J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S 5 (2012) 16–31

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/jmbbm

Research paper

Porous titanium materials with entangled wire structure for

load-bearing biomedical applications

Guo He ∗ , Ping Liu, Qingbiao Tan

State Key Lab of Metal Matrix Composites, School of Materials Science and Engineering, Shanghai Jiao Tong University,
Shanghai 200240, China

A R T I C L E I N F O A B S T R A C T

Article history: A kind of porous metal-entangled titanium wire material has been investigated in terms
Received 20 July 2011 of the pore structure (size and distribution), the strength, the elastic modulus, and the
Received in revised form mechanical behavior under uniaxial tensile loading. Its functions and potentials for surgical
30 September 2011 application have been explained. In particular, its advantages over competitors (e.g.,
Accepted 30 September 2011 conventional porous titanium) have been reviewed. In the study, a group of entangled
Published online 12 October 2011 titanium wire materials with non-woven structure were fabricated by using 12–180 MPa
forming pressure, which have porosity in a range of 48%–82%. The pores in the materials
Keywords: are irregular in shape, which have a nearly half-normal distribution in size range. The yield
Entangled titanium wire materials strength, ultimate tensile strength, and elastic modulus are 75 MPa, 108 MPa, and 1.05 GPa,
Porous titanium respectively, when its porosity is 44.7%. The mechanical properties decrease significantly as
Biomaterials the porosity increases. When the porosity is 57.9%, these values become 24 MPa, 47.5 MPa,
Implants and 0.33 GPa, respectively. The low elastic modulus is due to the structural flexibility of the
Elastic modulus entangled titanium wire materials. For practical reference, a group of detailed data of the
porous structure and the mechanical properties are reported. This kind of material is very
promising for implant applications because of their very good toughness, perfect flexibility,
high strength, adequate elastic modulus, and low cost.
⃝c 2011 Elsevier Ltd. All rights reserved.

1. Introduction fect. These advantages have facilitated the practical clini-

Metallic biomaterials are preferential choices in load-bearing
applications in the field of bone tissue engineering and re-
generative medicine due to their outstanding mechanical
strength and good toughness. Among them, titanium and its
alloys exhibit excellent biocompatibility, good corrosion re-
sistance, and relatively lower stiffness (compared with other
metallic biomaterials such as cobalt-based alloys, stainless
steel, etc.), which are beneficial to the durability of the load-
bearing implants and reduction of the stress-shielding ef-
cal applications of titanium and its alloys, such as artificial
titanium alloy joint (Long and Rack, 1998), dental titanium
implant (Bréme et al., 1993), coronary titanium–nickel stent
(Eigler et al., 1993), etc. In the past decades, attempts were
frequently made to seek optimal, mechanical and biocompat-
ible properties for titanium implants (Niinomi, 2008; Geetha
et al., 2009), so as to remedy the mismatch of the elastic mod-
uli (between titanium implant and cortical/cancellous bone)
and the bioinert properties of the titanium implant surface.
The former can lead to stress-shielding over the surround-

∗ Corresponding author. Tel.: +86 21 3420 2643; fax: +86 21 3420 2643.
E-mail address: ghe@sjtu.edu.cn (G. He).
c 2011 Elsevier Ltd. All rights reserved.
1751-6161/$ - see front matter ⃝
doi:10.1016/j.jmbbm.2011.09.016
 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S
ing bone, which results in resorption of the bone and asep-
tic loosening of the implant (Jacobs et al., 1993); the latter
may cause poor bone–implant interfacial bonding, because
the bulk titanium surface develops interfacial fibrous tissue
to form encapsulation that isolates the implants from their
surroundings (Chang et al., 1996). In order to solve these
problems, alloying is a metallurgical method to reduce the
elastic modulus via forming full beta structure. So far the
elastic modulus as low as thirty GPa has been achieved in
some beta titanium alloys (Geetha et al., 2009; Laheurte et al.,
2010). But compared with cortical bone (4.4–28.8 GPa) and can-
cellous bone (0.01–3.0 GPa) (Geetha et al., 2009), the beta ti-
tanium alloy is still stiffer. It seems that only a very limited
room for reduction of the elastic modulus of the bulk titanium
alloy remains. The surface modifications on titanium im-
plants have previously demonstrated that bioinert titanium
metal could be converted into bioactive material through
specific chemical and thermal treatments (Fujibayashi et al.,
2001; Kim et al., 2000; Nishiguchi et al., 1999), by which the
smooth titanium surface was processed into a specific porous
structure (Fujibayashi et al., 2004; Kim et al., 2000). These pre-
vious works implied the superiority of the porous titanium
materials as implants. In fact, porous biomaterials have re-
ceived a lot of attention in past ten years. The porous ti-
tanium can provide adequate macro/micro-pores for bone
ingrowth, vascularization, and flow transport of nutrients and
metabolic waste (Manukyana et al., 2010), which is supposed
to be osteoconductive, osteoinductive and capable of osseoin-
tegration (Albrektsson and Johansson, 2001; Stevens, 2008).
It is well accepted so far that the titanium material
with porous structure, which can imitate cancellous bone in
structure and properties, facilitates proliferation of cells into
the structure and provides space for bio-factor (cell, gene
and/or protein) delivery (Hollister, 2005). Such bone tissue
ingrowth through the pores forms stable long-term anchorage
for biological fixation of the implant (Garrett et al., 2006;
Krishna et al., 2007). It is important that the porous network
must interconnect and the appropriate pore size for in vivo
bone ingrowth should be in the range of 100–500 µm (Bungo
et al., 2006; Freyman et al., 2001; Garrett et al., 2006; Holy
et al., 2000; Li et al., 2007). Some investigations validated
that the porous titanium implants could be improved in
osteoconductive properties as porosity increased, and the
amount of new bone growth obviously increased as the
pore size increased (Bungo et al., 2006; Li et al., 2007).
Another benefit of porous titanium is that the effective elastic
modulus can be tuned to match the modulus of bone to
reduce the problems associated with stress-shielding. The
mechanical properties of porous titanium are dependent on
porosity, porous morphology and pore size distribution as
these determine the spatial shape and size of the struts or
cell-walls (for honeycombed structure) between the pores
which are bearing the load. The nature of the struts or
cell-walls together with the holistic architecture constructed
by the struts is crucial for the macro-mechanical behaviors
of the bulk porous titanium. All these key macroscopic
parameters are controlled by the fabrication methods of the
porous titanium materials.
There are many techniques to produce porous tita-
nium, and various methods may lead to different porosities
and morphologies. The earliest metal foaming techniques
5 (2012) 16–31 17
(Banhart, 2001; Körner and Singer, 2000) such as ‘foaming by
gas injection’ and ‘foaming with blowing agents’, which have
been widely used in the aluminum foaming industry, are in-
adequate to produce porous titanium for orthopedic applica-
tions because only closed-cell structure can be obtained. The
conventional metallurgical method by titanium powder sin-
tering was used to fabricate porous titanium successfully (Oh
et al., 2003), but the porosity was very limited. When space-
holding material was introduced in the powder sintering pro-
cessing the porosity could been increased up to 80% or even
more (Banhart, 2001; Garrett et al., 2006; Wen et al., 2001), and
this method could produce bimodal pore sizes by controlling
the space-holder size and the titanium particle diameter. Fur-
thermore, a modified powder metallurgical technique called
‘sponge replication’ (Cachinho and Correia, 2008; Lee et al.,
2009; Li et al., 2005) or ‘impregnating method’ (Zhao et al.,
2008) has been used in fabrication of the porous titanium
and its alloys, by which the pore size, shape, and distribution
can be well-adjusted in a large scale. Some interconnected
porous structures that perfectly mimic the spatial morphol-
ogy of the cancellous bone can be easily reproduced by this
method. However, the major drawbacks of the above metal-
lurgical methods are the difficulties to avoid contamination
and impurity phases in the titanium materials, which sig-
nificantly deteriorate the mechanical properties of the tita-
nium struts. In addition, the sintered titanium struts likely
contain cracks, unbonded contact nodes between the powder
particles, and other metallographic defects, thus, the porous
titanium materials cannot bear tensile stress and even ex-
hibit ‘brittle’ nature. This is the reason that almost all the
investigations did not report the tensile strength of the pow-
der sintered porous titanium materials in the literature. This
limitation implies a risk of the orthopedic applications of the
porous titanium in vivo where a tensile or impact load may
be applied on the implant.
In recent years a rapid prototyping technique has been uti-
lized to fabricate porous titanium by using laser power to sin-
ter/remelt titanium powders (Hollander et al., 2006; Krishna
et al., 2007, 2008; Mullen et al., 2008; Traini et al., 2008), which
is based on a CAD and layer-by-layer manufacturing process.
It is of great benefit to accurately controlling the pore shape,
size and distribution by beforehand design and programming,
so that an ideal porous architecture can be constructed. With
alternative fabrication approaches, this technique has been
developed into ‘direct laser sintering’ (Hollander et al., 2006;
Krishna et al., 2007; Traini et al., 2008) and ‘selective laser
melting’ (Krishna et al., 2008; Mullen et al., 2008). The rapid
prototyping technique provides good feasibility and excellent
flexibility for production of custom-built metallic implants
with complicated structures. However, the porous titanium
production also suffers from impurities, inclusions and other
defects associated with the laser sintering or melting process-
ing, which significantly deteriorate the mechanical properties
such as tensile-stress resistance and toughness. In the view
of physical metallurgy, the struts in the porous titanium with
as-sintered (laser sintering) or as-solidified (laser melting) mi-
crostructure usually exhibits much lower ductility than that
of the wrought or plastically deformed counterparts. It is
unlikely that the laser sintered/melted porous titanium im-
plants can stand against tensile and impact stress, though
18 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S
there are very few data published (Hollander et al., 2006; Kr-
ishna et al., 2007, 2008; Mullen et al., 2008; Traini et al., 2008).
Apart from laser rapid prototyping, a similar 3D deposition
approach was also used to fabricate porous titanium, in which
the deposited 3D porous structure was strengthened by dry-
ing in air and sintering in high vacuum (Li et al., 2006). This
approach does not need laser source and control system in-
volved. Despite the fact that a good 3D mesh structure was
constructed, its low tensile strength and low ductility can be
expected due to the similar above reasons.
The mechanical properties of the porous metallic
materials are closely related to the porosity and the porous
architecture (including pore size, shape, and distribution).
The relationship of the mechanical properties and the
porosity has already been established empirically which can
be expressed as the well-known Gibson–Ashby equation
(Gibson and Ashby, 1997):
 ∗ n 1
E∗ ρ This work will provide researchers and surgeons/clinicians
= C1
Es ρs
 ∗ n2
σ∗ ρ als for the load-bearing orthopedic applications.
= C2
σs ρs
where E, σ and ρ denote the elastic modulus, strength and
density, respectively. The superscript ‘∗’ indicates the value of
the porous material, and the subscript ‘s’ indicates the value
of the dense material. C1 and C2 are constants related to ma-
terial and experimental conditions. n1 and n2 are exponen-
tials related to porous structure. It is clear that both strength
and elastic modulus of the porous materials decrease as the
porosity increases. A contradiction is that low elastic modulus
(i.e., comparable to cortical bone) corresponds to low strength.
Considering the defects and contaminations in the struts, the
porous titanium with low elastic modulus is too weak to be
used in load-bearing implants. To enhance its strength, espe-
cially for resistance to tensile stress and impact load, will be
a challenge in the real orthopedic application.
In order to improve the toughness and tensile strength
of the porous titanium, a novel porous titanium material so
called ‘entangled wire material’ (Liu et al., 2009, 2010; Tan
et al., 2009) has been developed recently in our group. Its
prominent advantage is that the strong metal wires acting
as struts construct the spatial porous structure, so that the
drawbacks of the as-sintered/as-solidified struts (e.g., metal-
lurgical defects, contaminations, etc.) can be completely over-
come. Compared with the ‘brittle’ nature of the conventional
porous titanium, the entangled titanium wire porous material
is very tough and exhibits excellent superelasticity. Its supe-
riority in mechanical properties supposes the potentials for
orthopedic applications such as hip joint implant, lumber in-
terbody fusion device, dental implant, etc.
It should be noticed that the ‘entangled wire material’ is
a kind of fiber (wire) mesh, in which the entangled wire can
be either stochastic (like non-woven) (Liu et al., 2009, 2010)
or ordered (Liu et al., 2010; Tan et al., 2009). Originally, ti-
tanium wire (fiber) materials were introduced for producing
porous-surfaced implants (Ducheyne et al., 1974). The porous
titanium fiber layer bonded on the surface of the implant pro-
vides space for bone ingrowth to fix the implant. In order to
examine its validity, some works focused on the bone forma-
tion and wound healing in titanium fiber mesh (Chang et al.,
5 (2012) 16–31
1996; Paquay et al., 1997). And other works attempted to mod-
ify the mesh with bioactive coating (Vehof et al., 2000) and
load with growth factor (Vehof et al., 2002) or bone marrow
stromal cells (Dolder et al., 2003). They all had demonstrated
the good osteoconductive properties of porous titanium fiber
mesh. However, few works deal with the mechanical behav-
iors of the titanium fiber mesh. The relationship between the
mesh structure and the mechanical properties has not been
established. Furthermore, the basic aspects in terms of the
fabrication techniques (especially for the 3D porous titanium
wire materials), the structure characterization, and the me-
chanical properties tailoring, are still open for investigation.
A data base including all above essentials is urgently needed,
because it is necessary for the bio-applications. In this study,
we report a group of data on the structure and mechanical
properties of the 3D ‘entangled titanium wire material’. Spe-
cial emphasis is put on the tensile behavior and the rela-
tionship between the porosity and the mechanical properties.
(1)
with the valuable data of such tough porous titanium materi-
(2)
2. Materials and methods
2.1. Fabrication of the entangled titanium wire materials
The mechanical behavior of the porous titanium wire materi-
als strongly depends on raw material (titanium composition,
treatment status, and wire diameter), entangled wire archi-
tecture, and porosity. Our work (Liu, 2010) demonstrated that
a smaller diameter yields higher strength at the same poros-
ity. In this study, commercial pure titanium (Grade 1) wires
with diameters of 0.08 and 0.15 mm, respectively, are selected
as raw materials. The impurities in the titanium wires are
chemically determined to be N: 0.02%, O: 0.15%, H: 0.015%, C:
0.008% (all in weight percent). Their Young’s modulus, yield
strength, and ultimate strength are 116 GPa, 300 MPa, and
370 MPa, respectively. The spatial architecture constructed
with wires is controlled by the wire-twisting methods and
intertwisting procedures. Two styles (non-woven and quasi-
ordered) have been developed in our lab which show signifi-
cant difference in mechanical behavior between the two types
of the architecture. Their fabrication methods have been de-
scribed in the literature (Liu et al., 2009, 2010; Tan et al.,
2009). Generally, the porous titanium wire materials with
non-woven architecture exhibit higher strength and higher
stiffness than that with quasi-ordered architecture. In this
study, we focus on the non-woven titanium wire materials,
which are composed of entangled wires, wire cross points,
and pores between the wires. The shape, size, and morphol-
ogy of the pores in the material are determined by the spatial
architecture of the entangled wires, which is strongly decisive
for the mechanical behaviors and properties.
A simple intertwisting procedure following forming in
a special mold can be used to prepare the non-woven
entangled titanium wire material. The spatial architecture of
the entangled wires can be constructed via various winding
and forming methods. A general procedure for forming non-
woven titanium wire materials is as follows: firstly the wire
is entangled by winding and twisting to form a precursory
 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S
bundle of twisted wires. Then the bundle is put into a
specially designed steel mold for compaction and shaping.
A high pressure in a range of 12–180 MPa is necessary.
Generally, the porous materials with larger porosity need
lower forming pressure, but smaller porosity need higher
forming pressure. The holding time under certain pressure
and at high temperature (around 650 ◦ C) is about 10 min for
avoiding wire-rebound and shape-finalizing. At this stage, the
wires cross points in the material are mobile, so that the bulk
material exhibits flexible properties but low yield strength
and low stiffness. For improving its strength a sintering
processing is needed which can be performed in a furnace
in argon atmosphere. The sintering temperature is about
1200 ◦ C and holding time is 60 min. After the sintering a
step-cooling followed with annealing at 650 ◦ C is carried out
for plasticity recovering of the titanium wires. The optimized
sintering process has been described in detail elsewhere (Liu
et al., 2010). After sintering some wire cross points in the
material are fixed by forming sintered joints, so that the yield
strength and stiffness can be significantly improved.
So far, several shapes such as cylinders, rectangular
prisms, tubes, sheets, and flat dog-bones have been made in
our lab. The cylindrical samples with 10 mm in diameter and
20 mm in length are for compressive test. The flat dog-bones
with the gauge length of about 22 mm and the section size of
5 mm × 3 mm in the gauge are fabricated for tensile test.
2.2. Characterization of the entangled titanium wire
materials
The spatial architecture of the porous titanium wire materials
can be revealed by means of X-ray tomography which
allows visualizing the inner volume of the material without
sectioning or dismantling it. For quantitative analysis
the 3D architecture can be mathematically reconstructed
from 2D projection images collected by the tomography
device while the material is rotated around a defined axis
(Garcia and Kardjilov, 2005). Fig. 1 shows a reconstructed
3D porous material (left) and an arbitrary section (right).
The homogeneity and the inner porous structure can be
statistically evaluated from the section view. For clearly
showing the local porous structure and the wire joints a
scanning electron microscopy (SEM) JEOL-JSM6460 is used.
The porosity of the porous titanium wire materials can be
directly obtained by mass–volume calculation method with
the following formula:
M
 
P= 1− × 100% (3)
Vρs
where M is weight of the sample; V is volume of the sample;
ρs is the density of the titanium wire which is 4.51 g/cm3 .
Since all the pores in the material are open and connected,
the calculated porosity is very close to the measurement with
Archimedes method (Liu et al., 2010).
In order to measure the pore size quantitatively, an
equivalent pore diameter is defined as the interspace between
twisted wires. The measurements can be performed under
optical microscope by using the commercial Image Pro
Discovery software (the spatial pores are taken as two-
dimensional pores). Thus, the pore size distributions are
5 (2012) 16–31 19
determined by means of mathematical statistics of the
measurements. Five viewpoints are chosen from the cross
and longitudinal sections of each sample.
The uniaxial tensile stress–strain behavior of the porous
titanium wire materials has been performed by using
Shimadzu AG-100KNA testing machine at room temperature.
The tests are conducted under displacement control, with a
crosshead speed of 0.05 mm/s, which corresponds to a mean
strain rate of about 2.27 × 10−3 s−1 . The stress is calculated
by dividing the tensile force by the initial cross sectional
area of the sample, and the strain is the ratio between the
sample elongation and the initial gauge length. The elastic
modulus is determined from the initial near-elastic part of the
stress–strain curve. The deformation and failure mechanism
of the porous titanium wire materials are investigated by
using the SEM.
3. Results
3.1. The porous structure in the entangled titanium wire
materials
A macroscopical 3D image for the entangled titanium wire
material is shown in Fig. 1(a). By means of the tomography
reconstruction whole the sample can be revealed along any
cross section. As an example, Fig. 1(b) displays a section view
which clearly shows the inner macrostructure of the non-
woven entangled titanium wire material. Although the pores
in the material are approximately homogeneous, some of the
titanium wires tend to assemble into wire bundles during
the forming process and preferentially turn toward specific
orientation, e.g., along longitudinal axis of the sample. Thus,
the spatial architecture is usually inhomogeneous in the
non-woven entangled titanium wire material. Fig. 2 shows
a tube-like sample and a group of section views across the
longitudinal axis via the tomography. The pore distribution in
the material is roughly homogeneous, but the arrangement of
the entangled wires is inhomogeneous. This may lead to the
dispersion in the values of the mechanical properties.
In order to distinguish the entangled wire architecture
and the pore structure, a close look was taken at the
microstructure by using SEM. Fig. 3(a) and (d) show the
surface images of the entangled wire structures made of two
different diameter wires, and Fig. 3(b) and (e) show their
section images. It is clear that the wire diameter influences
the pore size. Larger wire diameter leads to larger pores
(compared (a) with (d), (b) with (e)) although both materials
are in very similar porosity. The sample made of 0.08 mm-
diameter wire with 56% porosity has about 60% pores smaller
than 100 µm and about 40% pores in range of 100–300 µm as
shown in Fig. 3(c). The other made of 0.15 mm-diameter wire
with 56.6% porosity has about 45% pores smaller than 100 µm
and about 55% pores larger than 100 µm. The largest pores are
above 600 µm as shown in Fig. 3(f). It is undoubted that the
entangled titanium wire materials with larger porosity can
provide more pores for bone ingrowth and vascularization.
The materials with smaller porosity can only provide a few
pores large enough for the implants according to Fig. 4 and
Table 1. Even though there are, so far, no data on how many
 20

Table 1 – Data of the porous structure for the entangled Ti wire materials (P1–P4: 0.08 mm wire, P3# : 0.15 mm wire).

No. Forming Green compact Sintering and heat treatment After sintering and heat treatment
pressure (MPa) porosity (%) parameters
Open porosity Total porosity Maximum Average Pore size
(%) (%) pore pore distribution
size (µm) size (µm) (µm)
P1-1 12.6 ± 0.8 82.0 ± 0.2 1100 ◦ C × 30 min + 650 ◦ C × 30 min 80.9 ± 0.1 81.2 ± 0.1 673 180 50–450
P1-2 12.6 ± 0.8 82.0 ± 0.2 1100 ◦ C × 60 min + 650 ◦ C × 30 min 80.4 ± 0.1 80.7 ± 0.1 656 176 50–450
P1-3 12.6 ± 0.8 82.0 ± 0.2 1100 ◦ C × 90 min + 650 ◦ C × 30 min 79.8 ± 0.1 80.1 ± 0.1 639 171 50–450
P1-4 12.6 ± 0.8 82.0 ± 0.2 1200 ◦ C × 30 min + 650 ◦ C × 30 min 79.2 ± 0.1 79.6 ± 0.1 623 167 45–425
P1-5 12.6 ± 0.8 82.0 ± 0.2 1200 ◦ C × 60 min + 650 ◦ C × 30 min 78.7 ± 0.1 79.1 ± 0.1 606 163 45–425
P1-6 12.6 ± 0.8 82.0 ± 0.2 1200 ◦ C × 90 min + 650 ◦ C × 30 min 78.1 ± 0.1 78.5 ± 0.1 589 159 45–400
P1-7 12.6 ± 0.8 82.0 ± 0.2 1300 ◦ C × 30 min + 650 ◦ C × 30 min 77.6 ± 0.1 78.0 ± 0.1 573 155 45–400
P1-8 12.6 ± 0.8 82.0 ± 0.2 1300 ◦ C × 60 min + 650 ◦ C × 30 min 77.0 ± 0.1 77.4 ± 0.1 556 151 45–375
P1-9 12.6 ± 0.8 82.0 ± 0.2 1300 ◦ C × 90 min + 650 ◦ C × 30 min 76.4 ± 0.1 76.8 ± 0.1 539 146 45–375
P2-1 35.9 ± 1.0 70.0 ± 0.2 1100 ◦ C × 30 min + 650 ◦ C × 30 min 68.8 ± 0.1 69.3 ± 0.1 523 142 40–350
P2-2 35.9 ± 1.0 70.0 ± 0.2 1100 ◦ C × 60 min + 650 ◦ C × 30 min 68.3 ± 0.1 68.8 ± 0.1 506 138 40–350
P2-3 35.9 ± 1.0 70.0 ± 0.2 1100 ◦ C × 90 min + 650 ◦ C × 30 min 67.8 ± 0.1 68.3 ± 0.1 489 134 40–350
P2-4 35.9 ± 1.0 70.0 ± 0.2 1200 ◦ C × 30 min + 650 ◦ C × 30 min 67.3 ± 0.1 67.8 ± 0.1 473 131 35–325
P2-5 35.9 ± 1.0 70.0 ± 0.2 1200 ◦ C × 60 min + 650 ◦ C × 30 min 66.8 ± 0.1 67.3 ± 0.1 457 128 35–325
P2-6 35.9 ± 1.0 70.0 ± 0.2 1200 ◦ C × 90 min + 650 ◦ C × 30 min 66.2 ± 0.1 66.7 ± 0.1 440 124 35–300
P2-7 35.9 ± 1.0 70.0 ± 0.2 1300 ◦ C × 30 min + 650 ◦ C × 30 min 65.7 ± 0.1 66.2 ± 0.1 424 121 35–300
P2-8 35.9 ± 1.0 70.0 ± 0.2 1300 ◦ C × 60 min + 650 ◦ C × 30 min 65.2 ± 0.1 65.7 ± 0.1 408 118 35–275
P2-9 35.9 ± 1.0 70.0 ± 0.2 1300 ◦ C × 90 min + 650 ◦ C × 30 min 64.6 ± 0.1 65.1 ± 0.1 391 115 35–275
P3-1 81.5 ± 1.2 59.0 ± 0.2 1100 ◦ C × 30 min + 650 ◦ C × 30 min 57.8 ± 0.1 58.4 ± 0.1 375 112 30–250
P3-2 81.5 ± 1.2 59.0 ± 0.2 1100 ◦ C × 60 min + 650 ◦ C × 30 min 57.3 ± 0.1 57.9 ± 0.1 358 109 30–250
P3-3 81.5 ± 1.2 59.0 ± 0.2 1100 ◦ C × 90 min + 650 ◦ C × 30 min 56.8 ± 0.1 57.4 ± 0.1 341 106 30–250
P3-4 81.5 ± 1.2 59.0 ± 0.2 1200 ◦ C × 30 min + 650 ◦ C × 30 min 56.4 ± 0.1 57.0 ± 0.1 325 103 25–225
P3-5 81.5 ± 1.2 59.0 ± 0.2 1200 ◦ C × 60 min + 650 ◦ C × 30 min 55.9 ± 0.1 56.5 ± 0.1 308 100 25–225
P3-6 81.5 ± 1.2 59.0 ± 0.2 1200 ◦ C × 90 min + 650 ◦ C × 30 min 55.4 ± 0.1 56.0 ± 0.1 291 96 25–200
P3-7 81.5 ± 1.2 59.0 ± 0.2 1300 ◦ C × 30 min + 650 ◦ C × 30 min 54.9 ± 0.1 55.5 ± 0.1 275 93 25–200
P3-8 81.5 ± 1.2 59.0 ± 0.2 1300 ◦ C × 60 min + 650 ◦ C × 30 min 54.4 ± 0.1 55.0 ± 0.1 258 90 25–175
J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

P3-9 81.5 ± 1.2 59.0 ± 0.2 1300 ◦ C × 90 min + 650 ◦ C × 30 min 53.9 ± 0.1 54.5 ± 0.1 241 87 25–175
P4-1 180.4 ± 1.5 48.0 ± 0.2 1100 ◦ C × 30 min + 650 ◦ C × 30 min 46.8 ± 0.1 47.5 ± 0.1 224 84 20–150
P4-2 180.4 ± 1.5 48.0 ± 0.2 1100 ◦ C × 60 min + 650 ◦ C × 30 min 46.4 ± 0.1 47.1 ± 0.1 206 81 20–150
P4-3 180.4 ± 1.5 48.0 ± 0.2 1100 ◦ C × 90 min + 650 ◦ C × 30 min 46.0 ± 0.1 46.7 ± 0.1 194 78 20–150
P4-4 180.4 ± 1.5 48.0 ± 0.2 1200 ◦ C × 30 min + 650 ◦ C × 30 min 45.6 ± 0.1 46.3 ± 0.1 182 75 15–135
P4-5 180.4 ± 1.5 48.0 ± 0.2 1200 ◦ C × 60 min + 650 ◦ C × 30 min 45.2 ± 0.1 45.9 ± 0.1 171 72 15–135
P4-6 180.4 ± 1.5 48.0 ± 0.2 1200 ◦ C × 90 min + 650 ◦ C × 30 min 44.8 ± 0.1 45.5 ± 0.1 159 69 15–135
5 (2012) 16–31

P4-7 180.4 ± 1.5 48.0 ± 0.2 1300 ◦ C × 30 min + 650 ◦ C × 30 min 44.4 ± 0.1 45.1 ± 0.1 148 67 15–120
P4-8 180.4 ± 1.5 48.0 ± 0.2 1300 ◦ C × 60 min + 650 ◦ C × 30 min 44.0 ± 0.1 44.7 ± 0.1 136 64 15–120
P4-9 180.4 ± 1.5 48.0 ± 0.2 1300 ◦ C × 90 min + 650 ◦ C × 30 min 43.5 ± 0.1 44.2 ± 0.1 125 61 15–120
P3# -1 81.5 ± 1.2 59.0 ± 0.2 1100 ◦ C × 90 min + 650 ◦ C × 30 min 57.3 ± 0.1 57.9 ± 0.1 686 131 35–400
P3# -2 81.5 ± 1.2 59.0 ± 0.2 1200 ◦ C × 90 min + 650 ◦ C × 30 min 56.0 ± 0.1 56.6 ± 0.1 645 122 30–350
P3# -3 81.5 ± 1.2 59.0 ± 0.2 1300 ◦ C × 90 min + 650 ◦ C × 30 min 54.5 ± 0.1 55.1 ± 0.1 605 116 30–300
 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S 5 (2012) 16–31 21

Fig. 1 – Tomography images of the entangled Ti wire materials with non-woven structure. (a) The reconstructed 3D image
of a rectangular prism-like sample, and (b) the sectioned 3D image of this sample.

Fig. 2 – Tomography image of a tube-like entangled Ti wire material. The section views show the porous structure and its
homogeneity.

pores are necessary for the implant for forming effective
fixing with the surrounding tissue, it is well accepted that
the pore sizes in range of 100–500 µm are favorable for the
surgical applications (Bungo et al., 2006; Freyman et al., 2001;
Garrett et al., 2006; Holy et al., 2000; Li et al., 2007). But
how these materials work in vivo needs further biological
experiments. Despite all that, one can be assured that the
0.15 mm-diameter titanium wire results in larger pores than
0.08 mm-diameter wire (Fig. 3), thus is preferential when the
porosity is the same.
To establish the relationship between the pore size and
the porosity, a group of samples made of 0.08 mm-diameter
titanium wire with different porosities were evaluated by
measuring the pore size. When the porosity is 48%, the pore
size is relatively homogeneous and more than 70% pores
are under 100 µm as shown in Fig. 4(a). For about 60%
porosity, near 50% pores are larger than 100 µm. A larger
porosity corresponds to a larger percent of the pores above
100 µm (Fig. 4). The sample with 82% porosity has about
25% pores under 100 µm and 75% pores above 100 µm as
shown in Fig. 4(d). A distinct relationship between the pore
size and the porosity has been shown in Fig. 5. For reference,
all the data of the porous structure for the entangled
titanium wire materials with various porosities are provided
in Table 1, including the forming pressure which is needed
for the specific porosity, and the sintering and heat treatment
parameters. The green compact porosity in the table was
calculated according to formula (3). The open porosity and
the total porosity were evaluated after sintering and heat
treatment by using the hydrostatic weighing method (Liu
et al., 2010; Taylor et al., 1999).
3.2. Tensile behavior and property of the entangled
titanium wire materials
Under uniaxial tensile stress the entangled titanium wire
materials behave partially like a rubber or biopolymer with
non-linearity, high elongation at break, and low elastic
modulus, because the entangled wire structure is similar to
the molecular networks in rubber or polymer. The typical
stress–strain curves are composed of initial near-elastic stage
(as indicated by OA in Fig. 6), the knee of the curve (A in Fig. 6),
plastic deformation stage (AB in Fig. 6), and failure start point
(B in Fig. 6). The elastic modulus is defined as the slope of
the initial OA stage. The yield strength is determined from
A point. It is easily understandable that the OA corresponds
22 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S 5 (2012) 16–31

Fig. 3 – Entangled wire architectures after sintering and the pore size distributions. (a)–(c): for wire diameter 0.08 mm; and
(d)–(f): for wire diameter 0.15 mm.

Fig. 4 – Pore size distributions in the entangled Ti wire materials with various porosities (wire diameter 0.08 mm).
 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S
Fig. 5 – Dependence of the pore size on the porosity.
Fig. 6 – Nominal tensile stress–strain curves of the
entangled Ti wire materials with various porosities.
to the structure elastic deformation, but AB corresponds to
the structure plastic deformation via wires sliding each other.
If the material is in high density (small porosity) the sliding
can be blocked, and then the structure plastic deformation
becomes difficult. Such effect has been clearly revealed in
Fig. 6 where the yield point moves to the failure start point
as the porosity decreases. For the sample with 44.7% porosity,
it behaves like a brittle alloy as shown in Fig. 6 for which there
is no distinct plastic deformation stage.
It should be noted that the entangled titanium wire
materials exhibit very low stiffness due to the pliable
spatial architecture. The entangled wire structure with
higher density (i.e. smaller porosity) should be more stable
under loading than that with lower density (larger porosity).
Therefore, the material with lower porosity has higher
strength and larger elastic modulus. For example, 44.7%
porosity corresponds to 75 MPa yield strength, 108 MPa
ultimate tensile strength, and 1.05 GPa elastic modulus. For
57.9% porosity, these values are only 24 MPa, 47.5 MPa, and
0.33 GPa, respectively. All the three values reveal a significant
dependence on the porosity, which can be plotted in Fig. 7.
The yield strength of 24–75 MPa and the ultimate tensile
5 (2012) 16–31 23
Fig. 7 – Effect of the porosity on the tensile strength and
the elastic modulus.
Fig. 8 – Effect of the Ti wire diameter on the tensile
behavior of the entangled Ti wire materials.
strength of 47.5–108 MPa can imitate that of the natural
cortical bone (ultimate tensile strength: 9.9–151 MPa) (Donald
et al., 1974). The elastic modulus in range of 0.04–1.2 GPa is
comparable to cancellous bone (0.01–3.0 GPa), but is too low
to compare with the cortical bone (4.4–28.8 GPa) (Geetha et al.,
2009). If they are used for replacement of the cortical bone,
strengthening must be needed. Another contradiction is that
the lower porosity corresponds to larger elastic modulus and
higher strength, but smaller pore size which is unfavorable
for the bone ingrowth. Further modification on the porous
structure is desirable.
The titanium wire diameters also influence the tensile
behaviors and properties as shown in Fig. 8. With the similar
porosity, the material made of 0.08 mm-diameter titanium
wire exhibits higher strength and larger elastic modulus
but smaller elongation than that of 0.15 mm-diameter
titanium wire. For reference, all the tensile properties of the
entangled titanium wire materials with various porosities are
summarized in Table 2. The sample numbers correspond to
that in Table 1.
In addition, good toughness of these materials is expected
because they exhibit a typical ductile manner with large
strain to failure under tensile loading. In order to evaluate
their toughness values, one can measure the area underneath
24 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S 5 (2012) 16–31

Table 2 – Tensile properties of the entangled Ti wire materials (P1–P4: 0.08 mm wire, P3∗ : without sintering; P3# :
0.15 mm wire).

No. Total porosity Yield strength UTS (MPa) Elastic modulus Elongation at F-max
(%) (MPa) (MPa) (%)
P1-1 81.2 ± 0.1 2.6 ± 0.1 5.7 ± 0.3 41.1 ± 0.6 54.7 ± 0.4
P1-2 80.7 ± 0.1 3.1 ± 0.1 6.8 ± 0.3 46.0 ± 0.5 53.4 ± 0.3
P1-3 80.1 ± 0.1 3.8 ± 0.1 8.0 ± 0.1 51.4 ± 0.3 51.9 ± 0.1
P1-4 79.6 ± 0.1 4.1 ± 0.1 8.6 ± 0.1 54.1 ± 0.3 51.2 ± 0.1
P1-5 79.1 ± 0.1 4.4 ± 0.1 9.2 ± 0.1 56.8 ± 0.4 50.5 ± 0.1
P1-6 78.5 ± 0.1 4.8 ± 0.1 9.9 ± 0.1 60.0 ± 0.4 49.7 ± 0.1
P1-7 78.0 ± 0.1 5.1 ± 0.1 10.5 ± 0.1 62.7 ± 0.5 49.0 ± 0.2
P1-8 77.4 ± 0.1 5.6 ± 0.2 11.5 ± 0.2 67.2 ± 0.6 47.8 ± 0.3
P1-9 76.8 ± 0.1 7.0 ± 0.3 12.7 ± 0.3 72.6 ± 0.9 46.3 ± 0.4
P2-1 69.3 ± 0.1 8.8 ± 0.2 19.0 ± 0.4 114.6 ± 0.8 44.2 ± 0.3
P2-2 68.8 ± 0.1 9.6 ± 0.2 20.7 ± 0.4 122.0 ± 0.8 43.2 ± 0.2
P2-3 68.3 ± 0.1 10.5 ± 0.2 22.8 ± 0.3 131.3 ± 0.9 41.9 ± 0.2
P2-4 67.8 ± 0.1 11.3 ± 0.1 24.7 ± 0.1 139.7 ± 0.6 40.8 ± 0.1
P2-5 67.3 ± 0.1 11.7 ± 0.1 25.5 ± 0.1 143.4 ± 0.6 40.2 ± 0.1
P2-6 66.7 ± 0.1 12.5 ± 0.1 27.4 ± 0.3 151.8 ± 0.8 39.1 ± 0.1
P2-7 66.2 ± 0.1 13.1 ± 0.1 28.8 ± 0.4 158.0 ± 1.0 38.3 ± 0.2
P2-8 65.7 ± 0.1 14.3 ± 0.3 31.4 ± 0.4 169.5 ± 1.6 36.7 ± 0.3
P2-9 65.1 ± 0.1 18.2 ± 0.5 32.8 ± 0.5 184.5 ± 3.5 35.9 ± 0.2
P3∗ 59.0 ± 0.2 18.2 ± 0.5 35.8 ± 0.9 305.5 ± 4.4 10.6 ± 0.3
P3-1 58.4 ± 0.1 23.0 ± 0.2 44.9 ± 0.4 316.1 ± 1.8 33.5 ± 0.2
P3-2 57.9 ± 0.1 24.1 ± 0.2 47.5 ± 0.6 334.7 ± 1.2 32.6 ± 0.1
P3-3 57.4 ± 0.1 25.5 ± 0.1 50.6 ± 0.3 356.9 ± 1.1 31.6 ± 0.1
P3-4 57.0 ± 0.1 26.4 ± 0.2 52.7 ± 0.3 371.7 ± 1.1 30.9 ± 0.1
P3-5 56.5 ± 0.1 27.6 ± 0.3 55.5 ± 0.5 391.8 ± 1.4 30.0 ± 0.2
P3-6 56.0 ± 0.1 29.2 ± 0.2 59.1 ± 0.4 417.6 ± 1.3 28.8 ± 0.1
P3-7 55.5 ± 0.1 30.3 ± 0.3 61.5 ± 0.5 434.8 ± 1.4 28.0 ± 0.1
P3-8 55.0 ± 0.1 32.0 ± 0.4 65.3 ± 0.5 462.0 ± 2.2 26.7 ± 0.3
P3-9 54.5 ± 0.1 33.8 ± 0.5 69.4 ± 0.6 491.3 ± 4.3 25.3 ± 0.5
P4-1 47.5 ± 0.1 52.6 ± 0.5 80.9 ± 0.4 609.3 ± 4.6 22.5 ± 0.1
P4-2 47.1 ± 0.1 54.9 ± 0.2 83.8 ± 0.5 656.8 ± 3.8 21.9 ± 0.1
P4-3 46.7 ± 0.1 57.8 ± 0.2 87.7 ± 0.4 720.6 ± 3.5 21.0 ± 0.1
P4-4 46.3 ± 0.1 59.8 ± 0.3 90.3 ± 0.5 763.1 ± 3.4 20.4 ± 0.1
P4-5 45.9 ± 0.1 62.7 ± 0.4 94.1 ± 0.6 825.3 ± 4.2 19.5 ± 0.2
P4-6 45.5 ± 0.1 66.2 ± 0.4 98.7 ± 0.6 900.6 ± 4.0 18.5 ± 0.1
P4-7 45.1 ± 0.1 69.0 ± 0.5 102.4 ± 0.7 961.1 ± 4.6 17.7 ± 0.2
P4-8 44.7 ± 0.1 74.9 ± 0.5 108.1 ± 0.7 1054.4 ± 5.1 16.4 ± 0.4
P4-9 44.2 ± 0.1 77.3 ± 0.7 116.8 ± 0.5 1222.6 ± 4.3 14.4 ± 0.3
P3# -1 57.9 ± 0.1 16.6 ± 0.3 46.2 ± 0.4 144.9 ± 5.5 49.1 ± 0.5
P3# -2 56.6 ± 0.1 20.7 ± 0.5 50.9 ± 0.5 243.8 ± 4.9 46.5 ± 0.4
P3# -3 55.1 ± 0.1 25.0 ± 0.6 56.4 ± 0.7 347.6 ± 5.8 43.5 ± 0.3

the tensile stress–strain curves (Fig. 6) by taking the integral.
The toughness value can be determined mathematically by:
 ε
energy f
= σdε (4)
volume 0 3.3.
where ε is strain, εf is the strain upon failure, and σ is
stress. Then, the toughness values of the entangled titanium
wire materials (porosities in the range of 44.7%–57.8% as
indicated in Fig. 6) are in the range of 1.19–1.56 kJ/cm3 .
They are far above that of the titanium foams, of which
the impact toughness is 2.4 kJ/m2 for 60% porosity and
1.3 kJ/m2 for 70% porosity (Kashef et al., 2010). Although there
is limited literature on the toughness of porous titanium, it
is undoubted that the toughness of the present materials is
superior to that of many other porous titanium materials.
If compared with natural bone, the entangled titanium wire
material is obviously advantageous in toughness because the
bone is almost brittle under tensile loading and has small
compressive strain to failure (less than 2%) (Donald et al.,
1974). The present materials exhibit 15%–40% tensile strain
to failure as shown in Fig. 6.
Failure of the entangled titanium wire materials
under tensile stress
The deformation of the entangled titanium wire materials
under tensile loading initiates from the structure stretching
and progresses accompanying local wires unwinding toward
the tensile load. How many wires unwind and to what
extent the unwinding attains depend on the porosity. For the
material made of 0.08 mm-diameter titanium wire with 57.9%
porosity, the sample behaves like a negative Poisson’s ratio
material (Lakes, 1993) with auxetic behavior in the process
of unwinding (corresponding to AB stage as shown in Fig. 6).
The auxeticity proceeds as the tension persists until most
of the wires are stretched toward the tensile load as shown
in Fig. 9. Under the ultimate tensile stress the stretched
 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S 5 (2012) 16–31 25

Fig. 9 – Deformation and failure during tension for the entangled Ti wire materials (wire diameter 0.08 mm,
porosity = 57.9%).

Fig. 10 – Fractographies of the entangled Ti wire materials (wire diameter 0.08 mm). (a) Failed samples with different
porosity after tension test, (b) and (c): magnified images of the zone of fracture for the sample with 45.9% porosity.

single wire fracture occurs and leads to rapid rupture of load and the compact entangled wire structure is completely
other neighboring wires due to the stress concentration. For destroyed.
the material with 45.9% porosity, the wires’ unwinding is
difficult because of the higher density (lower porosity). The
4. Discussion
entangled wire structure is remained in a high density until
failure occurs; as a result, the fracture is localized in a small 4.1. Mechanical property of the entangled titanium wire
zone as shown in Fig. 10(a)–(c). As the porosity increases the materials
fracture zone becomes indistinct. For the material with 78.5%
porosity, the sample becomes loosening after tension (Fig. 10), The entangled titanium wire material is a kind of porous
indicating that most of the wires unwind toward the tensile metal. Like other porous materials, their mechanical
26 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S 5 (2012) 16–31

properties strongly depend on the porosity. For conventional

open-cell foam a well-known relationship between the
mechanical properties and the porosity has been established
based on the initial linear elastic feature of the material by
using a simple procedure of structural mechanics (Gibson and
Ashby, 1997):
 ∗ 2
E∗ ρ
≈ (5)
Es ρs
 ∗ 3
σ∗ ρ 2
= 0.3 . (6)
σs ρs
These relationships have been validated in many open-cell
porous materials including polymer foams, sponge rubbers,
porous glasses, and porous metals (Gibson and Ashby,
1997). When we compare the mechanical properties of the
entangled titanium wire materials with the values predicted
by the above general formulas, distinct variances are found
between the measurements and the predictions as shown in
Fig. 11. At the side of small relative density (e.g., around 0.2),
the strengths are far below the predictions of Eq. (6), while
at the side of large relative density (e.g., around 0.55), the
strengths are far above the predictions (Fig. 11(a)). In addition,
a significant deviation can be found in the relative elastic
modulus as shown in Fig. 11(b). The measurements are about
two orders of magnitude less than the predictions. These
distinctions suggest very different deformation mechanism
and different mechanical behavior of the entangled titanium
wire materials compared to the open-cell foams.
The Gibson–Ashby equations were originally derived from
a simple stress relation considering an open-cell structure Fig. 11 – Relationship between the relative
under pressure as shown in Fig. 12(a). These relations were strength/modulus and the relative density for the
also considered being valid in the linear elastic stage for entangled Ti wire materials.
many porous materials under tension (Gibson and Ashby,
1997). In the nonlinear elastic stage, Warren and Kraynik
(1991) established a modified relation considering strain value
and the ‘strut’ rotation for the uniaxial tension as shown in
Fig. 12(b):
 ∗ 2  ∗ 2
σ∗ ρ ρ ρ
 ∗
= 1.1 ε + 3.74 ε2 + 0.0343 ε3 . (7)
Es ρs ρs ρs
For the entangled wire material, the structure can be
simplified as a 2D cell which is composed of wire segments
along various orientations and cross points as shown in
Fig. 12(c). The wire segments are usually curved and the
cross points may be free or fixed depending on the sintering
processing. The elastic deformation modes under low stress
can be the wire segments’ deflexion and unwinding if all Fig. 12 – Simplified structure models of the porous
the points are fixed. In this case the material should be in materials. (a) Linear elastic structure model of open-cell
relatively high stiffness and high yield strength. If all the foams under compression (Gibson and Ashby, 1997), (b)
cross points are free (without sintering) the structure should Nonlinear elastic structure model of open-cell foams under
be very flexible because the sliding can be triggered under tension (Warren and Kraynik, 1991), (c) Friction-slippage
very low stress. In such a case the material exhibits very structure model of the entangled Ti wire materials under
low stiffness and low yield strength. In as-sintered state, the tension.
free cross points and the sintered joints probably coexist in
the material. Although it is difficult to count the percentages
of the joints and the free cross points, one can affirm that materials have higher local contacted stress which is helpful
the joints are dominant at low porosity and the free cross for forming joints during sintering. Otherwise, the materials
points are in majority at high porosity. This can be easily with high porosity (loose structure) are difficult to form joints
understood because the wire cross points in the compact during sintering because of small local contacted stress in
 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S 5 (2012) 16–31 27

the cross points. It is undoubted that the joints contribute to

the structure stability and enhance the stiffness and the yield
strength.
Considering 2D situation, the entangled wire materials
would behave similar to the open-cell foams if the entire cross
points became ideal joints and the joints were strong enough
and did not break before the material failure. In this case both
the 2D structure models (Fig. 12(b) and (c)) were similar, and
the mechanical properties of the entangled wire materials
became comparable to the open-cell foams. Considering 3D
situation, the materials with such strong joints may behave
similar to the truss material (Wallach and Gibson, 2001) which
has much higher strength and stiffness than the metal foams
(Deshpande et al., 2001). In the real situation, the joints are
relatively weak and easily broken under loading. After joint
Fig. 13 – Relationship between the forming pressure and
breaking the sliding between cross wires occurs under the
the relative density for the entangled Ti wire materials.
elevated loading. The sliding mechanism raises an important
factor in this material: internal friction, which causes some
novel mechanical behaviors, such as elastic hysteresis (Liu
et al., 2010; Tan et al., 2009) and high capacity of energy
absorption (Tan et al., 2009).
Considering the entangled wire materials without sinter-
ing (joint does not form), the internal friction force for one
cross point can be expressed as:
f
Fi = µFni (8)
f
where Fi denotes the friction force of a cross point. µ is the
coefficient of friction. Fni denotes normal force exerted by
each wire on the other, which can be related to the forming
pressure (denoted as PF ) of the entangled titanium wire
materials as listed in Table 1. Since the forming pressure is
balanced by the total internal contacted stress during forming
processing, it is reasonable to assume:
Fig. 14 – Relationship between the yield strength and the
nFni ≈ PF (9) relative density for the entangled Ti wire materials.
where n is the numbers of the cross points per unit
area perpendicular to the stress direction. If the sliding
The strength predicted by this simple equation is in good
mechanism dominates (wire deformation resistance can be
agreement with the measured yield strength of the entangled
ignored in this case), the deformation resistance (equals to
titanium wire materials as shown in Fig. 14, indicating that
the strength) of the entangled wire materials can be simply
sliding mechanism and internal friction dominate during
taken as the total internal friction force:
deformation in this material.
f
σ ≈ nFi = nµFni . (10) In order to evaluate the mechanical properties of the
porous materials, Deshpande et al. (2001) draw Hashin–
By plotting the relative density against the forming
Shtrikman upper bounds (Hashin and Shtrikman, 1963) on
pressure based on the measurements (columns 2 and 3 in
the stiffness and the uniaxial yield strength obtained by
Table 1) and Eq. (3), an empirical relationship between the two
transforming the linear HS bound to the perfectly plastic
values can be obtained (as shown in Fig. 13):
case using the prescription of Suquet (1993). They found
ρ
 ∗
that the stiffness and strength of the octet-truss lattice
ln PF = 1.366 + 7.365 . (11)
ρs material were about half the theoretical upper bound values
Thus, a relationship between the strength and the relative for relative density between 0.01 and 0.1, and exceeded the
density can be established: corresponding values for metallic foams by a factor between 3
and 10. It is convenient to compare the mechanical properties
ρ
 ∗
ln σ = C + 7.365 (12) of the entangled titanium wire materials with those of other
ρs
porous materials by using the theoretical HS upper bounds as
where C is a constant determined by the coefficient of friction shown in Fig. 15. The strength of the entangled titanium wire
of titanium wires: C = ln µ + 1.366. For dry friction of titanium materials is about half the theoretical upper bound values for
wires, µ ≈ 0.2–0.3, then, C ≈ 0. Thus, relative density between 0.3 and 0.6, which are comparable to
ρ
 ∗ that of the octet-truss lattice material. But in lower relative
ln σ = 7.365 . (13) density (below 0.3), the strength is far below the theoretical
ρs
28 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S
Fig. 15 – Comparison among the stiffness and strength of
the entangled Ti wire materials, other porous materials
and the upper bounds for isotropic voided materials.
upper bound values. Even so, the relative strength of the
entangled titanium wire materials is higher than those of
metal foams if the relative density is not considered. Typically,
the relative elastic modulus of the entangled titanium wire
materials is far below the theoretical upper bound values, and
is also far below that of the octet-truss lattice material. This
is because of the very flexible entangled wire structure.
4.2. Strengthening of the entangled titanium
materials
Unlike other metallic porous materials, the entangled
titanium wire materials behave with sliding and friction
mechanism under loading. In order to strengthen the
materials one must block the wire sliding. Sintering leads
to form joints at the wire cross points which prevent wire
sliding, thus improve stiffness and strength. Fig. 12(c) has
visually illustrated this efficiency of the sintering.
The present work has confirmed that sintering strengthen-
ing is effective for the entangled wire materials. All the sin-
tering parameters (e.g., temperature, holding time, protective
argon atmosphere, and following heat treatment) are set for
forming strong joints. Even so, most of the joints are weak
in practical situation compared to the wire itself. They will
preferentially fail under loading. It is desired that most the
joints are as strong as possible, so that the entangled wire
materials are strong enough for applications. How to make
stronger joint is a problem in practice. The strength of the
sintered joints depends on the pre-pressure (local contacted
stress), contacted area, and the sintering parameters. Both
the pre-pressure and the contacted area are determined by
the forming pressure as listed in Table 1, which can be re-
lated to the porosity of the material. The high forming pres-
sure corresponds to the high pre-pressure and large contacted
area at the cross point. Fig. 16 shows cross points formed by
using two different forming pressures. It is evident that form-
ing pressure has caused local plastic deformation at the cross
points. A forming pressure of 81.5 MPa leads to small con-
tacted area (Fig. 16(a)), thus a relative weak joint is expected;
but a forming pressure of 180.4 MPa results in severe local
plastic deformation and large contacted area (Fig. 16(b)), thus
5 (2012) 16–31
Fig. 16 – Effect of the forming pressure on the wire cross
wire joints. (a) with 81.5 MPa, and (b) with 180.4 MPa.
a relative strong joint is expected. These observations help us
to understand why the strength and stiffness of the entangled
titanium wire materials increase significantly as the density
increases (Fig. 15).
It is undoubted that the spatial architecture of the en-
tangled titanium wire materials plays an important role in
the mechanical behavior and properties. We compared the
non-woven structure with a quasi-ordered entangled wire
structure, and found distinct difference in the mechanical
properties (Liu et al., 2010; Tan et al., 2009). The quasi-
ordered entangled wire materials are more flexible and well-
proportioned in the porous structure. These are favorable for
the biomedical applications. But their stiffness is still low.
Strengthening is highly needed when they are used for ar-
tificial hip joint (Garrett et al., 2006; Long and Rack, 1998) or
spine fusion (Imwinkelried, 2007; Müller et al., 2006; Singh
et al., 2010) where the stiffness must be comparable to those
of the cortical bone. Forming strong wire cross joints in the
entangled wire structure should be an effective strengthen-
ing method.
For the non-woven entangled titanium wire materials, the
microstructural homogeneity is another concern. The hetero-
geneous entangled wire structure leads to scattered mechan-
ical properties and undesirable pore distribution. Although
a great effort has been made in making homogeneous non-
woven entangled wire materials, some abnormal large voids
or local textures still exist in the materials (see section views
as shown in Fig. 2). Alternatively, the quasi-ordered entan-
gled wire materials are practically useful because they are
 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S
Fig. 17 – Entangled metallic wire materials with improved
homogeneity.
more homogeneous in porous structure and easily mass-
manufactured (Tan et al., 2009). Fig. 17 shows some cylindrical
samples of the quasi-ordered entangled wire materials with
different porosities. We will endeavor to strengthen them
with various methods in future study, so as to meet the de-
mand of the load-bearing biomedical applications.
5. Conclusions and remarks
The entangled titanium wire materials with non-woven
structure can be made under forming pressure of 12–180 MPa.
The porosity of the entangled titanium wire materials is
determined by the forming pressure. It decreases as the
forming pressure increases. Typically, the 12.6 MPa forming
pressure corresponds to 82% porosity, while the 180.4 MPa
corresponds to 48% porosity.
The pores in the entangled titanium wire materials
are irregular in shape, which have a nearly half-normal
distribution in size range. The average pore size depends on
the porosity and the wire diameter. The larger porosity and
the larger wire diameter lead to the larger average pore size.
More than 50% pores are in the range of 50–200 µm for all the
materials tested in this study, but only when the porosity is
larger than 60%, near 50% pores are larger than 100 µm for the
sample made of 0.08 mm-diameter wire. For example, the 82%
porosity corresponds to about 75% pores above 100 µm and
25% pores under 100 µm; while the 48% porosity corresponds
to about 70% pores under 100 µm.
The entangled titanium wire materials exhibit initial near-
linear elastic stage, nonlinear plastic deformation stage, and
failure stage under uniaxial tensile loading. The near-linear
elastic stage corresponds to the structure elastic deformation;
the nonlinear plastic deformation stage corresponds to the
structure plastic deformation via wires sliding over each
other. When the material is in small porosity the sliding
can be blocked, so that the structure plastic deformation
becomes unnoticeable. The failure mode of the materials can
be entangled wire loosening or wires broken, which depends
on the porosity. The material with large porosity fails via wire
loosening; otherwise it fails via wires broken.
The entangled titanium wire materials exhibit distinct
yielding that corresponds to startup of the wire sliding
in large scale. The yield strength significantly depends on
the porosity. Both the yield strength and ultimate strength
5 (2012) 16–31 29
increase as the porosity decreases. The elastic modulus is
very low for this kind of material because of the structural
flexibility. Typically, 44.7% porosity corresponds to 75 MPa
yield strength, 108 MPa ultimate tensile strength, and 1.05 GPa
elastic modulus; while 57.9% porosity corresponds to only
24 MPa yield strength, 47.5 MPa ultimate tensile strength, and
0.33 GPa elastic modulus.
The entangled titanium wire materials are very promis-
ing for implant applications because of their very good
toughness, perfect flexibility, high strength, adequate elastic
modulus, and low cost. But some modification and improve-
ment in the average pore size and/or the balanced mechanical
properties are needed when used for different applications.
For example, when they are used for bone ingrowth, a rel-
ative larger diameter titanium wire is recommended which
results in larger average pore size. Alternatively, the larger
porosity also leads to larger average pore size, but results in
low strength and small elastic modulus. Generally, the elastic
modulus of the entangled titanium wire materials is compa-
rable to that of the cancellous bone, but lower than the corti-
cal bone. It is possible to enhance the stiffness of this material
via the entangled wire structure design and the strength-
ening of the wire cross joints. All these need comprehen-
sive research on the structural mechanics and the fabrication
procedures of this kind of material.
Acknowledgments
We wish to thank Prof. John Banhart and Dr. García
Moreno, from Hahn–Meitner-Institut Berlin, Germany, for
their assistance in the tomography experiments and helpful
discussions.
REFERENCES
Albrektsson, T., Johansson, C., 2001. Osteoinduction, osteocon-
duction and osseointegration. Eur. Spine J. 10, S96–S101.
Banhart, J., 2001. Manufacture, characterisation and application of
cellular metals and metal foams. Prog. Mater. Sci. 46, 559–632.
Bréme, J., Biehl, V., Schulte, W., d’Hoedf, B., Donath, K., 1993.
Development and functionality of isoelastic dental implants
of titanium alloys. Biomaterials 14, 887–892.
Bungo, O., Mitsuru, T., Shunsuke, F., Masashi, N., Tadashi, K.,
Takashi, N., 2006. Pore throat size and connectivity determine
bone and tissue ingrowth into porous implants: three-
dimensional micro-CT based structural analyses of porous
bioactive titanium implants. Biomaterials 27, 5892–5900.
Cachinho, S.C., Correia, R.N., 2008. Titanium scaffolds for
osteointegration: mechanical, in vitro and corrosion behavior.
J. Mater. Sci. Mater. Electron. 19, 451–457.
Chang, Y.S., Oka, M., Kobayashi, M., Gu, H.O., Li, Z.L., Nakamura,
T., Ikada, Y., 1996. Significance of interstitial bone ingrowth
under load-bearing conditions: a comparison between solid
and porous implant materials. Biomaterials 17, 1141–1148.
Deshpande, V.S., Fleck, N.A., Ashby, M.F., 2001. Effective properties
of the octet-truss lattice material. J. Mech. Phys. Solids 49,
1747–1769.
Dolder, J., Farber, E., Spauwen, P.H.M., Jansen, J.A., 2003. Bone
tissue reconstruction using titanium fiber mesh combined
with rat bone marrow stromal cells. Biomaterials 24,
1745–1750.
30 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S
Donald, T.R., Albert, H.B., Cleveland, O., 1974. The mechanical
properties of cortical bone. J. Bone Joint Surg. Am. 56,
1001–1022.
Ducheyne, P., Martens, M., Aernoudt, E., Mulier, J., Meester, P.,
1974. Skeletal fixation by metal fiber coating of the implant.
Acta Orthop. Belg. 40, 799–805.
Eigler, N.L., Khorsandi, M.J., Forrester, J.S., Fishbein, M.C., Litvack,
F., 1993. Implantation and recovery of temporary metallic
stents in canine coronary arteries. J. Am. Coll. Cardiol. 22,
1207–1213.
Freyman, T.M., Yannas, I.V., Gibson, L.J., 2001. Cellular materials
as porous scaffolds for tissue engineering. Prog. Mater. Sci. 46,
273–282.
Fujibayashi, S., Nakamura, T., Nishiguchi, S., Tamura, J., Uchida,
M., Kim, H.M., Kokubo, T., 2001. Bioactive titanium: effect
of sodium removal on the bone-bonding ability of bioactive
titanium prepared by alkali and heat treatment. J. Biomed.
Mater. Res. 56, 562–570.
Fujibayashi, S., Neo, M., Kim, H.M., Kokubo, T., Nakamura, T.,
2004. Osteoinduction of porous bioactive titanium metal.
Biomaterials 25, 443–450.
Garcia, M.F., Kardjilov, N., 2005. Cellular architecture—overview of
X-ray and neutron radioscopy and tomography. Mater. World
6.
Garrett, R., Abhay, P., Panagiotis, A.D., 2006. Fabrication methods
of porous metals for use in orthopaedic applications.
Biomaterials 27, 2651–2670.
Geetha, M., Singh, A.K., Asokamani, R., Gogia, A.K., 2009. Ti based
biomaterials, the ultimate choice for orthopaedic implants—a
review. Prog. Mater. Sci. 54, 397–425.
Gibson, L.J., Ashby, M.F., 1997. Cellular Solids: Structure and
Properties, 2nd ed. Cambridge University Press, Cambridge,
UK.
Hashin, Z., Shtrikman, S., 1963. A variational approach to the
theory of the elastic behaviour of multi-phase materials. J.
Mech. Phys. Solids 11, 127–140.
Hollander, D.A., Walter, M., Wirtz, T., Sellei, R., Schmidt-Rohlfing,
B., Paar, O., Erli, H.J., 2006. Structural, mechanical and in
vitro characterization of individually structured Ti–6Al–4V
produced by direct laser forming. Biomaterials 27, 955–963.
Hollister, S.J., 2005. Porous scaffold design for tissue engineering.
Nature Mater. 4, 518–524.
Holy, C.E., Shoichet, M.S., Davies, J.E., 2000. Engineering three-
dimensional bone tissue in vitro using biodegradable scaffolds:
investigating initial cell-seeding density and culture period.
J. Biomed. Mater. Res. 51, 376–382.
Imwinkelried, T., 2007. Mechanical properties of open-pore
titanium foam. J. Biomed. Mater. Res. Part A 81, 964–970.
Jacobs, J.J., Sumner, D.R., Galante, J.O., 1993. Mechanisms of bone
loss associated with total hip replacement (Review). Orthop.
Clin. North Am. 24, 583–590.
Kashef, S., Asgari, A., Hilditch, T.B., Yan, W., Goel, V.K., Hodgson,
P.D., 2010. Fracture toughness of titanium foams for medical
applications. Mater. Sci. Eng. A 527, 7689–7693.
Kim, H.M., Kokubo, T., Fujibayashi, S., Nishiguchi, S., Nakamura,
T., 2000. Bioactive macroporous titanium surface layer on
titanium substrate. J. Biomed. Mater. Res. 52, 553–557.
Körner, C., Singer, R., 2000. Processing of metal foams—challenges
and opportunities. Adv. Eng. Mater. 2, 159–165.
Krishna, B.V., Bose, S., Bandyopadhyay, A., 2007. Low stiffness
porous Ti structures for load-bearing implants. Acta Biomater.
3, 997–1006.
Krishna, B.V., Xue, W., Bose, S., Bandyopadhyay, A., 2008.
Engineered porous metals for implants. JOM 60, 45–48.
Laheurte, P., Prima, F., Eberhardt, A., Gloriant, T., Wary, M., Patoor,
E., 2010. Mechanical properties of low modulus beta titanium
alloys designed from the electronic approach. J. Mech. Behav.
Biomed. Mater. 3, 565–573.
5 (2012) 16–31
Lakes, R., 1993. Advances in negative Poisson’s ratio materials.
Adv. Mater. 5, 293–296.
Lee, J.H., Kim, H.E., Koh, Y.H., 2009. Highly porous titanium
(Ti) scaffolds with bioactive microporous hydroxyapatite/TiO2
hybrid coating layer. Mater. Lett. 63, 1995–1998.
Li, J.P., Li, S.H., Van Blitterswijk, C.A., De Groot, K., 2005. A
novel porous Ti6 Al4 V: characterization and cell attachment.
J. Biomed. Mater. Res. Part A 73, 223–233.
Li, J.P., Pamela, H., Mirella, D., Clayton, E.W., Joost, R.W., Clemens,
A.B., Klaas, G., 2007. Bone ingrowth in porous titanium
implants produced by 3D fiber deposition. Biomaterials 28,
2810–2820.
Li, J.P., Wijn, J.R., Blitterswijk, C.A., Groot, K., 2006. Porous Ti6 Al4 V
scaffold directly fabricating by rapid prototyping: preparation
and in vitro experiment. Biomaterials 27, 1223–1235.
Liu, P., 2010. Fabrication and mechanical behavior of the high-
tough entangled metal wire materials and their application for
implant. Doctoral Dissertation. Shanghai Jiaotong University,
Shanghai.
Liu, P., He, G., Wu, L.H., 2009. Uniaxial tensile stress–strain
behavior of entangled steel wire material. Mater. Sci. Eng. A
509, 69–75.
Liu, P., Tan, Q.B., Wu, L.H., He, G., 2010. Compressive and
pseudo-elastic hysteresis behavior of entangled titanium wire
materials. Mater. Sci. Eng. A 527, 3301–3309.
Long, M., Rack, H.J., 1998. Titanium alloys in total joint
replacement—a materials science perspective. Biomaterials
19, 1621–1639.
Manukyana, K., Amirkhanyan, N., Aydinyan, S., Danghyan, V.,
Grigoryan, R., Sarkisyan, N., Gasparyan, G., Aroutiounian, R.,
Kharatyan, S., 2010. Novel NiZr-based porous biomaterials:
Synthesis and in vitro testing. Chem. Eng. J. 162, 406–414.
Mullen, L., Stamp, R.C., Brooks, W.K., Jones, E., Sutcliffe, C.J., 2008.
Selective laser melting: a regular unit cell approach for the
manufacture of porous, titanium, bone ingrowth constructs,
suitable for orthopedic applications. J. Biomed. Mater. Res. B
Appl. Biomater. 89B, 325–334.
Müller, U., Imwinkelried, T., Horst, M., Sievers, M., Graf-Hausner,
U., 2006. Do human osteoblasts grow into open-porous
titanium? Eur. Cells Mater. 11, 8–15.
Niinomi, M., 2008. Mechanical biocompatibilities of titanium
alloys for biomedical applications. J. Mech. Behav. Biomed.
Mater. 1, 30–42.
Nishiguchi, S., Nakamura, T., Kobayashi, M., Kim, H.M., Miyaji, F.,
Kokubo, T., 1999. The effect of heat treatment on bone-bonding
ability of alkali-treated titanium. Biomaterials 20, 491–500.
Oh, I.H., Nomura, N., Masahashi, N., Hanada, S., 2003. Mechanical
properties of porous titanium compacts prepared by powder
sintering. Scr. Mater. 49, 1197–1202.
Paquay, Y.C., Ruijter, J.E., Waerden, J.P., Jansen, J.A., 1997. Wound
healing phenomena in titanium fibre mesh: the influence of
the length of implantation. Biomaterials 18, 161–166.
Singh, R., Lee, P.D., Lindley, T.C., Kohlhauser, C., Hellmich, C.,
Bram, M., Imwinkelried, T., Dashwood, R.J., 2010. Character-
ization of the deformation behavior of intermediate poros-
ity interconnected Ti foams using micro-computed tomog-
raphy and direct finite element modeling. Acta Biomater. 6,
2342–2351.
Stevens, M.M., 2008. Biomaterials for bone tissue engineering.
Mater. Today 11, 18–25.
Suquet, P.M., 1993. Overall potentials and extremal surfaces of
power law or ideally plastic composites. J. Mech. Phys. Solids
41, 981–1002.
Tan, Q.B., Liu, P., Du, C.L., Wu, L.H., He, G., 2009. Mechanical
behaviors of quasi-ordered entangled aluminum alloy wire
material. Mater. Sci. Eng. A 527, 38–44.
Taylor, R.P., McClain, S.T., Berry, J.T., 1999. Uncertainty analysis
of metal-casting porosity measurements using Archimedes’
principle. Int. J. Cast Met. Res. 11, 247–257.
 J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S
Traini, T., Mangano, C., Sammons, R.L., Mangano, F., Macchi,
A., Piattelli, A., 2008. Direct laser metal sintering as a new
approach to fabrication of an isoelastic functionally graded
material for manufacture of porous titanium dental implants.
Dent. Mater. 24, 1525–1533.
Vehof, J.W.M., Haus, M.T.U., Ruijter, A.E., Spauwen, P.H.M., Jansen,
J.A., 2002. Bone formation in transforming growth factor beta-
1-loaded titanium fiber mesh implants. Clin. Oral. Implan. Res.
13, 94–102.
Vehof, J.W.M., Spauwen, P.H.M., Jansen, J.A., 2000. Bone formation
in calcium-phosphate-coated titanium mesh. Biomaterials 21,
2003–2009.
5 (2012) 16–31 31
Wallach, J.C., Gibson, L.J., 2001. Mechanical behavior of a three-
dimensional truss material. Internat. J. Solids Struct. 38,
7181–7196.
Warren, W.E., Kraynik, A.M., 1991. The nonlinear elastic behavior
of open-cell foams. J. Appl. Mech. 58, 376–381.
Wen, C.E., Mabuchi, M., Yamada, Y., Shimojima, K., Chino, Y.,
Asahina, T., 2001. Processing of biocompatible porous Ti and
Mg. Scr. Mater. 45, 1147–1153.
Zhao, J., Lu, X., Weng, J., 2008. Macroporous Ti-based
composite scaffold prepared by polymer impregnating
method with calcium phosphate coatings. Mater. Lett. 62,
2921–2924.