0% found this document useful (0 votes)
7 views11 pages

EBSD paper2

Uploaded by

PranjalGupta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
7 views11 pages

EBSD paper2

Uploaded by

PranjalGupta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Smart Materials and Structures

PAPER You may also like


- Role of surface energy in nanowire growth
Surface-energy-induced selective growth of (001) Xiaoming Yuan, Jiabao Yang, Jun He et
al.
grains in magnetostrictive ternary Fe–Ga-based - Effects of surface stability on the
morphological transformation of metals
alloys and metal oxides as investigated by first-
principles calculations
Juan Andrés, Lourdes Gracia, Amanda
To cite this article: S M Na and A B Flatau 2012 Smart Mater. Struct. 21 055024 Fernandes Gouveia et al.

- Surface energy of integrable quantum spin


chains
M T Batchelor and C J Hamer

View the article online for updates and enhancements.

This content was downloaded from IP address 103.37.201.174 on 22/07/2024 at 13:16


IOP PUBLISHING SMART MATERIALS AND STRUCTURES
Smart Mater. Struct. 21 (2012) 055024 (10pp) doi:10.1088/0964-1726/21/5/055024

Surface-energy-induced selective growth


of (001) grains in magnetostrictive ternary
Fe–Ga-based alloys
S M Na and A B Flatau
Department of Aerospace Engineering, University of Maryland, College Park, MD 20742, USA

E-mail: nsmv2k@umd.edu and aflatau@umd.edu

Received 27 January 2012, in final form 28 March 2012


Published 1 May 2012
Online at stacks.iop.org/SMS/21/055024
Abstract
Surface-energy-induced selective grain growth has been used to increase the presence of
preferred planes in polycrystalline ternary Fe–Ga based alloys and thereby maximize the
magnetostrictive performance of these alloys. In this study, alloys were either doped with
elemental sulfur or annealed under sulfur atmospheres to control the sulfur concentration on
the surface of samples during annealing. We show that the segregation of sulfur, which is
known to play an important role in controlling surface energy, can be correlated to the
selective growth of {001} grains and an increase in the saturation magnetostriction of the
samples. The results show that sulfur atoms are adsorbed (diffused) from the sulfur
atmosphere (bulk interior) and segregate on the sample surface. The correlation between
surface chemistry, texture development and magnetostriction is presented. The formation of
{001} grains occurred under slight surface segregation of sulfur, i.e. at levels of concentration
of surface sulfur between 0.5 and 1.35 at.%, for alloys of (Fe0.813 Ga0.187 )99 B1 and
(Fe0.813 Ga0.187 )99.5 B0.5 doped with 50 ppm S. In the case of (Fe0.81 Ga0.19 )99 (NbC)1 alloy
annealed under a H2 S atmosphere, abnormal growth of (001) grains resulted in 88.3% of the
sample area being covered with a (001) grain.
(Some figures may appear in colour only in the online journal)

1. Introduction magnetostriction (λ100 ) in single crystal galfenol. The


h110i direction, a medium magnetization direction, has a
Magnetostrictive Fe–Ga (galfenol) alloys are of great saturation magnetostriction, λ110 , that is given by (1/4)λ100 +
interest for use in actuation, sensing, and energy harvesting (3/4)λ111 [11]. For alloys with ∼19% Ga, λ111 ∼ 0 and
applications [1–4]. Single crystal galfenol, first introduced therefore λ110 is roughly one quarter of the magnetostriction
in 2000 [5], exhibits a maximum magnetostriction of that can be achieved along the h100i direction [7].
∼400 ppm along the h100i axes parallel to a magnetic easy Thus, to maximize both magnetostriction and ease of
axis [6, 7]. Highly textured polycrystalline rolled galfenol magnetization in highly textured rolled Fe–Ga sheet it is
sheet provides an attractive alternative to single crystal and desirable to develop textures for which at least one h001i
bulk polycrystalline galfenol, with potential for enhanced crystallographic axis is aligned along the plane of the rolled
mechanical robustness and cost-effective production. The sheet, i.e. to develop textures of the form {hkl} h001i.
development of Goss {011} h100i and cubic {001} h100i However, most of these orientations are hard if not impossible
textured rolled sheets shown in figure 1 has been investigated to develop. Of these, the {001} h100i orientation or cube
in ternary Fe–Ga–X alloys, where an additional element texture (figure 1(a)) is particularly desirable for sensor and
(X) such as B, Mo, Nb and NbC was introduced to actuator use in that it provides two h100i easy axes of
improve ductility and rollability [8–10]. The h100i is the magnetization in the plane of the sheet. The extra benefit of
direction of easy magnetization and of the largest saturation developing cubic texture is that it will make it feasible to use

0964-1726/12/055024+10$33.00 1 c 2012 IOP Publishing Ltd Printed in the UK & the USA
Smart Mater. Struct. 21 (2012) 055024 S M Na and A B Flatau

oxygen to a sample surface and/or at the grain boundary


interface has been studied and is known to affect surface
energy and recrystallization. It has been correlated to the
development of texture with specific surface planes [17–19].
Köhler and Kramer suggest that in Fe–Si alloys, surface-
energy-induced selective growth of {011} grains occurs under
very clean surface conditions, and that growth of {001} grains
occurs under slight surface contamination with sulfur due to
adsorption and segregation of the sulfur [17, 20]. However,
there is a threshold above which the selective growth of
{001} grains is changed to that of {111} grains. Their work
addresses the adsorption of sulfur using hydrogen sulfide
(H2 S) gas in an atmosphere of hydrogen. It has also been
suggested that surface-energy-induced selective grain growth
can be governed by the interfacial segregation of sulfur which
is contained in the bulk interior [21, 22]. This correlation
between the surface segregation of sulfur and selective grain
growth was observed in 3% Si–Fe alloys doped with 30 ppm
(0.003 at.%) sulfur. During final annealing, a {001}-grained
sample was obtained with relatively slight surface segregation
of sulfur.
Fe–Ga alloys have an α-iron structure that is similar to
the structure found in Fe–Si alloys. Because of this similarity,
we anticipated the potential for correlation between sulfur
segregation and selective grain growth that is similar to
what has been observed in Fe–Si alloys. To promote the
Figure 1. Schematic diagram of (a) cubic and (b) Goss textures in
polycrystalline rolled sheet. development and survival of {001} grains, we control the
segregation of sulfur at the surface of the rolled Fe–Ga sheet
by controlling annealing conditions such as time, temperature
magnetic field annealing to maximize performance [12, 13] and atmosphere. In this paper, results of chemical analysis
and thereby eliminate the need for stress annealing or use of on the surfaces of samples that had been annealed under the
prestress components in the design of devices that use these controlled sulfur environments and resultant crystallographic
materials [14, 15]. A {011} h100i or Goss texture (figure 1(b)) textures are presented. The development of a strong texture in
provides a h100i direction along the rolling direction (RD) ternary Fe–Ga-based alloys is shown to strongly correlate to
but has a h110i orientation perpendicular to the RD which an increase in the saturation magnetostriction values exhibited
is slightly less desirable because in-plane magnetization is in these samples.
not as easy, but with one in-plane h100i direction it is also
expected to exhibit magnetostriction that is useful for sensor 2. Experiments
and actuator applications.
Thermal annealing under atmospheric gas was employed Three kinds of raw materials were examined:
to develop cubic texture, {001} h100i, in ternary galfenol (Fe0.813 Ga0.187 )99 B1 , (Fe0.813 Ga0.187 )99.5 B0.5 doped with
sheets in two previous studies [10, 16]. Annealing of Fe85 Ga15 50 ppm sulfur, and (Fe0.81 Ga0.91 )99 (NbC)1 alloys. Both
alloy with 1 mol% NbC at 1300 ◦ C for ∼24 h changed Fe–Ga–B alloys were prepared at the Materials Preparation
the texture from a strong α-fiber texture h110i k RD in an Center, US DOE Ames Laboratory, IA, USA, by a standard
as-rolled sheet to a predominant texture of {001} h100i [16]; arc-melting technique and the Fe–Ga–NbC alloy was prepared
however, the authors did not report the magnetostriction by ETREMA Product Inc. by melting elemental Ga with pure
values before or after annealing and provide no explanation iron and chill casting into ingot form through an induction
for the mechanism responsible for the texture change. More melting technique. Elemental boron and niobium carbide
recently, Li et al reported that cubic textured galfenol alloys (NbC) particles with dimensions of 1–5 µm were used to
with added B and Nb exhibited a maximum magnetostriction improve ductility by suppressing grain boundary fracture
of 206 and 134 ppm, respectively [9, 10]. These values are less during the rolling process. The ingots were re-melted several
than 52% of saturation magnetostriction that can be achieved times to ensure homogeneity.
along the h100i direction in the single crystals. These authors The starting ingots were enclosed in a stainless steel
comment on that the mechanisms that drive the growth of cube can that was sealed to prevent oxidation on the surface
{001} grains are not well understood. of the specimen. To minimize diffusion bonding of the
Our work builds on prior work in which surface energy alloy specimen with the can during hot rolling, the can
control was used to promote {001} grains in iron-based alloys. was preoxidized at 1000 ◦ C in air for 1 h. Following hot
The segregation of surface-active elements such as sulfur and soaking at 1000 ◦ C, the canned specimen was hot rolled at

2
Smart Mater. Struct. 21 (2012) 055024 S M Na and A B Flatau

Table 1. Experimental parameters and sulfur segregation mechanisms on the sample surface.
Alloy Alloy Total Final Annealing Segregation Annealing
ID composite reduction (%) thickness (mm) atmosphere mechanism conditions
A (Fe0.813 Ga0.187 )99 B1 95.5 0.35 Elemental S in Vaporization and 1200 ◦ C for 1–5 h
argon adsorption
B (Fe0.813 Ga0.187 )99.5 B0.5 96.0 0.30 Argon Diffusion from 1200 ◦ C for 0.5–4 h
doped with 50 ppm S interior
C (Fe0.81 Ga0.19 )99 (NbC)1 94.0 0.45 0.5%H2 S in Decomposition 1200 ◦ C for 1–3 h
argon and adsorption

used for identifying the crystal structure and phase. Electron


backscattering diffraction (EBSD) patterns were captured
and analyzed using orientation imaging microscope (OIM)
data collection software to obtain pole figures (PF), inverse
pole figure (IPF), volume fraction of each orientation and
orientation distribution function (ODF) plots. The diffraction
spectra collected from the EBSD scans were analyzed
Figure 2. Photo of quartz sealing with a small amount of sulfur and using ODF software to quantify texture. This technique
rolled Fe–Ga–B sheet in an argon atmosphere. yields a three-dimensional (3D) representation of the overall
crystallographic texture. Each data point collected during the
scan is plotted in 3D space using its Euler angles (ϕ1 , 8, ϕ2 ) as
700–1000 ◦ C and subsequent warm rolling at 350–400 ◦ C X, Y, and Z coordinates. A section/plane is then isolated that
provided a thickness reduction of ∼90%. This sheet was represents a given texture. The h100i (η-fiber texture) is found
annealed at 800 ◦ C for 2 h under argon to relieve internal along the 8 at ϕ1 and ϕ2 = 0, and the h110i (α-fiber texture,
stress. After stress relief, cold rolling was undertaken until 8 = 45◦ and ϕ2 = 0) and h111i (γ -fiber texture, 8 = 55◦ and
the sample thickness was reduced to 0.30–0.45 mm. Samples ϕ2 = 45◦ ) are found along the ϕ1 coordinate line.
with a dimension of 12 mm × 12 mm were cut from the The saturation magnetostriction of each specimen was
rolled sheet and were subsequently annealed at 1200 ◦ C with measured using the directionally applied magnetic field
various annealing times and various atmospheres using a method, whereby strain gauged samples were rotated in a
sulfur atmosphere within a quartz ampoule for the sulfur-free DC field provided by Nd–Fe–B permanent magnets. A strain
Fe–Ga–B samples and using flowing argon for sulfur-doped gauge (Vishay Measurement Group model CEA-06-250UW-
Fe–Ga–B samples. A small amount (∼13.0 mg) of elemental 350) with an active gauge area of 4.57 mm × 6.35 mm (base
sulfur was placed in the ampoule with each specimen of area of 6.9 mm × 14.0 mm) was aligned along the rolling
sulfur-free Fe–Ga–B alloy sheet (figure 2). Fe–Ga–NbC direction in the center of each sample and attached to the
samples were annealed under a sulfur atmosphere using sample using M-bond 200 adhesive. The gauges themselves
0.5% hydrogen sulfide (H2 S) gas in argon. All annealed are insensitive to changes in magnetic fields. The strain
samples were quenched in water after annealing. Details of gauged sheet was centered within the 1.5 in gap between
the experimental parameters and mechanisms are shown in two 2 in diameter permanent magnets, where the nominal
table 1. magnetic field in air was uniform along the in-plane direction
Sulfur and other elements at the surface were char- of the rolled sheet specimen, with an amplitude of 3500 G.
acterized by x-ray photoelectron spectroscopy (XPS) and Samples were 0.3–0.45 mm thick; thick enough not to flex
Auger electron spectroscopy (AES). Here, XPS was used or bend in the magnetic field, yet thin enough (i.e. of a
to measure elemental composition and coupling states at sufficiently high aspect ratio) that the field from the magnets
the sample surface and AES was used to estimate the saturated the samples. Each specimen was rotated to orient
quantity of elemental concentrations with depth from the the rolling direction from parallel to perpendicular directions
surface. The AES instrument used in this study was a with respect to the DC field using a stepper motor. The strain
PHI-700 scanning Auger nanoprobe of ULVAC-PHI, INC., response follows a cosine squared function with the peak to
equipped with a cylindrical mirror analyzer and ion sputtering peak value of (3/2)λS = λk − λ⊥ .
Ar gun (500 eV) that could produce a beam size of
7 nm at a voltage of 20 kV and a current of 1 nA. The 3. Results and discussion
XPS experiments were carried out with a VG Microtech
ESCA 2000 spectrometer. The measurements were made 3.1. Sulfur-annealed Fe–Ga–B alloy (A alloy)
at room temperature under a vacuum of better than 1 ×
10−9 Torr with a x-ray source of non-monochromatized Sulfur-free (Fe0.813 Ga0.187 )99 B1 sheets (A alloy) were
Al Kα (1486.6 eV), x-ray power of 195 W, take-off angle annealed in quartz ampoules containing controlled amounts
of ∼57.7◦ , pass energy of 20 eV, and an investigated area of elemental sulfur. For this sulfur-anneal, elemental sulfur
of about 10 mm. Additionally, x-ray diffraction (XRD) was was chosen as an additive instead of H2 S gas because of ease

3
Smart Mater. Struct. 21 (2012) 055024 S M Na and A B Flatau

Figure 3. Magnetostriction as a function of annealing time in


(Fe0.813 Ga0.187 )99 B1 alloy sheets annealed at 1200 ◦ C in quartz
ampoules filled with argon and indicated mass of sulfur and then
water quenched.

of vaporization and adsorption in a closed system like the


ampoule. All samples were annealed and then quenched in
water. Figure 3 shows the measured magnetostriction values
as a function of annealing time. Sulfur-annealed samples that
used 2.5 and 6.4 mg of S exhibited the peak magnetostriction
values of about 200 ppm, while the peak value from samples
annealed with no sulfur in the ampoule is 160 ppm or
∼40 ppm lower than that of sulfur-annealed samples. As the Figure 4. XPS S 2p doublets of iron sulfide on the surface of
sulfur-annealed (Fe0.813 Ga0.187 )99 B1 alloy sheets. Samples were
sulfur content increases from zero to 6.4 mg, the peaks in the annealed at 1200 ◦ C for 2, 4 and 6 h each in quartz ampoules filled
convex profiles of magnetostriction versus duration-of-anneal with argon and 6.4 mg S and then water quenched.
shifted from 6 to 2 h and narrowed.
The sulfur was vaporized and then segregated onto the
surface of the samples during annealing. Analysis of surface
chemistry by XPS and AES was used to estimate the surface
condition by relating the quantity of sulfur bonds and other
chemical bonds, including alloying elements in the Fe–Ga–B
alloy. As shown in figure 4, XPS indicated contributions
centered at approximately 159.2 eV and 161.7 eV have
been assigned to elemental gallium and iron sulfide (Fe–S),
respectively. There is no evidence of coupling of Ga–S. The
two fitted peak positions of S 2p3/2 (S 2p1/2 ) at a binding
energy of around 161.5 (162.7) and 163.2 eV (164.4 eV) with
a full width at half maximum of 1.58 eV for three sheets
annealed at 1200 ◦ C represent the presence of stoichiometric
FeS and FeSn (polysulfide). The S 2p level decomposed by
means of a peak synthesis program is in good agreement
with segregated sulfur on a Fe(111) surface [23]. FeS was
clearly characterized by x-ray diffraction (XRD) patterns, but
no other iron sulfide was detected (figure 5). The main peaks Figure 5. X-ray diffraction patterns of sulfur-annealed
of FeS were observed at 2θ of 29.92◦ , 33.67◦ , 42.17◦ , 43.15◦ , (Fe0.813 Ga0.187 )99 B1 alloy sheets. Samples were annealed at
and 53.12◦ . The peak intensity of FeS in the 2 h-annealed 1200 ◦ C for 2 and 4 h each in quartz ampoules filled with argon and
6.4 mg S and then water quenched.
sample is much lower than that in the 4 h-annealed sample.
This corresponds to the relative areas occupied in XPS peaks.
It is likely that the surface structure may be composed of the surface, elemental sulfur (S◦ ) would be analyzed at a
FeS with additional sulfur stacking on the FeS layer at binding energy of around 164.0 eV [24]. As annealing time
the atomic scale. If more sulfur might be segregated on increased, the peak intensity of S 2p increased proportionally,

4
Smart Mater. Struct. 21 (2012) 055024 S M Na and A B Flatau

The surface segregation of sulfur was confirmed by AES


depth profile results in the 3 h-annealed (Fe0.813 Ga0.187 )99 B1
sample with the sulfur content of 6.4 mg in a quartz ampoule
(figure 6). The surface segregation of sulfur and boron was
observed in atomic concentrations of as much as 10% and
20%, respectively, while after ion sputtering for 2 min, both
elements were barely detected or were not detected at all.
It was also clear from the XPS analysis that boron was
mainly combined with nitrogen and oxygen to form BN (B 1s;
191.1 eV) and B2 O3 (B 1s; 192.7 eV) [25]. Both compounds
disappeared completely after polishing the surface of the
annealed sheet because they just existed at the surface. Some
oxidation of Ga and B was caused by quenching in water
because, while hot, the sample was exposed to air.
The demonstrated ability to control the segregation of
Figure 6. AES depth profile of a (Fe0.813 Ga0.187 )99 B1 sample sulfur on sample surfaces is promising for the next goal,
annealed at 1200 ◦ C for 3 h in a quartz ampoule filled with argon to demonstrate that surface-energy-induced selective grain
and 6.4 mg S and then water quenched. The sputter rate is growth can occur. As expected, the {011} grains are dominant
3.4 nm min−1 , based on the SiO2 . under a very clean surface, corresponding to an annealed sheet
with no sulfur. The {001} grains occur under a relatively
low surface contamination with sulfur due to the adsorption
and diffusion of sulfur. The {111} grains occur with greater
while that of Ga 3s is contrary to the sulfur trends. The
surface contamination due to the adsorption and diffusion of
adsorbed sulfur atoms appear to preferentially combine with
sulfur. Figure 7 shows ODF plots calculated from the EBSD
iron atoms on the sheet surface, whereas gallium is gradually data of as-rolled sheet, argon-annealed and sulfur-annealed
covered by adsorbed sulfur atoms and/or reacts with oxygen sheet (with 6.4 mg S), respectively. The 4.0 mm × 4.0 mm
to form Ga2 O3 (Ga 2p3/2 ; 1118.6 eV) during high temperature area of each scan is from the center region of the 12 mm ×
annealing. 12 mm samples. The ideal {001} h100i texture is located

Figure 7. ODF plots (ϕ2 = 45◦ ) of (Fe0.813 Ga0.187 )99 B1 alloy sheets: (a) as-rolled, (b) annealed in argon at 1200 ◦ C for 6 h and water
quenched, and (c) annealed at 1200 ◦ C for 2 h in a quartz ampoule filled with argon and 6.4 mg S and then water quenched.

5
Smart Mater. Struct. 21 (2012) 055024 S M Na and A B Flatau

Figure 9. Correlation between saturation magnetostriction, peak


concentration of sulfur from AES depth profiles and
surface-energy-induced selective grain growth in annealed
Figure 8. AES depth profiles of sulfur in sheets annealed at (Fe0.813 Ga0.187 )99.5 B0.5 doped with 50 ppm sulfur at 1200 ◦ C. The
1200 ◦ C with various annealing times, where the sputter rate is onset of increased {100} and {111} texture development correlates
2.0 nm min−1 . to concentration levels of sulfur contamination of C{100} and C{111}
of 0.5 and 1.35 at.%, for annealing times of t{100} and t{111} of 0.85
and 2.2 h, respectively.
within the upper center black line, while the acceptable
but less desirable {011} h100i texture appears as contours
concentrated in the lower right corner. Figure 7(a) from the of 4 h, are shifted by 0.5 eV towards higher binding
as-rolled sample shows {111} grains with {111} h110i and energies so that the bond strength of Fe–S appears to be
{111} h112i textures, corresponding to a γ -fiber texture with enhanced. It is considered that diffused sulfur atoms from
h111i parallel to the sheet normal. After annealing with no the interior segregate on the surface and then couple with
sulfur, the predominant texture changed to a texture that is iron atoms during annealing because the bond strength of a
close to a {011} h100i texture as shown in figure 7(b). The Fe–S molecule is three times higher than that of a Fe–Fe
texture shown in figure 7(c) approaches a {001} h100i texture. molecule [27]. These XPS results, while exhibiting trends that
This result is consistent with the formation of {011} h100i and are similar to those of the sulfur-annealed Fe–Ga–B alloy (A
{001} h100i in silicon steel [20, 26]. alloy), provide only an approximation of the relative intensity
The trends in peak magnetostriction measured along of surface segregation of sulfur achieved. In order to more
the rolling direction that are shown in figure 3 correlate to precisely estimate the amount of segregated sulfur on the sheet
the development of a preferred texture under given surface surface, AES depth profiles were carried out for as-rolled
environments. The peak value of 201 ppm is obtained from and annealed samples. Figure 8 shows AES depth profiles
the near {001} h100i texture (figure 7(c)), while a value of of sulfur from the surface to the interior as a function of
160 ppm was obtained from the material that exhibits the sputter time to analyze the segregated atoms, where the sputter
less desirable near {011} h100i texture (figure 7(b)). However, rate is 2 nm min−1 . As the annealing time increases, the
the contour peak intensity of the results in figure 7(c) is peaks of sulfur depth profiles are sharper and closer to the
much lower than what one would obtain from a single crystal surface. A peak in sulfur content of 3.7 at.% was observed
of the same Ga content. This indicates misalignment of the in the 4-h annealed sample. Considering the initial sulfur
developed texture from the rolling direction, which would concentration of 0.005 at.%, the sulfur atoms in the interior
contribute to magnetostriction values being lower than those definitely diffused and segregated onto the free surfaces of the
obtained from single crystal samples. sample. In addition, the peak intensity is greatly diminished
at a sputter time of around 3 min, corresponding to a depth
3.2. Argon-annealed Fe–Ga–B alloy doped with sulfur (B of 6.0 nm.
alloy) Figure 9 shows the correlation between saturation
magnetostriction, peak concentration of sulfur from an AES
Next, we introduce an alternative approach for achieving depth profile (data from figure 8) and surface-energy-induced
surface-energy-induced selective grain growth. The effect selective grain growth. ODF plots for analyzing and
of diffusion and segregation of sulfur atoms from the bulk determining the predominant texture corresponding to each
interior in (Fe0.813 Ga0.187 )99.5 B0.5 doped with 50 ppm sulfur magnetostriction value were reported in previous work [28].
(0.005 at.% S) was examined through annealing at 1200 ◦ C The sulfur concentration in the samples annealed for 1 and 2 h
for dwell times ranging from 0.5 to 4 h under flowing argon. shows a slight increase with time. After ∼2 h, the trend for
The XPS fitted peaks for a sheet annealed at 1200 ◦ C for sulfur concentration increases steeply, appearing to saturate
2 h provides evidence of the presence of stoichiometric after ∼4 h at a concentration of ∼3.5%. Textures with {011}
FeS and FeSn (polysulfide). The fitted peaks, in the case grains were dominant for segregation of sulfur to the surface

6
Smart Mater. Struct. 21 (2012) 055024 S M Na and A B Flatau

Figure 10. Inverse pole figure images along normal direction of the sheet surface for both annealed (Fe0.81 Ga0.19 )99 (NbC)1 samples at
1200 ◦ C for 3 h under 0.5% H2 S in argon, where red, green, and blue colors indicate (001), (011), and (111) grains, respectively.

at concentration levels of below ∼0.5% and an annealing as described section 3.2 of this study. According to these
duration of less than ∼1 h. A maximum magnetostriction results of Fe–Ga–B in sections 3.1 and 3.2, surface energy
of 198 ppm was obtained in the 2 h-annealed sheet which control by sulfur segregation on the sample surface played
had a texture close to a {001} h100i cubic texture, and an important role in the development and growth of {001}
occurred with only a slight increase of sulfur concentration grains, accompanied by an increase of magnetostriction up to
(∼0.5–0.7%). However, an increase in annealing time to ∼200 ppm. Although near {001} h100i a cubic texture was
3 h resulted in a steep increase in the concentration of developed in Fe–Ga–B alloy sheet annealed under a sulfur
surface segregated sulfur, a dramatic texture change to atmosphere, it shows multi-grains with an average size of
{111} h110i and {011} h211i and a steep drop in saturation ∼282 µm and the magnetostrictive value of ∼200 ppm is 58%
magnetostriction, down to 84 ppm. of saturation magnetostriction along the h001i direction in the
In figure 9, estimates can be made of the bounds on single crystal of this composition ((3/2)λ100 = 342 ppm for
annealing time, t{hkl} , and of the concentration of surface 18.7 at.% Ga).
sulfur contamination, C{hkl} , that are associated with the With the goal of reaching the magnetostrictive capability
onset of a {hkl} texture. For development of a {001} texture, of single crystal, the desirable texture needs to be more fully
annealing for t{001} with a surface sulfur contamination developed and sharper along the RD, so as to more closely
concentration of C{001} is required. For development of a resemble the (001) h100i texture found in the single crystal.
{111} texture, an anneal time of t{111} and a surface sulfur According to Kramer’s results, it is not easy to achieve a
contamination level of C{111} is needed. In figure 9, the (001) single-crystal-like texture because the optimum surface
surface concentrations of sulfur, C{001} and C{111} , that seem sulfur contamination condition needed to form a high fraction
to bound selective {001} grain growth are 0.5 and 1.35 at.%, of (001) grains (10–25 ppm) is very sensitive to multiple
respectively. This is also supported by that for the sample parameters [20]. The approach we employed for surface
in figure 6, which was annealed in a sulfur atmosphere for segregation of sulfur involved using H2 S gas in argon during
longer than the time associated with the peak magnetostriction annealing at 1200 ◦ C for 3 h. This annealing schedule was
(indicated in figure 3), the surface concentration of sulfur is used for (011) Goss grain growth in a (Fe0.81 Ga0.19 )99 (NbC)1
greater than 1.5 at.%. sheet annealed under argon. Figure 10 shows that a red (001)
grain was abnormally grown to as much as 74.5% and 88.3%
3.3. Sulfur-annealed Fe–Ga–NbC alloy (C alloy) in the scanned areas from two different samples (the scanned
area is equal to half the surface area of the whole sample,
In our previous work [29] we demonstrated abnormal growth i.e. 6 mm×12 mm). The same time and temperature annealing
of a (011) Goss grain texture in Fe–Ga–NbC alloy sheets protocol was used for both of the samples in figure 10;
annealed at a temperature of 1200 ◦ C for 3 h under an however, the gas flow rate was allowed to vary. Interestingly,
argon atmosphere. The Goss-textured sample exhibited one noticeably different (100) grain growth patterns are evident,
abnormally large (011) grain relative to the size of the suggesting that differences in H2 S flow rate (0–30 sccm) cause
(001) and (111) grains, with grain size ranging from 42 to different patterns of abnormal grain growth (AGG). In order
4200 µm and an average grain size of around 130 µm. to verify this, we are acquiring a capability that will allow for
The observed magnetostriction of 163 ppm is similar to that control of the gas flow rate and content using two mass flow
of the Goss-textured Fe–Ga–B alloy annealed under argon controllers for 0.5% H2 S+Ar and argon gas. The combination

7
Smart Mater. Struct. 21 (2012) 055024 S M Na and A B Flatau

Figure 12. SEM image of the distribution of NbC particles in


grains and at grain boundaries between (111) and matrix (001)
grains, where the particles are interacting with grain boundaries.

3.4. The role of surface energy in (001) cubic grain growth in


ternary Fe–Ga alloys

Both (011) Goss and (001) cubic textures with h100i orienta-
tion are produced by AGG during secondary recrystallization.
The AGG phenomenon in Goss grain-oriented silicon steels
and Fe–3% Si alloys is known to be dependent on the presence
of a precipitation or dispersion of fine second phase particles
such as MnS and AlN, inhibiting normal grain growth (NGG),
i.e. primary recrystallization with continuous growth and
random orientation [30, 31]. The dispersion of NbC particles
as second phase particles has been shown to play a major
role in producing Goss-textured ternary Fe–Ga alloys [29].
However, the presence of these particles is not sufficient
for fully developed AGG over the entire sample surface,
because dispersed NbC particles will ultimately retard grain
boundary movement by acting as a pinning site that acts as
a barrier to movement during grain growth. Figure 12 shows
the distribution of NbC particles in Fe–Ga matrix grains and
at grain boundaries between (001) and (111) grains. When
Figure 11. (a) Pole figure, and (b) observed magnetostriction along a moving boundary encounters a second phase particle, the
the rolling direction and h100i orientation of the annealed particle exerts a restraining force upon the boundary, which
(Fe0.81 Ga0.19 )99 (NbC)1 sample at 1200 ◦ C for 3 h under 0.5% H2 S causes the boundary to be pulled back at the particle location
in argon, respectively, corresponding to the right image in figure 10.
as indicated by arrows in figure 12. The grain boundary is
finally pinned and stabilized by the particles which lower
their interface energy and grain boundary energy. In addition,
of the two controllers will allow for exact control of the partial
the boundaries intersecting the sample surface tend to form
pressure of H2 S to examine the effect of gas flow rate and
grooves. These grooves also impede grain boundary motion.
content on abnormal grain growth.
Thus, an additional driving force is required for the continuous
The growth of (001) grains relative to small (110) and grain boundary movement needed to form highly textured
(111) grains is dominant in the left image, corresponding to and even single-crystal-like rolled sheets. Surface energy
magnetostriction of 193 ppm and possessing a [100] direction differences associated with the misorientation angle between
off from the RD by as much as 17◦ . In the right image, the grains have also been considered as an explanation of the
matrix (001) cubic grain with an area fraction (f001 ) of 88.3% observed extent of single (001) grain as a driver for AGG [17,
is tilted by as much as 24◦ from the RD, calculated from the 20].
(001) pole figure as shown in figure 11(a). In figure 11(b), A summary of observed cubic texture and magne-
the observed magnetostriction value along the RD is 222 ppm tostriction values in ternary Fe–Ga alloys annealed under
and along the [100] orientation it is 314 ppm. This matches various conditions is shown in table 2. Even though the
quite well with an estimate of 309 ppm calculated along segregation process for surface-active elements such as sulfur
the [100] direction based on an 88.3% area fraction of a and oxygen is different, the predominant development of
single crystal of the same composition of 19% Ga (350 ppm {001} h100i cubic texture was observed in ternary Fe–Ga
((3/2)λ100 ) × 0.883(f001 )). alloys where the presence of Fe2 B precipitates, Nb-dissolved

8
Smart Mater. Struct. 21 (2012) 055024 S M Na and A B Flatau

Table 2. Observed texture and magnetostriction for various annealing conditions in ternary Fe–Ga alloys.
Max. contour
Alloy Annealing Annealing Predominant intensity of Magnetostriction
composition condition atm. texture PF|ODF (λk − λ⊥ )
(Fe0.83 Ga0.17 )99 B1 a 1100 ◦ C for 5 h O2 (40 ppm) in argon {001} h100i — 10.00 206 ppm
(Fe0.813 Ga10.187 )99 B1 1200 ◦ C for 2 h Elemental S in argon Near {001} h100i 4.34 12.49 201 ppm
(Fe0.813 Ga0.187 )99.5 B0.5 1200 ◦ C for 2 h Argon Near {001} h100i 4.74 10.20 198 ppm
doped with 50 ppm S
(Fe0.81 Ga0.19 )99 Nb1 b 1200 ◦ C for 5 h Argon {001} h100i — 6.00 134 ppm
(Fe0.85 Ga0.15 )99 (NbC)1 1300 ◦ C for 2 h
c
Argon Near {001} h100i 3.15 — N/A
(Fe0.85 Ga0.15 )99 (NbC)1 c 1300 ◦ C for 24 h Argon {001} h100i 3.97 — N/A
(Fe0.81 Ga0.19 )99 (NbC)1 1200 ◦ C for 3 h 0.5% H2 S in argon (001) h210i matrix 32.18 158.3 222 ppm, k RD
grain 314 ppm, k h100i
Single crystal As-grown — (001) h100i 39.90 203.6 350 ppm
Fe81.2 Ga18.8
a Li et al [9].
b Li et al [10].
c Srisukhumbowornchai and Guruswamy [16].

precipitates and NbC particles which exist in the matrix of and shown to correlate to changes in sample texture under
starting materials acts as a primary grain growth inhibitor. high temperature anneals. A theory is postulated to explain
Cubic-textured Fe–Ga–B alloys, which are independent the observed results in which changes in surface energy due to
of surface-active elements and segregation mechanisms, slight sulfur segregation correlate to the preferential formation
exhibited magnetostriction values of around 200 ppm and the of {001} cubic grain growth. Three separate approaches to
almost same values for maximum intensity of PF and ODF producing surface segregation of sulfur yielded similar results.
images. In the case of Nb-added alloy, the observed properties In one approach, (Fe0.813 Ga0.187 )99 B1 alloy was annealed in a
(e.g. magnetostriction, sharpness of texture) are lower than in sulfur atmosphere using elemental sulfur to be vaporized and
the other cases, although it has cubic texture. Of interest, the adsorbed. In the second approach, (Fe0.813 Ga0.187 )99.5 B0.5
(Fe81 Ga19 )99 (NbC)1 alloy shows a large area fraction of AGG doped with 50 ppm sulfur was annealed in a pure argon
with (001) plane, accompanied by a high magnetostriction of atmosphere and thereby sulfur atoms diffused from the bulk
314 ppm along the [100] orientation. Compared to the results interior to the surface. Hydrogen sulfide (0.5% H2 S) gas
for a single crystal, 90% of magnetostriction was achieved in argon was used for sulfur decomposition and adsorption
and the maximum contour intensities are close to those of during annealing of a (Fe0.81 Ga0.19 )99 (NbC)1 rolled sheet as
a single crystal. From these results and EBSD analysis, the a third approach.
magnetostriction strongly depends on the coverage of (001) These sets of samples were studied by surface chemistry
grains and the deviation angle of preferred orientation from analysis (AES and XPS), texture analysis (EBSD) and
h100i orientation, which are proportional to the observed magnetostriction measurement. The results showed that sulfur
contour intensity [32]. was adsorbed (diffused) from a sulfur atmosphere (the bulk
In conclusion, recrystallized (011) Goss grains in ternary interior) then segregated at the surface. The segregated
Fe–Ga alloys may be replaced by (001) cubic grains, after sulfur was preferentially combined with iron and iron sulfide
which AGG of (001) grains will be facilitated if a difference (FeS) formed at the surface. Although the mechanism of
in surface energy, γ , provides a driving force for the AGG of sulfur segregation is different for these three cases, sulfur
adsorption in two cases and sulfur diffusion in the other,
(001) grains. This will occur when the surface energy of (001)
the surface-energy-induced selective grain growth with a
surface planes is lower than that of (011), i.e. γ011 > γ001 .
preference for a specific plane/texture is very similar.
These results demonstrate that variations in the segregation
Observations indicate that {011} grains were dominant
of active elements on the surface of a sample influences
under very clean surface conditions, corresponding to a low
surface energy. Furthermore, under a very clean surface
sulfur atmosphere (limited adsorption) or short annealing
condition γ001 > γ011 , AGG of (011) grains arises because the
times (limited diffusion). The {001} grains occurred under
closed-packed (011) plane is the plane with the lowest energy
‘slight’ surface contamination with sulfur, at concentration
in body-centered cubic (bcc) structures. The qualitative
levels of between ∼0.5 and ∼1.35 at.%, due to adsorption
surface chemistry analysis of Fe–Ga–B doped sulfur indicates or diffusion of sulfur to the sample surface, then maximum
a range of 0.5–1.35 at.% for surface segregation of sulfur magnetostriction values of approximately 200 ppm were
correlated to selective {001} grain growth. obtained in the both Fe–Ga–B alloys. Moreover, abnormal
growth of a (001) grain that covered up to 88.3% of the sample
4. Conclusions area was developed and promoted in Fe–Ga–NbC alloy which
displayed a magnetostriction of 314 ppm, approaching that of
The effect of sulfur contamination on the surface of single crystal Fe–Ga alloy of similar Ga composition, along
polycrystalline ternary Fe–Ga-based alloys has been studied the [100] direction.

9
Smart Mater. Struct. 21 (2012) 055024 S M Na and A B Flatau

Finally, these results offer both a theory for the [14] Restorff J B, Wun-Fogle M, Clark A E and
mechanism of how texture is influenced and specific Hathaway K B 2006 Induced magnetic anisotropy in
guidance for the development of procedures for maximizing stress-annealed galfenol alloys IEEE Trans. Magn.
42 3087–9
magnetostriction in textured polycrystalline alloys through [15] Summers E, Meloy R and Restorff J B 2009 galfenol alloying
surface-energy-induced selective grain growth caused by additions and the effects on uniaxial anisotropy generation
surface segregation of sulfur. J. Appl. Phys. 106 024914
[16] Srisukhumbowornchai N and Guruswamy S 2004
Crystallographic textures in rolled and annealed Fe–Ga and
Acknowledgments Fe–Al alloys Metall. Mater. Trans. A 35A 2963–70
[17] Köhler D J 1960 Promotion of cubic grain growth in 3%
This work was supported by ONR MURI silicon iron by control of annealing atmosphere
grant N000140610530. The authors acknowledge Miji Seo at composition J. Appl. Phys. 31 408–409S
Sungkyunkwan University in Korea, for XPS operation and [18] Foster K and Kramer J J 1960 Effect of directional orientation
on the magnetic properties of cube-oriented magnetic
analysis. sheets J. Appl. Phys. 31 233–234S
[19] Dunn C G and Walter J L 1962 Surface energies and other
References surface effects relating to secondary recrystallization
textures in high-purity iron, zone-refined iron, and 0.6 pct
[1] Zhao P, Zhao Z, Hunter D, Suchoski R, Gao C, Mathews S, Si–Fe Trans. TMS-AIME 224 518–32
Wuttig M and Takeuchi I 2009 Fabrication and [20] Kramer J J 1992 Nucleation and growth effects in thin
characterization of all-thin-film magnetoelectric sensors ferromagnetic sheets: a review focusing on
Appl. Phys. Lett. 94 243507 surface-energy-induced secondary recrystallization Metall.
[2] Atulasimha J and Flatau A B 2008 Experimental actuation and Trans. A 23 1987–98
sensing behavior of single-crystal iron–gallium alloys [21] Heo N H, Chai K H and Na J G 2000 Correlation between
J. Intell. Mater. Syst. Struct. 19 1371–81 interfacial segregation and surface-energy-induced selective
[3] Ueno T, Summers E, Wun-Fogle M and Higuchi T 2008 grain growth in 3% silicon–iron alloy Acta Mater.
Micro-magnetostrictive vibrator using iron–gallium alloy 48 2901–10
Sensors Actuators A 148 280–4 [22] Chai K H, Heo N H, Na J G and Lee S L 2000 Tertiary
[4] Ueno T and Yamada S 2011 Performance of energy harvester recrystallization and magnetic induction under various
using iron–gallium alloy in free vibration IEEE Trans. annealing atmospheres in thin-gauged 3% Si–Fe strip
Magn. 47 2407–9 J. Appl. Phys. 87 5233–5
[5] Clark A E, Restorff J B, Wun-Fogle M, Lograsso T A and [23] Arabczyk W, Baumann T, Müssig H-J and Storbeck F 1990
Schlagel D L 2000 Magnetostrictive properties of Interaction between sulphur and oxygen on Fe(111)
body-centered cubic of Fe–Ga and Fe–Ga–Al alloys IEEE Vacuum 41 79–81
Trans. Magn. 36 3238–40 [24] Nesbitt H W, Bancroft G M, Pratt A R and Scaini M J 1998
[6] Clark A E, Wun-Fogle M, Restorff J B, Lograsso T A and Am. Mineral. 83 1067–76
Cullin J R 2001 Effect of quenching on the [25] Dehlinger A S, Pierson J F, Roman A and Bauer Ph 2003
magnetostriction of Fe1−x Gax (0.13 < x < 0.21) IEEE Properties of iron boride films prepared by magnetron
Trans. Magn. 37 2678–80 sputtering Surf. Coat. Technol. 174/175 331–7
[7] Clark A E, Hathaway K B, Wun-Fogle M, Restorff J B, [26] Heo N H, Kim S B, Choi Y S, Cho S S and Chai K H 2003
Lograsso T A, Keppens V M, Petculescu G and Interfacial segregation, nucleation and texture development
Taylor R A 2003 Extraordinary magnetoelasticity lattice in 3% silicon steel Acta Mater. 51 4953–64
softening in bcc Fe–Ga alloys J. Appl. Phys. 93 8621–3 [27] Lin S-S and Kant A 1969 Dissociation energy of Fe2 J. Phys.
[8] Na S M and Flatau A B 2008 Deformation behavior and Chem. 73 2450–1
magnetostriction of poly crystalline Fe–Ga–X (X = B, C, [28] Na S M and Flatau A B 2007 Secondary recrystallization,
Mn, Mo, Nb, NbC) alloys J. Appl. Phys. 101 07D304 crystallographic texture and magnetostriction in rolled
[9] Li J H, Gao X X, Zhu J, Bao X Q, Xia T and Zhang M C 2010 Fe–Ga basedalloys J. Appl. Phys. 101 09N518
Ductility, texture and large magnetostriction of [29] Na S M, Yoo J H and Flatau A B 2009 Abnormal (110) grain
Fe–Ga-based sheets Scr. Mater. 63 246–9 growthand magnetostriction n recrystallized galfenol with
[10] Li J H, Gao X X, Zhu J, He C, Qiao J and Zhang M 2009 dispersed niobiumcarbide IEEE Trans. Magn. 45 4132–5
Texture evolution and magnetostriction in rolled [30] Sakakura A 1969 Effect of AlN on the primary
(Fe81 Ga19 )99 Nb1 J. Alloys Compounds 93 8621–3 recrystallization textures in cold-rolled-(110)[001]-oriented
[11] Cullity B D 1972 Introduction of Magnetic Materials single crystals of 3% silicon iron J. Appl. Phys. 40 1534–8
(Reading, MA: Addison-Wesley) [31] Maazi N and Penelle R 2009 Introduction of preferential
[12] Yoo J H, Restorff J B, Wun-Fogle M and Flatau A B 2008 The Zener drag effect in Monte Carlo simulation of abnormal
effect of magnetic field annealing on single crystal iron Goss grain growth in the Fe–3% Si magnetic alloys Mater.
gallium alloy J. Appl. Phys. 103 07B325 Sci. Eng. A 504 135–40
[13] Yoo J H, Na S M, Restorff J B, Wun-Fogle M and [32] Na S M and Flatau A B 2012 Single grain growth and large
Flatau A B 2009 The effect of field annealing on highly magnetostriction in secondarily recrystallizated Fe–Ga thin
textured polycrystalline galfenol strips IEEE Trans. Magn. sheet with sharp Goss (011) [100] orientation Scr. Mater.
45 4145–8 66 307–10

10

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy