(1)

The effects of annealing temperature and sputtering power on the structure

and magnetic properties of the Co-Fe-Zr-B thin

films

Guo-Ju Chen

a,*

, Sheng-Rui Jian

a,*

, Jason Shian-Ching Jang

b

, Yung-Hui Shih

a

, Yuan-Tsung Chen

a

,

Shien-Uang Jen

c

, Jenh-Yih Juang

d

aDepartment of Materials Science and Engineering, I-Shou University, Kaohsiung 840, Taiwan

bInstitute of Materials Science and Engineering, Department of Mechanical Engineering, National Central University, Chung-Li 320, Taiwan cInstitute of Physics, Academia Sinica, Taipei 115, Taiwan

dDepartment of Electrophysics, National Chiao Tung University, Hsinchu 300, Taiwan

a r t i c l e i n f o

Article history:

Available online 26 April 2012 Keywords: B. Glasses, metallic F. Calorimetry B. Thermal stability F. Microscopy B. Magnetic properties

a b s t r a c t

The microstructure and magnetic properties of the amorphous Co-Fe-Zr-B thinfilms grown on glass substrates by dc magnetron sputtering are investigated using differential scanning calorimetry (DSC), transmission electron microscopy (TEM), and superconducting quantum interference device (SQUID) techniques. The Co-Fe-Zr-B thinfilms deposited at room temperature were annealed at temperatures ranged from 683 K to 773 K. Experimental results indicated that the coercivity (Hc) of the Co-Fe-Zr-B thin

films is significantly influenced by residual stress and crystalline phases within the films. The correlation of the coercivity and the microstructure of Co-Fe-Zr-B thinfilms are discussed. After annealed at 683 K, the coercivity of the Co-Fe-Zr-Bfilm was as low as 1.2 Oe.

Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Recently, Co-Fe based metallic glasses and nanocrystalline materials have attracted much interest owing to the excellent soft magnetic properties[1e3]. To extend the application of BMG in the high frequency power converters, the soft magnetic BMG in the films form, i.e., thin film metallic glass (TFMG), can be used in inductors, current transformers and other devices requiring high permeability and low core loss at low frequencies. In order tofit the requirement, soft metallic glass should exhibit a high susceptibility and large saturation[4]. Meanwhile, the magneticfilms generally required high electrical resistivity to suppress eddy currents and also increase the skin depth in high frequency applications. Thin film metallic glasses (TFMGs) are thought to be a suitable candidate for MEMS applications because they are isotropic and homoge-neous as well as free from defects originating from the crystal structure[5].

In this study, we prepared Co-Fe-Zr-Bfilms by DC sputtering method. It was found that the microstructures of the annealedfilms as well as the associated magnetic properties are strongly

dependent on the sputtering power delivered to the Co-Fe-Zr-B target and the subsequent annealing temperatures.

2. Experiment

The Co-Fe-Zr-Bfilms were prepared by DC sputtering method. The Co52Fe20Zr8B20 target was placed on sputtering gun as the

source materials. The films were grown on Coring 7059 glass substrates at room temperature for 20 min. The base pressure was 5 106 torr. Ar was used as the working atmosphere and the

working pressurefixed at 4 mtorr during growth. The sputtering power delivered to the Co-Fe-Zr-B target ranged from 10 to 50 W. The Co-Fe-Zr-Bfilms were annealed by rapid thermal annealing at a heating rate of 50 K/s, and the annealing temperatures were kept within the range of 683 Ke773 K for 3 min. The main reasons of annealing the Co-Fe-Zr-Bfilms in such a short time period were to release the residual stress in thefilms and to control the grain sizes of the crystalline phase formed in the Co-Fe-Zr-Bfilms in the nano-scale.

The crystal structure and the microstructure of thefilms were investigated by X-ray diffraction and transmission electron microscopy operated at 200 kV (FEI Tecnai G2). The compositions of thefilms were examined by electron probe microanalyzer (EPMA). Magnetic properties were studied using a superconducting

* Corresponding authors.

E-mail addresses:[email protected](G.-J. Chen),[email protected](S.-R. Jian).

Contents lists available atSciVerse ScienceDirect

Intermetallics

j o u rn a l h o m e p a g e : w w w . e l s e v ie r . c o m / l o c a t e / i n t e r m e t

0966-9795/$e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.intermet.2012.03.017

(2)

quantum interference device (SQUID) at room temperature with a maximum magneticfield of 1.2 kOe.

3. Results and discussions

DSC traces of the as-deposited Co-Fe-Zr-B films at different heating rates are used to evaluate the crystallization of the crys-talline phase, as shown inFig. 1. From DSC curves, the glass tran-sition temperature (Tg),first crystallization temperatures (Tx), and

exothermic peak temperature (Tp) can be respectively determined.

The DSC curves exhibit two exothermic procedures indicating two stages of crystallization. The crystallization temperature of thefirst crystalline phase, Tx, for the Co-Fe-Zr-Bfilms with various

sput-tering power is in the temperature range of 753e762 K, which does not show apparent dependence with sputtering power. However, the onset glass transition temperature Tgof the Co-Fe-Zr-Bfilms is

slightly increased with the sputtering power as listed inTable 1. For bulk metallic glass, the glass transition temperature is related to the content of the free volume in the materials. The relaxation annealing, on the other hand, would lead to the escape of the free volume entrapped during the alloy solidification from the amor-phous matrix[6]. During thinfilms deposition, the sputtering may cause some inhomogeneity and defects in the films, which may

lead to some empty volume, i.e., some fraction of matter is having a lower atomic coordination than that in a reference region with a denser atomic packing. These sites may be the preferred regions for initiating the glassy structure destabilization caused by glass transition[7]. Therefore, the glass transition temperature of the thinfilms might be influenced by deposition parameters.

The activation energy for crystallization of the ordering Co-Fe-Zr-Bfilms was determined by means of the Kissinger plot[8], ln  4 T2  ¼ Ea RTþ const (1)

where4 is the heating rate, T is the specific temperature, R is the gas constant, and Eais the activation energy. The plot of the ln(b/T2)

versus 1/Tp(Tpis the peak temperature of crystallization) yielded

a straight line and the slope is related to the activation energy of crystallization. The Kissinger plot for Co-Fe-Zr-B films grown at various sputtering power is shown asFig. 1(b). From the slope, the activation energy of thefilm was estimated. The activation energy of the first crystallization of the Co-Fe-Zr-B thin films increases from 242 kJ*mole1to 276 kJ*mole1with the increasing sputter-ing power. While the second crystallization was only observed in the Co-Fe-Zr-Bfilms grown at 40 W and 50 W, and the activation energy of the second crystallization of the Co-Fe-Zr-B thinfilms increases enormously from 335 kJ*mole1 to 412 kJ*mole1with the increasing sputtering power.

The compositions of the Co-Fe-Zr-B films are identified by EPMA. By varying the sputtering power, there is no obvious difference between the composition of the Co-Fe-Zr-Bfilms and that of the target.Fig. 2shows the XRD patterns of the as-deposited and annealed Co-Fe-Zr-B thin films. Since the time duration of annealing quite limited, the content of crystalline phase may have

a

b

Fig. 1. (a) DSC curves with various heating rates of the Co-Fe-Zr-Bfilms grown at 40 W; (b) The Kissinger plots for Co-Fe-Zr-Bfilms grown at various sputtering powers.

Table 1

The Tg, Tx,DTxand activation energies of the Co52Fe20Zr8B20films grown at various

sputtering powers. Sputtering power (W) Tg(K) Tx(K) DTx(K) QTp1a(kJ/mole) QTp2(kJ/mole) 20 686 753 67 242 e 30 693 762 69 231 e 40 702 759 57 276 335 50 698 758 60 253 412 aQ

Tp1and QTp2are the activation energies of thefirst and second stages of

crystalline phases in the amorphous Co52Fe20Zr8B20films after annealing.

(3)

been still lower than the detection limits of XRD. As a result, no trace of any crystallized phase can be identified by the result of X-ray diffraction.

For the as-deposited Co-Fe-Zr-B thinfilms, the TEM observation revealed the morphology of a uniform phase, as shown inFig. 3(a). The corresponding electron diffraction pattern also confirmed the amorphous feature of the Co-Fe-Zr-B films, which indicates an excellent glass forming ability and homogeneous composition distribution.Fig. 3 (b)w(d) show the conventional transmission electron microscopy photos of the Co-Fe-Zr-Bfilms grown at 40 W and then annealed at 713 K, 743 K and 773 K, respectively. For the Co-Fe-Zr-Bfilms grown at 40 W, the Tgand Txare 702 K and 759 K,

respectively. The Co-Fe-Zr-B films exhibited amorphous feature, while the annealing temperature was below the glass temperature (not shown here). When the annealing temperature was increased to 713 K, which is higher than Tg, nanocrystalline phases started to

homogeneously precipitate in the amorphous matrix. The grains size of these nanocrystals ranges from 3 nm to 12 nm with an average size of about 6 nm. From the corresponding electron diffraction and composition analysis, thefirst crystalline phase was identified as the fcc-(Co, Fe) phase. With further increasing the annealing temperature to 773 K, the nanocrystals grew to a larger average grain size of about 9 nm. The corresponding electron diffraction pattern indicated the existences of the second crystal-line phases as CoZr2and Co23Zr6.

According to the TEM results, the evolution of the crystallization of the Co-Fe-Zr-Bfilms is consistent with the DSC results, which showed two exothermic events indicating two crystallization

stages. Briefly, the crystallization was triggered when the annealing temperature was raised to higher than glass transition tempera-ture. The fcc-(Co, Fe) nanocrystals with an average size of 6 nm was first to appear at the early stage of crystallization. Afterwards, nanocrystals of fcc-(Co, Fe), CoZr2 and Co23Zr6 were observed to

precipitate from the amorphous matrix upon the subsequent stage of crystallization. The crystallization sequence can be briefly summarized below.

a/

T>Tg

a

0þfccðCo;FeÞ/a00þfccðCo;FeÞþCoZr

2þCo23Zr6 (2)

where

a

is the amorphous phase below glass transition tempera-ture,

a

’ and

a

" are amorphous phases when the annealing temperature is above glass transition temperature.

Fig. 4 shows the hysteresis loops of the Co-Fe-Zr-B films annealed at (a) 683 K, (b) 713 K, (c) 743 K, and (d) 773 K, respec-tively. The magnetic measurement was performed along thefilm direction with a maximum applied field of 1200 Oe. All the as-deposited Co-Fe-Zr-Bfilms are amorphous for various sputtering powers. Owing to the lack of long-range order in atomic arrange-ment, the crystal anisotropy is absent in the amorphousfilms. The coercivefield (Hc) of the as-deposited Co-Fe-Zr-Bfilms grown at

40 W is about 6.8 Oe. With the annealing treatment, the Hcvalue of

the Co-Fe-Zr-Bfilms initially decreased with increasing tempera-ture from 6.8 Oe to 1.2 Oe, presumably due to the relief of residual stresses. However, with the crystallization of fcc-(Co, Fe), CoZr2and

Co23Zr6 phases, the Hc increases with increasing annealing

temperature when it is above the glass transition temperature.

(4)

It has been pointed out that the residual stress as well as microstructure irregularities existed in thefilms might destroy the soft magnetization behavior of films [9]. According to Mayr’s model, the intrinsic stress in the amorphousfilm is closely related to the surface morphology andfilm thickness[10]. The annealing-induced initial reduction in the coercivity field at low annealing temperatures described above thus might directly correlate to the elimination of the residual stress in the as-depositedfilms. This could be the primary reason why the film annealed at 683 K exhibited extremely soft magnetizing behavior with Hcas low as

1.2 Oe. When the annealing temperature was elevated to 713 K, the dramatic increase in Hcmay, on the other hand, originate from the

crystallization of Co-Fe-Zr-Bfilms. As confirmed in the TEM results, the (Co, Fe), CoZr2and Co23Zr6phase formed during annealing at

high temperatures. Moreover, the average grain size of the ferro-magnetic crystalline (Co, Fe) phase increased with annealing temperature to about 9 nm. Herzer et al. reported on the grain size dependent coercive force Hcof the nanocrystalline soft magnetic

materials, which predicted a dependence of Hcf D6, where D is the

mean diameter of the crystallites in the nanocrystalline material [11]. The present results, thus, consistently confirmed that the magnetic properties of the Co-Fe-Zr-B films are significantly influenced by the detailed nanostructure and the residual stress existing infilms.

4. Conclusions

In this study, the effects of the sputtering power and annealing temperature on the nanostructure and magnetic properties of the Co-Fe-Zr-B thinfilms have been investigated. Experimental results showed that the sputtering power has some influences on the glass transition temperatures, which in turn enhances the GFA of the Co-Fe-Zr-Bfilms. In addition, the magnetization behavior of the Co-Fe-Zr-B thinfilms was found to strongly dependent on the detailed nanostructure of the amorphous film as well as the appearance of the crystalline phases. Below the glass transition temperature, increasing the annealing temperature tends to relieve the intrinsic stress of the amorphous Co-Fe-Zr-Bfilms leading to an extremely low coer-civefield of 1.2 Oe for films annealed at 683 K. The appearance of nanocrystalline phases, however, would lead to dramatic increase in the coercivefield.

Acknowledgement

This work was partially supported by National Science Council of Taiwan and I-Shou University, under Grant No.: NSC98-2221-E-214-015, NSC100-2221-E-214-024 and ISU100-02-16. JYJ is partially supported by the National Science Council

(5)

and the MOE-ATU program operated at NCTU. We also appreci-ated the help of the MANALAB at ISU for experiments.

References

[1] Inoue A, Shen B, Nishiyama N. Development and applications of Late transition metal bulk metallic glasses. USA: Springer; 2007.

[2] Johnson WL. MRS Bull 1999;24:42. [3] Wang WH. Prog Mater Sci 2007;52:540.

[4] Song JS, Min BK, Jeong DY, Kim HS, Heo JS. J Magn Magn Mater 2001;234:494. [5] Hata S, Sakurai J, Shimokohbe A. 18Th IEEE Inter Conf Micro Electro Mech

Syst; 2005:479.

[6] Jiang WH, Jiang F, Green BA, Liu FX, Liaw PK, Choo H, et al. Appl Phys Lett 2007;91:041904.

[7] Chen M. Ann Rev Mater Res 2008;36:445. [8] Kissinger HE. Analyst Chem 1957;29:1702.

[9] Chen LH, Chen HK, Hsieh CT, Shih YH, Chen IG, Chen SY, et al. J Appl Phys 2002;91:8450.

[10] Mayr SG, Samwer K. Phys Rev Lett 2001;87:036105. [11] Herzer G. J Magn Magn Mater 1992;112:258.

數據

Fig. 1. (a) DSC curves with various heating rates of the Co-Fe-Zr-B films grown at 40 W; (b) The Kissinger plots for Co-Fe-Zr-B films grown at various sputtering powers.
Fig. 4 shows the hysteresis loops of the Co-Fe-Zr-B films annealed at (a) 683 K, (b) 713 K, (c) 743 K, and (d) 773 K,  respec-tively
Fig. 4. The hysteresis loops of the Co-Fe-Zr-B films (a) as-deposited, and annealed at (b) 683 K, (c) 743 K, and (d) 773 K.

參考文獻

相關文件

 Promote project learning, mathematical modeling, and problem-based learning to strengthen the ability to integrate and apply knowledge and skills, and make. calculated

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

Then, we tested the influence of θ for the rate of convergence of Algorithm 4.1, by using this algorithm with α = 15 and four different θ to solve a test ex- ample generated as

Particularly, combining the numerical results of the two papers, we may obtain such a conclusion that the merit function method based on ϕ p has a better a global convergence and

Then, it is easy to see that there are 9 problems for which the iterative numbers of the algorithm using ψ α,θ,p in the case of θ = 1 and p = 3 are less than the one of the

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>

compounds, focusing on their thermoelectric, half-metallic, and topological properties. Experimental people continue synthesizing novel Heusler compounds and investigating

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix