METALLURGICAL AND MATERIALS TRANSACTIONS A VOLUME 28A, JULY 1997—1429

Amorphization Reaction of Ni-Ta Powders during Mechanical

Alloying

PEE-YEW LEE, JU-LUNG YANG, CHUNG-KWEI LIN, and HONG-MING LIN

This study examined the amorphization behavior of Ni

1002x

alloy powders synthesized by me-

chanically alloying (MA) mixtures of pure crystalline Ni and Ta powders with a SPEX high energy

ball mill. According to the results, after 20 hours of milling, the mechanically alloyed powders were

amorphous for the composition range between Ni

and Ni

. A supersaturated nickel solid

solution formed for Ni

, as well. X-ray diffraction analysis reveals two different types of amor-

phization reactions. Through an intermediate solid solution and by direct formation of amorphous

phase. The thermal stability of the amorphous powders was also investigated by differential thermal

analysis. As the results demonstrated, the crystallization temperature of amorphous Ni-Ta powders

increased with increasing Ta content. In addition, the activation energy of amorphous Ni-Ta powders

reached a maximum near the eutectic composition.

I. INTRODUCTION

PREPARING amorphous alloys based on IVa to VIa

group refractory transition metals is highly attractive, since

amorphous phases of these high melting temperature alloys

possess high crystallization temperatures.

[1]

However, the

high cooling rate necessary to bypass the nucleation and

growth of crystalline phases in the undercooled alloy melts

always severely restricts, for instance, amorphization of the

alloy with a high melting temperature by the liquid quench-

ing method.

[2]

As previous investigations demonstrated,

amorphous materials can be obtained by destabilizing a

solid phase’s crystallinity by a solid-state reaction.

[3,4]

Var-

ious techniques to synthesize amorphous alloys via solid-

state amorphization include hygrogenation, multilayer

interdiffusion, and mechanical alloying (MA).

[5,6]

Mechanical alloying was developed by Benjamin in the

early 1970s to produce superalloy powders strengthened by

oxide dispersions.

[7]

It is a high energy ball milling process

consisting of repeated mechanical mixing, cold welding,

fracturing, and rewelding of powders during ball-powder

collision events.

[8]

Owing to this unique novel milling be-

havior, MA allows materials scientists to circumvent ma-

terial limitations and manufacture alloys that are difﬁcult

or impossible to produce by conventional melting and cast-

ing techniques. For instance, other investigators used MA

to successfully synthesize many equilibrium and/or non-

equilibrium phases in many alloy systems, including amor-

phous materials, quasicrystals, nanocrystalline materials,

rare earth magnets, and intermetallics.

[9,10]

Since the discovery of amorphous phase in mechanically

alloyed Ni

powder,

[11]

this technique has been applied

to numerous amorphous materials.

[6,9,10]

However, the Ni-

Ta binary system has received only limited attention.

PEE-YEW LEE, Professor and Director, and CHUNG-KWEI LIN,

Visiting Associate Professor, are with the Institute of Materials

Engineering, National Taiwan Ocean University, Keelung, Taiwan 202.

JU-LUNG YANG, Engineer, is with Sino-American Silicon Products,

Inc., Hsinchu, Taiwan 300. HONG-MING LIN, Professor and Chairman,

is with the Department of Materials Engineering, Tatung Institute of

Technology, Taipei, Taiwan 104.

Manuscript submitted November 22, 1996.

Therefore, this study investigates the feasibility of prepar-

ing amorphous Ni-Ta powders by MA, starting from ele-

mental Ni and Ta powders and using a shaker ball mill.

II. EXPERIMENTAL DETAILS

A. MA Processes

Elemental powders of Ni (99.98 pct, ,300 mesh) and

Ta (99.98 pct, ,325 mesh) were weighed to yield the de-

sired compositions, Ni

1002x

(x 5 10, 20, 30, 40, 50, 60,

70, 80, and 90), and then canned into an SKH 9 high speed

steel vial (40 mm in diameter and 50 mm in height) to-

gether with Cr steel balls (7 mm in diameter with ball-to-

powder ratio 5 2:1) under an argon-ﬁlled glove box, where

a SPEX* 8000 shaker ball mill (1200 rpm) was employed

*SPEX is a trademark of SPEX Industries, Edison, NJ.

for MA.

The overall MA processes lasted 20 hours and were

interrupted every 15 minutes for the ﬁrst hour and every

30 minutes thereafter. Each interruption was followed by

an equal length of time (;30 minutes) to cool down the

vials. An appropriate amount of the mechanically alloyed

powders was extracted to examine the progress of amor-

phization.

B. Characterization of Mechanically Alloyed Powders

Techniques used to examine the status of amorphiza-

tion include X-ray diffraction, scanning electron micros-

copy (SEM), Vickers microhardness, and differential

thermal analysis (DTA). The X-ray analysis was per-

formed using a SIEMENS* D-5000 diffractometer with

a monochromatic Mo K

radiation (40 kV and 30 mA).

The microstructure and the morphology of the mechani-

cally alloyed powders were examined using a Hitachi

S-4100 scanning electron microscope. Next, the Vickers

microhardness (ﬁve tests for each specimen) was mea-

sured with a Matsuzawa MXT50-UL machine using a

load of 25 g. The thermal stability of the as-milled pow-

1430—VOLUME 28A, JULY 1997 METALLURGICAL AND MATERIALS TRANSACTIONS A

ders was examined using a DU PONT* 2000 differential

*DU PONT is a trademark of Du Pont de Nemours & Co., Inc.,

Wilmington, DE.

thermal analyzer, where the sample was heated from room

temperature to 1300 7C in a puriﬁed argon atmosphere at

a rate of 40 K/min.

III. RESULTS AND DISCUSSION

A. Crystalline-to-Amorphous Transformation

was chosen as a prototype to investigate the

amorphization behavior of Ni-Ta system during MA. The

powder morphology (Figures 1(a) through (d)) and particle

cross sections (Figures 1(e) through (h)) of mechanically

alloyed powders vs milling time were examined using

SEM. More speciﬁcally, the MA process for preparing

involves three stages: cold welding (0 to 1 hours),

fracturing (1 to 5 hours), and amorphization (5 to 20 hours).

During the cold welding stage, the mean particle size

sharply increased (starting from ,20 to ;1000

m, Fig-

ures 1(a) and (b)) and a typical lamellar structure formed

(Figures 1(e) and (f)). This phenomenon may be attributed

to the predominance of repetitive cold welding of Ni and

Ta elemental powders at the early stage of milling. The

individual layer thickness measured from Figure 1(f) (i.e.,

1 hour of milling) ranged from 0.4 to 4

m. As the milling

time increases, the particles are stressed continuously.

Eventually, fracturing dominates, thereby causing a decrease

in particle size and lamellar thickness; the fracturing stage

lasted 1 to 5 hours. Interestingly, during the fracturing stage,

the morphology of the milled powders becomes more spher-

ical (Figure 1(c), 5 hours of milling) than the original plate-

like powders (Figures 1(a) and 1(b)). A balance between cold

welding and fracturing of powders is gradually achieved,

leading to a relatively constant particle size (17 5 5

measured from Figure 1(d)) at the end of the amorphization

stage. In addition, the reﬁned lamellar structure obtained af-

ter 5 hours of milling (Figure 1(g)) became indistinguishable

at the end of processing (20 hours, Figure 1(h)).

Figure 2(a) depicts the variation in particle sizes (mean

5 one standard deviation) for mechanically alloyed

powders as a function of milling time. In addition

to the change in particle size and the reﬁnement of layer

thickness, microhardness measurements can be used to

monitor the progress of MA.

[12]

Figure 2(b) shows that mi-

crohardness increases throughout the cold welding and frac-

turing stages, and no signiﬁcant change occurs during the

amorphization stage. This shows the same trend as reported

in the literature.

[12,13]

The reﬁnement of layer thickness and

the increase in hardness have been used to model the MA

processing.

[13]

X-ray diffraction analysis is another conventional tech-

nique for monitoring the progress of amorphization. Figure

3 displays the X-ray diffraction patterns of as-milled

powders as a function of milling time. As this ﬁg-

ure reveals, most Bragg peaks from pure Ni already dis-

appeared and the intensities of the Bragg peak from Ta

decreased after 5 hours of milling. The peaks of bcc Ta

also broadened asymmetrically toward the high-angle side

because of the dissolution of the smaller Ni atoms in the

bcc lattice. A decreasing intensity and broadening diffrac-

tion peaks are common during the early stage of the MA

process.

[10]

After 20 hours of ball milling, only a broad dif-

fraction peak appears around 2

5 18.7 deg, indicating that

fully amorphous powders have formed.

Furthermore, the crystalline-to-amorphous transforma-

tion behavior of Ni

was examined by differential ther-

mal analyses (DTA). Figure 4 presents the DTA traces of

the as-milled powders vs milling time, where two or three

exothermic peaks (indicating phase transformations) ap-

pear. Interestingly, as the milling time increases, the ﬁrst

exothermic peak (850 to 1000 K) decreases and vanishes

after 5 hours; meanwhile, the second exothermic peak

(1000 to 1100 K) continuously increases. To further ex-

amine the implications of these peaks, two Ni

speci-

mens milled for 5 and 20 hours were annealed for 5 minutes

at 1000 and 1100 K, respectively, quenched to room tem-

perature, and then subjected to X-ray diffraction analysis.

Figure 5 shows the X-ray diffraction patterns of Ni

as milled for 5 hours and annealed powders. This ﬁgure

indicates that after annealing at 1000 K (marked as ‘‘I’’ in

Figure 4; i.e.,Ni

powders milled for 5 hours and an-

nealed at 1000 K for 2 minutes), the intensities for Ta (110)

and other high angle peaks decrease. This observation im-

plies that the percentage of amorphization increases after

annealing at 1000 K. Recall the reﬁned lamellar structure

in Figure 1(g) after 5 hours of milling; the crystalline-to-

amorphous transformation has not yet been completed at

this stage. This layer structure, however, is still too ‘‘thick’’

for the interdiffusion of Ni or Ta to be completed. In ad-

dition, a previous study noted that the diffusivity of Ni

through amorphous Ni

ﬁlm is at least 10,000 times

lower than that for Ni in Zr.

[14]

Therefore, the formation of

thin amorphous ﬁlms at the layer boundaries may inhibit

any further diffusion of Ni or Ta. Annealing of the as-

milled (5 hours) powders at a temperature of 1000 K, which

is below the crystallization temperature of 1048 K of amor-

phous Ni

powder, may enhance the interdiffusion of

Ni-Ta layers and subsequently increase the proportion of

amorphous phase. Consequently, the phase transformation

of the ﬁrst exothermic peak is a crystalline-to-amorphous

transition. This thermally activated solid-state amorphiza-

tion reaction exhibits the same trend as several rod-milled

(TM: transition metals) alloys.

[15,16]

On the other hand, crystalline phases form with powders

heat treated at 1100 K (marked as ‘‘II’’ in Figure 4), and

the second exothermic peak denotes an amorphous-to-crys-

talline transition.

Similar procedures were followed for mechanically al-

loyed Ni

powders after 20 hours of processing. As

Figure 6 indicates, the X-ray diffraction patterns of as-

milled (20 hours) and annealed (at 1000 K, marked as

‘‘III’’ in Figure 4) powders exhibit an amorphous phase;

however, crystallization occurs after annealing at 1100 K

(marked as ‘‘IV’’ in Figure 4). This ﬁnding reconﬁrms that

an amorphous-to-crystalline transition occurs during the

second exothermic peak.

B. Mechanical Alloying of Ni-Ta System

1. Amorphization reaction

Weeber and Bakker

[10]

reported three different types of

amorphization reactions by MA of binary elemental powder

mixtures (A

). Type I denotes that the ‘‘effective crystal-

METALLURGICAL AND MATERIALS TRANSACTIONS A VOLUME 28A, JULY 1997—1431

Fig. 1—SEM micrographs of (a) through (d) powder morphologies and (e) through (h) particle cross sections of as-milled Ni

powders vs milling

time. Please note that different magniﬁcations were used and the milling times were (a) and (e) 15 min, (b) and (f) 1 h, (c) and (g) 5 h, and (d) and (h)

20 h.

line size’’ continuously decreases and the peaks’ positions

shift, ultimately resulting in an amorphous alloy; i.e., the

solid solution is an intermediate stage. The second type

(type II) directly forms an amorphous alloy from the start-

ing elemental powders; i.e., the diffraction peak intensities

of the elements decrease and a broad peak’s intensity of

1432—VOLUME 28A, JULY 1997 METALLURGICAL AND MATERIALS TRANSACTIONS A

Fig. 2—The variations in (a) particle size and (b) microhardness of as-

milled Ni

powders vs milling time.

Fig. 4—DTA traces of mechanically alloyed Ni

powders as a

function of milling time.

Fig. 5—X-ray diffraction patterns of Ni

powders as milled for 5 h

and annealed at 1000 and 1100 K for 10 min.

Fig. 3—X-ray diffraction patterns of mechanically alloyed Ni

powders for different milling times.

the amorphous alloy increases. For the ﬁnal type, inter-

metallics form as intermediate products and further milling

results in an amorphous alloy.

In this study, we observed the ﬁrst two types of amor-

phization reactions. As Figure 7 illustrates, type I amor-

phization reaction (i.e., through a solid solution) occurs at

Ni- or Ta-rich compositions (Ni

; Ni

; and

; Ni

), where only a slight amount or no

amorphous phase appeared after 5 hours of MA processing.

However, for the composition range between Ni

and

, most Ni diffraction peaks disappeared, and a

broad diffraction peak, indicating the formation of amor-

phous alloy, was superimposed on the Ta (110) diffraction

METALLURGICAL AND MATERIALS TRANSACTIONS A VOLUME 28A, JULY 1997—1433

Fig. 6—X-ray diffraction patterns of Ni

powders as milled for 20 h

and annealed at 1000 and 1100 K for 10 min.

Fig. 7—X-ray diffraction patterns of Ni

1002x

powders after 5 h of

milling.

Fig. 8—X-ray diffraction patterns of amorphous Ni

1002x

powders.

peak at around 2

5 18 to 20 deg (Figure 7). This obser-

vation suggests that the 5 hour ball-milled powders were

mixtures of Ta deg solid solution and amorphous phases.

Such an amorphization reaction can be referred to as type

II.

As reported earlier, binary amorphous alloys form by

MA easier for compositions with a large negative heat of

mixing (H

[10]

The values of H

, as calculated by the Mie-

dema model,

[17]

indicate H

to be large between Ta-40 to

60 at. pct Ni and small at both Ni- and Ta-rich sides. Such

a discrepancy implies that the amorphization rate should be

higher in the middle compositions than at the Ni- or Ta-

rich compositions, possibly accounting for why type I is

preferred at the Ni- or Ta-rich compositions and type II

occurs in the middle composition ranges.

2. Glass formation and extended solid solution

Further milling causes amorphous alloys to form over a

wide composition range, i.e.,Ni

to Ni

. Figure 8

presents the corresponding X-ray diffraction patterns. Ac-

cording to this ﬁgure, this range is larger than for amor-

phous alloys produced by the rapid-quenching technique

(40 to 70 at. pct Ni by Giessen et al.,

[18]

35 to 65 at. pct

Ni by Rohr et al.,

[19]

and 50 to 70 at. pct Ni by Prokoshina

and Belousov

[20]

) or by the electron-gun-deposition tech-

nique (20 to 55 at. pct Ni by Schafer and Menzel

[21]

). To

our knowledge, amorphous Ni-Ta alloys have never formed

at compositions between 10 and 30 at. pct Ni, using the

rapid-quenching technique. The absence of a eutectic point

and a high liquidus temperature in this composition range

[22]

are presumably responsible for the inability of an amor-

phous phase to form by rapid solidiﬁcation.

The Ni

powder, however, forms an fcc nickel-rich

solid solution and no signiﬁcant difference arises after 25

hours of milling (Figure 9). According to the phase dia-

gram, the solubility at 800 7Cis;3 at. pct Ta,

[22]

although

the maximum solubility of nickel solid solution was found

to be 17.2 at. pct Ta at 1360 7C.

[23]

The lattice parameter

increases from 0.352 to 0.358 nm, which follows Vegard’s

law;

[24]

i.e., there is a linear relationship between the solu-

bility and expanded lattice parameters. An extended solid

solution can be prepared by MA, as reported in several

binary systems, such as Ni-Nb

[25]

and Cu-Ti.

[26]

3. Thermal stability of amorphous Ni-Ta alloys

Finally, the thermal stability of the Ni-Ta amorphous

powders was examined and the crystallization temperatures

were deﬁned from the ﬁrst peak’s position in the DTA

1434—VOLUME 28A, JULY 1997 METALLURGICAL AND MATERIALS TRANSACTIONS A

Fig. 9—X-ray diffraction patterns of Ni

powders vs milling time.

Fig. 10—The crystallization temperatures and activation energies of

amorphous Ni

1002x

powders.

scans of amorphous powders after 20 hours of milling. Fig-

ure 10 exhibits the crystallization temperature T

, corre-

sponding to a scanning rate of 40 K min

vs composition,

where T

generally increases with increasing Ta content.

Prokoshina and Belousov

[20]

noted a similar crystallization

temperature-composition dependence at compositions be-

tween 30 to 50 at. pct Ta for melt-spun samples, but with

lower crystallization temperatures. For most amorphous

metallic alloys, T

ranges between 0.4 and 0.6 of the melt-

ing temperature,

[27]

which suggests that the alloy composi-

tions with a high liquidus temperature are favorable for a

high T

. Examination of a Ni-Ta phase diagram reveals a

general monotonic increase in liquidus temperature for

large Ta compositions. Rohr et al.

[19]

measured T

of their

splat-quenched Ni

1002x

(x 5 35 to 60) amorphous sam-

ples by measuring resistivity. The T

values exceeded 950

K. Figure 10 reveals that T

lies between 1050 and 1150 K

at compositions between 40 and 90 at. pct Ta, in agreement

with Rohr’s results.

In addition, Figure 10 depicts the activation energy (E

)

of the Ni-Ta amorphous powders as a function of compo-

sition, where a maximum appears at Ni-40 at. pct Ta. In-

vestigations addressing the thermal stability of TM-Ni (TM:

Fe, Hf, Ti, or Zr) have indicated that a high value in

activation energy is associated with a eutectic composi-

tion.

[28,29,30]

Our results support this argument, since the eu-

tectic composition in the Ni-Ta system is Ni-36 at. pct Ta.

IV. SUMMARY

1. Complete amorphization is feasible for the composition

range between Ni

and Ni

. This range is larger

than for amorphous alloys prepared by the splat-cooling

process or by the electron-gun-deposition technique.

2. A supersaturated nickel solid solution formed for Ni-rich

compositions, such as Ni

3. Two different types of amorphization reactions were

identiﬁed: through an intermediate solid solution and by

direct formation of amorphous phase.

4. The crystallization temperatures of amorphous Ni-Ta

powders were proportional to the Ta content.

5. The activation energy of amorphous Ni-Ta powders ex-

hibited a maximum near the eutectic composition.

6. The amorphization rate at the early stage of MA is faster

for intermediate compositions and slower at both Ni-

and Ta-rich compositions.

ACKNOWLEDGMENT

The authors would like to thank the National Science

Council of the Republic of China for ﬁnancially supporting

this work under Grant No. NSC 83-0405-E-019-002.

REFERENCES

1. S. Davis, M. Fischer, B.C. Giessen, and D.E. Polk: in Rapidly

Quenched Metals III, B. Cantor, ed., The Metal Society, London,

1978, vol. 2, pp. 425-30.

2. D. Turnbull: Metall. Trans. B, 1981, vol. 12B, pp. 217-30.

3. X.L. Yeh, K. Samwer, and W.L. Johnson: Appl. Phys. Lett., 1983,

vol. 42, pp. 242-44.

4. R.B. Schwarz and W.L. Johnson: Phys. Rev. Lett., 1983, vol. 51, pp.

415-18.

5. W.L. Johnson: Progr. Mater. Sci., 1986, vol. 30, pp. 81-134.

6. L. Schultz and J. Eckert: in Topics in Applied Physics, vol. 72, Glassy

Metals III H. Beck and H.-J. Gu¨therodt, eds., Springer-Verlag, New

York, NY, 1994, pp. 69-120.

7. J.S. Benjamin: Metall. Trans., 1970, vol. 1, pp. 2943-51.

8. J.S. Benjamin: Sci. Am., 1976, vol. 234, pp. 40-48.

METALLURGICAL AND MATERIALS TRANSACTIONS A VOLUME 28A, JULY 1997—1435

9. C.C. Koch: Mater. Sci. Technol., vol. 15, Processing of Metals and

Alloys, R.W. Cahn, ed., VCH, Weinheim, 1991, pp. 193-246.

10. A.W. Weeber and H. Bakker: Phys. B, 1988, vol. 153, pp. 93-135.

11. C.C. Koch, O.B. Cavin, C.G. McKamey, and J.O. Scarbrough: Appl.

Phys. Lett., 1983, vol. 43, pp. 1017-19.

12. J.S. Benjamin and T.E. Volin: Metall. Trans., 1974, vol. 5, pp. 1929-

34.

13. D. Maurice and T.H. Courtney: Metall. Mater. Trans. A, 1995, vol.

26A, pp. 2437-44.

14. J.C. Barbour, F.W. Saris, M. Nastasi, and J.W. Mayer: Phys. Rev. B,

1985, vol. 32, pp. 1363-65.

15. El-Eskandarany, K. Aoki, and K. Suzuki: J. Appl. Phys., 1992, vol.

72 (7), pp. 2665-72.

16. El-Eskandarany, K. Aoki, and K. Suzuki: Metall. Trans. A, 1992, vol.

23A, pp. 2131-40.

17. A.R. Miedema, P.F. Dechatel, and F.R. de Boer: Phys. B, 1980, vol.

100, pp. 1-28.

18. B.C. Giessen, M. Madhava, and D.E. Polk: Mater. Sci. Eng., 1976,

vol. 23, pp. 145-150.

19. L. Rohr, P. Reimann, P.T. Richmond, and H.-J. Guntherodt: Mater.

Sci. Eng., 1991, vol. A133, pp. 715-17.

20. G.F. Prokoshina and O.K. Belousov: Russ. Metall. Met., 1991, No. 6,

pp. 170-72.

21. A. Schafer and G. Menzel: Thin Solid Films, 1978, vol. 52, pp. 11-

21.

22. A. Nash and P. Nash: Bull. Alloy Phase Diagram, 1984, vol. 5 (3),

pp. 259-65.

23. R.C. Ruhl, B.C. Giessen, M. Cohen, and N.J. Grant: J. Less-Common

Met., 1967, vol. 13, pp. 611-18.

24. B.D. Cullity: Elements of X-ray Diffraction, Addison-Wesley

Publishing Co., Inc., Reading, MA, 2nd ed., p. 350.

25. P.Y. Lee and C.C. Koch: J. Non-Cryst. Solids, 1987, vol. 94, pp.

88-100.

26. C. Politis and W.L. Johnson: J. Appl. Phys., 1986, vol. 60, pp.

1147-51.

27. A. Calka, M. Madhava, D.E. Polk, B.C. Giessen, and J. Vandersande:

Scripta Metall., 1977, vol. 11, pp. 65-70.

28. Z. Altounian, G.-H. Tu, and J.O. Strom-Olsen: J. Appl. Phys., 1983,

vol. 54, pp. 3111-16.

29. Z. Altounian, C.A. Volkertand, and J.O. Strom-Olsen: J. Appl. Phys.,

1985, vol. 57, pp. 1777-82.

30. K.H. Buschow: J. Phys. F, 1984, vol. 14, pp. 593-607.