Available online at www.sciencedirect.

com

Acta Biomaterialia 4 (2008) 1536–1544

www.elsevier.com/locate/actabiomat

Cellular Ti–6Al–4V structures with interconnected macro porosity

for bone implants fabricated by selective electron beam melting
Peter Heinl a,*, Lenka Müller b, Carolin Körner a, Robert F. Singer a, Frank A. Müller b,c
a
University of Erlangen-Nuremberg, Department of Materials Science – Institute of Science and Technology of Metals,
Martensstrasse 5, 91058 Erlangen, Germany
b
University of Erlangen-Nuremberg, Department of Materials Science – Biomaterials, Henkestrasse 91, 91054 Erlangen, Germany
c
Friedrich-Schiller – University of Jena, Institute of Materials Science and Technology (IMT), Loebdergraben 32, 07743 Jena, Germany

Received 29 October 2007; received in revised form 15 February 2008; accepted 20 March 2008
Available online 10 April 2008

Abstract

Selective electron beam melting (SEBM) was successfully used to fabricate novel cellular Ti–6Al–4V structures for orthopaedic appli-
cations. Micro computer tomography (lCT) analysis demonstrated the capability to fabricate three-dimensional structures with an inter-
connected porosity and pore sizes suitable for tissue ingrowth and vascularization. Mechanical properties, such as compressive strength
and elastic modulus, of the tested structures were similar to those of human bone. Thus, stress-shielding effects after implantation might
be avoided due to a reduced stiffness mismatch between implant and bone. A chemical surface modification using HCl and NaOH
induced apatite formation during in vitro bioactivity tests in simulated body fluid under dynamic conditions. The modified bioactive
surface is expected to enhance the fixation of the implant in the surrounding bone as well as to improve its long-term stability.
Ó 2008 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Keywords: Titanium alloys; Surface modification; Bioactivity; Selective electron beam melting; Porous structures

1. Introduction Bone consists of the outer cortical bone, a dense structure

Pure titanium and some of its alloys have been exten-
sively used as ‘‘load-bearing” implants for biomedical appli-
cations, due to their high strength-to-weight ratio, corrosion
resistance in the physiological environment, fatigue resis-
tance and low elastic modulus [1–3].
Several studies have demonstrated that a biomechanical
mismatch between an implant and the surrounding tissue
may lead to stress-shielding phenomena which induce an
unfavourable stress distribution at the bone–implant inter-
face, retarding both bone healing and remodelling [4,5].
Thus, a major goal within the biomedical community is
to develop new metallic biomaterials for load-bearing
orthopaedic and dental applications with an elastic modu-
lus similar to that of human bone.
*
Corresponding author. Tel.: +49 9131 85 27768; fax: +49 9131 85
27515.
E-mail address: peter.heinl@ww.uni-erlangen.de (P. Heinl).
with high mechanical strength, and the inner trabecular
bone, a network of struts enclosing large voids with 55–
70% interconnected porosity. Porous metals with an inter-
connected pore structure are of particular interest for
orthopaedic implant applications due to their potential abil-
ity to facilitate tissue ingrowth [6]. An interconnected pore
system with pore diameters in excess of 100 lm is required
for cell penetration, tissue ingrowth, vascularization and
nutrient delivery to the centre of the regenerating tissue
[7,8]. In addition, porous metals represent a promising
means of reducing stiffness mismatch and avoiding stress-
shielding effects [6]. Open-cell titanium structures can be
fabricated by a number of different techniques, e.g. con-
trolled sintering of powder preforms [5], solid-state foaming
by superplastic expansion of argon-filled pores [9] or poly-
meric foam replication [10]. However, these manufacturing
techniques have limitations concerning the control of the
outer shape and the pore structure. In addition, porous
titanium foams have been prepared from titanium powders

1742-7061/$ - see front matter Ó 2008 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.actbio.2008.03.013
 P. Heinl et al. / Acta Biomaterialia 4 (2008) 1536–1544
with help of space holders [11]. According to micro com-
puter tomography (lCT), this processing technique resulted
in porous structures with a large spread of pore sizes in the
range 200–500 lm [12]. Recently, Li et al. published a rapid
prototyping technique to create porous Ti–6Al–4V implants
with controlled size, pore shape and distribution using a
three-dimensional (3D) printing of slurries [13,14]. In this
case the preparation of the slurry and the shrinkage of the
porous structures after sintering seem to be critical steps.
Selective electron beam melting (SEBM) is a new addi-
tive manufacturing technique with high capability for the
fabrication of porous metals with well-defined cellular
structures [15]. The basic principle of this technology is
the generation of structures by the selective melting of dis-
crete powder layers by an electron beam under vacuum.
With this approach a huge variety of cellular metal struc-
tures with different densities and cell morphologies is feasi-
ble. Due to its high affinity for atmospheric gases such as
oxygen and nitrogen, fabrication under vacuum is of par-
ticular importance for titanium. The absorption of atmo-
spheric gases should be effectively prevented because this
might lead to a reduction in the ductility of titanium.
It is well established that the osseointegration process is
affected by surface modifications in terms of chemical and
physical properties [16]. Two directions have been reported
for titanium and its alloys in order to improve their bone-
bonding ability. Among coating techniques, plasma spray-
ing of hydroxyapatite is the most frequently used in clinical
practice. However, coating methods still encounter some
general problems, including the lack of adherence to the
substrate and non-uniformity of the layer thickness.
Another possibility to modify titanium surfaces is repre-
sented by chemical treatments in order to obtain OH-
groups that have been described as favourable to enhanced
osseointegration [17,18].
The aim of this study was to fabricate bioactive cellular
Ti–6Al–4V structures with a controllable interconnected
porosity by SEBM. The mechanical properties of these
structures were examined under compression load and
compared with the properties of human bone. Surface
modifications were performed by a wet chemical treatment
in HCl and NaOH, and the bioactivity of the thus prepared
structures was evaluated by testing the in vitro apatite for-
mation in dynamic simulated body fluid (SBF).
2. Materials and methods
2.1. Fabrication of cellular titanium structures
Samples of two different cellular titanium structures
with a nominal layer thickness of 100 lm were fabricated
in-house by SEBM on a machine supplied by a commercial
vendor (EBM S12, Arcam AB, Sweden).
In Fig. 1 the general procedure of the additive manufac-
turing by SEBM is depicted. Firstly, a 3D computer-aided
design (CAD) model is sliced into layers with constant
thicknesses to provide layer information. The SEBM pro-
1537
cess starts with the homogeneous application of a layer
of metal powder on a process platform. After a preheating
step, an electron beam scans the powder layer and creates a
cross-section of the part by fusing the loosely joined pow-
der particles. Subsequently, the process platform is lowered
by the thickness of one layer, a new powder layer is applied
and the process is repeated until the whole part has been
built. The process is performed under vacuum (10 4 to
10 5 mbar). Since the generation process takes place in a
powder bed, the metal powder that is not molten during
the process acts as a support material for the parts.
The two structures used in this study vary with respect
to manufacturing method, porosity and cell structure.
The first structure, named the diamond structure, is based
on the CAD model of diamond lattice, where each atom is
surrounded tetrahedrally by four other atoms. The second
structure, named the hatched structure, was generated by
scanning the powder layers with the electron beam in par-
allel lines with constant spacing (here 1.0 mm). The scan-
ning direction was altered by 90° every eighth layer. Due
to the small spot size of the beam, the molten lines do
not overlap and a porous 3D structure is produced. In this
case the cell structure is not determined by the CAD model,
which only defines the outer specimen dimensions, but by
the process parameters. Both manufacturing methods for
cellular titanium and the SEBM process in general are
explained in detail by Heinl et al. [15].
Spherical titanium powder of the alloy Ti–6Al–4V with
a mean particle size of 70 lm was used for the generation of
the samples. The chemical composition of the titanium
powder used and of the generated structures corresponds
to the specification ASTM F 1108 for Ti–6Al–4V alloy
castings for surgical implants. Two different sample geom-
etries were fabricated to achieve the requirements for the
conducted compression and bioactivity tests. The samples
for the compression tests were generated oversized, cleaned
by blasting with titanium powder and machined to cuboids
with coplanar faces and dimensions of 15  15  23 mm3.
The samples for the chemical surface modification and
the subsequent bioactivity tests were cubic with an edge
length of 10 mm. They were also cleaned by powder blast-
ing but not post-worked.
2.2. Chemical surface modification
Prior to chemical pretreatments the generated cellular
titanium samples were sonicated with isopropanol, then
rinsed with double-distilled water and dried in air. The sur-
face modification was performed by etching of the samples
in 37% HCl under an argon atmosphere for 90 min at 50 °C
and subsequently for 60 min at 40 °C. Afterwards all spec-
imens were soaked in 10 M NaOH aqueous solution at
60 °C for 24 h, washed with distilled water and dried at
100 °C [17].
Cellular titanium samples with and without treatment in
HCl and NaOH were immersed in a mixture of 1.6 M
ammonium chloride solution and 0.13 M zinc chloride
1538 P. Heinl et al. / Acta Biomaterialia 4 (2008) 1536–1544

Fig. 1. Schematic explaining the additive manufacturing by selective electron beam melting used in the present paper to generate titanium bodies with a
cellular structure.

Table 1
Comparison of the ionic concentrations in blood plasma and the theoretical one in SBF10 (mmol l 1)
Na+ K+ Mg2+ Ca2+ Cl HCO3 HPO24 SO24
Blood plasma 142.0 3.6–5.5 1.0 2.1–2.6 95.0–107.0 27.0 0.65–1.45 1.0
SBF10 142.0 5.0 1.0 2.5 126.0 10.0 1.0 1.0

Fig. 2. Schematic of the dynamic in vitro test used in the present paper.

solution at ambient temperature for 15 min in order to esti-
mate the amount of hydroxyl groups at the surface. During
immersion, active hydroxyl groups on the titanium surface
are exchanged for zinc ions in the ratio of Zn2+:OH = 1:2.
The rinsed and dried samples were subsequently immersed
in 2.42 M HNO3 to release the zinc ions [19]. The concen-
tration of zinc ions in the solution was measured using
inductively coupled plasma optical emission spectroscopy
(ICP-OES; dula, Spectro, Germany).
2.3. Bioactivity test
The in vitro bioactivity of chemically pretreated samples
was evaluated in terms of apatite-forming ability by soak-
ing the samples in SBF10 for 6 days. SBF10 solution with
ionic concentrations nearly equal to human blood plasma
was prepared according to Ref. [20]. Briefly, concentrated
solutions of KCl, NaCl, NaHCO3, MgSO4
7H2O, CaCl2
and KH2PO4 were pipetted into double-distilled water
and buffered with tris-hydroxymethyl aminomethane and
HCl to pH 7.4 at 37 °C. Sodium azide (NaN3) was added
to the solution to inhibit the growth of bacteria. The com-
position of the applied SBF10 solution compared to that
of the inorganic part of human blood plasma is given in
Table 1. The samples were exposed to SBF10 under
dynamic conditions as illustrated in Fig. 2. The sample
chamber with a volume of 6 ml and 1 l polyethylene storage
container with SBF10 were placed in a waterbath at
37 ± 0.1 °C for 6 days. A peristaltic pump controlled the
flow rate (0.3 ml min 1) of SBF10 from the storage tank
 P. Heinl et al. / Acta Biomaterialia 4 (2008) 1536–1544
through the sample chamber. The SBF10 storage container
was equipped with a magnetic stirrer in order to guarantee
a homogeneous concentration throughout the solution
volume.
2.4. Characterization
After each processing step, starting with the generated
and cleaned samples up to soaking in SBF10, the structures
were analysed with regard to changes in the structural
design and in particular the surface by scanning electron
microscopy equipped with EDX analysis (SEM/EDX, Phi-
lips XL 30, France). In addition, structural investigations
using micro computer tomography (lCT; MicroCT 40,
Scanco Medical, Swiss) were conducted prior to the chem-
ical treatments to determine the mean pore size and the
porosity of the structures. The resolution was 20 lm.
Raman spectra were collected on the samples before
soaking in SBF10 in the range from 1200 to 200 cm 1 using
a dispersive Raman spectrometer (Nicolet Almega
Thermo, USA) equipped with a 780 nm laser.
Fourier transform infrared (FT-IR) spectra were mea-
sured in transmission using the KBr technique in the range
from 4000 to 400 cm 1 at a resolution of 4 cm 1 (Impact
420, Nicolet Instruments, USA). Approximately 1 mg of
the precipitate on the sample surfaces was removed from
the substrate after 6 days in SBF10, mixed with 300 mg
of dry KBr powder and ground using an agate mortar
and pestle. The resulting mixture was pressed into trans-
parent pellets with a diameter of 13 mm applying a force
of 105 N (Perkin Elmer Hydraulic Press, Germany).
Compression tests were conducted to determine the
mechanical properties of the untreated cellular titanium
structures. For both structure types the tests were carried
out with the loading direction parallel as well as perpendic-
ular to the building direction and with five specimens
per loading direction. The tests were performed using a
universal testing machine (Inspekt Retrofit 100, Hegewald
and Peschke, Germany) with a crosshead speed of
0.5 mm min 1, where the compressive strain was measured
with an extensometer. The elastic modulus E was calcu-
lated from the slope of the compressive stress–strain curve
in the linear elastic region. The compressive yield strength
ry, 0.2 was determined from the stress–strain curve accord-
ing to the 0.2% offset method, while the maximum strength
rmax was calculated by dividing the highest load sustained
by the specimen before fracture by the initial cross-sec-
tional area (ASTM E9). The porosity of the samples was
evaluated from the weight and the apparent volume of
the specimens. All measurements were carried out at room
temperature (20 ± 2 °C).
3. Results and discussion
Fig. 3 shows reconstructed 3D images from lCT mea-
surements and micrographs from SEM examination of
the investigated structures in the untreated state after fab-
1539
rication. Due to the layer-by-layer generation process, the
appearance of the structures differs in top and lateral view.
Both structures exhibit an interconnected porosity, which is
one of the most important requirements for tissue
ingrowth. For the diamond structure the tetrahedron-like
arrangement of the struts is visible. In contrast, the hatched
structure is characterized by parallel and orthogonal
arranged straight struts. lCT measurements give a porosity
of 80.5% and a mean pore size of 1.23 mm for the diamond
structure, and 61.3% and 0.45 mm for the hatched struc-
ture, respectively. Thus, the structures exhibit porosities
comparable to that of trabecular bone.
The results of the compression tests of the untreated
titanium structures are summarized in Table 2. As expected
for cellular solids, the relative density and the porosity of
the structures, respectively, have an important influence
on their stiffness and strength [24]. Due to its lower poros-
ity, the hatched structure exhibits enhanced mechanical
properties. Because of the generation of the structures
layer-by-layer the mechanical properties are anisotropic.
Thus, the loading direction has a strong influence on the
structure properties. Both structure types reach consider-
ably higher stiffness and strength values if loaded parallel
to the building direction and perpendicular to the layers,
respectively.
Comparing the mechanical properties of the structures
and human bone, the diamond structure exhibits properties
comparable with trabecular bone, whereas the hatched
structure’s properties lie in between those of trabecular
and cortical bone. It should be mentioned that the mechan-
ical properties of bone depend on numerous factors, such
as the location where the sample was taken, the age of
the bone tissue, the loading direction or whether the spec-
imen was wet or dry stored [25]. Hence, the properties vary
over a wide range depending on the study considered, espe-
cially for trabecular bone. For this reason the listed values
for trabecular and cortical bone in Table 2 should only rep-
resent the dimension of a bone’s properties. Nevertheless,
the comparison shows that the elastic moduli of bone
and of cellular Ti–6Al–4V processed by SEBM are in a
similar range, which should minimize the stress-shielding
effect. This is a very important result with respect to the
application of cellular titanium structures as biomaterial
for bone replacement. The SEBM technology exhibits the
possibility to adjust stiffness and strength by varying the
porosity of the structure over a wide rage, whereas stiffness
and strength are strongly connected [15]. Thus, the
mechanical properties of cellular Ti–6Al–4V processed by
SEBM can be tailored for different applications in the field
of implantology.
The chemical surface modification and the bioactivity
tests were carried out for the hatched and the diamond
structures. In spite of considerable differences between
these two structures in porosity and pore size no significant
differences among both structures were observable. Thus,
the influence of the different chemical treatment steps and
the results of the bioactivity tests will be discussed for the
1540 P. Heinl et al. / Acta Biomaterialia 4 (2008) 1536–1544

Fig. 3. 3D image from lCT measurement of Ti–6Al–4V diamond and hatched structure with the corresponding SEM micrographs in top and lateral view.

Table 2
Mechanical properties of the untreated cellular Ti–6Al–4V structures tested in compression with the loading direction parallel and perpendicular to the
building direction of the specimens, where E is elastic modulus, ry, 0.2 the yield strength, rmax the maximum strength (standard deviation in parentheses)
Material Loading direction E (GPa) ry, 0.2 (MPa) rmax (MPa) Porosity (%)
Diamond structure Parallel 1.6 (±0.3) 22.0 (±1.1) 29.3 (±0.8) 81.1 (±0.4)
Perpendicular 0.9 (±0.1) 16.1 (±0.4) 21.0 (±0.7) 80.8 (±0.3)
Hatched structure Parallel 12.9 (±0.9) 107.5 (±3.6) 148.4 (±3.5) 59.5 (±0.6)
Perpendicular 3.9 (±2.1) 49.6 (±20.6) 127.1 (±29.2) 59.5 (±0.6)
Trabecular bone Unknown 1.08 (±0.86)a 15.2 (±8.0)a 25.0 (±8.1)a –
Cortical bone (human femur) Paralleld 18.2 (±0.85)b 181 (±13.1)c 205 (±12.6)c –
Perpendiculare 11.7 (±1.01)b – 131 (±20.7)b –
a
Ref. [21].
b
Ref. [22].
c
Ref. [23].
d
Loading direction parallel to the long axis of the bone.
e
Loading direction perpendicular to the long axis of the bone.

hatched structure as an exemplar. Fig. 4 shows the appear- eral view. The SEM micrographs evince the layer-by-layer
ance of the hatched structure in lateral view after each pro- generation of the structures in a powder bed. The surface in
cessing step, starting with the untreated state at different the untreated state appears corrugated with adherent,
magnifications. The building direction is parallel to the lat- partly molten powder particles (Fig. 4a). A comparison
 P. Heinl et al. / Acta Biomaterialia 4 (2008) 1536–1544 1541

Fig. 4. SEM micrographs of Ti–6Al–4V hatched structure in lateral view: (a) starting state, (b) HCl treated, (c) HCl and subsequently NaOH treated, (d)
after 6 days in SBF10; inset 1 and 2, corresponding EDX analyses.

of the different treatment steps at the lowest magnification Treatment in HCl causes corrosion of the surface, lead-
shows that the cell morphology itself is not influenced by ing to small craters with a mean diameter of 5–10 lm and
the chemical treatments. sharp edges (Fig. 4b). In addition, the amount of adherent
1542 P. Heinl et al. / Acta Biomaterialia 4 (2008) 1536–1544
powder particles is considerably reduced. As a consequence
of this treatment step the thin passive titanium oxide layer,
which normally covers the titanium surface, was dissolved
and a titanium hydride interface layer was formed. This
layer is recovered with titanium oxide due to exposure with
air moisture [17].
After treatment with NaOH aqueous solution the crater-
like appearance still remains but with rounded and fattened
edges (Fig. 4c). The EDX analysis of the alkali-modified
surface shows titanium and the alloying elements alumin-
ium and vanadium, and in addition sodium (Fig. 4; inset
1). Weak and broad vibrations were detected in the spectra
of the Raman analysis (Fig. 5), which support the observa-
tion that the acid-etched surface exposed to a subsequent
alkali treatment is not well crystalline but nearly amor-
phous. However, the bands centred at 276, 442 and
693 cm 1 could be associated with sodium hexatitanate
Na2Ti6O13, showing the most intensive vibrations at 277,
460 and 685 cm 1 [26]. Sol–gel derived Na2Ti6O13 struc-
tures have recently been described to induce apatite forma-
tion [27].
After dynamic bioactivity tests in SBF10 solution for
6 days a new layer is partially visible on the surface
(Fig. 4d). A representative EDX analysis of this layer is
shown as inset 2 in Fig. 4. Besides titanium, magnesium,
calcium, sodium and phosphorus were detected. The phase
composition of the layer was analyzed in detail using FT-
IR (Fig. 6). The spectra shows that the layer is composed
of carbonated hydroxyapatite (CHA), which is supported
by the presence of the bands originating from stretching
vibrations of CO23 at 1450 cm 1 [28]. A bending mode
at 1645 cm 1 originating from H2O could be observed in
the spectra of the precipitated CHA films. A broad phos-
phate band associated with the P–O asymmetric stretching
mode (m3) of the (PO4)3 group was found in the region
Raman Intensity
Ti-6Al-4V
Ti-6Al-4V+HCl+NaOH
1200 1000
Fig. 5. Raman analysis of chemically pretreated Ti–6Al–4V hatched structure.
from 1200 to 960 cm 1, indicating a deviation of the phos-
phate ions from their ideal tetrahedral structure [29]. The
bending modes of the O–P–O bonds in the phosphates were
found in the spectral range from 603 to 565 cm 1. It is pos-
sible that HPO24 is also incorporated in the crystal lattice,
due to very small vibrations at 958 and 875 cm 1. How-
ever, the band detected at 875 cm 1 could also be assigned
to the CO23 group in carbonated apatite.
The detected CHA layer is also visible in the inner part
of the structure, which is shown in Fig. 7. From this result
it can be supposed that the chemical treatment was success-
ful throughout the whole structure. The CHA layer that
forms demonstrates that due to the chemical pretreatment
used here it is possible to provide cellular Ti–6Al–4V struc-
tures with bioactive behaviour.
It was described that the OH groups, which are always
present on TiO2 surfaces in an aqueous environment, could
assist the induction of calcium phosphates from supersatu-
rated solutions [30]. Therefore, amounts of active hydroxyl
groups on Ti–6Al–4V were estimated. Results showed that
the density of OH groups on chemically treated Ti–6Al–4V
(17.48 ± 2.22 (OH) nm 2) significantly exceeds that one
of non-treated surfaces (1.55 ± 0.03 (OH) nm 2). Thus,
a negatively charged surface was produced by samples
placed in a NaOH solution. According to the experimental
findings reported previously by Zhu et al. [31], solution
conditions greatly influence the formation of calcium phos-
phate (CaP) on charged surfaces. In stable supersaturated
solutions (pH < 7.5), negatively charged surfaces were
shown to have a more powerful induction capability for
the heterogeneous nucleation of CaP. Furthermore, the
accumulation of Ca2+ near the liquid/solid interface, as a
result of electrostatic attraction, increases supersaturation
with respect to CaP and thus the initial nucleation is
triggered.
800 600 400 200
-1
Raman Shift (cm )
 P. Heinl et al. / Acta Biomaterialia 4 (2008) 1536–1544 1543

3-
(PO4)

3-
(PO4)
H2O

Transmission %

2-
(CO3)
H2O

2- 2-
(CO3) (HPO4)

4000 3600 3200 2800 2400 2000 1600 1200 800 400
-1
Wavenumber (cm )

Fig. 6. FT-IR analysis of chemically pretreated Ti–6Al–4V hatched structure soaked for 6 days in SBF10. The characteristic vibrations are assigned to the
respective groups.

Fig. 7. SEM micrographs of a cross-section from the middle of a chemically pretreated Ti–6Al–4V hatched structure soaked for 6 days in SBF10.

4. Conclusions [2] Okazaki Y, Nishimura E, Nakada E, Kobayashi K. Surface analysis

Cellular Ti–6Al–4V structures with interconnected
porosity suitable for bone ingrowth were manufactured
by SEBM. The mechanical properties of these cellular
structures were similar to those of human bone, which
might minimize stress-shielding effects. The suggested
chemical surface modification using HCl and NaOH
induce in vitro apatite formation and thus is believed to
provide a better fixation of the implant in the surrounding
bone and improve the long-term stability of the implant.
Acknowledgement
Deutsche Forschungsgemeinschaft (DFG) is thankfully
acknowledged for the financial support.
References
[1] Okazaki Y, Rao S, Ito Y, Tateishi T. Corrosion resistance,
mechanical properties and cytocompatibility of new Ti alloys without
Al and V. Biomaterials 1998;19:1197–215.
of Ti–15Zr–4Nb–4Ta alloy after implantation in rat tibia. Biomate-
rials 2001;22:599–607.
[3] Eisenbarth E, Velten D, Müller M, Thull R, Breme J. Biocompat-
ibility of b-stabilizing elements of titanium alloys. Biomaterials
2004;25:5705–13.
[4] Moyen BJ, Lahey Jr PJ, Weinberg EH, Harris WH. Effects on intact
femora of dogs of the application and removal of metal plates. A
metabolic and structural study comparing stiffer and more flexible
plates. J Bone Joint Surg Am 1978;60:940–7.
[5] Oh IK, Nomura N, Hanada S. Mechanical properties of porous
titanium compacts prepared by powder sintering. Scripta Mater 2003;
49:1197–202.
[6] Ryan G, Pandit A, Apatsidis DP. Fabrication methods of porous metals
for use in orthopaedic applications. Biomaterials 2006;27:2651–70.
[7] van Blitterswijk A, Grote JJ, Kuijpers W, Daems WT, de Groot K.
Macropore tissue ingrowth: a quantitative and qualitative study on
hydroxyapatite ceramic. Biomaterials 1986;7:137–43.
[8] Schliephake H, Neukam FW, Klosa D. Influence of pore dimensions on
bone ingrowth into porous hydroxylapatite blocks used as bone graft
substitutes. A histometric study. Int J Oral Maxillofac Surg 1991;20:53–8.
[9] Davis NG, Teisen J, Schuh C, Dunand DC. Solid-state foaming of
titanium by superplastic expansion of argon-filled pores. J Mater Res
2001;16:1508–19.
[10] Li JP, Li SH, de Groot K, Layrolle P. Preparation and character-
ization of porous titanium. Key Eng Mater 2002;218:51–4.
1544 P. Heinl et al. / Acta Biomaterialia 4 (2008) 1536–1544
[11] Wen CE, Mabuchi M, Yamada Y, Shimojima K, Chino Y, Asahina
T. Processing of biocompatible porous Ti and Mg. Scripta Mater
2001;45:1147–53.
[12] Otsuki B, Takemoto M, Fujibayashi S, Neo M, Kokubo T,
Nakamura T. 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 2006;27:5892–900.
[13] Li JP, de Wijn JR, van Blitterswijk CA, de Groot K. Porous Ti6Al4V
scaffold directly fabricating by rapid prototyping: preparation and
in vitro experiment. Biomaterials 2006;27:1223–35.
[14] Li JP, Habibovic P, van de Doel M, Wilson CE, de Wijn JR, van
Blitterswijk CA, et al. Bone ingrowth in porous titanium implants
produced by 3D fiber deposition. Biomaterials 2007;28:2810–20.
[15] Heinl P, Rottmair A, Körner C, Singer RF. Cellular titanium by
selective electron beam melting. Adv Eng Mater 2007;9:360–4.
[16] Lemons JE. Biomaterials, biomechanics, tissue healing, and immedi-
ate-function dental implants. J Oral Implantol 2004;30:318–24.
[17] Jonasova L, Müller FA, Helebrant A, Strnad J, Greil P. Biomimetic
apatite formation on chemically treated titanium. Biomaterials
2004;25:1187–94.
[18] Kokubo T. Formation of biologically active bone-like apatite on
metals and polymers by a biomimetic process. Thermochim Acta
1996;280/281:479–90.
[19] Hanawa T, Kon M, Doi H, Ukai H, Murakami K, Hamanaka H,
et al. Amounts of hydroxyl radical on calcium-ion-implanted
titanium and point of zero charge of constituent oxide of the
surface-modified layer. J Mat Sci: Mat Med 1998;9:89–92.
[20] Müller L, Müller FA. Preparation of SBF with different HCO3
content and its influence on the composition of biomimetic apatites.
Acta Biomater 2006;2:181–9.
[21] Sevilla P, Aparicio C, Planell JA, Gil FJ. Comparison of the
mechanical properties between tantalum and nickel-titanium foams
implant materials for bone ingrowth applications. J Alloys Com-
pounds 2007;439:67–73.
[22] Reilly DT, Burstein AH. The elastic and ultimate properties of
compact bone tissue. J Biomech 1975;8:393–405.
[23] Cezayirlioglu H, Bahniuk E, Davy DT, Heiple KG. Anisotropic yield
behavior of bone under combined axial force and torque. J Biomech
1985;18:61–9.
[24] Gibson LJ, Ashby MF. Cellular solids – structure and proper-
ties. 2nd ed. Cambridge: Cambridge University Press; 1997, p.
175–234.
[25] Currey JD. Bone – structure and mechanics. Princeton and
Oxford: Princeton University Press; 2002, p. 54–62.
[26] Bamberger CA, Begun GM. Sodium titanates: stoichiometry and
Raman spectra. J Am Ceram Soc 1987;70:C-48–51.
[27] Becker I, Hofmann I, Müller FA. Preparation of bioactive sodium
titanate ceramics. J Eur Cer Soc 2007;27:4547–53.
[28] Koutsopoulus S. Synthesis and characterization of hydroxyapatite
crystals: a review study on the analytical methods. J Biomed Mat Res
2002;62:600–12.
[29] Stoch A, Jastrzebski W, Brozek A, Stoch J, Szatraniec J, Trybalska B,
et al. FTIR absorption–reflection study of biomimetic growth
of phosphates on titanium implants. J Mol Struct 2000;555:375–
82.
[30] Van Kemenade KJJM, de Bruyn PL. A kinetic study of precipitation
from supersaturated calcium phosphate solutions. J Colloid Interf Sci
1987;118:564–85.
[31] Zhu P, Masuda Y, Yonezawa T, Koumoto K. Investigation of apatite
deposition onto charged surfaces in aqueous solutions using a quartz-
crystal microbalance. J Am Ceram Soc 2003;86:782–90.