~

Mechanical properties and biomechanical

compatibility of porous titanium for
dental implants

Kenzo Asaoka and Norihiko Kuwayama

Department of Dental Engineering, School of Dentistry, Tokushima University,
Tokushima, 770 Japan

Osamu Okuno and Ishi Miura

Division of Metallurgy, lnstitute for Medical and Dental Engineering, Tokyo Medical and
Dental University, Tokyo, 101 Japan

Titanium powder with a granule diameter
of 420-500 pm was prepared and porous
titanium specimens were made from this
powder. The mechanical properties of these
specimens were examined. The compres-
sive strength and low cyclic compressive
fatigue strength were 182 and 40 MPa, re-
spectively. Fractography was also observed
by scanning electron microscopy. Typical
fatigue characteristics of the bonding areas
of the powder were observed. In addition,
porous-titanium-coated dental implants
with pure titanium cores were prepared.
The compressive strength of the material
used was 230 MPa, fatigue strength not
being improved. Biomechanical stress cal-
culations using the finite element method
were made using a model that employed
the use of the material implanted in alveo-
lar bone. Shear stress at the implant-bone
interface as well as compressive stress con-
centrations in the bone was calculated. The
most suitable elastic modulus for the dental
implant was then estimated from these cal-
culations. Finally, based on these results,
the use of porous titanium for dental im-
plants was assessed.

INTRODUCTION

Porous biomaterials such as aluminum oxide, hydr~xyapatite,~ methyl-

methacrylate,6cobalt-chromium and titani~m'~-'~ have been stud-
ied for future applications as dental implants.I6 The use of porous materials
may overcome some of the problems faced by the clinician. Retention and
stabilization have been enhanced by fibrous tissue ingrowth through such
implants. Adjustment of the modulus of elasticity of implants close to host
bone supports a stable host site environment. The use of porous metals
prevent brittleness and subsequent fracture to which ceramics are prone. The
surface of titanium coated with titanium oxide is passive and is protected
from corrosion and tarnishing. Titanium also has good mechanical properties
and excellent biocompatibility. In anticipation of these values, porous ti-
tanium dental implants were fabricated and tested.

Journal of Biomedical Materials Research, Vol. 19, 699-713 (1985)

0 John Wiley & Sons, Inc. CCC0021-9304/85/060699-15$04.00
700 ASAOKA ET AL.

MATERIALS AND METHODS

Titanium powder preparation

Titanium has a high melting point and is a very reactive metal. It is difficult
to obtain the powder by ordinary atomization. Inve~tigatorsl~ have tried to
make spherical powder grains by the rotation electrode process (REP).In this
procedure, pure titanium rod was made from sponge titanium using an
arc-melting furnace. The titanium rod connected to the anode terminal was
connected to a high-speed motor. A tungsten rod was used for the negative
pole. The terminals were placed in a vessel evacuated to 1 x lop3Pa and
with high-purity argon gas later introduced. The anode rod rotating at high
speed was melted in an argon atmosphere by arc fusion. The melting metal
flew about in the direction of centrifugal force and formed spherical powder
granules. The particle diameter was controlled by adjusting the rotational
speed of the anode rod.

Fully porous specimen

The titanium powder fabricated by the above-mentioned method was

sifted out. Particles of 420-500 pm diameter only were used. The compacted
particles had weak bonding one with another, after sintering with an alum-
ina mold made by the lost wax method, in a 1 x Pa vacuum at 1000°C.
The presintered specimen was sintered further in a vacuum without the
mold for 8.6 X lo4 s (24 h) at 1400°C. The specimen after sintering has a
height of 10 mm and a diameter of 4 mm.
Porosity and average pore diameter were measured on the cross section of
the specimen using the feature analysis system (Omunicom, Baush & Lomb,
NY 14625).
The specimen was polished, and micro-Vicker’s hardness number was
measured at the center of the particle before and after sintering. Compressive
strength, tensile strength, bending strength, and low-cycle fatigue com-
pressive strength were tested using an Instron-type universal testing ma-
chine. Fractography was observed by scanning electron microscopy.

Surface porous implant material

Pure titanium rod was inserted into an alumina mold, and the titanium
powder was used to fill the space surrounding the core. Sintering was
performed under the same conditions as with the fully porous metal speci-
mens. Figure 1 shows the titanium powder, the titanium core, and the sin-
tered surface porous implant. The surface porous implant has a height of
10 mm and a diameter of 4 mm. The core has a wide variety of lengths and
diameters of 0.8,1.2, and 1.6 mm, respectively. In addition, a surface porous
POROUS TITANIUM FOR DENTAL IMPLANTS 701

Figure 1. Titanium powder granules of 420-500 pm diameter, titanium core

metal, and porous-titanium-coated dental implant.

specimen as shown in Figure 2 was made to determine the bonding strength

along the interface of the two materials. The push-out test was then per-
formed. Bonding strength was determined from the relation between coated
length L and push-out force P.

@ . .
I J

1
.. . .. .
. .
:.. . .. .
. .. .
..
. ..
:_. .
.. .. . . .. .

dp ---4
Figure 2. Push-out test method, and finite element analysis model. d, =
2.0 mm.
702 ASAOKA ET AL.

RESULTS

1. Characterization of porous titanium

The hardness number of the particles fabricated by REP was 140 Hv. After
sintering, the hardness of the same part of the particle was 245 Hv.
The mechanical properties of titanium are considerably influenced by a
small amount of impurity. In particular, the amount of oxygen and nitrogen
present with titanium is important for its mechanical properties such as
hardness, strength, and ductility. Heated to high temperature, titanium
derives its strength from solution hardening by such elements as oxygen and
nitrogen. These elements also confer the property of brittleness to titanium.
Since the hardness number was 245 Hv after sintering, the level of solute
atoms in the titanium particle was below 0.4% for oxygen and below 0.2%
for nitrogen.
Various pore diameters were, of course, observed at the surface of the
sintered specimen, as shown in Figure 3. On the cross section, loose bonding
was also observed. The average diameter of the channel voids was 220 pm.
The density of the specimen was 2.6 g/cm3, and the porosity was 41.5%. The
reason why the particles were not able to form a closely packed structure may
be due to the differences in the diameters, from 420 to 500 pm, of the packed
spherical particles. The specimen then has a porosity lying between that of
body-centered cubic (b.c.c.) type structure and simple cubic-type structure.
The diameter of the connecting void channels for b.c.c. structure theo-
(77 pm) and 0.58~(145 pm). Here, Y is the radius
retically lies between 0 . 3 1 ~
of the lattice point particle and is estimated at 250 pm. If the shape of the

Figure 3. Scanning electrograph of porous titanium after sintering.

POROUS TITANIUM FOR DENTAL IMPLANTS 703

ingrowth bone can be easily changed, bone of 160 pm diameter can pass the
narrowest channel on the closest plane in the b.c.c. structure. But the di-
ameter of the channel is actually a little less because of the bonding areas of
the particles. Single cubic structure is looser than b.c.c. structure, and bone
of 200 pm diameter can easily pass the narrow void.
Klawitter and Hulbert” demonstrated experimentally that interconnecting
pores of 150 pm diameter provide an optimum setting for the ingrowth of
osseous tissue. Predecki et aI.l9 showed that ingrowth of bone was most
rapid in titanium with a pore diameter of 500 pm. Therefore, it might be
expected from these results that this porous titanium promotes bone in-
growth and retention of the implant.

2. Mechanical properties of porous titanium

Compressive strength after sintering of the fully porous titanium was

184 2 34 MPa for 10 specimens at a cross-head speed of 0.5 mm/min.
Shrinkage to ultimate strength was 12 & 2%, and modulus of elasticity was
5.2-5.5 GPa. An example of the stress-strain curves obtained is shown in
Figure 4. The compressive tests were also carried out at various cross-head
speeds from 0.5 to 500 mm/min, and there was little difference between the
results. Bite frequency is about 10 times per second at most. The maximum
bite speed is then approximately 1000 mm/min. The mechanical properties
of the material do not change at these strain rates.
The bending strength of the specimen of 20 mm length, 4 mm width, and
3 mm thickness was 73.0 ? 9.0 MPa. Tensile strength was 45.6 2 6.4 MPa.
Compressive fatigue tests at a cross-head speed of 1.0 mm/min were per-
formed. Figure 5 shows the relation between the number of loaded cycles
and the shrinkage of the specimen. In the early testing cycles, shrinkage of
about 100 p m was observed, but there was little dimensional change after the
completion of these cycles. The shrinkage of the specimen after 8 x 10’

a
Lz
200

0 10 20
STRAIN %
Figure 4. Compressive stress-strain curve of porous titanium.
704 ASAOKA ET AL.

103 104 105 106

NUMBER O F CYCLES
Figure 5. Results of fatigue tests at various loads.

cycles on 500 N was approximately 1%.After 9 x lo5 cycles, the specimen

rapidly shrank and fractured. But, on 1 kN, fatigue failure occurred after
about 1 x lo5 cycles.
Figure 6 shows the falling off surface of the titanium particle after com-
pressive fracture, and the tear surface after fatigue fracture. A river pattern
was observed from the fractography of the compressive test (Figure 6a) and
a striation-like marking was observed from the fractography of the fatigue
test (Figure 6c). Figures 6b and d show the features of the fracturing bonded
areas. In the case of the compression test, the bonded area was torn from
both surfaces of the particles. The river pattern was marked during
the fracture process. The striation after fatigue testing was marked from the
next process. In the early cycles of the test, a few cracks were generated in
the bonded areas. These cracks may then grow and spread, as shown in
Figure 6d.

3. Mechanical properties of the porous titanium-coated dental implant

Push-out testing was performed. A compressive load was applied to the

core metal as shown in Figure 2, and bonding strength at the interface was
examined. Figure 7 shows the relation between push-out fractured load P
and porous titanium-coated length L. The calculation of the stress distribu-
tion using the finite element method is shown in Figure 8. Here, the moduli
of elasticity of the core and of the porous metal were estimated at 118 and
5 GPa, respectively. Point A, as shown in Figure 2, denotes greatly increased
stress concentration. Cracks initiated near point A may then pass through
the interface, resulting in the failure of the specimen. If the shear strength
at the core-porous-metal interface is estimated at 180 MPa from the results
of compressive strength testing, push-out force P is theoretically the broken
line shown in Figure 7.
The surface porous titanium implant used in the compression test is sche-
matically shown in Figure 9. The relationship between compressive strength
and core length was examined. The core has a diameter of 1.2 mm and a
length of 0.0-10.0 mm. The powder immediately under the core greatly
increased for every core length used by a factor of 2.1 times stress concen-
POROUS TITANIUM FOR DENTAL IMPLANTS 705

Figure 6 . Scanning electron micrograph of porous-titanium-fractured sur-

face. (a) River pattern on compressive fracture surface; (b) features of com-
pressive fracture at the bonding areas; (c) striation-like marking on fatigue
fracture surface; (d) features of fatigue fracture at the bonding areas.

tration from the finite element calculations. Table I shows the results of
compression tests for specimens of core diameter 1.2 mm. One such speci-
men, with a core length of 10 mm, was fractured by buckling and had a low
compressive strength. Figure 10 shows the stress-strain curve and macro-
fractography of the surface porous titanium dental implant with a core length
of 8 mm. The fracture was caused by slippage along 30" from the loading
direction. The cores of all these specimens were bent by buckling.
706 ASAOKA ET AL.

4
z
Y

3
n
a
3 2
W
L1:
3
z
a 1
w
LL

0
0 1 2 3 4 5
L/d,
Figure 7. Results of push-out test. (of, experimental data; (- - -) theoretical
results.

The relationship between compressive strength and core thickness was

examined just as before. The core has a length of 8 mm and diameters of 0.8,
1.2, and 1.6 mm, respectively. Table I1 shows the results for compressive
strength and ductility of specimens with cores embedded in the titanium
powder. It is clear from the finite element calculations that there are no
D

Figure 8. Stress distribution along m, m, Dl3 by push-out test.

POROUS TITANIUM FOR DENTAL IMPLANTS 707

~ .. . ..

.. ..
. .. .
... .
0 0 0
4
I ....
.. . .

1'
. :.
.. ... ..
.. .
..... *
' .. .. '
.. ... .
. . ....
. ... . .._.
.

Figure 9. Porous-titanium-coated dental implant model.

remarkable differences between the stress concentrations found under the

core among the different thickness used. This is also true for compressive
strength.

DISCUSSION

Various implant materials with a wide variety of modulus of elasticity,

from porous PMMA with 0.5 GPa to alumina with 340 GPa, have been used.
When occlusive force generates compression on a protruding post, bio-

TABLE I
Results of the Compressive Tests on the Porous-Titanium-Coated
Dental Implants with Various Core Length
Core length Compressive strength Shrinkage
(mm) (MW ("/.I
0.0 234 15
0.0 158 9
0.8 175 16
2.0 195 18
2.6 195 15
3.2 189 15
4.9 176 15
5.9 178 15
6.7 205 16
9.4 238 15
10.0 234 15
10.0 159 4
708 ASAOKA ET AL.

300
d
a
4
200

v)
v)
w
lx
100
I-
v)

0
0 10 20
STRAIN 2

Figure 10. Compressive stress-strain curve of porous-titanium-coated den-

tal implant, and before and after testing of a specimen.

mechanical compatibility, i.e., suitability of the material to the bone, was

studied with finite element simulation. The material with the most desirable
modulus of elasticity for dental implant was determined from the results.
Figure 11 shows the simulation model. The stress distribution in the model,
which was composed of 159 nodes and 144 elements, was calculated using
the two-dimensional axisymmetric finite element method. The modulus of

TABLE I1
Results of the Compressive Tests on the Porous-Titanium-Coated
Dental Implants with Various Core Thickness
Core thickness Compressive strength Shrinkage
(mm) (MPa) (%)
0.8 237 19
228 17
1.2 218 16
228 13
1.6 232 13
232 12
POROUS TITANIUM FOR DENTAL IMPLANTS 709

i Ii
DENTAL
IMPLANI

=r
4
O
4
M

IALVEOLAR
BONEB

Figure 11. Finite element calculation model and mesh pattern.

elasticity of the alveolar bone was estimated to be 14 GPa, and the same value
was applied over the entire alveolar bone region.
Figure 12 shows the relationship between the modulus of elasticity of the
implant materials and the displacement of the gingival line under loading.
The material with a modulus of under 100 GPa has great shrinkage. In view
of the strain mismatch at the interface, it would not seem unreasonable to
suppose that implant material of low elastic modulus such as polymer is of
limited use. When occlusive force generates compression on a protruding
post, KB immediately beneath the implanted material in the alveolar bone
decreases in compressive stress. The interface CD has a distribution of shear
stress. Points C and D show greatly increased shearing stress concentration;
the central part of the interface has almost no shearing stress concentration.
Interfacial flaking may then be initiated from the points C and D. Figure 13
shows the relationship between the modulus of elasticity of the material and
maximum compressive stresses at the points A and B. The stress values are
also divided by the compressive stress on the protruding post. The material
with a modulus of over 20 GPa has minimally higher stress concentration at
710 ASAOKA ET AL.

10 lo2 103
ELASTIC MODULUS /GPA
Figure 12. Relationship between the elastic modulus of dental implant mate-
rial and displacement between points C and D.

point A than material with a modulus of exactly 20 GPa. However, at point B

there is increased stress concentration. Material such as alumina with a high
modulus of elasticity is used for implant, so there is greatly increased stress
concentration at the part corresponding to point B.
Figure 14 shows the changes in interfacial shear stresses at the points C
and D corresponding to different moduli of elasticity of implanted materials.
If the material has a soft elastic modulus, eruption of the interface may be
initiated from the neck of the implant. On the other hand, hard material may
cause a tear by separating from the bone near the bottom when exposed on
loading. Material with a modulus of elasticity of 70-200 GPa is favorable for
adhesion to tissue on the grounds of good biomechanical compatibility.
From these results, it is feasible to use surface porous titanium as dental
implant material. Material with thick core and with tissue ingrowth into
pores perhaps has an increased modulus of elasticity. Therefore such mate-

a
z
V
0 10 102 103
ELASTIC MODULUS /GPA
Figure 13. Relationship between the elastic modulus of dental implant mate-
rial and compressive stress concentration factor at points A and B.
POROUS TITANIUM FOR DENTAL IMPLANTS 711

?,O

0,s

0,6

0.4

0,2

0,o
10 102 103
ELASTIC MODULUS /GPA
Figure 14. Relationshipbetween the elastic modulus of dental implant mate-
rial and maximum stress concentration factor at points C and D.

rial has favorable biomechanical compatibility together with retention and

stabilization of the implant, depending on ingrowth of bone.
Another problem commonly observed in dental implant materials is
strength; the implants must be strong enough to withstand biting force. A
number of studies on biting force have been made. However, it varies
markedly between individuals. In the molar region it may range from 370 to
890 N; in the premolar region, from 230 to 450 N; from 140 to 330 N on
cuspids and from 88 N to 250 N on incisors.20,21 The level of compressive
stress at the neck of a tooth is about 13-17 MPa, as calculated from the biting
force and the area of the neck. The proportional limit of the tooth structure
is 125-224 MPa, and its compressive breaking strength is 232-305 MPa,20,22
this value being 15-18 times that of the stress developed during mastication.
Compressive strength of the porous titanium in this study was almost of the
same value as the proportional limit of the natural tooth. The fatigue strength
was 40 MPa, two or three times the stress developed during mastication. The
engineer always multiplies these expected stress values by a "safety factor"
in order that a particular structure may be able to withstand a certain amount
of overstress. In this regard, porous titanium undoubtedly requires higher
strength. Pore size, porosity, and the strength of the particle-to-particle inter-
face are important factors contributing to the strength of porous materials
and improvements are required on these points. On the other hand, porous
titanium may be stronger in alveolar bone because of bone ingrowth.The
compression test in this study constitutes a stricter inspection than in an oral
environment, since the test was performed between ridge jigs.

CONCLUSIONS

Titanium spherical powder granules of 420-500 pm diameter were fabri-

cated by use of the rotating electrode process. Porous titanium was made
712 ASAOKA ET AL.

from this powder by sintering in a highly evacuated furnace with a controlled

temperature of 1400°C. The density of this porous titanium was 2.6 g/cm3,
and the porosity was 41.5%. The mechanical properties of the specimen were
investigated. The compressive strength of the porous titanium was 182 MPa.
The tensile and bending strength were 45.6 and 73.0 MPa, respectively. The
compressive strength of this material is almost the same as the proportional
limit of the natural tooth structure. From the results of low cyclic fatigue test,
it was shown that the specimen had little deformation with cyclic com-
pressive force of 500 N. However, it was found that at lo5cycles at 1 kN, the
specimen was deformed by shrinkage of about 2%.
Titanium powder-coated titanium dental implant was sintered and tested.
The compressive strength was about 230 MPa; however, fatigue strength
was not improved.
The possibility of using porous titanium on dental implant can be consid-
ered with respect to biomechanical compatibility and strength. Material with
an elastic modulus of 70-200 GPa has been used most frequently as dental
implant material. Porous titanium has a lower elastic modulus than this
value. However, a thick core with ingrowth of tissue into void channels
raises the elastic modulus of the material.
It was concluded that porous titanium dental implant has better bio-
mechanical compatibility than low modulus material such as polymer and
high modulus material such as ceramics. An improvement in the strength of
this material may be necessary for safe use in dental implants.

References
1. W.B. Hammer, R.G. Topazian, R. V. Mckinney, and S.F. Hulbert,
”Alveolar Ridge Augmentation with Ceramics,” ]. Dent. Res., 52(6),
356-361 (1973).
2. K. N. Pedersen, H. R. Haanaes, and S. Lyng, “Tissue Ingrowth into
Mandibular Intrabony Porous Ceramic Implants,” Int. ]. Oral. Surg., 3,
158-165 (1974).
3. K. N. Pedersen, and H. R. Haanaes, “Experimental Subperiosteal Im-
plantation of Porous A1203 Ceramic for Mandibular Ridge Augmen-
tation,’’ Actu Odontol Scand., 37, 181-187 (1979).
4. K.N. Pedersen, “Porous Ceramic Implants in the Alveolar Ridge of
Humans,” Int. I. Oral Surg., 9, 49-54 (1980).
5. J. F. Piecuch, R. G. Topazian, S. Skoly, and S. Wolfe, ”Experimental
Ridge Augmentation with Porous Hydroxyapatite Implants,” I. Dent.
Res., 62(2), 148-154 (1983).
6. J. J. Klawitter, A. M. Weinstein, and L. J. Peterson, ”Fabrication and
Characterization of Porous-Rooted Polymethyl-Methacrylate (PMMA)
Dental Implants,” 1. Dent. Res., 56(4), 385-393 (1977).
7. J. J. Klawitter, A. M. Weinstein, and L. J. Peterson, “Fabrication and
Characterization of Porous-Rooted Cobalt-Chromium-Molybdenum
(Co-Cr-Mo) Alloy Dental Implants,” 1. Dent. Res., 56(5), 474-480 (1977).
8. L. J. Peterson, R. V. Mckinney, B. M. Pennel, J. J. Klawitter, and A.M.
Weinstein, ”Clinical, Radiographic, and Histological Evaluation of Po-
rous Rooted Cobalt-Chromium Alloy Dental Implants,” ]. Dent. Res.,
59(2), 99-108 (1980).
POROUS TITANIUM FOR DENTAL IMPLANTS 713

9. S. D. Cook, A. M. Weinstein, and J. J. Klawitter, “A Three-dimensional

Finite Element Analysis of a Porous Rooted Co-Cr-Mo Alloy Dental
Implant,” J , Dent. Res., 61(1), 25-29 (1982).
10. H.U. Cameron, R. M. Pilliar, and I. Macnab, ”The Effect of Movement
on the Bonding of Porous Metal to Bone,” J. Biomed. Muter. Res., 7,
301-311 (1973).
11. H. U. Cameron, R. M. Pilliar, and I. Macnab, ”Porous Vitallium in Im-
plant Surgery,“ J. Biomed. Muter. Res., 8, 283-289 (1974).
12. S. D. Cook, J. J. Klawitter, and A.M. Weinstein, ”A Model for the
Implant-Bone Interface Characteristics of Porous Dental Implants,” J.
Dent. Res., 61(8), 1006-1009 (1982).
13. F. A. Young, M. Spector, and C. H. Kresch, ”Porous Titanium Endos-
seous Dental Implants in Rhesus Monkeys: Micro-radiography and His-
tological Evaluation,” J. Biomed. Muter. Res., 13, 843-856 (1979).
14. A.M. Weinstein, J. J. Klawitter, J. Anderson, A. Clemow, and J.
Koeneman, “Interface Mechanics of Porous Ti-6A1-4V Implants and
Bone,” presented at the l l t h International Biomaterials Symposium,
Clemson, SC, April 28-May 1, 1979.
15. C. Johnson, J. 0. Galante, and W. Rostoker, “Metallic Concentration
in Tissue Surrounding Porous Fiber Titanium Implants,” presented at
the l l t h International Biomaterials Symposium, Clemson, SC,
April 28-May 1, 1979.
16. D. D. Gronoble, and R. Voss, ”Materials and Designs for Implant Den-
tistry,” Biomat. Med. Dm.,Art Org., 4(2), 133-169 (1976).
17. 0. Okuno and I. Miura, “The Properties of Co-Cr-Mo Alloy Powder
Prepared by the Rotating Electrode Process,” Rep. lnst. Med. Dent. Eng.,
Tokyo Med. Dent. Univ., 14, 1-5 (1980).
18. J.J. Klawitter and S. F. Hulbert, ”Application of Porous Ceramics for the
Attachment of Load Bearing Internal Orthopedic Applications,” 1. Bio-
med. Muter. Res. Symp., No. 2 (part l), 161-229 (1971).
19. P. Predecki, J. E. Stephan, B. A. Auslaender, V. L. Mooney, and K. Kirk-
land, ”Kinetics of Bone Growth into Cylindrical Channels in Aluminum
Oxide and Titanium,” J. Biomed. Muter. Res., 6, 375-400 (1972).
20. R. W. Phillips, Science of Dental Materials, Saunders, Philadelphia, 1982.
21. A. H. Howell and R. S. Manly, “An Electronic Strain Gauge for Mea-
suring Oral Forces,” J. Dent. Xes., 27, 705-712 (1948).
22. J. W. Stanford, K. V. Weigel, G. C. Paffenbarger, and W. T. Sweeney,
”Compressive Properties of Hard Tooth Tissue,” J . A m . Dent. Assoc., 60,
746-756 (1960).

Received December 10, 1984

Accepted January 21, 1985