tribochemical.pdf

(381 KB) Pobierz
PII: S0928-4931(02)00322-3
Materials Science and Engineering C 23 (2003) 461–465
www.elsevier.com/locate/msec
Tribochemical structuring and coating of implant metal surfaces with
titanium oxide and hydroxyapatite layers
U. Gbureck a, * , A. Masten a , J. Probst b , R. Thull a
a Chair of functional Materials in Medicine and Dentistry, Universit¨tofW¨rzburg, Am Pleicherwall 2, D-97070 W¨rzburg, Germany
b Fraunhoferinstitut f¨r Silicatforschung, Neunerplatz 2, D-97070 W¨rzburg, Germany
Abstract
The sandblasting process with corundum is used for cleaning, roughening and activating of metal surfaces in dentistry and orthopaedics.
The high local energy transfer at the impact point originates the displacement of particles in the surface. In principle, this method can be used
for coating surfaces by sandblasting. In this work, we present a newly developed technique, which allows the coating of metal surfaces with
titanium dioxide (TiO 2 ) and hydroxyapatite (HA) using a sandblasting process. The blasting material is a composite ceramic consisting of an
alumina core (carrier material) covered with a porous shell of titanium dioxide or hydroxyapatite. The technique is applied to titanium
substrates; the surface roughness, morphology and composition of the samples are analysed. The procedure results in an averaged surface
roughness of 10–15 Am. Energy dispersive X-ray analysis (EDX) indicates the formation of a thin layer consisting of coating material on the
metal surface. Furthermore, the traces of corundum crystals, which are inevitable by using the common technique, i.e. sandblasting with
single-component grains, are clearly decreased. X-ray diffraction analysis (XRD) indicates mainly the existence of crystalline rutile and
hydroxyapatite/h-tricalcium phosphate (h-TCP) on the surface. Therefore, the presented method would be suitable for simultaneously
roughening, coating and optimizing the biocompatibility of metal implant surfaces in dentistry and endoprosthetics.
D 2002 Elsevier Science B.V. All rights reserved.
Keywords: Sandblasting; Titanium oxide; Hydroxyapatite; Tribochemical
1. Introduction
The impact not only affects the metal surface but addi-
tionally leads to the deformation and in some cases to the
total disintegration of the grain itself. Thereby the aluminum
oxide partly gets incorporated within the metal surface, i.e.
corundum sandblasted surfaces become Al 2 O 3 -contami-
nated [4–6] .
In our method, this contamination effect is utilised for
coating the metal surface during the sandblasting process.
The pure corundum grains are therefore replaced with
composite ceramics consisting of a massive core, the base
material, and a porous shell, i.e. a layered material. This
results in a tribochemical coating of the metal due to the
sandblasting procedure. Thus, simultaneous roughening,
activation and coating of the substrate becomes feasible.
In the field of dental prosthesis, this procedure is already in
use. Here the metal surfaces get conditioned with a silicon–
ceramic layer serving as a hydrolytic stable interface be-
tween metal and polymer [7] .
The main goal of the presented work was the development
of a reproducible method which allows the conditioning of
titanium and CoCr-alloys via sandblasting with titanium
dioxide or hydroxyapatite. Therefore, the surface treatment
Sandblasting with corundum is a common method for
cleaning, roughening and activating metal surfaces for
applications in dentistry and orthopaedics [1–3] . The
impact of a grain on the metal surface results in the transfer
of its momentum and kinetic energy, respectively. The
kinetic energy is partly absorbed by the crystal lattice
causing surface melting within a microscopic range. More-
over, a larger area of lattice defects is observed. The
dimensions of the melting zone and the lattice defect zone,
respectively, depend on the material properties of the metal
substrate and on the energy of the blasting grain. Local
changes of the metal surface structure initiated by the grain
impact are schematically shown in Fig. 1 . Note that the
particle momentum shown and the corresponding kinetic
energy are restricted to the surface area.
* Corresponding author. Tel.: +49-931-201-73550; fax: +49-931-201-
73500.
E-mail address: uwe.gbureck@fmz.uni-wuerzburg.de (U. Gbureck).
0928-4931/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
PII: S0928-4931(02)00322-3
4977302.003.png
462
U. Gbureck et al. / Materials Science and Engineering C 23 (2003) 461–465
Fig. 1. Variations of a metal surface at the impact point of a grain during sandblasting process.
of metal implants, which will be in contact with bone tissue, is
scheduled to be the area of application. The generated
tribochemical layers are characterized with regards to surface
morphology, composition, roughness and adhesion.
avoid texture effects, measurements were performed on
rotating samples.
3. Results
2. Materials and methods
3.1. Composite blasting material
Composite ceramics for sandblasting were synthesized
by sintering aqueous titanium dioxide (TiO 2 ) or hydroxya-
patite (HA) suspensions on corundum (110 Am medium
particle size) at a mass ratio of 1:5. Typically 200 g of fine
powdered coating material was suspended in 300 ml of
water and stirred for 1 h. Afterwards, 1000 g of corundum
were added to the suspension. After drying the suspension at
100 jC, the materials were sintered at a temperature of 1250
jC for 1 h and subsequently sieved. In order to achieve an
optimized covering of the corundum grains with TiO 2 or
HA, the coating procedure was repeated. The sandblasting
of commercial pure (cp) titanium plates (d = 16 mm) was
performed by using a sandblaster Topstar 3 (Bego, Bremen,
Germany) at a blasting pressure of 0.4 MPa for 20 s/cm 2 .
The surface roughness was measured with a profilometer
Surftest 211 (Mitutoyo, Japan); five measurements were
carried out at various positions on the surface; the average
and the deviation were calculated.
SEM-analyses were performed with a DSM 940 (Zeiss,
Oberkochen, Germany). Energy dispersive X-ray analysis
(EDX) of the sandblasted surfaces were recorded with a
QX2000 system (Link, England) at a working distance of 25
mm, an acceleration voltage of 10 kV. To determine the
adhesion of the coatings, the samples were cleaned accord-
ing to custom [8,9] ultrasonically for 15 min in deionised
water and once more characterized by EDX.
The phase composition of the coatings was investigated
with an X-ray diffraction system D5005 (Siemens, Karls-
ruhe, Germany) in the range of 2Theta = 20–50j. In order to
Twofold sintering of aqueous TiO 2 or HA suspensions
onto corundum in a mass ratio of 1:5 revealed as a proper
method to synthesize the composite ceramics. After the
coating procedure, the surface of the corundum grains is
covered by a porous shell of coating material. It was proved
by X-ray diffraction analyses that no aluminium titanate or
calcium aluminate was formed at the sintering temperature
of 1250 jC.
Fig. 2. SEM-micrograph of a titanium surface sandblasted with a TiO 2 /
corundum composite ceramic.
4977302.004.png
U. Gbureck et al. / Materials Science and Engineering C 23 (2003) 461–465
463
Fig. 3. SEM-micrograph of a titanium surface sandblasted with a TiO 2 /
corundum composite ceramic; ultrasonically cleaned in deionised water for
15 min.
3.2. Surface morphology
In a macroscopic scale, the metal surface shows a dark
grey colour after the sandblasting process with the compo-
site ceramics. In contrast, sandblasting with pure corundum
normally retains the colour of the untreated metal surface.
Figs. 2 and 3 show SEM micrographs of tribochemical
coated titanium, sandblasted with the TiO 2 composite
ceramic.
The coating appears to be strongly adhesive to the metal
substrate. After the ultrasonic cleaning procedure, the sur-
face morphology remained unchanged (Fig. 3) .
The results of the surface roughness measurements are
listed in Table 1 . The average roughness is in the range of
10–15 Am. Compared to pure corundum, the surface rough-
ness is slightly increased for the composite ceramics. This
may be due to the larger average grain mass.
Fig. 4. EDX-analysis of tribochemically coated titanium surfaces; surface
sandblasted with (A) corundum, (B) TiO 2 composite ceramic, (C) TiO 2
composite ceramic (ultrasonically cleaned), (D) hydroxyapatite composite
ceramic and (E) hydroxyapatite composite ceramic (ultrasonically cleaned).
corundum, the use of composite ceramics yields to a
significantly decreased alumina embedding within the sur-
face. The titanium dioxide and hydroxyapatite layers are
mostly resistant against further surface cleaning. The result
is a strong adherence of the coatings. Element distribution
micrographs also indicate a homogeneous coating of the
metal surface (Fig. 5) .
3.3. EDX analysis
The EDX analyses of a titanium surface sandblasted with
corundum and TiO 2 or hydroxyapatite blasting materials
before and after the ultrasonic cleaning procedure are shown
in Fig. 4 . Compared to the sandblasting procedure with pure
Table 1
Surface roughness of titanium sandblasted with various blasting materials
Blasting material
Average surface
roughness r z (Am)
Maximum surface
roughness r max (Am)
Corundum, 110 Am 8.54F0.8
9.68F0.8
TiO 2 /corundum
composite ceramic
12.3F0.7
14.8F2.5
HA/corundum
composite ceramic
13.3F1.2
15.3F2.2
Fig. 5. Calcium and phosphate element distribution of a titanium surface
sandblasted with a hydroxyapatite/corundum composite ceramic.
4977302.005.png 4977302.006.png 4977302.001.png
464
U. Gbureck et al. / Materials Science and Engineering C 23 (2003) 461–465
Fig. 6. X-ray diffraction-analysis of tribochemically coated titanium
surfaces; surface sandblasted with (A) hydroxyapatite composite ceramic,
(B) TiO 2 composite ceramic and (C) corundum.
As deduced from EDX analyses and SEM micrographs,
the coatings show a strong adherence to the metal surface
and are almost resistant against ultrasonic cleaning. There
are three adhesion mechanisms, which can be considered:
physical bonding, incorporation of the coating material
within the surface and thermochemical reactions among
substrate and deposited material. An indication for a chem-
ical reaction could be the formation of aluminium or
calcium titanates. Such compounds were not detected,
maybe due to the detection limit of 1–2 wt.% of XRD
surface analyses. The most likely adhesion mechanism is
considered to be a strongly enhanced interdiffusion at the
impact point of a grain due to local surface melting. A
similar mechanism was already discussed by Ishikawa et al.
[10] who observed a sintering of fine hydroxyapatite grains
with titanium and among themselves during the sandblast-
ing process.
The methods used in this work (EDX and XRD) give
information about element and phase composition of the
surface with an analytical depth of about 2–20 Am. The
bone implant interaction is mostly affected by the properties
of the implant material near the interface. Previous studies
on tribochemical TiO 2 and SiO 2 coatings deposited on
stainless steel by means of electron spectroscopy for chem-
ical analysis (ESCA) showed a TiO 2 content on top of the
surface of 65% which decreased to about 48% in 1.1 Am
analytical depth. The thickness of the layer was calculated to
be 5–6 Am. No sharp interface between coating and metal
surface is expected, but a gradient-like composition [11] .
Therefore, future works will present XPS analyses of the
here considered TiO 2 and HA coatings. This method can
give additional informations about the composition of the
surface layer with an analytical depth of a few nanometers.
As mentioned before, the tribochemical TiO 2 and
hydroxyapatite coatings are of great interest with regard to
the required modifications of dental and orthopaedic implant
surfaces. State of the art is sandblasting with pure corun-
dum. Using this method, our investigations show a content
of about 30–35% corundum within the surface which is not
removable by ultrasonic cleaning neither in aqueous nor in
alcoholic solutions. Instead of this, our presented novel
tribochemical coating technique emerged as a proper as
well as a simple method in order to reduce the contami-
nation with corundum and to reinforce the native oxide layer
of titanium. Furthermore, the surface modification with
hydroxyapatite promotes bone ingrowth. The method
should also be applicable to other materials, e.g. stainless
steel or CoCr-alloys. Future works should be engaged in the
optimization of the composite ceramics with respect to the
resulting TiO 2 or HA covering and the achievable roughness
of the sandblasted surface. Variations of the grain size as
well as the TiO 2 (HA)/corundum ratio appear to be promis-
ing. In order to avoid any corundum contamination, TiO 2
could be used as core material of the composite ceramic
grains, but note that a biphasic grain composition is still
required to achieve a homogenous surface coating. Single
3.4. X-ray diffraction analysis
The X-ray diffraction analysis (XRD) patterns of tita-
nium after sandblasting with corundum, TiO 2 or HA-coated
corundum are shown in Fig. 6 . The peaks observed corre-
spond to titanium, corundum, rutile and hydroxyapatite/h-
tricalcium phosphate (h-TCP). The existence of the h-TCP
phase is traced back to a slight impureness of the hydrox-
yapatite raw material. Other phases as titanium aluminate or
calcium titanate, which would point at mechanochemical
reactions between blasting material and titanium substrate,
are not detected.
4. Discussion and conclusions
In this work was shown that a novel sandblasting
technique with composite ceramics makes the coating of
cp titanium with TiO 2 or HA possible. As observed for
corundum blasted surfaces, the result is a micro-mechanical
structured and activated implant surface with an average
surface roughness of 10–15 Am. The advantages of the
novel technique are the reduction of corundum contamina-
tion as well as the deposition of thin biocompatible TiO 2 or
calcium phosphate layers, which was shown via EDX and
XRD analyses.
4977302.002.png
U. Gbureck et al. / Materials Science and Engineering C 23 (2003) 461–465
465
phase blasting materials, as currently used for medical
applications, simply lead to a spot-like embedding.
[5] J.L. Ricci, F.J. Kummer, H. Alexander, R.S. Casar, Technical note:
embedded particulate contaminants in textured metal implant surfa-
ces, J. Appl. Biomater. 3 (1992) 225–230.
[6] B.W. Darvell, N. Samman, W.K. Luk, R.K. Clark, H. Tideman,
Contamination of titanium castings by aluminium oxide blasting, J.
Dent. 23 (1995) 319–322.
[7] R. Guggenberger, Das Rocatec system—Haftung durch tribochemi-
sche Beschichtung, Dtsch. Zahn¨rztl. Z. 44 (1989) 874–876.
[8] M. Kern, V.P. Thompson, Sandblasting and silica-coating of dental
alloys: volume loss, morphology and changes in the surface compo-
sition, Dent. Mater. 9 (1993) 155–161.
[9] U. Gbureck, R. Thull, Tribochemische TiO 2 -Beschichtungen auf Den-
tallegierungen, BioMaterialien 2 (2–3) (2001) 93–98.
[10] K. Ishikawa, Y. Miyamoto, N. Nagayama, K. Asaoka, Blast coating
method: new method of coating titanium surface with hydroxyapa-
tite at room temperature, J. Biomed. Mater. Res. 38 (2) (1997)
129–134.
[11] U. Gbureck, G. Neumann, R. Thull, Oberfl¨chenmodifikation vonDen-
tallegierungen mit tribochemischen TiO 2 -Beschichtungen, Biomed.
Tech. (1995) 40; Erg¨nzungsband 1 (1995) 35–36.
References
[1] E. Wintermantel, S.-W. Ha, Biokompatible Werkstoffe und Bauweis-
en, Springer, Berlin, 1996, p. 290.
[2] H.-J. Tiller, B. Magnus, R. G¨bel, R. Musil, A. Garschke, Der Sand-
strahlprozess und seine Auswirkungen auf den Oberfl¨chenzustand
von Dentallegierungen (I), Quintessenz 36 (10) (1985) 1927–1934.
[3] H.-J. Tiller, B. Magnus, R. G¨bel, R. Musil, A. Garschke, P. Lock-
owandt, A. Oden, Der Sandstrahlprozess und seine Auswirkungen auf
den Oberfl¨chenzustand von Dentallegierungen (II), Quintessenz 36
(11) (1985) 2151–2158.
[4] M. B¨hler, F. Kanz, B. Schwarz, I. Steffan, A. Walter, H. Plenk,
K. Knahr, Adverse tissue reactions to wear particles from Co-alloy
articulations, increased by alumina-blasting particle contamination
from cementless Ti-based total hip implants, J. Bone Jt. Surg. 84-
B (2002) 128–136.
Zgłoś jeśli naruszono regulamin