Influence of stearic acid on mechanochemical reaction between Ti and BN powders

Article abstract of DOI:10.1016/S0925-8388(03)00638-8

The influence of stearic acid as a process control agent on the mechanochemical reaction between Ti and BN to form TiN/TiB₂ nanocomposite powder by high energy ball milling has been investigated. A powder mixture of pure Ti and hexagonal BN powders with a molar ratio of 3:2 was milled for up to 40 h with up to 1.75 wt.% of stearic acid. The XRD analysis shows that an intimate mixture of TiN and TiB₂ crystalline powders was formed during milling by a displacement reaction. It has been observed by monitoring the temperature of the vial surface during the milling process that the addition of over 1.5 wt.% of stearic acid changed the reaction mode from a mechanically induced self-propagating reaction (MSR) to a gradual reaction. It has also been found by XRD and TEM analyses that the MSR produces TiN and TiB₂ particles initially of sub-micron size, which are reduced gradually by further milling. However, there still remain some coarse TiB₂ particles larger than a few hundred nanometers even after 16-h milling due to their extremely high hardness. On the other hand, the gradual reaction induced by stearic acid successfully inhibits the formation of coarse TiB₂ particles and eventually decreases the crystallite size of the products below 40 nm after 16 h of milling.

Full text of DOI:10.1016/S0925-8388(03)00638-8

Journal of Alloys and Compounds 365 (2004) 149–156
Influence of stearic acid on mechanochemical reaction between
Ti and BN powders
∗
Jung-Soo Byun, Jae-Hyeok Shim, Young Whan Cho
Nano-Materials Research Center, Korea Institute of Science and Technology, Seoul 136-791, South Korea
Received 8 May 2003; received in revised form 2 June 2003; accepted 2 June 2003
Abstract
The influence ofstearic acidasa process control agent on the mechanochemical reactionbetweenTi andBNto form TiN/TiB
powder by high energy ball milling has been investigated. A powder mixture of pure Ti and hexagonal BN powders with a molar ratio of 3:2
was milled for up to 40 h with up to 1.75 wt.% of stearic acid. The XRD analysis shows that an intimate mixture of TiN and TiB crystalline
2
nanocomposite
2
powders was formed during milling by a displacement reaction. It has been observed by monitoring the temperature of the vial surface during
the milling process that the addition of over 1.5 wt.% of stearic acid changed the reaction mode from a mechanically induced self-propagating
reaction (MSR) to a gradual reaction. It has also been found by XRD and TEM analyses that the MSR produces TiN and TiB
of sub-micron size, which are reduced gradually by further milling. However, there still remain some coarse TiB particles larger than a few
hundred nanometers even after 16-h milling due to their extremely high hardness. On the other hand, the gradual reaction induced by stearic
acid successfully inhibits the formation of coarse TiB particles and eventually decreases the crystallite size of the products below 40 nm after
6 h of milling.
2003 Elsevier B.V. All rights reserved.
2
particles initially
2
2
1
©
Keywords: Transition metal alloy; Nanostructured materials; Mechanical alloying; Powder metallurgy; TEM
1
. Introduction
Recently, it has been reported that nanocrystalline
TiN/TiB2 composite powder was synthesized from Ti and
There has been increasing interest in refractory materi-
BN powders by high energy ball milling [3]. TiB exhibits
2
als such as carbides, nitrides, borides and silicides owing
to their exceptional hardness and stability at high tem-
peratures. However, single-phase refractory materials have
reached their optimum stages and attention has now turned
to composite refractory materials. On the other hand,
nanocrystalline refractory materials have recently received
much attention owing to their improved mechanical prop-
erties [1]. A variety of techniques for preparing nanocrys-
talline materials have been proposed over a decade. Among
them, high energy ball milling has been recognized as an
effective way of producing nanocrystalline, amorphous and
other non-equilibrium structured powders mainly through a
solid-state reaction [2]. Particularly, high energy ball milling
accompanying chemical reactions, i.e. mechanochemical
processing, is considered to be very effective for preparing
nanocomposite powders including more than two phases.
a high elastic modulus and high hardness. However, pure
TiB is not deformed plastically even at very high tem-
2
peratures due to its intrinsically high Peierls barrier to the
dislocation movement [4]. On the other hand, TiN has a
relatively low elastic modulus and low hardness compared
to TiB and can be deformed at high temperatures. There-
2
fore, it is expected that a nanocomposite of TiN and TiB2
might provide a desired combination of high-temperature
hardness and stability with adequate ductility and fracture
toughness [5]. The mechanochemical synthesis from Ti and
BN powders to form TiN/TiB nanocomposite powder is
2
completed through a mechanically induced self-propagating
reaction (MSR) [6,7] and a nanometer scale mix of the
product powders by further milling. However, there still re-
mains a significant portion of coarse TiB particles because
2
TiB is much harder than TiN. Therefore, it is desirable to
2
reduce coarse TiB2 particles during the milling process.
∗
The purpose of the present study is to mechanochemically
synthesize nanocrystalline TiN/TiB2 composite powder
Corresponding author. Fax: +82-2-958-5379.
E-mail address: oze@kist.re.kr (Y.W. Cho).
0
925-8388/$ – see front matter © 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0925-8388(03)00638-8

1
50
J.-S. Byun et al. / Journal of Alloys and Compounds 365 (2004) 149–156
using high-energy ball milling with the addition of stearic
acid as a process control agent (PCA) to efficiently re-
duce the crystallite size by suppressing MSR. The effect of
stearic acid on the ignition of MSR is investigated and the
structures of TiN/TiB2 powder by both MSR and gradual
reactions are compared.
2
. Experimental procedure
The starting powders used in this study were 99% pure Ti
with a particle size of less than 45 ␮m and 99% pure hexag-
onal BN of average 4 ␮m. A mixture of Ti and BN was
prepared with a molar ratio of 3:2. The Ti/BN mixture was
charged together with stearic acid and Cr-steel balls (two
1
2.7 mm and four 6.35 mm in diameter) into a tool steel
vial with an O-ring seal under Ar atmosphere. The amount
of stearic acid was controlled from zero up to 1.75 wt.%.
The ball-to-powder weight ratio (BPR) was fixed at 5:1. The
high-energy ball milling was carried out for up to 40 h us-
ing a SPEX 8000 mixer/mill. During the milling process,
the temperature of the vial surface was monitored using a
non-contact infrared thermometer. The as-milled powders
were investigated by X-ray diffraction (XRD) using Cu K␣
radiation, and transmission electron microscopy (TEM) us-
ing Philips CM30 operating at 200 kV.
3
. Results
3
.1. MSR without stearic acid
Fig. 1, reproduced from our previous article [3] where all
experimental procedures were the same except that the BPR
was 10:1, shows XRD patterns of Ti/BN mixtures milled
without stearic acid for different milling times. After some
incubation time, the Ti/BN mixture abruptly reacted to form
TiN/TiB2 powder, which resulted in a sudden increase in the
Fig. 2. (a,b) Bright field images of the powder milled without stearic acid
for 105 min, (c) micro diffraction pattern of TiN [112] zone axis obtained
¯
from crystallite A and (d) micro diffraction pattern of TiB2 [1100] zone
axis obtained from crystallite B in (b).
vial surface temperature. Fig. 2 shows bright field (BF) im-
ages of the powder whose milling was interrupted 1 min af-
ter the MSR. The crystallites just after MSR are very coarse
and the size is in the order of 100 nm. Using micro diffrac-
tion as shown in Fig. 2c,d, the products could be indexed
as TiN and TiB2. The displacement reaction during milling
could be represented as:
3Ti + 2BN = 2TiN + TiB2
(1)
Munir [8] proposed a simple guideline to decide whether
or not a self-propagating reaction might occur for a certain
Fig. 1. XRD patterns of the powders milled without stearic acid for (a)
0
, (b) 1.5, (c) 1.75, (d) 4, (e) 8 and (f) 16 h.

J.-S. Byun et al. / Journal of Alloys and Compounds 365 (2004) 149–156
151
Fig. 4. Variation in vial surface temperature with milling time.
3.2. Gradual reaction
In order to control MSR, stearic acid was added as a PCA
and in order to detect MSR the vial surface temperature dur-
ing milling was measured as shown in Fig. 4. Fig. 5 shows
that the incubation time for the MSR depends on the con-
centration of stearic acid. The incubation time is ∼200 min
without stearic acid and there is little change up to 1 wt.%
of stearic acid. The incubation time is rather longer than
that in the previous article [3] because the BPR is reduced
from 10:1 to 5:1. The addition of stearic acid of between 1
and 1.5 wt.% slightly increases the incubation time, but the
MSR still occurred. When 1.5 wt.% of stearic acid is added,
however, the MSR was suppressed in one experiment while
it occurred in another one. With over 1.5 wt.% of stearic
acid, reaction (1) always proceeded gradually. Therefore, the
critical concentration of stearic acid to inhibit the MSR of
reaction (1) is ∼1.5 wt.%. Fig. 6 shows the XRD patterns
of the powders with different amounts of stearic acid af-
ter 16-h milling. Up to 1.375 wt.% of stearic acid, there is
Fig. 3. TEM micrographs of the powder milled without stearic acid for
6 h: (a) dark field image and (b) high resolution image.
1
system. According to his proposal, the reaction can start
without additional energy from an exterior source when
−
ꢀH/Cp is above 2000 K. The −ꢀH/Cp value of
reaction (1) calculated by Thermo-Calc [9] is ∼3800
K, which indicates that reaction (1) can progress via
MSR.
With further milling, the peaks of the products in the
XRD patterns broaden while no other phases are ob-
served. Fig. 3 shows TEM images of the powder milled
for 16 h. The dark field (DF) image shows that the crys-
tallite size is below 10 nm. In this article, all DF images
are obtained from the diffraction rings of TiN {111}
¯
¯
and {200}, and TiB2 {1010} and {1011}. As shown in
Fig. 3b, however, there still remains a significant portion of
coarse TiB2 crystallites, probably because of its very high
hardness.
Fig. 5. Incubation time for MSR with different amount of stearic acid.

1
52
J.-S. Byun et al. / Journal of Alloys and Compounds 365 (2004) 149–156
Fig. 6. XRD patterns of the powders milled for 16 h with (a) 0, (b) 0.5,
c) 1, (d) 1.25, (e) 1.375, (f) 1.5 and (g) 1.75 wt.% stearic acid.
Fig. 7. XRD patterns of the powders milled with 1.75 wt.% stearic acid
for (a) 4, (b) 6, (c) 8, (d) 16 and (e) 40 h.
(
no distinct difference in the XRD patterns of the powders.
However, with over 1.5 wt.% of stearic acid, both TiN and
TiB2 peaks broaden much more, which means that the sup-
pression of the MSR resulted in the reduction of the average
crystallite sizes. The crystallite sizes of TiN and TiB2, esti-
mated by Scherrer’s formula [10], are 9 and 19 nm without
stearic acid, and 5 and 8 nm with 1.75 wt.% of stearic acid,
respectively.
Fig. 8. (a) Bright and (b) dark field image of the powder milled with 1.75 wt.% stearic acid for 6 h, and (c,d) selected area diffraction patterns (SADPs),
respectively, of particles A and B in (b).

J.-S. Byun et al. / Journal of Alloys and Compounds 365 (2004) 149–156
153
Fig. 7 shows the XRD patterns of Ti/BN powders with
.75 wt.% of stearic acid milled for different times. Af-
3.3. Microstructure
1
ter milling for 6 h, TiN peaks appears with broadened Ti
peaks. Further milling increases the intensities of TiN and
TiB2. After milling for 16 h, the displacement reaction
Fig. 8 shows images and selected area diffraction patterns
(SADPs) of the powder milled with 1.75 wt.% of stearic
acid for 6 h. Particle A in Fig. 8b represents the products of
reaction (1), which consist of TiN and TiB2 as can be seen
from SADP, while particle B seems to be unreacted. It may
(1) seems to be almost completed judging from the XRD
data.
Fig. 9. (a,d) Bright field image and (b,e) dark field image of the area marked by rectangle in (a) and (d), and (c,f) SADPs, respectively, of the powder
milled for 16 h without and with 1.75 wt.% stearic acid.

1
54
J.-S. Byun et al. / Journal of Alloys and Compounds 365 (2004) 149–156
Fig. 10. (a,c) Dark field image and (b,d) SADP of the powder milled for 8 h with 1.25 and 1.75 wt.% stearic acid, respectively.
also be inferred from Fig. 8 that reaction (1) is completed
as a whole within a particle triggered by the impacts during
milling even though the transfer of the reaction between
particles by exothermic heat is difficult. The crystallite size
of the TiN/TiB2 products is ∼20 nm as shown in Fig. 8b,
which seems to be typical among several DF images, even
though there are a few particles whose crystallite size is
above 40 nm. Compared with the results right after the MSR
as shown in Fig. 2, the gradual reaction results in much
smaller crystallites.
Fig. 9 shows TEM micrographs and SADPs of the pow-
ders milled for 16 h without and with 1.75 wt.% stearic
acid. The DF image of the as-milled powder without stearic
acid shows that the crystallite size is ∼10 nm. However, the
SADP shows the spots diffracted from coarse TiB2 particles
besides the rings diffracted from the fine particles. It could be
observed by tilting the specimen that there are some coarse
TiB2 particles of over 100-nm diameter, two of which are in-
dicated by the arrows in Fig. 9a. On the other hand, there are
no such coarse TiB2 particles when milled with 1.75 wt.%
stearic acid as shown in Fig. 9d,f. Furthermore, the size of
the fine TiN and TiB2 crystallites is even smaller than that
without stearic acid as compared in Fig. 9b,e. The reduction
in crystallite size observed by TEM is in agreement with the
X-ray result as shown in Fig. 6.
There is no effect of stearic acid on the crystallite size un-
less the amount of stearic acid is enough to result in a grad-
ual reaction. Fig. 10 shows TEM micrographs and SADPs
of the powders milled for 8 h with 1.25 or 1.75 wt.% of
stearic acid. While the powder with 1.25 wt.% stearic acid
shows still coarse crystallites together with fine crystallites,
the powder with 1.75 wt.% of stearic acid induced a gradual
reaction which results in a fine mixture of nanocrystalline
TiN/TiB2 and unreacted Ti/BN.
4. Discussion
The structural evolution from a Ti and BN mixture to
TiN/TiB2 nanocomposite powders can be divided into three
stages when the MSR occurs. At the first stage, crystallite
sizes decrease, reactants mix on a fine scale, and chemically
active defect sites form during the incubation time. The MSR
abruptly occurs to produce TiN and TiB2 of sub-micron
sizes. Finally, further milling reduces the crystallite size and
mixes the products on a nanometer scale. The milling mode

J.-S. Byun et al. / Journal of Alloys and Compounds 365 (2004) 149–156
155
though the crystallite size of TiN is more reduced as shown
in Fig. 11.
As shown in Section 3, ∼0.2 mol.% (1.5 wt.%) of
stearic acid successfully changed reaction (1) from MSR
to a gradual reaction type. Such an amount is too small to
consider the possibility that the change of reaction mode
could be induced by other reactions such as that between
Ti and C from the stearic acid. In general, the change of
reaction mode by an inert additive is attributed primarily
to two mechanisms, i.e. increasing the heat capacity and
decreasing the contact area between the reactants [6,7]. The
adiabatic temperature rise (−ꢀH/Cp) of reaction (1) will
decrease from ∼3800 to ∼3650 K due to the increase in
specific heat by adding 1.5 wt.% of stearic acid. Therefore,
it seems that the suppression of the MSR by stearic acid is
primarily due to the decrease in contact area between the
reactants.
Gradual reaction results in a fine crystallite size of TiN
and TiB2, which can be easily understood because there is
no large temperature increase as the reaction proceeds. In
addition, the TiN and TiB2 products formed in a gradual
reaction seem to be present as a composite as shown in
Fig. 8. Further, it can be inferred from Figs. 8 and 10b
that reaction (1) and mixing the products with unreacted
amorphous Ti/BN occur simultaneously, which is advan-
tageous for forming nanocomposite powders compared
with simply milling two brittle phases of TiN and TiB2
together.
5
. Conclusion
Fig. 11. Dark field images obtained from the TiN/TiSi2 nanocomposite
powders milled for 16 h (a) without and (b) with 1.5 wt.% stearic acid.
The arrows in (b) indicate some unreacted Si3N4.
Nanocrystalline TiN/TiB2 powders are synthesized from
a mixture of Ti and BN powders by high-energy ball milling
with the addition of stearic acid as a process control agent
of TiN and TiB2 products can be classified as a brittle-brittle
system. It is known that TiB2 is much harder than TiN.
Therefore, coarse TiB2 crystallites are difficult to reduce in
size compared with TiN, which causes a portion of coarse
TiB2 particles to remain as shown in Figs. 2 and 9a. There-
fore, it is necessary to control the crystallite size of TiB2
during milling by suppressing the MSR and subsequently
preventing the formation of coarse TiB2 particles. It is worth
comparing the aforementioned results with the structure of
TiN/TiSi2 nanocomposite mechanochemically synthesized
from Ti and Si3N4 mixture [11]. This displacement reaction
which can be expressed as
(PCA). The XRD analysis shows that Ti and BN powders
form TiN and TiB2 powders at the early stage of milling
by a displacement reaction. Up to 1.5 wt.% of stearic acid,
an abrupt increase in the temperature of the vial surface
is detected, which indicates that the reaction progresses in
an MSR form. Over 1.5 wt.% of stearic acid, MSR is sup-
pressed and a gradual reaction occurs. It is found by XRD
and TEM analyses that the MSR produces TiN and TiB2
particles of sub-micron size while the gradual reaction pro-
duces nanocrystalline TiN and TiB2 composite powders.
Therefore, coarse TiB2 particles which still remain even af-
ter 16 h when milling without stearic acid could be success-
fully inhibited by inducing a gradual reaction with stearic
acid.
1
1Ti + 2Si3N4 = 8TiN + 3TiSi2(C54)
(2)
also progresses as a kind of MSR but TiSi2 is softer than
TiN. As milling progresses, the TiN and TiSi2 products
successfully form nanocomposite powder without coarse
particles, where the crystallite size of TiN is ∼10 nm and
TiSi2 seems to be amorphous. In such a system, there
seems to be no big advantage in a gradual reaction against
MSR compared with that in the TiN/TiB2 system, even
References
[
[
1] K. Niihara, J. Ceram. Soc. Jpn. 99 (1991) 974.
2] C. Suryanarayana, Prog. Mater. Sci. 46 (2001) 1.
[3] J.-H. Shim, J.-S. Byun, Y.W. Cho, Scripta Mater. 47 (2002) 493.

1
56
J.-S. Byun et al. / Journal of Alloys and Compounds 365 (2004) 149–156
[
[
4] J. Ramberg, C. Woolfe, W. Williams, J. Am. Ceram. Soc. 68 (1985) [8] Z.A. Munir, Am. Ceram. Soc. Bull. 67 (1988) 342.
C78.
[9] B. Sundman, B. Jansson, J.-O. Andersson, CALPHAD 9 (1985)
153.
[10] B.D. Cullity, S.R. Stock, in: , Elements of X-Ray Diffraction, Prentice
Hall, London, 2001, p 170.
5] F. Olevsky, P. Mogilevsky, E.Y. Gutmanas, I. Gotman, Metall. Mater.
Trans. A 27 (1996) 2071.
[6] L. Takacs, Prog. Mater. Sci. 47 (2002) 355.
7] C. Chakurov, V. Rusanov, J. Koichev, J. Solid State Chem. 71 (1987)
[
[11] J.-H. Shim, J.-S. Byun, Y.W. Cho, J. Metastable Nanocrystalline
Mater. 15–16 (2003) 557.
522.