Content uploaded by Martin Balog
Author content
All content in this area was uploaded by Martin Balog on Oct 12, 2018
Content may be subject to copyright.
1
Bioactive Ti + Mg composites fabricated by powder metallurgy: the relation between the
microstructure and mechanical properties.
Martin Balog1,*, Ahmed Mohamed Hassan Ibrahim1,2, Peter Krizik1, Oto Bajana1, Alena
Klimova1, Amir Catic3, Zdravko Schauperl4
1Institute of materials and machine mechanics, The Slovak academy of sciences, Dubravska cesta
9, 84513 Bratislava, Slovakia
2Slovak University of Technology, Faculty of Materials Science and Technology, Jana Bottu
2781/25, 917 24 Trnava, Slovakia
3School of dental medicine, University of Zagreb, Gunduliceva 5, 10000 Zagreb, Croatia
4Faculty of mechanical engineering and naval architecture, University of Zagreb, Ivana Lucica 5,
10000 Zagreb, Croatia
* corresponding author: martin.balog@savba.sk
Abstract
Metallic implant materials are biomaterials that have experienced major development over the
last fifty years, yet some demands posed to them have not been addressed. For the
osseointegration process and the outcome of endosseous implantation, it is crucial to reduce the
stress shielding effect and achieve sufficient biocompatibility. Powder metallurgy (PM) was
utilized in this study to fabricate a new type of titanium (Ti) + magnesium (Mg) bioactive
composite to enable stress-shielding reduction and obtain better biocompatibility compared with
that of the traditional Ti and Ti alloys used for dental implants. Such composites are produced by
well-known cost-effective and widely used PM methods, which eliminate the need for complex
and costly Ti casting used in traditional implant production. The relation between the
microstructure and mechanical properties of as-extruded Ti + (0 - 24) vol.% Mg composites was
investigated with respect to the Mg content. The microstructure of the composites consisted of a
biodegradable Mg component in the form of filaments, elongated along the direction of extrusion,
which were embedded within a permanent, bioinert Ti matrix. As the Mg content was increased,
the discrete filaments became interconnected with each other and formed a continuous Mg
network. Young`s modulus (E) of the composites was reduced to 81 GPa, while other tensile
mechanical properties were maintained at the values required for a dental implant material. The
corrosion behavior of the Ti + Mg composites was studied during immersion in a Hank's
balanced salt solution (HBSS) for up to 21 days. The elution of Mg pores formed at former Mg
sites led to a further decrease of E to 74 GPa. The studied compositions showed that a new Ti +
Mg metallic composite should be promising for load-bearing applications in endosseous dental
implants in the future.
Keywords
bioactive; composite; dental implant; magnesium; powder metallurgy; titanium;
2
1. Introduction
Increase in human life expectancy has led to increasing demand for different types of bone
implantation surgical procedures in diverse biomedical areas ranging from dentistry to
orthopedics. Although commercially available pure titanium (CP Ti) implants show good clinical
performance, there are two issues that have not been fully addressed: i) the stress-shielding effect,
and ii) the insufficient bioactivity of the implants [1]. The mechanical compatibility of metallic
implants with bones is ensured to a large extent by the similarity of an implant material to the
bone stiffness [2 - 5]. Many studies investigated mechanical conditions such as duration, intensity
and type of loading that showed to have a crucial influence on the formation of the bone-to-
implant bond [6]. Dynamic loading had the strongest impact on the growth of a tissue at the
implantation site [7]. Many recent studies [8 - 15] focused on finding out how Young’s modulus
(E) of Ti implants of 105 GPa could be lowered while preserving sufficient values of other
mechanical properties, such as the fatigue strength, tensile strength, hardness, and wear
resistance. The studies were mostly oriented on low-E β-type Ti alloys. With the addition of non-
toxic stabilizers, including Nb, Zr, Ta, Sn, and Mo, the monophasic alloys, such as TiMo6Zr2Fe,
Ti15Mo5ZrAl, Ti13Nb13Zr, exhibited quite dramatic stiffness reduction of E values between 60
and 85 GPa [16, 17]. However, the precious alloying elements are quite costly and they have high
melting points [17, 18]. It was also shown that structures made from the β-type alloys were often
inhomogeneous. To overcome implant-bone elasticity mismatch difficulties, one can take another
approach: introduce porosity to an implanted Ti-based material. Porous Ti not only showed to
have a beneficial impact on the stress shielding effect, but it also contributed to bone ingrowth
and attachment to the surface of implants. However, as residual porosity leads to a substantial
deterioration of other mechanical properties, such porous implant materials have so far been
applied only under low loading conditions [19].
Magnesium (Mg) is a naturally occurring metallic element in the human body, and it is relatively
affordable. It was researched in recent years as the basis for new types of biodegradable implants
[20, 21]. Along with its very good biodegradation potential, Mg has a low E of 45 GPa, which is
very similar to that of the bones [22]. However, Mg implants suffer from the difficulty to control
their degradation rate after implantation. The implants corrode fast due to chloride ions that are
present at high concentrations in the human body. The corrosion inhibits full recovery from
injuries [20]. Also, as pure Mg undergoes corrosion, toxic products are released, eventually
leading to the necrosis of the surrounding tissue [23]. Thus, previous studies on Mg as an implant
material concentrated on Mg corrosion rates under exposure to bodily fluids and how the
corrosion could be controlled [24]. A few studies were aimed at boosting the bioactivity of Ti and
Ti alloy implants by finishing their surface with Mg or Mg alloys [25].
Conventional metallurgical methods are ineffective in the case of Ti, as it is highly reactive at
elevated temperatures [26]. Moreover, Ti and Ti alloys are classified as difficult-to-machine
materials [27]. Therefore, there is growing interest in powder metallurgy (PM) as cost-effective
technology for the production of complex Ti parts with improved chemical and microstructural
3
homogeneity [28]. The press-and-sinter approach is economically the most attractive approach to
the production of near-to-shape parts, but it results in a relatively high residual porosity [29]. By
contrast, CP Ti, fabricated by an economic PM approach realized by the warm extrusion of a Ti
powder, showed mechanical properties and fatigue life superior to the Ti Grade 4 reference
material, which is commonly used in dentistry for prosthodontics parts and implants [30].
This motivated the idea to produce a two-phase Ti + Mg material that would selectively exploit
the advantages of both biometals. While Ti would serve as the matrix material providing the
implant with strength, Mg would bring about the reduction of E thus minimizing the stress-
shielding effect during loading, and moreover, gradient porosity would form with time as the
result of selective Mg dilution into the surface and volume of the implant. The dilution of the Mg
phase should improve the osseointegration process thanks to the porosity. Some pioneering
studies on Ti + Mg composites with reduced E were published [31 – 34]. On the one hand, the
authors described the benefits of such composites in terms of minimizing the stress-shielding
effect and improving osseointegration by their application as scaffold implants. On the other
hand, the composites showed rather poor mechanical properties, while only compressive modulus
and strength were reported. Such Ti + Mg composites do not even reach the challenging
mechanical properties needed for dental implants applications, namely the appropriate function
without deterioration under expected clinical loading conditions. In our leading paper, we
introduced the promising potential of a novel Ti + 12 vol.% Mg composite that exhibited
mechanical properties meeting all demands on dental implants including the dynamic fatigue test.
The composite was fabricated by a warm PM consolidation approach [35]. This work is devoted
to a systematic study on the development of the Ti + Mg composite and optimization of its
microstructure and mechanical properties.
2. Experimental
To achieve Ti + Mg composites with homogeneous microstructure, we used a narrow particle
size distribution fraction of Ti and Mg powders. The raw materials used in the experiment were
two size fractions of hydrogenation–dehydrogenation (HDH)-processed Ti 99.4 wt.% powder,
labeled as coarse Ti(C) and fine Ti(F), and one size fraction of gas atomized Mg 99.8 wt.%
powder (Tab. 1 and Fig. 1). The particle size distributions of Ti and Mg powders were
determined by a FRITSCH Analysette 22 laser diffraction system. The O content of the loose Ti
powders was determined by gas fusion analysis (GFA) using a LECO OHN836 analyzer.
Simultaneous differential scanning calorimetry (DSC) and thermal gravimetric analysis (TGA) of
the loose Ti and Mg powders were conducted in the air, using a Q600 TA Instrument at a heating
rate of 5 °C·min-1. The mixtures of the Ti + (12, 17 and 24) vol.% Mg powders were
homogenized in a Turbula shaker for 30 min. The powder mixtures were cold compacted by cold
isostatic pressing (CIP) at 200 MPa. CIP blanks were then pressed into Cu sleeves with a wall
thickness of 3 mm by uniaxial hot vacuum pressing (HVP) at a pressure of 500 MPa, temperature
of 410 °C, and vacuum of 8 Pa. The HVP blanks including the Cu sleeves were finally
consolidated by direct extrusion (DE) into rods with a diameter of 7.5 mm using a flat face die at
4
a reduction ratio of 16:1. The extrusion temperature was varied from 400 to 500 °C depending on
the Mg content and Ti powder fraction size. The ram speed during the extrusion was set to 10
mm.s-1.
Table 1 Properties of the HDH coarse Ti(C) and fine Ti(F), and atomized Mg powders.
powder
size fraction
μm
d50
μm
O
wt.%
Ti(C)
63 - 150
76
0.182 ± 0.008
Ti(F)
20 - 40
26
0.477 ± 0.003
Mg
20 - 40
30
-
Fig. 1 SEM images of feedstock Ti coarse (a), Ti fine (b), and Mg (c) powders.
The density of the Ti + Mg composites was measured using Archimedes’ principle according to
the ASTM B962-08 standard. The microstructure of the powders and composites was
characterized using scanning electron microscopy (SEM, JEOL 7500 machine) equipped with an
energy dispersive X-ray spectrometer (EDS, Oxford). Tensile specimens with a gauge diameter
of 3 mm and a gauge length of 21 mm were machined along a longitudinal direction of the
extruded Ti + Mg composites. Tensile properties were evaluated using a Zwick Roell 1474
machine at a strain rate of ~5·10-4 s-1 according to the ASTM E8 standard. The E of the Ti + Mg
a
b
a
b
100 µm
c
100 µm
100 µm
5
composites was measured by dynamic mechanical analysis (DMA, TAQ800 machine) using 4 ×
2.5 × 55 mm3 bars and the three-point bending method. The dimensions of the DMA specimens
were measured with an accuracy of 0.005 mm. The DMA testing was conducted at a room
temperature at a frequency of 1 Hz, preload force of 0.5 N, and sinusoidal force amplitude of 7 N
for 4 min. The E was measured using two samples, while each sample was measured four times
on each side.
To evaluate the degradation rate of the Ti + Mg composites, the DMA and tensile specimens
were immersed into a Hank's balanced salt solution (HBSS) for various times extending from 1 to
21 days. The chemical composition of the HBSS is summarized in Tab. 2. The temperature and
pH of the HBSS were maintained constant at 37 °C and 7.4. The HBSS liquid was changed every
3 days. The ratio between the exposed surface area of a specimen and a liquid volume of 0.2 –
0.4 ml·mm-2 was outlined by the ASTM G31–72 standard for determining the mass loss of
structural metals. Before being weighed, the diluted samples were cleansed in a chromate acid
water solution (200 g·L-1 CrO3) for 1 min to remove the corrosion products. They were then
ultrasonically cleaned in deionized water for 1 min and dried in an oven at 37 °C for 10 min. The
degradation rate (DR) of the diluted specimens was calculated following:
𝐷𝑅 = 𝑊
𝐴∙𝑡 (1)
, where W is the weight loss, A is the exposed area, and t is the exposure time. The mechanical
properties of the diluted specimens were determined by DMA and tensile testing, and the surface
was characterized by SEM and EDS. The corrosion product was evaluated by ray diffraction
(XRD, Philips X‘Pert machine).
Table 2 Chemical composition of a Hank's balanced salt solution.
component
NaCl
KCl
CaCl2
MgSO4.
7H2O
MgCl2.
6H2O
Na2HPO4
KH2PO4
glucose
NaHCO3
fraction, mg·L-1
8000
400
140
100
100
60
60
1000
350
3. Results and discussion
Highly reactive Ti and Mg elements, especially when present in the form of discrete powders, are
prone to oxidation at relatively low temperatures. The pronounced oxidation of Ti and Mg
powders may prevent a successful consolidation, and hence it may be detrimental to the
mechanical properties of a final composite material. Furthermore, O as an impurity deteriorates
the elongation of metallic Ti. Simultaneous DSC - TGA analysis performed on the as-received Ti
and Mg powders in the air confirmed the onsets of oxidation at ~550 and ~480 °C, respectively
(Fig. 2). In a powdered state, Ti and Mg can possibly react with an ambient atmosphere only
during the HVP step. In addition to the HVP being realized in the vacuum, a maximum
temperature during the HVP reaches only 410 °C. Moreover, during the DE, a reaction with the
6
ambient is restricted as already quite dense Ti + Mg precursors with a limited porosity pressed
into Cu sleeve are processed. As confirmed by GFA, the O content in the loose coarse and fine Ti
powders was 0.182 ± 0.008 and 0.477 ± 0.003 wt.%, respectively (Tab. 1). The content of O in
as-extruded Ti(C) and Ti(F) was maintained, and it increased only marginally to 0.193 ± 0.001
and 0.576 ± 0.003 wt.%, respectively. Due to the GFA principle determination of O in Mg
powder and Ti + Mg powder mixtures was not feasible. Nevertheless, it all suggests that
oxidation of Ti and Mg components is suppressed to a large extent during the entire processing.
The level of O determined in Ti(C) was within the ASTM B265 and B988 limit (<0.4 wt.%)
defined for the cast and PM Ti Grade 4, respectively.
Fig. 2 Simultaneous DSC-TGA curves of the as-received Ti (a) and Mg (b) powders upon heating in the air. For DSC
curves, the exothermic reaction goes upwards.
The powder metallurgy consolidation approach resulted in sound and well-compacted Ti(C),
Ti(F) and Ti(C) + (12, 17 and 24) vol.% Mg profiles. The maximum breakthrough pressure
a
b
7
determined with the laboratory scale DE of the Ti(C) + Mg composites was ~790 MPa. As the
pressure limit of industrial extrusion presses is typically at ~900 MPa, a future up-scaling of the
applied consolidation process is feasible. Ti(C), Ti(F) and Ti(C) + (12, 17 and 24) vol.% Mg
profiles had an apparent density higher than 99.2% of the theoretical density. A porosity of 0.8%
and less was fully acceptable assuming that the extruded materials were manufactured by a PM
approach and that the porosity of the composites would increase later due to Mg dilution upon
implantation. Mg powder particles deformed severely during extrusion by significantly stronger
Ti powder particles with higher deformation resistance. The microstructure of the as-extruded
composites was characteristic of Mg filaments embedded in a Ti matrix (Fig. 3). The longitudinal
cross-sectional images showed the Mg filaments arrayed along the direction of extrusion, while
the images in the transversal direction confirmed a homogenous dispersion of the Mg component
in the Ti matrix. No distinct formation of Mg conglomerates was observed for all composites. On
the one hand, interconnection between neighboring Mg filaments increased significantly as the
Mg content increased. For Ti(C) + 24 vol.% Mg, one can refer to a continuous Mg network in the
Ti matrix. On the other hand, the aspect size ratio of the Mg filaments decreased, and they
become thicker as the Mg content increased. The coefficient of thermal expansion (CTE) of Mg
is approximately three times higher than that of Ti. Rather large CTE mismatch could lead to
disintegration and formation of porosity at Ti – Mg interfaces during cooling after extrusion due
to a pronounced thermal shrinkage of the Mg phase. Nevertheless, a detailed microstructural
characterization revealed that the interfaces were generally free of porosity and voids (Fig. 4).
EDS line analyses confirmed Ti – Mg interfaces which were essentially not enriched in O (Fig.
5). This supports GFA data on the O content proving that an intense formation of oxide layers on
the Ti and Mg powders was avoided during the entire consolidation process.
Conversely, selected processing parameters resulted in imperfect Ti(F) + Mg composites with hot
surface cracks. Since sound Mg free Ti(F) profiles were fabricated and had reasonable
mechanical properties, as shown in the next paragraph, the formation of hot cracks points to an
improper size ratio of Ti and Mg powders. A difference in deformation resistance between the
fine Ti and Mg powders became more pronounced, which prevented a successful consolidation of
the Ti + Mg mixtures.
8
Fig. 3 SEM micrographs of as-extruded Ti(C) + 12 vol.% Mg in the longitudinal (a) and transversal (b)
directions, Ti(C) + 17 vol.% Mg in the longitudinal (c) and transversal (d) directions, and Ti(C) + 24 vol.% Mg in
the longitudinal (e) and transversal (f) directions. The red arrow indicates the extrusion direction.
Fig. 4 EDS mapping of Ti and Mg elements performed for as-extruded Ti(C) + 24 vol.% Mg shown in the
transversal direction.
a
100 µm
b
c
d
e
f
Ti
Mg
_20 µm_
Ti
Mg
100 µm
100 µm
100 µm
100 µm
100 µm
9
Fig. 5 EDS line analysis performed across Ti - Mg interfaces in as-extruded Ti(C) + 12 vol.% Mg shown in the
transversal direction.
The tensile tests revealed good reproducibility of the obtained mechanical properties of the as-
extruded Ti(C) profiles, as shown in Fig. 6 and Tab. 3. The Ti(C) matrix material showed a
reasonably high 0.2% strain offset yield stress (YS0.2) of 511 ± 1 MPa and ultimate tensile
strength (UTS) of 598 ± 3 MPa, accompanied with a large elongation (ε) of 31.3 ± 1.8%. The
UTS and ε of Ti(C) was improved significantly over the standardized properties required for the
reference: i) cast Ti Grade 4 and ii) PM fabricated Ti Grade 4 PM90 (Tab. 3). At the expense of a
high ε, the YS0.2 of Ti(C) was slightly below the limits defined by the ASTM B265 (483 MPa)
and B988 (435 MPa) standards. Owing to a reasonable strain hardening ability of the as-extruded
Ti(C), an increase of the YS0.2 can be easily managed by the introduction of extra deformation
strengthening. Owing to a finer grain structure and an increased O content, the strengths of Ti(F)
significantly increased and the ε decreased compared to Ti(C). It is apparent that the mechanical
properties of the Ti matrix could be tailored and varied to a large extent by the selection of the
HDH Ti powder size fraction.
10
Fig. 6 Tensile stress-strain curves of the Ti(C) and Ti(F) materials extruded from the coarse and fine Ti powders,
respectively.
The tensile test results of the as-extruded Ti(C) + Mg composites are summarized in Tab. 3. As
was expected, the introduction of the Mg component into the Ti matrix resulted in the
deterioration of all mechanical properties. While the strengths of Ti(C) + 12 vol.% Mg decreased
only slightly, the ε deteriorated quite significantly as it decreased by more than one order of
magnitude compared with the values for Ti(C), Fig. 7. With an increase of the Mg content the
strengths and the ε decreased gradually. The results suggest that the Mg phase did not contribute
effectively to load transfer in the plastic region and the strength of the composite originated
mostly from the Ti matrix contribution. It implies that there is no perfect metallurgical bonding at
clean Ti – Mg interfaces free of voids and pores, i.e. no strong chemical interlayer was formed
between the virtually immiscible Ti and Mg elements. Analysis of the surfaces of fractured
tensile bars of the Ti(C) + Mg composites revealed dimpled areas of the Ti matrix (Fig. 8). In
spite of imperfect Ti – Mg interfaces, no pull up or disintegration of the Mg filaments from the Ti
matrix occurred. Lower ε was attributed to the presence of the Mg phase, with inferior
metallurgical bonding to the Ti matrix, which acted as a stress concentrator. A gradual decrease
of the ε with the increase of the Mg content was due to: i) a smaller fraction of the Ti matrix,
which provided the strength, and ii) a gradual increase of interconnectivity between discrete Mg
filaments and the formation of a spatial network. Nevertheless, obtained ε was measured with a
reasonable reproducibility.
The presence of the Mg component led to a valuable decrease of E from 99.7 down to 81 GPa for
Ti(C) and Ti(C) + 24 vol.% Mg, respectively (Tab. 3, Fig. 7). If the Mg component (E = 45 GPa)
is assumed contributing to elastic load transfer following a simple rule of mixture, the E of the
composite should decrease at a rate of 0.55 GPa per 1 vol.% Mg. Another boundary condition
defines that if Mg contributes to no elastic load transfer, the E of the composite should decrease
theoretically at a rate of 0.99 GPa per 1 vol.% Mg. The actual E of the composites in this study
11
decreased gradually at a rate of 0.78 GPa per 1 vol.% Mg. These assumptions point out to an
inferior but desirable metallurgical bonding at the Ti and Mg interfaces, which limited load
transfer in the elastic region. A reduction of E significantly minimized the critical stress-shield
phenomenon, and the mechanical incompatibility between the implant and bone improved
eventually. It needs to be emphasized that a lower E of all Ti(C) + Mg composites was
accompanied by acceptably high tensile strengths and namely by reproducibly measured
elongation in tension. There is no relevant standard which specifies a minimum ε value for a
fabricated dental implant. However, the ε becomes crucial for load-bearing applications subjected
to dynamic loading e.g., dental implants. The dental implants must comply with the ISO14801
standard for the fatigue testing of endosseous dental implants. The ISO 14801 standard is
designed to reproduce potential overload conditions due to unfavorable but possible clinical
conditions of implant placement not perpendicular to the occlusal plane, resulting in non-axial
implant loading, i.e. a bending that must be successfully endured by the dental implant. The
standard also simulates a total of 3 mm of bone resorption, which effectively increases the stress
levels by up to 4 times compared to the stresses that result in true axial implant loading [36]. In
order to endure such intense fatigue testing a combination of reasonably high YS0.2 and ε is
required. This is in a strict contrast with the works published on various Ti + Mg composite
materials, in which only compression testing was included, presumably due to a brittle nature of
the composites [25, 34].
Table 3 Ultimate tensile strength (UTS), 0.2% strain offset yield stress (YS0.2), elongation (ε), and Young`s modulus
(E) of the as-extruded Ti materials produced from fine (F) and coarse (C) Ti powders, and Ti(C) + Mg composites.
Data for a Ti Grade 4 PM90 reference material fabricated by the powder metallurgy approach according to the
ASTM B988 standard and minimum requirements for the reference a cast Ti Grade 4 material according to ASTM
B265 are also included.
E
GPa
ε
%
YS0.2
MPa
UTS
MPa
material
101.7 ±0.3
16.5 ± 3.5
782 ± 17
945 ± 19
Ti(F)
99.7 ± 0.2
31.3 ± 1.8
411 ± 1
598 ± 3
Ti(C)
92.6 ± 0.2
2.7 ± 0.6
400 ± 19
498 ± 26
Ti(C) + 12 vol.% Mg
88.6 ± 0.6
1.5 ± 0.1
367 ± 5
443 ± 1
Ti(C) + 17 vol.% Mg
81 ± 1
1.1
340
409
Ti(C) + 24 vol.% Mg
-
16
435
495
Ti Grade 4 PM90
-
15
483
550
Ti Grade 4 minimum
12
Fig. 7 Ultimate tensile strength (UTS) and 0.2% strain offset yield stress (YS0.2) (a), and elongation (ε) and Young`s
modulus (E) (b) as functions of the Mg content in the Ti(C) + Mg composites.
Fig. 8 Fracture surface of a tensile bar fabricated from the Ti(C) + 17 vol.% Mg composites.
To achieve an insight into changes of the mechanical properties due to the elution of Mg, the
corrosion behavior of the Ti(C) + Mg composites was studied. The corrosion of the as-extruded
Ti(C) + Mg composites in conditions similar to human body environment was evaluated as a
function of exposure duration in the HBSS. Fig. 9 displays weight losses and DR of the Ti(C) +
12 and 24 vol.% Mg composites. The elution of Mg proceeded gradually with the exposure time
and became saturated after 7 and 14 days for the Ti(C) + 12 and 24 vol.% Mg composites,
respectively. The volume loss of Mg at a saturation point for the Ti(C) + 12 and 24 vol.% Mg
composites represented 3.2 and 9.9% of all Mg available in the as-extruded state, respectively.
Weight loss was dependent on sample geometry and dimensions. Moreover, the weight loss was
_10 µm_
Mg
Ti
13
affected by the texture of the Mg component oriented along the extrusion direction. The corrosion
behavior of the Ti(C) + Mg composites was evaluated using rectangular specimens machined in
parallel with the extrusion direction. Thus, it can be expected that the weight loss should be much
pronounced if specimens are machined perpendicularly with respect to the extrusion direction.
Fig. 10 shows the subsurface microstructure of the Ti(C) + Mg composites diluted in the HBSS.
Reaching a saturation point, the dilution of Mg in the transversal direction proceeded down to
~50 and ~100 µm below a specimen surface for the Ti(C) + 12 and 24 vol.% Mg composites,
respectively. A larger penetration depth may be expected in the longitudinal cross-section. The
typical diameter of a dental implant is ~4.5 mm, it features a threaded inner hole for an abutment
be screwed in and has a threaded outer area. Furthermore, it is expected that implant will be
machined longitudinally from an extruded raw composite material. Thus, in spite of a relatively
small penetration depth in small cross-section thread areas of Ti + Mg dental implants, Mg may
become even fully released across the cross-section. The pore size in the presented Ti + Mg
composites was in the order of tens of µm and the pores were interlinked with arms with a typical
thickness in the order of a few µm.
14
Fig. 9 Weight loss from a total composite weight and a degradation rate (DR) of the Ti(C) + 12 vol.% Mg (a) and
Ti(C) + 24 vol.% Mg (b) composites after dilution in a Hank's balanced salt solution for different time periods.
a
b
15
Fig. 10 EDS elemental maps of Ti and Mg shown in a transversal cross-section of the Ti(C) + 12 vol.% Mg (a) and
Ti(C) + 24 vol.% Mg (b) composite diluted in a Hank's balanced salt solution for 14 days. The dashed lines represent
the surface of the samples.
Unlike the weight loss, the values of DR should be independent of sample geometry and
dimensions. Owing to a higher portion of Mg present on sample surface, a thicker cross-sectional
area of Mg arms of a continuous network, and interconnectivity between Mg areas, the values of
DR were much higher in the case of the Ti(C) + 24 vol.% Mg composite. Even after 21 days of
elution the value of DR for the Ti(C) + 24 vol.% Mg composite equaled roughly the one of Ti(C)
+ 12 vol.% Mg after the third day of elution (~0.1 mg.cm-2.day-1). Thus, the elution of Mg in
Ti(C) + 24 vol.% Mg proceeded more deeply into the bulk of the composite. DR for the Ti + Mg
composites slowed down with increased exposure time. Following the progress of the weight loss
curves, it was assumed that DR reached virtually a zero value after exposure during 14+ and 21+
days for the Ti(C) + 12 and 24 vol.% Mg composites, respectively. The decrease of DR was
attributed to the precipitation of insoluble corrosion products mainly, amorphous magnesium
phosphate [37], carbonated apatite species and brucite, as was confirmed by EDS point analysis
on the sample surface (Fig. 11) and XRD spectra of corrosion products (Fig. 12). The formation
of corrosion products is expected to promote bone ingrowth after the implantation process.
_100 µm_
b
a
_100 µm_
16
Fig. 11 SEM image of corrosion product precipitates, marked by red arrows, on the surface of the Ti(C) + 12 vol.%
Mg composite dilution in a Hank's balanced salt solution for 2 days. EDS elemental analysis of the spots yielded the
following composition: 12.12 wt.% Mg, 13.89 wt.% Ca, 55.62 wt.% O, 10.13 wt.% P, and 8.24 wt.% C.
Fig. 12 XRD lines of a corrosion product from the surface of the Ti(C) + 12 vol.% Mg composite diluted in a Hank's
balanced salt solution for 2 days. The broad peak marked by arrows corresponds to amorphous magnesium
phosphate.
The E, UTS, YS0.2, and ε of as-extruded Ti(C) + Mg composites were investigated as functions of
dilution time. Tab. 4 summarizes the E of the Ti(C) + 12 and 24 vol.% Mg composites eluted in
the HBSS for up to 21 days. The E decreased with the increase of elution time and reached the
saturation points, which were in accord with the weight loss data. A decrease in the E of the
composite after immersion was due to the dilution of Mg from the surface, which left pores at
former Mg sites. The decrease in E became apparent namely for the Ti(C) + 24 vol.% Mg
composite where E was further reduced by 8.6% compared with that of the as-extruded state. In
contrast, E decreased only marginally for the Ti(C) + 12 vol.% Mg composite. This may be
rationalized by the difference in a total volume of eluted Mg in the Ti(C) + 12 and 24 vol.% Mg
_20 µm_
17
composites and at the same time by a poor contribution of Mg to the E of the composite. It needs
to be pointed out that obtained slight variations of E for the Ti(C) + 12 vol.% Mg composite were
at the limit of the DMA method. The decrement in E strongly depended on a cross-sectional area
of the specimen and the orientation with respect to the extrusion direction. A decrease in E would
escalate profoundly with smaller cross-sectional dimensions of DMA specimens machined
perpendicularly to the extrusion direction. Since the contribution of the Mg component to load
transfer in a plastic region was limited, the porous structure of the Ti matrix with released Mg
sites still provided required mechanical strengths and elongation in tension (Tab. 5). There was
no major detrimental effect of Mg dilution on the tensile properties determined for the Ti(C) + 12
vol.% Mg composite. For the Ti(C) + 24 vol.% Mg composite diluted for 14 days, the tensile
strengths and ε decreased compared with those of the as-extruded sample. However, the strengths
after dilution decreased only by ~10% for the composite with the highest content of Mg, and thus
it remained sufficient for the specific application.
Table 4 Evolution of Young`s modulus (E) of the Ti(C) + 12 and 24 vol.% Mg composites after immersion in a
Hank's balanced salt solution for 3, 7, 14, and 21 days.
E
GPa
21 days
14 days
7 days
3 days
as-extruded
material
92.3 ± 0.6
92.3 ± 0.6
92.2 ± 0.7
92.5 ± 0.6
92.6 ± 0.2
Ti(C) + 12 vol.% Mg
74 ± 2.6
74 ± 2.6
75.9 ± 2.3
77.7 ± 1.9
81 ± 1
Ti(C) + 24 vol.% Mg
Table 5 Tensile mechanical properties of the Ti(C) + 12 and 24 vol.% Mg composites determined after the
immersion of the tensile bars in a Hank's balanced salt solution for 7 and 14 days, respectively.
ε
%
YS0.2
MPa
UTS
MPa
material
4.8 ± 0.6
361 ± 28
490 ± 8
Ti(C) + 12 vol.% Mg
0.4 ± 0.2
330 ± 27
359 ± 39
Ti(C) + 24 vol.% Mg
A general concept of the presented Ti + Mg composites is not necessarily based on the
dissolution of all Mg component from the entire volume of the composite. Mg enhances the
osteoinduction, osteoconduction and osseointegration of the implanted material by interacting in
important intra- and intercellular osteogenic processes in the alveolar bone during a new,
perimplant bone formation. Moreover, the presence of two different metals (Ti and Mg) in a
direct mechanical and electrical contact with different electrode potentials is expected to lead to a
galvanic corrosion where Mg acts as anode and undergoes corrosion while Ti remains relatively
intact. In that way, the controlled dissolution of Mg could be used to manage a favorable
biological effect on the processes of osteoinduction, osteoconduction and osseointegration in a
18
living tissue. However, due to a small penetration depth of Mg dilution measured from the
surface, it can be expected that the ingrowth of a bone in the Ti + Mg composites will be limited
mostly to the surface of the implant. Furthermore, it must be further emphasized that depending
on a bone an optimal pore size, which promotes bone intergrowth into porous Ti, is in the order
of hundreds µm [38]. A typical pore size in the Ti + Mg composites in this study was less by one
order of magnitude, which may hinder bone intergrowth into the volume of an implant. In hand
with a reduced Young`s modulus, the Ti + Mg composites in this study reached reasonable
tensile mechanical properties needed for dental implants application, which is to be compared
directly with the Ti Grade 4 or 5 reference materials. Such attractive set of mechanical properties
is very difficult to attain when a pore or Mg size, which represent critical locations to crack
initiation, is in the order of hundreds µm. Polished Ti is bioinert with no bonding ability to a
bone. On the contrary, the dilution of Mg from the surface contributes to the attachment of a bone
to the surface of an implant by the increase of the microroughness of the implant.
4. Conclusions
A new type of Ti + 12, 17 and 24 vol.% Mg bimetallic composites were successfully produced
using a multi-step low temperature PM process, which included blending, CIP, HVP, and DE.
The microstructure of the composites was formed by a biodegradable Mg component with a low
E in the form of filaments homogeneously dispersed within a permanent Ti matrix. The filaments
were arrayed along the extrusion direction. With the increase of the Mg content discrete filaments
became interconnected with each other thus forming a continuous Mg network embedded in the
Ti matrix. The PM process applied proved to be adequate to avoid an undesirable oxidation of the
Ti and Mg powders. The role of Mg as a modulator of mechanical and corrosion properties was
also confirmed by the reduction of E of the as-extruded composites down to 81 GPa,
accompanied with a reasonably high UTS of at least 409 MPa and a sufficient ε of at least 1.1%.
In-vitro testing in the HBSS confirmed the elution of a biodegradable Mg component and
formation of the pores in former Mg sites. In this way a macro and micro surface roughness
enhanced, which provided beneficiary conditions during the osseointegration process. The elution
of Mg proceeded gradually with time and became saturated after 7 and 14 days for the
composites with 12 and 24 vol.% Mg, respectively. The volume loss of Mg at a saturation point
represented dissolution of 3.2 and 9.9% of all Mg content available in the as-extruded composites
with 12 and 24 vol.% Mg, respectively. The E of the composites exposed to the HBSS further
decreased to 74 GPa, while other mechanical properties of the composites with dissolute Mg
were mostly maintained. Ti(C) + 17 vol.% Mg composite showed the best trade-off of properties,
with E reduced by ~15%, and with a reasonably high YS0.2 and a sufficient ε after Mg dilution
from implant’s surface, both needed to endure fatigue testing of implants. The mechanical and
bioactive properties proved an immense application potential of the new Ti + Mg metallic
composite as a material for dental implants subjected to load-bearing conditions.
Acknowledgments
19
Financial support from the SRDA APVV-16-0527 project is gratefully acknowledged. The
authors thank to Dr. P. Svec Jr. for the help with XRD analyses.
References
[1] P. Lichte, H.C. Pape, T. Pufe, P. Kobbe, H. Fischer, Scaffolds for bone healing: Concepts,
materials and evidence, Injury 42 (2011) 569-573.
[2] J.M. Anderson, A. Rodriguez, D.T. Chang, Innate and adaptive immune responses in tissue
engineering, Elsevier seminars in immunology 20 (2008) 86–100.
[3] T. Albrektsson, P.I. Brånemark, H.A. Hansson, J. Lindström, Osseointegrated titanium
implants: requirements for ensuring a long-lasting, direct bone-to-implant anchorage in
man, Acta Orthop. Scand. 52 (1981) 155-170.
[4] T.J. Blokhuis, J.C. Arts, Bioactive and osteoinductive bone graft substitutes: definitions,
facts and myths, Injury 42 (2011) S26-29.
[5] P. Moshayedi, G. Ng, J.C. Kwok, G.S. Yeo, C.E. Bryant, J.W. Fawcett, K. Franze, J. Guck,
The relationship between glial cell mechanosensitivity and foreign body reactions in the
central nervous system, Biomaterials 35 (2014) 3919-3925.
[6] H.M. Frost, Wolff's Law and bone's structural adaptations to mechanical usage: an
overview for clinicians, Angle Orthod. 64 (1994) 175-188.
[7] R.L. Duncan, C.H. Turner, Mechanotransduction and the functional response of bone to
mechanical strain, Calcif. Tissue Int. 57 (1995) 344-358.
[8] S. Bahl, A.S. Krishnamurthy, S. Suwas, K. Chatterjee, Controlled nanoscale precipitation to
enhance the mechanical and biological performances of a metastable β Ti-Nb-Sn alloy for
orthopedic applications, Mater. Des. 126 (2017) 226-237.
[9] B. Piotrowski, A.A. Baptista, E. Patoor, P. Bravetti, P. Laheurte, Interaction of bone–dental
implant with new ultra-low modulus alloy using a numerical approach, Mater. Sci. Eng. C
38 (2014) 151-160.
[10] A. Revathi, A.D. Borrás, A.I. Muñoz, C. Richard, G. Manivasagam, Degradation
mechanisms and future challenges of titanium and its alloys for dental implant applications
in oral environment, Mater. Sci. Eng. C 76 (2017) 1354-1368.
[11] J.M. Cordeiro, V.A.R. Barão, Is there scientific evidence favoring the substitution of
commercially pure titanium with titanium alloys for the manufacture of dental implants?,
Mater. Sci. Eng. C 71 (2017) 1201-1215.
[12] P. Chui, Near β-type Zr-Nb-Ti biomedical alloys with high strength and low modulus,
Vacuum 143 (2017) 54-58.
[13] S.X. Liang, X.J. Feng, L.X. Yin, X.Y. Liu, R.P. Liu, Development of a new β Ti alloy with
low modulus and favorable plasticity for implant material, Mater. Sci. Eng. C 61 (2016)
338-343.
[14] M.T. Mohammed, Z.A. Khan, A.N. Siddiquee, Beta titanium alloys: the lowest elastic
modulus for biomedical applications: a review, Int. J. Chem. Mol. Nucl. Mater. Metall.
Eng. 8 (2014) 788-793.
20
[15] J.I. Qazi, H.J. Rack, Metastable Beta Titanium Alloys for Orthopedic Applications, Adv.
Eng. Mater. 7 (2005) 993–998.
[16] C. Zhao, X. Zhang, P. Cao, Mechanical and electrochemical characterization of Ti–12Mo–
5Zr alloy for biomedical application, J. Alloy Compd. 509 (2011) 8235-8238.
[17] M. Niinomi, M. Nakai, J. Hieda, Development of new metallic alloys for biomedical
applications, Acta Biomater. 8 (2012) 3888-3903.
[18] R. Narayan, ASM Handbook, Volume 23, Materials for Medical Devices. Materials Park:
ASM International. 2012.
[19] B.V. Krishna, S. Bose, A. Bandyopadhyay, Low stiffness porous Ti structures for load-
bearing implants, Acta Biomater. 3 (2007) 997-1006.
[20] M.P. Staiger, A.M. Pietak, J. Huadmai, G. Dias, Magnesium and its alloys as orthopedic
biomaterials: a review, Biomaterials 27 (2006) 1728-1734.
[21] R. Zeng, W. Dietzel, F. Witte, N. Hort, C. Blawert, Progress and challenge for magnesium
alloys as biomaterials, Adv. Eng. Mater. 10 (2008) 3-14.
[22] Z. Esen, B. Dikici, O. Duygulu, A.F. Dericioglu, Titanium–magnesium based composites:
Mechanical properties and in-vitro corrosion response in Ringer's solution, Mater. Sci. Eng.
A 573 (2013) 119-126.
[23] N.T. Kirkland, Magnesium biomaterials: past, present and future, Corrosion Engineering,
Sci. Tech. 47 (2012) 322-328.
[24] G. Song, Control of biodegradation of biocompatible magnesium alloys, Corros. Sci. 49
(2007) 1696–1701.
[25] O. Duygulu, A.A. Kaya, G. Oktay, F.C. Sahin, Diffusion bonding of magnesium, zirconium
and titanium as implant material, Mater. Sci. Forum 546–549 (2007) 417–420.
[26] W. Kroll, Verformbare Legierungen des Titans, Z. Metallkunde 29 (1937) 189-192.
[27] J.E. Barnes, W. Peter, C.A. Blue, Mater. Evaluation of Low Cost Titanium Alloy Products,
Sci. Forum 618-619 (2009) 165-168.
[28] M. Qian, F.H. Froes, Titanium Powder Metallurgy: Science, Technology and Applications,
Elsevier Science (2015) ISBN: 9780128009109
[29] P.J. Anderson, V.M. Svoyatkovsky, F.H. Froes, Y. Mahajan, D. Eylon, Modern
Developments in Powder Metallurgy, Metal Powder Industries Federation, Princeton NJ
(1981) 537-549.
[30] M. Balog, J. Viskic, P. Krizik, Z. Schauperl, M. Snajdar, Z. Stanec, A. Catic, CP Ti
fabricated by low temperature extrusion of HDH powder: application in dentistry, Key Eng.
Mater. 704 (2016) 351-359.
[31] Y. Liu, K. Li, T. Luo, M. Song, H. Wu, J. Xiao, Y. Tan, M. Cheng, B. Chen, X. Niu, R. Hu,
X. Li, H. Tang, Powder metallurgical low-modulus Ti–Mg alloys for biomedical
applications, Mater. Sci. Eng. C 56 (2015) 241–250.
[32] Y.L. Xi, D.L. Chai, W.X. Zhang, J.E. Zhou, Titanium alloy reinforced magnesium matrix
composite with improved mechanical properties, Scr. Mater. 54 (2006) 19–23.
21
[33] G. Jiang, Q. Li, C. Wang, J. Dong, G. He, Fabrication of graded porous titanium–
magnesium composite for load-bearing biomedical applications, Mater. Des. 67 (2015)
354-359.
[34] S. Jiang, L.J. Huang, Q. An, L. Geng, X.J. Wang, S. Wang, Study on titanium-magnesium
composites with bicontinuous structure fabricated by powder metallurgy and ultrasonic
infiltration, J. Mech. Behav. Biomed. Mater. 81 (2018) 10-15.
[35] M. Balog, M. Snajdar, P. Krizik, Z. Schauperl, Z. Stanec, A. Catic, Titanium-Magnesium
Composite for Dental Implants (BIACOM), Chapter TMS 2017 146th Annual Meeting &
Exhibition Supplemental Proceedings, Part of the series The Minerals, Metals & Materials
Series (2017) 271-284.
[36] P. Sevilla, C. Sandino, M. Arciniegas, J. Martínez-Gomis, M. Peraire, F.J. Gil, Evaluating
mechanical properties and degradation of YTZP dental implants, Mater. Sci. Eng. C 30
(2010) 14-19.
[37] Y.C. Xin, K.F. Huo, H. Tao, G.Y. Tang, P.K. Chu, Influence of aggressive ions on the
degradation behavior of biomedical magnesium alloy in physiological environment, Acta
Biomater. 4 (2008) 2008–2015.
[38] N. Taniguchi, S. Fujibayashi, M. Takemoto, K. Sasaki, B. Otsuki, T. Nakamura, T.
Matsushita, T. Kokubo, S. Matsuda, Effect of pore size on bone ingrowth into porous
titanium implants fabricated by additive manufacturing: An in vivo experiment, Mater. Sci.
Eng. C 59 (2016) 690–701.
