University of Birmingham
Optimization of SLM Process Parameters for
Ti6Al4V Medical Implants
El-Sayed, Mahmoud; Ghazy, Mootaz; Yehia, Youssef; Essa, Khamis
DOI:
10.1108/RPJ-05-2018-0112
License:
Other (please specify with Rights Statement)
Document Version
Peer reviewed version
Citation for published version (Harvard):
El-Sayed, M, Ghazy, M, Yehia, Y & Essa, K 2018, 'Optimization of SLM Process Parameters for Ti6Al4V
Medical Implants', Rapid Prototyping Journal. https://doi.org/10.1108/RPJ-05-2018-0112
Link to publication on Research at Birmingham portal
Publisher Rights Statement:
This is the Accepted Author's Manuscript for the following article: Mahmoud Elsayed, Mootaz Ghazy, Yehia Youssef, Khamis Essa, (2018)
"Optimization of SLM process parameters for Ti6Al4V medical implants", Rapid Prototyping Journal, https://doi.org/10.1108/RPJ-05-2018-
0112
General rights
Unless a licence is specified above, all rights (including copyright and moral rights) in this document are retained by the authors and/or the
copyright holders. The express permission of the copyright holder must be obtained for any use of this material other than for purposes
permitted by law.
•Users may freely distribute the URL that is used to identify this publication.
•Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private
study or non-commercial research.
•User may use extracts from the document in line with the concept of ‘fair dealing’ under the Copyright, Designs and Patents Act 1988 (?)
•Users may not further distribute the material nor use it for the purposes of commercial gain.
Where a licence is displayed above, please note the terms and conditions of the licence govern your use of this document.
When citing, please reference the published version.
Take down policy
While the University of Birmingham exercises care and attention in making items available there are rare occasions when an item has been
uploaded in error or has been deemed to be commercially or otherwise sensitive.
If you believe that this is the case for this document, please contact UBIRA@lists.bham.ac.uk providing details and we will remove access to
the work immediately and investigate.
Download date: 10. Aug. 2022
Optimization of SLM Process Parameters for Ti6Al4V Medical Implants
Abstract
Ti6Al4V alloy has received a great deal of attention in medical applications due to its
biomechanical compatibility. However, the human bone stiffness is between 10 and 30 GPa
while solid Ti6Al4V is significantly stiffer, which would cause stress shielding with the
surrounding bone which can lead to implant and/or the surrounding bone’s failure. In this
work, the effect of SLM process parameters on the characteristics of Ti6Al4V samples, such
as porosity level, surface roughness, elastic modulus and compressive strength (UCS), has
been investigated using Response Surface Method (RSM). The examined ranges of process
parameters were 35-50 W for laser power, 100-400 mm/s for scan speed and 35-120 µm for
hatch spacing. The results showed that the porosity % of a SLM component could be
increased by reducing the laser power and/or increasing the scan speed and hatch spacing. It
was also shown that there was a reverse relationship between the porosity level and both the
modulus of elasticity and UCS of the SLM part. In addition, the increased laser power
resulted in a substantial decrease of the surface roughness of SLM parts. The process
parameters have been optimized to obtain structures with properties very close to that in
human bones. Results from the optimization study revealed that the interaction between laser
process parameters (i.e. laser power, laser speed, and the laser spacing) have the most
significant influence on the mechanical properties of fabricated samples. The optimized
values for the manufacturing of medical implants were 49 W, 400 mm/s and 99 m for the
laser power, laser speed and laser spacing, respectively. The corresponding porosity, surface
roughness, modulus of elasticity and UCS were 23.62%, 8.68 µm, 30 GPa and 522 MPa,
respectively.
Keywords: Selective laser melting (SLM); Design of Experiment; Ti-6Al-4V; Medical
Implants
1. Introduction
Selective laser melting (SLM) is an additive manufacturing technique that produces near
fully dense metal parts directly from a CAD design by adding layer upon layer [1-4]. The
main concept is based on a laser beam that passes over a thin layer of powder and diffuses it
selectively to the desired shape. Next, a new layer of powder is spread, the platform is
lowered according to the required layer thickness and then the melting process is repeated
until the full part is obtained [5,6]. SLM has many advantages such as producing complex
shapes that are difficult to fabricate via conventional methods, short time from design to
market, and near net shape production which minimizes waste of materials [7,8]. For these
reasons, the SLM process is used in aerospace and biomedical applications such as implants
and prostheses [9,10]. Examples of metal powder used in SLM processes are: titanium alloys,
steels, cobalt, chromium and aluminum alloys [11]. On the other hand, SLM has some
limitations that include the stair step effect which increases surface roughness, and balling
phenomenon which increases both the surface roughness and the porosity of SLM parts [12].
T
h
and m
o
affect
s
defect
s
Ψ, wh
i
Wher
e
thickn
e
sampl
e
identi
f
[15].
O
desig
n
Analy
s
Comp
o
desig
n
know
n
param
e
varied
Fig 1
proces
qualit
y
A
t
micro
s
contro
l
manu
f
Ti6Al
4
b
ecau
s
high c
Furth
e
medic
a
S
o
charac
b
een
s
h
e quality o
orphology
o
s
the degre
e
s
. One of th
i
ch could b
e
e
P is the l
a
e
ss. Many
e
s with the
f
ying an op
O
n the oth
e
n
of experi
m
s
is of Vari
a
o
site Desig
n
n
is a com
b
n
as axial p
o
e
ter called
over 5 lev
e
below. Th
e
s
s paramete
r
y
and poros
i
t
tar et al. [1
s
tructure a
n
lled manne
r
f
acturing co
m
4
V alloy i
s
s
e of its bio
c
orrosion re
s
e
rmore, stre
n
a
l applicati
o
o
ng et al.
c
teristics of
s
uccessfull
y
f the SLM
f
o
f the pow
d
e
of consoli
e approach
e
e
expresse
d
a
ser power,
researcher
s
heat input,
timum ene
r
e
r hand sev
e
m
ents (Do
E
a
nce (AN
O
n
(CCD). I
n
b
ination of
o
ints) and c
e
α. For Cen
t
e
ls (- α, -1,
0
e
se techniq
u
r
s such as l
a
i
ty content
i
Fig 1. C
e
9] stated th
a
n
d mechani
c
r
. This ma
k
m
plex sha
p
s
among th
e
c
ompatible
s
istance an
d
n
gth, stiffn
e
o
ns.
[23] hav
e
Ti6Al4V
S
y
manufact
u
f
abricated
p
d
er used.
A
dation of t
h
e
s to repres
e
according
t
v is the s
c
s
[14] appl
i
but with a
c
r
gy densit
y
e
ral studie
s
E
) techniqu
e
O
VA). One
n
this desig
n
two-level
f
e
ntre point
s
t
ral Comp
o
0
, 1 and α) [
u
es were s
u
a
ser power,
i
n Selectiv
e
ntral comp
o
a
t titanium
a
c
al propert
i
k
es the tech
n
p
es of funct
i
e
most co
m
nature [20
-
d
relativel
y
e
ss, corrosi
o
e
studied
t
S
LM parts.
u
red by sel
e
p
arts depen
d
A
nother im
p
h
e powder
p
e
nt the lase
r
t
o equation
c
an speed,
h
ied this fu
n
c
ommon ai
m
y
level corr
e
s
suggested
e
s such as
of the mo
s
n
the numbe
f
actorial (k
n
s
. The axial
o
site Desig
n
16]. Desig
n
uccessfull
y
scan speed
e
Laser Sint
o
site desig
n
a
lloys are v
i
es of part
s
n
ology suit
a
i
onal impla
n
m
monly us
e
-
22]. It has
l
y
low Youn
g
on behavio
r
t
he effect
They repo
r
e
ctive laser
d
s upon ma
n
p
ortant fact
o
p
articles as
r
heat input
1 as follo
w
h
is the ha
t
n
ction to c
m
to fabric
a
e
sponding
t
the use st
a
the Respo
n
s
t favourite
r of factors
n
own as c
u
points are
c
n
, α is larg
e
n
s for k = 2
a
y
applied t
o
and scan s
p
ering (SLS
)
n
s for k = 2
ery compat
i
produced
a
ble for bio
m
n
ts from bi
o
e
d Ti mate
r
l
ow densit
y
g
’s modul
u
r
and proc
e
of the pr
o
r
ted that f
u
melting u
s
n
y paramet
e
o
r is the la
s
well as th
e
is the ener
g
w
s [13]:
(1)
t
ch spacing
orrelate th
e
a
te fully so
l
t
o minimu
m
a
tistical an
a
n
se Surfac
e
RSM desi
g
examined i
s
u
be points)
,
c
ontrolled t
h
e
r than one
a
nd k = 3 f
a
o
investigat
e
p
acing on t
h
)
and SLM
p
a
nd k = 3 [
1
i
ble with S
L
via SLM c
m
edical ap
p
o
compatibl
e
r
ials for i
m
y
, good mec
h
u
s of appro
x
e
ss accurac
y
o
cessing p
a
lly-solid T
i
ing the fol
l
e
rs such as
t
s
er heat in
p
e
formatio
n
g
y density
fu
and t is t
h
e
density
o
l
id compo
n
m
porosity
a
lysis by m
e
Method,
a
gns is the
s noted as "
,
face poin
t
h
rough a st
a
and each
fa
a
ctors are s
h
e
the influ
e
h
e resulting
p
rocesses [
1
6].
L
M techniq
u
c
an be gra
d
p
lications i
n
e
metals is
m
plant appl
i
hanical pro
x
imately 1
1
y
were suit
a
arameters
i
6Al4V pa
r
l
owing par
a
t
he size
p
ut as it
of any
fu
nction
h
e layer
o
f SLM
n
ents by
content
e
ans of
a
nd the
Central
k
". The
t
s (also
a
tistical
fa
ctor is
h
own in
e
nce of
surface
17,18].
u
e. The
ed in a
n
which
c
rucial.
i
cations
perties,
0 GPa.
a
ble for
on the
r
ts have
a
meters
(laser power = 110 W, scan speed = 400 mm/s, scan spacing = 40 µm, and layer thickness =
50 µm). Sun et al. [24] used the Taguchi method to optimize four process parameters: layer
thickness, linear energy density, hatch spacing and scanning strategy. They reported that 80
W laser power, 200 mm/s scan speed, 60 μm hatch spacing, a 20 μm layer thickness and X-Y
inter-layer for scanning strategy was sufficient to achieve fully dense, good quality Ti6Al4V
components. In another study Murr et al. [25] have produced Ti6Al4V parts via SLM for
biomedical implants. It was indicated that SLM was capable of producing good quality parts
with mechanical properties better than wrought and cast Ti6Al4V parts. Vandenbroucke and
Kruth [26] also produced medical and dental parts fromTi6Al4V alloy and tested their
mechanical and chemical properties. The Ti6Al4V produced had achieved 99.98 % density.
However, it should be noted that in the earlier studies such as those by Murr [25] and
Vandenbroucke and Kruth [26], the objective was mainly to produce SLM parts with
minimum porosity in order to achieve mechanical properties that could reach, or even
exceed, those of bulk material. In the work reported by Vandenbroucke and Kruth [26], a
tensile young's modulus of about 94 GPa was obtained. Nevertheless, the elastic modulus of
bones in human body ranges from 10 to 30 GPa. The large difference in moduli between
titanium implants and bones, known as stiffness mismatch, can result in stress shielding,
which has been held responsible for implant loosening and consequently could cause the
patients to require a revision surgery. Two solutions were found to this problem: the first one
was developing new types of titanium alloys that have modulus closer to bones and the
second one was developing porous structure instead of solid structures which reduces
material modulus [27-30].Titanium alloys that have 30% volume porosity can have modulus
similar to human bones. One problem of porous structures is that it decreases toughness and
creates stress concentration around the pores [31].
Furthermore, a medical implant should have high compressive strength to prevent
fractures and improve functional stability. High strength is also required to impede
spring-back both during and after the operation procedure [32,33]. Finally, an implant should
have sufficient surface roughness to improve the ingrowth of the human tissues into it.
Compared to smooth surfaces, textured implants surfaces exhibit more surface area for
integrating with bone via osseointegration process. It was suggested that a surface roughness
in the range from 1 to 10 microns would be required to enhance both the osteoconduction
(in-migration of new bone), and osteoinduction (new bone differentiation) processes [34-36].
Previous investigations related to additive manufacturing of Ti alloys have focused on
producing fully dense and high integrity structures. There is a clear gap in literature regarding
the simultaneous enhancement and adjustment of pore fraction, surface and mechanical
properties of Ti6Al4V SLM components towards biomedical implants. In the present work,
artificial pores have been created in Ti6Al4V parts fabricated via SLM by controlling the
process parameters to achieve surface and mechanical properties suitable for biomedical
applications. The influence of processing parameters by means of laser power, scan speed
and hatch spacing on the surface roughness, porosity content and mechanical properties of
Ti6Al4V components produced by SLM will be investigated. Statistical analysis by means of
Design of Experiments (DoE) and Analysis of Variance (ANOVA) will be adopted to
optimise the SLM process parameters and fabricate custom parts with elastic modulus, UCS
and surface roughness sufficiently close to that of human bones.
2. Experimental Methods
2.1 Materials
Ti6Al4V gas atomized alloy powder was supplied by LPW Technology. Most of the
powder particles had a size range between 19-45m as measured using a laser diffraction
analyzer (Microtrac) following the ASTM B822 standard. The size distribution of powder
used is shown in Table 1.
Table 1. Ti6Al4V powder size distribution
Particle size
(m)
<16 16-22 22-31 31-44 >45
Percentage
(%)
5 10 28 46 11
2.2. Statistical design of experiment (DoE) using response surface
In this study the design of experiment RSM was carried out to generate an experimental
plan with minimum possible trials. ANOVA was utilized to find a relationship between the
input and output parameters, identify the most significant parameters, and find the optimal
setting of those parameters that can achieve the intended objective function. The response
surface “Y” can be expressed by a second order polynomial (regression) equation as shown
in Equation 2:
Yb
∑
b
x
∑
b
x
∑
b
x
x
(2)
where x
i
are the factors input parameters. The terms b
0,
b
i
, b
ii
, and b
ij
are the model
coefficients that depend on the main and interaction effects of the process parameters.
Method of least squares is used to determine the constant coefficients. To perform the design
of experiment, Design-Expert Software Version 7.0.0 (Stat-Ease Inc., Minneapolis, USA)
was used.
The procedure adopted in this study was as the following:
1. Identification of the key process parameters, and setting the upper and lower bound for
each.
2. Selection of the output response.
3. Developing the experimental design matrix.
4. Carrying out the experiments according to the design matrix, and recording the output
response.
5. Developing a mathematical model to correlate the process parameters to the output
response.
6. Optimizing that model using genetic algorithm.
In the current study three factors (process parameters) were considered which are the laser
power, scan speed and hatch spacing. According to the central composite design, and as
described above, each parameter was varied over 5 levels (-α, -1, 0, 1 and α). See Fig 1. In
this work α was considered to be 2 in order to change each factor over five equal levels. Table
2 shows the levels of each factor in this investigation. As shown -α and α represent the
minimum and maximum levels respectively, of each factor. Also, three center points (at the 0
level (middle) of all factors, see Fig1) were considered. The center points are used to provide
