Crystallization behavior of the Zr63Al7.5Cu17.5Ni10B2 amorphous alloy during isothermal annealing

Crystallization behavior of the Zr63Al7.5Cu17.5Ni10B2 amorphous alloy during isothermal annealing

Intermetallics 13 (2005) 907–911 www.elsevier.com/locate/intermet Short communication Crystallization behavior of the Zr63Al7.5Cu17.5Ni10B2 amorphou...

314KB Sizes 0 Downloads 107 Views

Intermetallics 13 (2005) 907–911 www.elsevier.com/locate/intermet

Short communication

Crystallization behavior of the Zr63Al7.5Cu17.5Ni10B2 amorphous alloy during isothermal annealing J.S.C. Janga,*, L.J. Changa, G.J. Chena, J.C. Huangb a

Department of Materials Science and Engineering, I-SHOU University, q, Sec. 1, Shiuecheng Rd, KTashu Hsiang, Kaohsiung 840, Taiwan, ROC b Institute of Materials Science and Engineering, National Sun Yat-Sen University, Kaohsiung, Taiwan, ROC Received 30 October 2003; accepted 17 December 2004 Available online 23 March 2005

Abstract The crystallization behavior of Zr63Al7.5Cu17.5Ni10B2 amorphous Alloy was studied by means of scanning differential calorimetry (DSC), X-ray diffraction (XRD), and transmission electron microscopy (TEM). A single stage transformation of the amorphous phase forming a Zr2Cu-type crystalline phase was observed. Kinetics for such single stage crystallization was analyzed by means of Johnson–Mehl–Avrami equation and discussed regarding to the value of Avrami exponent. In addition, the cube of crystal size presents a linear relationship with isothermal annealing time. This indicates that the crystallization process of Zr63Al7.5Cu17.5Ni10B2 amorphous alloy belongs to thermally activated process of Arrhenius type. From the HRTEM analysis, small amount of Zr2Cu-type crystals in the nano-scale dimension (10– 20 nm) were observed to precipitate from the amorphous matrix upon the early stage of isothermal annealing the amorphous alloy at the temperature between the glass transition (Tg) and the onset crystallization temperature (Tx). q 2005 Elsevier Ltd. All rights reserved. Keywords: A. Nanostructured intermetallics; B. Glasses, metallic; B. Grain growth; B. Phase transformations; B. Thermal stability

1. Introduction In recent years, a great of attention has been devoted to the mechanism of the high stability of supercooled liquid in the bulk metallic glasses (BMGs), such as the Mg- [1,2], the La- [3], the Ti- [4], and the Zr-based [5–7] systems. Among these metallic glasses, the Zr-based metallic glass is considered to be one of the most promising materials and has much attractive attention due to its exceptional glassforming ability (GFA) with a critical cooling rates as low as 1 k/s and an extremely wide super cooled liquid region exceeding 100 K [6,8]. The supercooled region, DTx, is defined as the difference between the glass transition temperature Tg and the onset crystallization temperature Tx. Another attractive characteristic of the BMG is its unique properties, such as high corrosion resistance, excellent elastic and mechanical properties [9–11] for Zrbased amorphous alloys. In addition, the tensile fracture * Corresponding author. E-mail address: [email protected] (J.S.C. Jang).

0966-9795/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.intermet.2004.12.025

strength of the Zr-based amorphous alloy reaches 1600– 1800 MPa at room temperature and their densities are around 5.9–6.7 Mg/m3 [12]. Therefore, the high specific strength of the BMG is expected to extend the application of bulk amorphous alloys. Based on the fact that this series of new amorphous alloy posses a wide supercooled liquid region, the glass forming ability (GFA) of an alloy which Inoue [13] proposed to define as the critical cooling rate Rc (the minimum cooling rate to obtain a BMG) can be evaluated by both values of Trg and DTx, where Trg is defined as Tg/Tm and Tm is the onset melting temperature. Lu and Li [14] also showed recently that the best forming composition for BMG alloys is at or the near eutectic composition, at which the reduced glass transition temperature Trg given by Tg/Tl reaches the highest value, where Tl is the offset melting temperature. Since Tl plays the most influential role in glass formation, one must look for regions of exceptionally low-lying Tl in the alloy phase diagram, such as the eutectic region. Inoue and Masumoto [13,15,16] also revealed the relationship between the thermal stability of amorphous phase and the topological short-range ordering. It is generally thought that the increase in packing density would cause an increase in thermal stability and mechanical

J.S.C. Jang et al. / Intermetallics 13 (2005) 907–911

strength of an amorphous phase. Therefore, the low values of Tx and mechanical strength for the binary amorphous alloys are presumably because their amorphous structure has a looser atomic configuration as compared with that for the ternary and the quaternary amorphous alloys. In this study, the Zr63Al7.5Cu17.5Ni10B2 alloy with a high activation energy of crystallization (about 360 kJ/mol) [17] is selected to study the crystallization behavior during isothermal annealing process between Tg and Tx.

693K 690K

100

Fraction of crystallization (%)

908

681K

687K

80 60 40 20 0

2. Experimental procedure

0

1000

2000

3000

4000

Time (sec) Fig. 1. Isothermal transformation curves of the Zr63Al7.5Cu17.5Ni10B2 alloy at different temperature.

(a)

8.0

7.5

7.0

lnt

The chemical composition of the alloy used in this study is base on Zr63Al7.5Cu17.5Ni10B2 by atomic percentage. The pre-alloyed ingots were prepared by arc melting of the appropriate mixture of pure elements, including Zr (99.8 wt% purity), Ni (99.9 wt% purity), Cu (99.99 wt% purity), Al (99.99 wt% purity), and B (99.99% purity), in an argon atmosphere. The alloy ribbon samples of about 0.1 mm in thickness and 5 mm in width were prepared by the single roller melt-spinning technique in an argon atmosphere. The thermal property of the as-quenched samples was characterized by TA Instruments DSC 2920 differential scanning calorimeter (DSC). All amorphous ribbons were isothermally annealed at a temperature between Tg and Tx in vacuum at different time. The as quenched and the annealed structure were examined by Scintag X-400 X-ray diffraction (XRD) with monochromatic CuZ Ka radiation and transmission electron microscopy (Joel 300CX TEM) with nano beam electron diffraction.

6.5

6.0 30% 50% 80%

5.5

3. Result and discussion

5.0 1.43

The kinetic study of crystallization was performed by using the Johnson–Mehl–Avrami (JMA) [18] isothermal analysis for volume fraction x, which is transformed as a function of time t based on the following Eq. (1):

ln½lnð1=ð1 K xÞÞ Z n lnðkÞ C n lnðtÞ:

(2)

1.46

1.47

(b) 0.5

0.0

(1)

These samples were annealed isothermally at several temperatures between Tg and Tx. The volume fraction transformed versus the annealing time plots for Zr63Al7.5 Cu17.5Ni10B2 alloy is shown in Fig. 1. For each annealing temperature (Ta), the time needed to reach 30, 50, and 80% crystallinity was individually recorded to plot curves of ln t versus 1/Ta, as shown in Fig. 2(a). The activation energy was determined to be 380 kJ/mol for the Zr63Al7.5Cu17.5Ni10B2 alloy by the slope. This value is about 20% higher than the activation energy of crystallization for the Zr65Al7.5Cu17.5 Ni10 alloy (314 kJ/mol). This indicates that the additions of boron can improve the thermal stability of the Zr-based alloy. In additions, Eq. (2) can be rewritten as:

1.45

1/T(10–3K–1)

ln(-ln(1-f))

xðtÞ Z 1 K exp½KðktÞn :

1.44

–0.5

–1.0 681 687 690 693

–1.5

–2.0 4.0

4.5

5.0

5.5

6.0

6.5

7.0

n=1.71 n=1.69 n=1.7 n=1.91 7.5

8.0

lnt Fig. 2. (a) Plots of ln t versus 1/Ta for different fractions of crystallization to derive the activation energy for Zr63Al7.5Cu17.5Ni10B2 alloy. (b) JMA plots constructed for 0.1!0.9 from the isothermal transformation curves depicted in Fig. 2.

Arbitrary intensity

J.S.C. Jang et al. / Intermetallics 13 (2005) 907–911

Fig. 3 shows the XRD pattern of the as-spun Zr63Al7.5 Cu17.5Ni10B2 alloy. A broad maximum with two small peaks of Zr2Ni (211) and Zr2Cu (103) was observed of this alloy in this study. In parallel, the TEM examination revealed that the structure of the as-quenched ribbon was not quite uniform. This can be illustrated by the HRTEM (high resolution TEM) as shown in Fig. 4. The majority of the matrix was amorphous phase (as shown in Fig. 4(a)) but

Zr2Cu (103) Zr2Cu (110)

Zr2Ni (211)

20

30

40

50

909

60

70

80

2£c Fig. 3. XRD pattern of as-quenched Zr63Al7.5Cu17.5Ni10B2 alloy.

where k is the effective rate constant and n is the Avrami exponent. Avrami plots of ln[Kln(1Kx)] versus ln(t), constructed for 0.1!0.9 from the curves at different temperatures in Fig. 2(b). The average value of the Avrami exponent n was 1.75 for Zr63Al7.5Cu17.5Ni10B2 alloy at the temperature range from 694 to 700 K, which is similar to the value (1.96G0.25) of Zr65Al7.5Cu17.5Ni10 alloy. This indicates that these two alloys present similar crystallization behavior, with a decreasing nucleation rate [19].

Fig. 4. HRTEM images of the as-quenched alloy 4Si: (a) bright field image, (b) SAD of the amorphous matrix, (c) dark field image from the (103) reflection of Zr2Cu phase.

Fig. 5. HRTEM lattice image of the Zr63Al7.5Cu17.5Ni10B2 alloy annealed at 681 K at different time; (a) 0 s, (b) 300 s, and (c) 1000 s.

910

J.S.C. Jang et al. / Intermetallics 13 (2005) 907–911

where D0 is the particle size in diameter at taZ0, Dt is the particle size at taZt, and the rate constant k depends on temperature in the Arrhenius relation. This result is analogous to the growth behavior with thermal activation process of Arrhenius type reported by Saida [22].

(a) 30

crystal size (nm)

25 20 15

4. Conclusion

10 5 0

250

500

750

1000

1250

1500

1750

Aging time (sec.)

(b)

According to the results of DSC, X-ray diffraction, HRTEM observation, and SEM observation for the Zr63 Al7.5Cu17.5Ni10B2 alloy, the crystallization behavior and mechanical properties of Zr63Al7.5Cu17.5Ni10B2 alloy can be summarized as:

cube of crystal size ( D3/nm3 )

35000 30000 25000 20000 15000 10000 5000 0 0

250

500

750

1000

1250

1500

1750

Aging time (sec) Fig. 6. Relationship between the (a) grain size and (b) cube of grain size of Zr63Al7.5Ni10Cu17.5B2 alloy during isothermal annealing at 681 K.

some ultra fine crystalline particles with random orientation, of size about 5 to 10 nm, were embedded in the amorphous matrix (as shown in Fig. 4(b)). Each single particle examined by the HRTEM image is confirmed to be single crystals of tetragonal-structured Zr2Cu type crystalline phase. This is analogous to the previous reported by Massalski et al. [20] and Jang et al. [21]. Both research teams discovered that a single metastable crystalline phase with a size about 10 nm formed near the glass transition temperature during rapid solidification from the melt. When the Zr63Al7.5Cu17.5Ni10B2 alloy annealed at 681 K for 300 s, the average size of Zr2Cu type nano-crystals grows from 5 to 12 nm, as shown in Fig. 5(a) and (b). In addition, the average size of Zr2Cu type nano-crystals increases to 20 nm with annealing time up to 1000 s, as shown in Fig. 5(c). The crystal size measured by TEM exhibits increases monotonously with increasing ta and shows a good linear relationship between the annealing time ta and the cube of particle size of the Zr2Cu type crystalline phase, as shown in Fig. 6. This can be described by the following Eq. (3): D3t K D30 Z kta

(3)

(1) The activation energy of crystallization for the Zr63 Al7.5Cu17.5Ni10B2 alloy was about 380 kJ/mol as determined by the Arrhenius plot. These values are 20% higher than the activation energy of crystallization for the Zr65Al7.5Cu17.5Ni10 alloy (314 kJ/mol). This indicates that the boron addition exhibits the effect on improving the thermal stability of the Zr65Al7.5Cu17.5 Ni10 alloy. (2) The average value of the Avrami exponent n was calculated to be 1.75 for Zr63Al7.5Cu17.5Ni10B2, which is similar to the value (1.96) of Zr65Al7.5Cu17.5Ni10 alloy. This indicates that these two alloys present the similar crystallization process, with a decreasing nucleation rate. (3) The changes in crystal size with increasing annealing time shows a good linear relationship between annealing time and the cube of crystal size of the Zr2Cu type crystalline phases. This indicates that the crystal growth of the alloy belongs to thermally activated process of Arrhenius type.

Acknowledgements The authors would like to gratefully acknowledge the sponsorship from the National Science Council of ROC under the project NSC92-2216-E-214-015.

References [1] Inoue A, Ohtera K, Kita K, Masumoto T. Jpn J Appl Phys 1988;27: L2248. [2] Kim SG, Inoue A, Masumoto T. Mater Trans JIM 1990;31:929. [3] Inoue A, Yamaguchi H, Zhang T, Masumoto T. Mater Trans JIM 1990;31:104. [4] Inoue A, Amiya K, Yoshii I, Nishiyama N, Masumoto T. Mater Sci Eng 1994;A179/180:692. [5] Inoue A, Zhang T, Masumoto T. Mater Trans JIM 1990;31:177. [6] Zhang T, Inoue A, Masumoto T. Mater Trans JIM 1991;32:1005. [7] Johnson WL. MRS Bull 1999;24:42. [8] Kim YJ, Busch R, Johnson WL, Rulison AJ, Rhim WK, Isheim D. Appl Phys Lett 1994;65:2136.

J.S.C. Jang et al. / Intermetallics 13 (2005) 907–911 [9] Inoue A, Zhang T, Masumoto T. Mater Trans JIM 1995;36:391. [10] Inoue A. Mater Trans JIM 1995;36:886. [11] Inoue A, Koshiba M, Zhang T, Masumoto T. J Appl Phys 1998;83: 1967. [12] Inoue A, Negishi T, Kimura HM, Zhang T, Yavari AR. Mater Trans JIM 1998;39:318. [13] Inoue A. Acta Mater 2000;48:279. [14] Lu ZP, Tan H, Li Y, Ng SC. Scr Mater 2000;42:667. [15] Inoue A, Zhang T, Masumoto T. J Non-Cryst Solids 1993;156-158: 473.

911

[16] Inoue A, Zhang T, Chen MW, Sakurai T. J Mater Res 2000;15:2195. [17] Jason SC, Jang LJ, Chang YT, Jiang PW. Wong. Mater Sci Forum 2003;426-432:1867. [18] Christian: JW. The Theory of transformations in metals and alloys. 2nd ed. New York, NY: Pergamon Press Inc.; 1975 p. 15–19. [19] Christian: JW. The Theory of transformations in metals and alloys. 2nd ed. New York, NY: Pergamon Press Inc.; 1975 p. 542. [20] Massalski TB, Woychik CG. Acta Metall 1985;33:1873. [21] Jang JSC, Hung TH, Chang LJ. Mater Sci Eng 2004;A375-377:307. [22] Saida J, Li C, Matsushita M, Inoue A. J Mater Sci 2000;35:4143.