SAMPE Journal - July/August 2017 - 41

Article
(Shown in Figure 1c) from original
SEM-EBSD data sets (with Inverse
Pole Figure (IPF) key parallel to
Z, build direction) during the
solidification of Ti6Al4V EBM
samples. In general two different
types of primary b-grain structures
can be seen namely (a) a complex
skin and (ii) a core, primary b-grain
structure with its inner and outer
skin layers labelled as shown in
Figure 2a-b. The detailed description
of evolution of microstructure
during AM of Ti6Al4V using EBM
is presented by Antonysamy et
al. in3, 7-9. There were five different
primary b-grain structures evolved
during rapid solidification of the
EBM Ti6Al4V samples from outer
surface towards core of the samples
namely: (1) fine grains within a thin
layer of partially melted powder
attached to the wall surface with
a random orientation (since it was
heterogeneously nucleated from the
randomly oriented partially melted
powders from the bed of adjacent
wall), from which (2) fine inwardgrowing curved grains develop as
far as the centre line of the contour
pass. The columnar grains are curved
and growing along Z direction since
the grain growth happens always
perpendicular to the curved Solid/
Liquid interface of the melt pools and
against maximum thermal gradient
in Z (See the curved melt pool
interface as shown in (Figure 2c);
(3) some of these curved grains are
favourably orientated and continue
to grow long distances up to the
wall at the contour pass centrelineproducing a long 'axial <001>II Z
b-grains(red)'; (4) this is followed
by regular lath or half-herringbone
morphology
columnar
grains
growing upwards in the inner half of
the contour pass (appear with a blue
tint grains and are titled away from
<001>b // Z towards <111>b // Z) and,
finally (5) vertically-growing coarse
irregular and epitaxial columnar
grains are seen within the centre of
thick sections where cross hatching
is employed due to high cooling rate
of the order of 105 K/s3. Solidification
SAMPE Journal, Volume 53, No. 4, July/August 2017

of all these five types of grains is
shown schematically in Figure 2c.
The residual stresses measured on
both direction of the sample in XY
and XZ planes by d(Sin^2 y) method
were summarised in Table 1. In order
to measure the homogeneity or
anisotropy of the materials, residual
stresses have been measured in both
Phi (y) =0° and, Phi (y) =90°. The
residual stress has been measured
at three different locations in this
study namely; (1) Multiple exposure
at a centre point (both Phi0 and
Phi90) i.e., stresses are measured at
the single point more than 10 times,
(2) Square map of 10x10 mm2 at one
area (Phi0 and Phi90); and (3) Square
map of 10x10 mm2 at another area
(Phi0 and Phi90) i.e., in both (2) &
(3) stress data's are measured at two
different area of 10 x 10 mm2 for the
better understanding of materials
homogeneity.
The amount of residual stresses

measured in Ti6Al4V EBM built
samples from top surface to interior
of sample is shown in Figure 3 (in
XY plane) for up to ~ 0.5 mm at an
interval of approx. 100 mm. It has
been expected that the preheating of
powder bed to 740°C in EBM would
result in components being free
from any residual stresses in it since
the general residual stresses relief
temperature for Ti6Al4V is 690°C
for 1-2 hours10. However, it can be
seen that the amount of residual
stresses present on the top layer of
the as-deposited sample (without
machining) varies from -175.9 MPa
to -192.13 MPa for Phi = 0° and,
from -69.9 MPa to -90.86 MPa for
Phi = 90°, respectively. However, the
subsequent sub-surfaces measured
from 110 mm to 403 mm depths has
showed a variations of residual
stresses from -17.5 MPa to +29.7 MPa
for Phi=0° and, from -17.16 MPa to
+26.8 MPa for Phi=90°, respectively,

41
http://www.expanded-materials.com
Table of Contents for the Digital Edition of SAMPE Journal - July/August 2017
