Materials Science and Engineering C 48 (2015) 337–346

Contents lists available at ScienceDirect

Materials Science and Engineering C

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

Mixed zirconia calcium phosphate coatings for dental implants: Tailoring

coating stability and bioactivity potential☆
Karoline Pardun a, Laura Treccani a,⁎, Eike Volkmann a, Philipp Streckbein b, Christian Heiss c,d,
Giovanni Li Destri e, Giovanni Marletta e, Kurosch Rezwan a
a
University of Bremen, Advanced Ceramics, Am Biologischen Garten 2, 28359 Bremen, Germany
b
University Hospital, Justus-Liebig-University Giessen, Department of Cranio-Maxillo-Facial Surgery, Klinikstrasse 33, 35385 Giessen, Germany
c
University Hospital of Giessen-Marburg, Department of Trauma Surgery, Rudolf-Buchheim-Strasse 7, 35385 Giessen, Germany,
d
Laboratory of Experimental Surgery, Kerkraderstrasse 9, 35392 Giessen, Germany
e
Laboratory for Molecular Surfaces and Nanotechnology (LAMSUN), Department of Chemistry, University of Catania and CSGI, Viale A. Doria 6, 95125 Catania, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Enhanced coating stability and adhesion are essential for long-term success of orthopedic and dental implants. In
Received 4 August 2014 this study, the effect of coating composition on mechanical, physico-chemical and biological properties of coated
Received in revised form 4 November 2014 zirconia specimens is investigated. Zirconia discs and dental screw implants are coated using the wet powder
Accepted 5 December 2014
spraying (WPS) technique. The coatings are obtained by mixing yttria-stabilized zirconia (TZ) and hydroxyapa-
Available online 9 December 2014
tite (HA) in various ratios while a pure HA coating served as reference material. Scanning electron microscopy
Keywords:
(SEM) and optical profilometer analysis confirm a similar coating morphology and roughness for all studied coat-
Calcium phosphate ings, whereas the coating stability can be tailored with composition and is probed by insertion and dissections
Zirconia experiments in bovine bone with coated zirconia screw implants. An increasing content of calcium phosphate
Coating stability (CP) resulted in a decrease of mechanical and chemical stability, while the bioactivity increased in simulated
Bioactivity body fluid (SBF). In vitro experiments with human osteoblast cells (HOB) revealed that the cells grew well on
Dissolution all samples but are affected by dissolution behavior of the studied coatings. This work demonstrates the overall
Cellular biocompatibility good mechanical strength, the excellent interfacial bonding and the bioactivity potential of coatings with higher
TZ contents, which provide a highly interesting coating for dental implants.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction
Dental implants undergo several load-bearing conditions, such as
chewing pressure, and thus require high mechanical reliability. They
must be able to withstand both, peak loads and millions of low level
cyclic stresses occurring during lifetime while chewing [1]. This can be
achieved by the use of metallic materials; the vast majority of dental im-
plants currently used are made of titanium or titanium alloy. Both mate-
rials have been established for years and are characterized by a high
degree of biocompatibility, favorable mechanical properties and low
corrosion rates [2]. Nevertheless, several reports about questionable
biostability have been stated [3,4].
Ceramics, usually yttria-stabilized zirconia (TZ), receive increased
attention as dental materials due to their low toxicity and beneficial me-
chanical properties like high fracture toughness and bending strength as
well as satisfying esthetic requirements. Despite of the biocompatible
properties, TZ is known to be bioinert and thereby no stimulation of
☆ The authors declare no conflict of interest.
⁎ Corresponding author at: Am Biologischen Garten 2, 28359 Bremen, Germany.
E-mail address: treccani@uni-bremen.de (L. Treccani).
bone ongrowth is achieved [5–7]. As a result, fibrous layer may form
around the implant and induce aseptic loosening [4,8]. Besides surface
roughness inorganic coatings on the material surface have been
proposed in literature to overcome the instability and promote direct
attachment to bone tissue [9,10]. Calcium phosphate (CP) based mate-
rials, such as hydroxyapatite (HA) and tricalcium phosphate (TCP) are
considered to be bioactive and thus stimulate bone regeneration [11].
However, pure CP coatings exhibit a poor stability and provide a weak
bonding strength to the substrate [12–15]. Therefore, one challenge is
to control the adhesion strength of the coating together with a good bio-
activity but without affecting the bulk properties of the substrate, so
that long-term stability can be achieved.
To overcome these drawbacks of pure CP coatings several studies
used TZ reinforced HA coatings on titanium to combine enhanced me-
chanical properties with an increased osseointegration. The addition
of TZ to HA and subsequent sintering provides an enhanced coating sta-
bility as well as better coating–substrate adhesion caused by improved
particle-substrate interactions when applied on zirconia substrates
[16–18]. Sintering, a thermally driven process, gives the opportunity
for diffusion processes between the particles of coating and substrate,
which in turn may have a great influence on the adhesion [19].

http://dx.doi.org/10.1016/j.msec.2014.12.031
0928-4931/© 2014 Elsevier B.V. All rights reserved.
338 K. Pardun et al. / Materials Science and Engineering C 48 (2015) 337–346
Both, bone-implant bond strength and coating-implant bond
strength, are two important factors for successful and strong anchorage
of the implant to bone tissue [20]. Thus, a poor stability within the coat-
ing and a weak bonding of the coating to the substrate could cause fail-
ure or delamination due to the occurring forces during insertion and
after implantation [21,22]. Therefore, coating composition and the coat-
ing characteristics involved are of great interest to achieve a high effec-
tiveness and long-term stability of the implant.
Regarding the significance of coating stability and adhesion the aim
of this study was to assess the effect of coating composition on mechan-
ical, physico-chemical and biological properties of coated zirconia spec-
imens. Different mixed coatings consisting of TZ and CP were prepared
by wet powder spraying (WPS) and subsequently investigated for mi-
crostructure, phase analysis, coating adhesion and mechanical strength.
In vitro dissolution behavior, and surface morphology in simulated
body fluid (SBF) as well as proliferation and differentiation of human
osteoblasts (HOB) were examined to understand and evaluate the me-
chanical and chemical stability, and bioactivity of different coating
compositions.
2. Materials and methods
2.1. Coating preparation
The details of substrate and suspension preparation, as well as the
coating process are described in Pardun et al. [18].
Briefly, zirconia discs (diameter (Ø) = 15 mm, height (h) =
1.7 mm) were fabricated using commercially available TZ-3YSB-E pow-
der (Tosoh, Tokyo, Japan) and uniaxially pressed at 38 MPa. After press-
ing, the substrates were isostatically densified at 1200 bar for 5 min and
pre-sintered at 1100 °C for 2 h. Dental implants (length (l) = 17 mm,
Ø = 4.5 mm) prepared from a zirconia feedstock (INMAFEED K1012,
INMATEC Technologies GmbH, Rheinbach, Germany) were fabricated
via injection molding and subsequently pre-sintered at 950 °C for 2 h
(Fraunhofer IFAM, Bremen, Germany).
The different mixed suspensions for coating were prepared of HA
(Sigma-Aldrich Chemie GmbH, Munich, Germany) and TZ-3YS-E pow-
der (TZ, Tosoh, Tokyo, Japan), coating compositions are shown in
Table 1. The aqueous ceramic suspensions consisted of different amounts
of HA and TZ and were stabilized with polyacrylic acid-based dispersant
(PAA, 12 mg/g ceramic powder, Syntran® 8220, Interpolymer GmbH,
Hassloch, Germany) and at pH 10 by the addition of ammonium hydrox-
ide solution (25% NH3 basis, Sigma-Aldrich). As reference material a
pure HA coating was used. All suspensions were homogenized by
ultrasonication (Sonifier 450, Branson, Dietzenbach, Germany) for
10 min (mixed coatings) and 3 min (pure HA coating).
The coating of pre-sintered substrates was done with WPS. The sub-
strates were fixed and coated using a double-action airbrush spray gun
(BD 183-K, Artistic Life, Boenen, Germany) with a working distance of
200 mm. Coating deposition was performed with an air pressure of
2 bar, a relative humidity of ~ 60% and an airbrush nozzle with Ø
0.8 mm. After coating the samples were dried for 24 h at room temper-
ature and afterwards sintered at 1500 °C for 2 h. All subsequent charac-
terizations were performed on sintered samples.
Table 1
Coating composition of the different mixed coatings.
Sample name TZ HA
wt.% vol.% wt.%
CP – – 26.9
TZCP 1:2 12.9 3.0 26.9
TZCP 1:1 21.1 5.1 21.1
TZCP 2:1 29.5 7.5 15.4
2.2. Coating characterization
The surface morphology of the coatings was analyzed using a scan-
ning electron microscopy (SEM, Camscan Series 2, Obducat CamScan
Ltd., Cambridgeshire, United Kingdom) at 20 kV operating in secondary
electron mode and previously sputter coated with gold for 60 s (K550,
Emitech, West Sussex, UK).
Measurements of the surface roughness were characterized with an
optical profilometer (Plμ2300, Sensofar, Terassa, Spain). Areas of 477
× 636 μm2 were scanned and the average surface roughness was calcu-
lated according to ISO 25178.
Crystal phase analysis was carried out by grazing incidence X-ray
diffraction (GI-XRD) using an Ultima IV type III diffractometer (Rigaku,
Tokyo, Japan) equipped with Cross Beam Optics (CBO) and Cu Kα radi-
ation. Each run was performed with 2 theta (2θ) values between 20°
and 70° carried out in parallel beam mode with a fixed incident angle
of 0.5°.
2.3. Characterization of mechanical properties
Coating adhesion was determined with scratch tests performed by a
pencil hardness tester (PH-5800, BYK-Gardner GmbH, Geretsried,
Germany) according to ISO 15184. The mechanical adhesion between
substrate and coating was qualitatively examined with a pencil (Vickers
hardness 90.6 ± 2.5 HV 0.2) and a pencil-shaped bovine femur (Vickers
hardness 89.8 ± 5.2 HV 0.2), which were fixed in a sclerometer and
moved over the surface with a constant load of 7.5 N and a velocity of
1 mm/s along a 12 mm track. Prior to SEM investigation graphite and
bone leftovers on the sample surface were burned out at 1000 °C and
1400 °C, respectively.
Biaxial flexural strengths of the coated specimens (Ø = 15 mm, h =
1.7 mm) were tested with the ball on three balls (B3B) test using a uni-
versal testing machine (Zwick/Roell Z005, Ulm, Germany) with a con-
stant speed of 0.5 mm/min and metal balls of 4 mm radius. The
flexural strength was calculated applying the following equation [23]:
3  F  ð1 þ νÞ
σ max ¼
"4  π  t

2
!
 #
Ra 1−ν ðt=3Þ2 Ra 2
 1 þ 2 ln þ  1−  ð1Þ
t=3 1þν 2  Ra R
where: F is the applied load, ν the Poisson's ratio of the substrate mate-
rial (here 0.32 for TZ), t the sample thickness, Ra the support radius and
R the sample radius. The strength was tested on 30 samples for each
coating. Furthermore, fracture surfaces of the B3B specimens were ob-
served under SEM to analyze the coating–substrate interface.
2.4. Insertion and dissection of coated zirconia dental implants in bovine rip
bone
The surface coated zirconia screws (n = 10) were inserted into
predrilled holes of five fresh bovine rip bones. The insertion was carried
out with drilling and insertion tools from the similarly macrodesigned
implant system (BEGO Semados® RI; BEGO Implant Systems, Bremen,
Germany) and performed according to the clinically approved protocols
provided by the manufacturer.
After careful dissection of the implants SEM was used to assess coat-
ing morphology and stability. To further examine the coating stability at
the thread tips, the same samples were embedded in resin (EpoFix,
Struers GmbH, Willich, Germany), grounded to the middle of the im-
plant and examined by SEM.
vol.%
12.0 2.5. Dissolution behavior
12
9.9 To test the dissolution behavior of the various coatings, specimens
7.5
were placed into sterile polystyrene culturing dishes (24-well
 K. Pardun et al. / Materials Science and Engineering C 48 (2015) 337–346
multidish, Nunc, Wiesbaden, Germany) and immersed in 1 ml Tris–HCl
buffer up to 21 days. Tris–HCl buffer solution was prepared by dissolv-
ing 13.25 g of Tris(hydroxymethyl)aminomethane (Sigma-Aldrich) in
500 mL double deionized water and buffered at pH 7.4 with 1 M hydro-
chloric acid (HCl, Sigma-Aldrich) at 37 °C according to ISO 10993-14.
The samples were placed in an incubator at 37 °C (Inkubator 1000,
Heidolph, Schwabach, Germany) with an integrated shaker (160 rpm,
Unimax 1010, Heidolph). The supernatant was replaced by fresh Tris–
HCl solution at 12 different sampling points. The release of calcium
ions (Ca2 +) into the supernatant was determined photometrically
using the ortho-cresolphthalein complexone method (Fluitest Ca CPC,
Analyticon Biotechnologies AG, Lichtenfels, Germany) in accordance to
the manufacturer's instruction. Ortho-cresolphthalein complexone
forms a purple-colored complex in alkaline solution. The color intensity
is proportional to the Ca2+ concentration and was measured at 578 nm
(Multiscan GO, Thermo Scientific, Schwerte, Germany). Experiments
were carried out in replicates (n = 4).
2.6. Immersion in simulated body fluid
Conventional simulated body fluid (cSBF) was prepared by dissolv-
ing the following salts NaCl, NaHCO3, KCl, K2HPO4·3H2O, MgCl2·6H2O,
CaCl2 and Na2SO4 (all salts were purchased from Sigma-Aldrich). cSBF
was prepared according to Oyane et al. and ISO 23317. The solution
was buffered at pH 7.4 with Tris–HCl. Each sample was immersed in
40 mL of cSBF solution and incubated at 37 °C under static conditions
in a climate chamber (KBF 115, Binder, Tuttlingen, Germany). After im-
mersion for 4, 7, 10, 14 and 21 days the samples were gently rinsed with
double deionized water and dried at room temperature. The newly
formed mineral layer was observed by SEM (Carl Zeiss AG, Oberkochen,
Germany) and the chemical composition was determined by energy-
dispersive X-ray spectroscopy (EDX, INCA PentalFETx3, Oxford Instru-
ments, Tubney Woods, UK). In addition, the supernatant was collected
after each sampling day and Ca2+ concentration was measured photo-
metrically using Fluitest Ca CPC (Analyticon), as discussed in Section 2.5.
2.7. In vitro biological investigation
In vitro investigations with human osteoblast cells (HOB, Provitro
GmbH, Berlin, Germany) were conducted up to 9 days to assess the bio-
compatibility of the different coated samples. All samples were heat
sterilized at 200 °C for 2 h prior cell seeding. Thermanox® (∅ =
15 mm, surface roughness 0.03 ± 0.0 μm, Thermo Fisher Scientific
Inc., Bonn, Germany) was used as reference material.
HOBs were grown in Dulbecco's modified Eagle's minimal essential
medium (D-MEM, Gibco, Life Technologies GmbH, Darmstadt,
Germany) containing 10% heat-inactivated fetal calf serum (FCS,
Sigma-Aldrich) and 1% antibiotic–antimycotic (Gibco) in an incubator
(C200, Labotect Labor-Technik-Göttingen GmbH, Göttingen, Germany)
at 37 °C, 9.3% CO2 and 95% relative humidity [24]. The culture medium
was changed every two days. Cells were seeded onto the different ce-
ramic samples at a density of 2 × 104 cells/ml in 24-well multidishes.
To evaluate the cell morphology and growth immunofluorescence
staining was carried out at days 1, 4 and 7 of culture. Samples (n = 3)
were rinsed with 37 °C PBS twice and fixed with 4% paraformaldehyde
(PFA) (Riedel-de Haën, Seelze, Germany) for 15 min at room tempera-
ture. After washing with PBS, the cells were permeabilized for 3 min
with 0.5% Triton X-100 (Sigma-Aldrich). Cell nuclei and cytoskeletal
actin were stained with 4,6-diamidino-2-phenylindole (DAPI, 1:2500,
Sigma-Aldrich) and Alexa Fluor® 488-Phalloidin (1:100, Life Technolo-
gies), respectively, for 45 min at room temperature in the dark. After ad-
ditional washing steps with PBS, microscopy was carried out using an
AX-10 fluorescence microscope (Axio Vision Imager M.1, Carl Zeiss
AG, Jena, Germany).
The metabolic activity of cells was determined with a colorimetric
water-soluble tetrazolium salt (WST-1, Roche Diagnostics GmbH,
339
Mannheim, Germany), which is enzymatically cleaved to formazan
only by living cells. The WST-1 reagent was added to each well after me-
dium change (n = 6 for each kind of specimen and day) and incubated
at 37 °C and 9.3% CO2 for 2.5 h. At days 1, 4, 7 and 9 of culture the amount
of formazan was measured at 450 nm using a microplate reader
(Chameleon, HIDEX, Turku, Finland).
The investigation of cell number, alkaline phosphatase activity (ALP)
and protein concentration was performed with cell lysates (n = 5 for
each kind of specimen and day) after 4, 7 and 9 days of culture. Cell
lysis was achieved with 1% Triton X-100 (Sigma-Aldrich) in 0.9% NaCl
(Sigma-Aldrich) and incubation for 1 h on ice.
The cell number on the ceramic samples was assessed by DNA anal-
ysis using the Quant-iT™ PicoGreen® dsDNA Kit (Life Technologies). An
aliquot of the cell lysate was diluted with 1× TE buffer (10 mM Tris–HCl,
1 mM EDTA, pH 7.5) prior to addition of PicoGreen working solution.
Sample fluorescence was measured after an incubation of 5 min with
an excitation wavelength of 485 nm and an emission wavelength of
535 nm using a plate reader (Chameleon, HIDEX). The DNA concentra-
tion was correlated with the cell number using a calibration curve with
defined cell number.
For the determination of the ALP activity an aliquot of the cell
lysate was incubated with ALP substrate buffer, containing 6 mM
paranitrophenyl phosphate in 0.1 M glycine, 1 mM MgCl2, 0.1 mM
ZnCl2 at 37 °C for 30 min. The reaction was stopped with 1 M NaOH
and released 4-nitrophenol was determined photometrically at
405 nm. A standard curve was generated with different concentrations
of 4-nitrophenol. All chemicals for the ALP activity were purchased from
Sigma-Aldrich. The protein contents were assessed with the Pierce®
BCA Protein Assay Kit (Thermo Scientific). An aliquot of the cell lysate
was incubated with working reagent at 37 °C for 30 min and afterwards
measured at a wavelength of 570 nm. The specific ALP activity was cal-
culated by normalization to the protein content.
2.8. Statistical analysis
All collected data are expressed as mean values and standard devia-
tion. Statistical analyses were performed using one-way ANOVA follow-
ed by Tukey's test (Minitab 16, Minitab Inc., Pennsylvania, USA) for
multiple comparisons between the different coated groups. A confi-
dence level of 5% was considered to indicate a statistically significant
difference.
3. Results
3.1. Coating characterization
Surface morphology of all studied coatings was analyzed by SEM to
understand the influence of TZ:CP ratio on the resulting microstructure
(Fig. 1). SEM micrographs of all sample surfaces revealed an undulated,
rough and porous morphology. CP, TZCP 1:2, TZCP 1:1 and TZCP 2:1
showed an average surface roughness of 9.0 ± 1.1 μm, 7.4 ± 0.8 μm,
6.4 ± 0.5 μm, and 5.8 ± 0.4 μm, respectively. Furthermore, pores sizes
between 2 and 20 μm were observed. Formation of cracks along the
coating surface was also detected on samples CP and TZCP 1:2; the
width of the cracks was less than 0.5 μm. The thickness of the coating
varied between 33 and 55 μm.
GI-XRD analyses performed after sintering at 1500 °C are given in
Fig. 2. The bottom diffractogram exhibits XRD pattern of the pure CP
coating where all peaks are consistent with the presence of β-TCP and
α-TCP. TZCP 1:2 consisted of cubic calcium zirconium oxide and β-
TCP. TZCP 1:1 showed the same peaks as TZCP 1:2. Moreover, additional
peaks for monoclinic zirconia were also detected. On TZCP 2:1 charac-
teristic peaks for α-TCP were observed, as well as monoclinic and
cubic zirconia. Thus, spectra of all sintered specimens showed phase de-
composition of HA to TCP due to the high sintering temperature.
340 K. Pardun et al. / Materials Science and Engineering C 48 (2015) 337–346
Fig. 1. SEM micrographs depicting the surface morphology of various coatings on TZ obtained by WPS after heat treatment at 1500 °C. A: pure CP coating, B: TZCP 1:2, C: TZCP 1:1, D: TZCP
2:1. B–D: light gray areas represent TZ and dark gray areas represent CP.
3.2. Coating stability and adhesion
The biaxial flexural strength of the coated samples was determined
via B3B as shown in Fig. 3A. The mean strength values of CP, TZCP 1:2,
TZCP 1:1 and TZCP 2:1 were 790 ± 55 MPa, 850 ± 63 MPa, 939 ±
64 MPa and 1023 ± 80 MPa, respectively (Fig. 3B). The lowest strength
Fig. 2. GI-XRD analysis of pure CP and various mixed coatings on zirconia after heat treat-
ment at 1500 °C.
value was observed for pure CP and the highest value for the mixed
coating with the highest amount of zirconia (TZCP 2:1). TZCP 1:1 and
TZCP 1:2 exhibited intermediate strength values indicating a linear de-
pendence between biaxial flexural strength and amount of CP in the
composition. All samples showed statistically significant differences.
Investigations of fracture surfaces (Fig. 3C) support the B3B results.
TZCP 2:1 showed a smooth and continuous morphology along the
interbonding. Whereas CP, TZCP 1:2 and TZCP 1:1 exhibited a distinct
interface layer between substrate and coating, whose thickness in-
creased with enhanced CP content in the coating mixture (Fig. 3C
arrows).
Fig. 4 shows the sample surfaces before (A–D) and after scratch test
with a pencil (E–H) and a bovine femur (I–L). The black dashed lines
correspond to the pencil or bone scratch path. The results of the pencil
scratch test showed a completely damaged CP coating (Fig. 4E) that
was largely removed due to a poor coating adhesion. TZCP 1:2
(Fig. 4F) was only partially removed by the pencil indicating an inter-
mediate coating–substrate adhesion. In TZCP 1:1 (Fig. 4G) only minor
coating failures were observed that resulted solely in a pore enlarge-
ment. In contrast, TZCP 2:1 was completely resistant to the pencil sug-
gesting a firm attachment and a strong interbonding between coating
and substrate.
The bovine femur scratch test showed different results although the
pencil and bone exhibited the same Vickers hardness of about ~90 HV.
The pure CP coating (Fig. 4I) was again completely removed from the
substrate. Whereas all mixed coatings (Fig. 4J–L) demonstrated a good
mechanical integrity, no damage of the coating was observed after
bone burn out.
Zirconia screw implants (Fig. 5A) were tested with the same coating
compositions to evaluate the coating stability during implant place-
ment. Homogenously and fully coated implants were achieved by
WPS (Fig. 5B top row) and the coating morphology was similar to that
obtained on TZ discs. The implants were carefully dissected after inser-
tion into bovine rip bone (Fig. 5B, middle row). TZCP 2:1 exhibited a
high coating stability; the thread tips as well as the thread sides showed
 K. Pardun et al. / Materials Science and Engineering C 48 (2015) 337–346 341

Fig. 3. Mechanical strength behavior of various coated samples determined via B3B. A: Experimental set-up. B: Box plot diagram of the biaxial flexural strength. The black rectangle displays
the mean value and the horizontal line the median. All tested samples were significantly different from each other (p b 0.05). C: Corresponding fracture surfaces; the black arrows point the
interface between coating and substrate.

an intact coating confirming a strong adhesion between substrate and implants after dissection support these results (Fig. 5C bottom row).
coating, as no damages were detected. The other coatings possessed The SEM images depict coating failures after insertion particularly at
varying degrees of defects, especially at the thread tips. In areas with the thread tips. These findings are in agreement with the results from
no direct bone contact, the coating was still intact. Cross-sections of the scratch test (Fig. 3).

Fig. 4. SEM micrographs of the coating morphology before scratch test (A–D), after pencil scratch (E–H) and after bone scratch (I–L). The black dashed lines represent the scratch path.
342 K. Pardun et al. / Materials Science and Engineering C 48 (2015) 337–346
Fig. 5. Coating adhesion and stability analysis of zirconia implant-like screws. A: Implant overview of a non-coated (left) and coated implant (right). B: SEM micrographs of the implant
coating morphology before insertion into bone (top row) and after dissection (middle row) with the appropriate cross section of the thread tips after dissection of bone and embedding in
resin (bottom row).
3.3. In vitro bioactivity assessment
Fig. 6 shows the cumulative calcium ion release in Tris–HCl solution
represented as a function of time for all tested coatings. During the first
4 days of immersion the ionic concentration of Ca2+ increased rapidly
Fig. 6. Dissolution behavior of pure CP and various TZCP coatings sintered at 1500 °C after
immersion at pH 7.4 for a period of 21 days as a function of immersion time.
over time for all studied coatings followed by a gradual increase on
the following days. TZCP 1:2 showed the highest calcium ion release
followed by CP and TZCP 1:2. The lowest dissolution was recognized
for TZCP 2:1, indicating that Ca2+ release was predominantly decreased
with increasing zirconia content.
SBF was used to investigate the biological activity of all studied
coatings in vitro, reflected in their capability to form a mineral
layer onto the surface. Fig. 7A shows the top view of all tested sam-
ples after 21 days in SBF. SEM micrographs of CP and TZCP 1:2 dem-
onstrate a completely covered surface by a newly formed mineral
layer with the typical globular morphology of apatite precipitates.
Cracks were also found in the mineral layer and are possibly formed
during the drying process. The higher magnification of the newly
formed layer on the CP surface evidenced a detailed layer structure
(inset in Fig. 7A). The surface of TZCP 1:2 was almost completely cov-
ered and TZCP 2:1, in contrast, showed only low mineral formation
ability. EDX analysis (Fig. 7B) of the mineral layer grown on the coat-
ed samples depicted the typical peaks of Ca, phosphor (P), magne-
sium (Mg), sodium (Na) and oxygen (O), with a Ca/P ratio of 2.3,
close to the theoretical value for apatite (2.15) [25]. However, the
carbon (C) peak is related to sputter coating with carbon. Fig. 7C
shows the Ca2 + concentration of the SBF supernatant on different
sampling days of all tested coatings. A significant decrease in Ca2 +
concentration over time was observed for all coatings in comparison
to TZCP 2:1. The reactivity of precipitate formation increased with
increasing CP content in the coating.
 K. Pardun et al. / Materials Science and Engineering C 48 (2015) 337–346
Fig. 7. Behavior of CP, TZCP 1:2, TZCP 1:1 and TZCP 1:2 coatings after immersion in simulated body fluid for 21 days at 37 °C. A: SEM micrographs of the newly formed layer on the tested
coatings with the typical apatite structure (inset), B: EDX elemental analysis of newly formed layer on the CP surface, C: Time dependent calcium ion concentration in the supernatant of
simulated body fluid solution at various sampling days.
3.4. In vitro biological characterization
Immunofluorescence staining (Fig. 8) performed with human osteo-
blasts revealed cell morphology, spreading and growth after 1, 4 and
7 days of culture on all studied coatings. Cells cultured on the reference
material Thermanox® were flattened and widespread already 1 day
after seeding with many filopodia and well-organized actin fibers. In
the following the cells grew further, until after 7 days of culture the sur-
face was completely covered by a dense and confluent cellular layer. On
the coated samples the cells spread and grew to a different extent. After
1 day of culture the cells displayed a round or elongated shape. But on
TZCP 1:1 and TZCP 2:1 slightly more cells were found with a more flat-
tened and spread morphology compared to CP and TZCP 1:2. At day 3
the cells were spread well on all coatings except on TZCP 1:2 where
most of the cells remained rounded and small. After 7 days of culture
proliferation was more pronounced on TZCP 2:1 and TZCP 1:1, indicat-
ing better cell spreading and growth with enhanced content of TZ.
WST-1 assay was used to determine HOB metabolic activity on all
samples after days 1, 4, 7 and 9 as shown in Fig. 9A. The metabolic activ-
ity increased continuously during 9 days. After 1 day of culture, all test-
ed samples showed a nearly similar metabolic activity. At day 4 cells had
a significant greater metabolic activity on TZCP 1:2 and TZCP 2:1 com-
pared to CP and TZCP 1:2. However, the highest metabolic activity was
detected on Thermanox®. The same issue becomes more evident after
7 and 9 days of incubation.
Cell proliferation analyzed via the quantity of DNA (Fig. 9B) and pro-
tein content (Fig. 9C) confirmed these observations. Within the coated
samples, the cell proliferation was retarded on TZCP 1:2, whereas the
significant highest amount of cells was found on TZCP 2:1. In compari-
son with the coated samples, Thermanox® exhibited a significantly
higher proliferation.
343
To evaluate the cell differentiation on the tested samples intracellu-
lar ALP activity of HOB cells was characterized for 4, 7 and 9 days
(Fig. 9D). In general, the ALP activity increased with time on all tested
coatings. ALP activities of cells cultured on CP and TZCP 2:1 were signif-
icantly pronounced in comparison to TZCP 1:2 and TZCP 1:1 coated
samples on all sampling days. As expected, the highest ALP activity
was detected on Thermanox®.
4. Discussion
Pure HA or CP coatings on metallic dental implants were widely in-
vestigated for an improved tissue integration because of their good
osteoconductivity, but these coatings are also known to have a poor sta-
bility. One of the major challenges in fabrication of coated implants is
the maintenance of a suitable adhesion and stability during insertion ac-
companied with a favorable bioactivity. The present study demon-
strates the fabrication of mixed coatings with various ratios of CP and
TZ. The coatings were applied with the WPS technique, which suitability
was already shown in previous works [18,26]. During heat treatment
the starting materials HA and TZ decomposed and transformed to
other phases in all studied coatings due to the high sintering tempera-
ture. The resulting phases were dependent on the initial TZ and HA con-
tent, which either leads to a co-existence of β-TCP and α-TCP or to one
of both phases. Furthermore, a transformation of tetragonal (t) to cubic
(c) zirconia caused by CaO diffusion [27,28] and the formation of mono-
clinic zirconia were also detected and could compromise the implant
strength. Therefore, understanding the mechanical as well as the biolog-
ical relevance in combination with the coating composition is of crucial
importance.
Mechanical stability and coating adhesion were tested in this study
by B3B, scratch test and dissection experiments of inserted zirconia
344 K. Pardun et al. / Materials Science and Engineering C 48 (2015) 337–346

Fig. 8. Fluorescence micrograph of HOBs seeded on different coated samples over a culture period of 7 days compared to Thermanox®. The cells were fixed and stained with AF488-
phalloidin (green) for actin filaments and with DAPI for nuclei (blue).

implants. The results of the biaxial flexural strength test showed a major
impact of the coating composition on the mechanical strength. While
for CP the flexural strength was strongly reduced, TZCP 2:1 nearly had
no effect and is comparable with uncoated TZ, which has strength of
around 1100 MPa [18]. The mechanical stability showed a linear depen-
dence between strength and coating composition, which significantly
correlated with TZ:CP ratio. This behavior can be explained with in-
creasing thickness of the interface layer between coating and substrate
that increased with enhanced content of CP. During sintering released
CaO from HA decomposition can also diffuse into zirconia substrates,
probably resulting in a transformation of tetragonal to cubic zirconia.
With decreasing amount of TZ in the coating CaO is predominantly in-
corporated into the zirconia substrate and leads very likely to lower
strength values, because cubic zirconia has poor mechanical properties
compared to tetragonal zirconia [5].
The adhesion depends on the coating–substrate bond strength and
plays an important role for the use of coated load-bearing implants.
Other studies have demonstrated the application of interfacial layers
to overcome the mismatch between coating and substrate due to differ-
ent thermal expansion coefficients [29,30]. In this study a firm adhesion
was reached by TZ addition to the coating and a coating–substrate co-
sintering process in order to provide a sufficient bonding between coat-
ing and substrate [28,31]. The adhesion was studied via qualitative
scratch test, performed with a pencil and bovine femur at constant
load and speed. However, coated zirconia screws were inserted into
fresh bovine rip bone, which is similar to jaw bone and characterized
with respect to their coating integrity. The pure CP coating showed
the weakest resistance in both experiments indicating a poor interfacial
bonding between coating and substrate. The mixed coatings exhibited
different degrees of damage, which were more pronounced with
 K. Pardun et al. / Materials Science and Engineering C 48 (2015) 337–346
Fig. 9. A) Quantification of metabolic activity, (B) cell proliferation, (C) protein content and (D) ALP activity of HOB cultured on different coated samples and on Thermanox® over a culture
period of 9 days. Different symbols above the columns mark the significance level (p b 0.05) tested by ANOVA.
increased CP content. In other words, the higher the TZ content the
stronger the interbonding between coating and substrate. Slightly dif-
ferent results obtained from the pencil and bone scratch test can prob-
ably be explained by different stiffness and elasticity characteristics of
the scratching materials. These results show that the scratch test can
give a first indication of the coating stability. But insertion and dissec-
tion experiments, which are similar to a clinical situation, provide
more reliable data.
Dissolution experiments showed that the Ca2 + release was very
similar for all samples at the beginning of the experiment and differ
only slightly in the further course of time. TZCP 1:2 exhibited the
highest Ca ion release and TZCP 2:1 the lowest. However, the trend of
the cumulative release curves is the same for all tested specimens. The
dissolution behavior is directly influenced not only by the phase compo-
sition on the one hand, but also by the microstructure on the other hand.
With increased TZ content in the coating less CP is directly exposed at
the surface as can be seen in Fig. 1, thus less calcium and phosphate
can dissolve [32].
SBF bioactivity results indicated significant mineral layer formation
abilities. After 21 days of incubation in SBF solution, all coatings showed
a newly formed layer with typical cauliflower morphology except for
TZCP 2:1. The corresponding EDX analyses revealed calcium and phos-
phorus as main components, as well as magnesium and sodium in
small amounts, which might have similarities to bone apatite. Changes
of Ca2+ concentration in SBF revealed the tendency for mineralization
processes with increasing content of CP in the coating. The higher the
mineral forming ability, the lower the Ca2+ concentration in SBF solu-
tion was. Thus, precipitation of bone-like minerals reflects the dissolu-
tion behavior of the studied coatings. The same observation was made
by Weng et al. who showed that no precipitation occurred on surfaces
with low dissolution [33].
345
Bioactive properties of material surfaces play a crucial role in the
evaluation of cell-biomaterial interactions in order to assess the efficacy
of their application in human medicine. Therefore, cellular response of
HOBs, such as cell growth, proliferation and differentiation was investi-
gated in order to assess the influence of surface chemistry and physical
features. Fluorescence microscopic images revealed that cells on coat-
ings with increasing TZ contents showed faster attachment, a more
spread-out morphology, taking a polygonal shape with several cyto-
plasmic extensions and an apparently higher cell density. Higher CP
contents, in contrast, seemed to hamper cell flattening and spreading.
Metabolic activity, cell number and protein content were significantly
greater on TZCP 1:1 and TZCP 2:1 at days 4–9 compared to the other
coatings, indicating a good cytocompatibility. ALP activity followed
almost the same trend with the exception of an elevated value on CP.
These results show that the cells respond differently to the investi-
gated coatings, indicating an influence of the coating properties.
The highest cell proliferation, viability and differentiation were
found on Thermanox®. The cells were flattened, well spread and
reached confluence at day 9, which is due to the sample's physical
properties and consistent with the work of others [34]. These
in vitro investigations showed that all studied coatings were non-
toxic. Nevertheless, due to similar coating surface topography and
roughness, the varying extent of cell behavior can be mainly described
by different coating dissolution. Coatings with a higher amount of CP
dissolve more rapidly and thus lead to an alkaline pH of the cell culture
medium which affected the initial cell viability [35,36].
Nevertheless, further work is clearly needed to characterize the
in vivo bioactivity of the coating, as well as the long-term stability. Con-
sequently, it is imperative to know the coating stability after insertion
and the dissolution rate in a living organism under physiological condi-
tions. Moreover, the suspension composition can be altered, by
346 K. Pardun et al. / Materials Science and Engineering C 48 (2015) 337–346
introducing further additives, in order to get closer to the composition
of natural bone.
5. Conclusion
The present study examined the influence of different TZ and CP ra-
tios on the mechanical and biological properties of coatings on zirconia
substrates. Coating adhesion and mechanical stability were consider-
ably increased with increasing addition of TZ. SBF test results, however,
revealed a higher bioactivity potential of coatings with enhanced CP
content due to different dissolution behaviors. Despite similar surface
morphology and average roughness in vitro investigations showed
that the cells respond differently with regard to proliferation and differ-
entiation, most likely affected by dissolution of the studied coatings.
Considering the overall good mechanical strength, the excellent interfa-
cial bonding and the bioactivity potential, TZCP 1:1 and TZCP 2:1 pro-
vide a highly interesting coating for dental implants.
Acknowledgments
Authors would like to acknowledge financial support from the
Deutsche Forschungsgemeinschaft, DFG (Grant No. RE 2735/7-1). We
thank Petra Witte of Historical Geology–Palaeontology, University of
Bremen for the performance of SEM and EDX analysis. We are also
grateful to Martin Ellerhorst from BEGO Implant Systems for providing
implant geometries and Andreas Reindl from Fraunhofer IFAM Bremen
for preparation of zirconia implants, as well as Michael Teck for his help
with experimental investigations.
References
[1] B. Reinhardt, T. Beikler, Dental implants, in: J.Z. Shen, T. Kosmač (Eds.), Adv. Ceram.
Dent, Elsevier Inc., Amsterdam, 2014, pp. 51–76.
[2] M. Niinomi, Mechanical properties of biomedical titanium alloys, Mater. Sci. Eng. A
243 (1998) 231–236.
[3] H. Tschernitschek, L. Borchers, W. Geurtsen, Nonalloyed titanium as a bioinert metal
— a review, Quintessence Int. 36 (2005) 523–530.
[4] C. Wu, Y. Ramaswamy, D. Gale, W. Yang, K. Xiao, L. Zhang, et al., Novel sphene coat-
ings on Ti–6Al–4V for orthopedic implants using sol–gel method, Acta Biomater. 4
(2008) 569–576.
[5] C. Piconi, G. Maccauro, Zirconia as a ceramic biomaterial, Biomaterials 20 (1999)
1–25.
[6] J. Chevalier, What future for zirconia as a biomaterial? Biomaterials 27 (2006)
535–543.
[7] P.F. Manicone, P. Rossi Iommetti, L. Raffaelli, An overview of zirconia ceramics: basic
properties and clinical applications, J. Dent. 35 (2007) 819–826.
[8] G. Balasundaram, T.J. Webster, Increased osteoblast adhesion on nanograined Ti
modified with KRSR, J. Biomed. Mater. Res. A 80A (2007) 602–611.
[9] L. Le Guéhennec, A. Soueidan, P. Layrolle, Y. Amouriq, Surface treatments of titanium
dental implants for rapid osseointegration, Dent. Mater. 23 (2007) 844–854.
[10] V. Borsari, G. Giavaresi, M. Fini, P. Torricelli, A. Salito, R. Chiesa, et al., Physical char-
acterization of different-roughness titanium surfaces, with and without hydroxyap-
atite coating, and their effect on human osteoblast-like cells, J. Biomed. Mater. Res. B
Appl. Biomater. 75 (2005) 359–368.
[11] S.V. Dorozhkin, Calcium orthophosphates in dentistry, J. Mater. Sci. Mater. Med. 24
(2013) 1335–1363.
[12] M. Jarcho, Calcium phosphate ceramics as hard tissue prosthetics, Clin. Orthop.
Relat. Res. (1981) 259–278.
[13] L.L. Hench, Bioceramics: from concept to clinic, J. Am. Ceram. Soc. 74 (1991)
1487–1510.
[14] X. Liu, P. Chu, C. Ding, Surface modification of titanium, titanium alloys, and related
materials for biomedical applications, Mater. Sci. Eng. R. Rep. 47 (2004) 49–121.
[15] E. Mohseni, E. Zalnezhad, A.R. Bushroa, Comparative investigation on the adhesion
of hydroxyapatite coating on Ti–6Al–4V implant: a review paper, Int. J. Adhes.
Adhes. 48 (2014) 238–257.
[16] S. Yugeswaran, C.P. Yoganand, a. Kobayashi, K.M. Paraskevopoulos, B. Subramanian,
Mechanical properties, electrochemical corrosion and in-vitro bioactivity of yttria
stabilized zirconia reinforced hydroxyapatite coatings prepared by gas tunnel type
plasma spraying, J. Mech. Behav. Biomed. Mater. 9 (2012) 22–33.
[17] T.J. Matsumoto, S.-H. An, T. Ishimoto, T. Nakano, T. Matsumoto, S. Imazato, Zirconia–
hydroxyapatite composite material with micro porous structure, Dent. Mater. 27
(2011) e205–e212, http://dx.doi.org/10.1016/j.dental.2011.07.009.
[18] K. Pardun, L. Treccani, E. Volkmann, G. Li Destri, G. Marletta, P. Streckbein, et al.,
Characterization of wet powder-sprayed zirconia/calcium phosphate coating for
dental implants, Clin. Implant. Dent. Relat. Res. (2013), http://dx.doi.org/10.1111/
cid.12071.
[19] I. Lhermitte-Sebire, R. Colmet, R. Naslain, J. Desmaison, G. Gladel, The adhesion be-
tween physically vapour-deposited or chemically vapour-deposited alumina and
TiC-coated cemented carbides as characterized by Auger electron spectroscopy
and scratch testing, Thin Solid Films 138 (1986) 221–233.
[20] L. Montanaro, C.R. Arciola, D. Campoccia, M. Cervellati, In vitro effects on MG63
osteoblast-like cells following contact with two roughness-differing
fluorohydroxyapatite-coated titanium alloys, Biomaterials 23 (2002) 3651–3659.
[21] M. Stefanic, K. Krnel, T. Kosmac, Novel method for the synthesis of a β-tricalcium
phosphate coating on a zirconia implant, J. Eur. Ceram. Soc. 33 (2013) 3455–3465.
[22] D. Barnes, S. Johnson, R. Snell, S. Best, Using scratch testing to measure the adhesion
strength of calcium phosphate coatings applied to poly(carbonate urethane) sub-
strates, J. Mech. Behav. Biomed. Mater. 6 (2012) 128–138.
[23] A. Börger, P. Supancic, R. Danzer, The ball on three balls test for strength testing of
brittle discs: stress distribution in the disc, J. Eur. Ceram. Soc. 22 (2002) 1425–1436.
[24] M.G. Holthaus, J. Stolle, L. Treccani, K. Rezwan, Orientation of human osteoblasts on
hydroxyapatite-based microchannels, Acta Biomater. 8 (2012) 394–403.
[25] E.M. White, L.A. Hannus, Chemical weathering of bone in archaeological soils, Am.
Antiq. 48 (1983) 316.
[26] A. Ruder, H.P. Buchkremer, H. Jansen, W. Malléner, D. Stöver, Wet powder spraying
— a process for the production of coatings, Surf. Coat. Technol. 53 (1992) 71–74.
[27] K. Ioku, M. Yoshimura, S. Somiya, Microstructure and mechanical properties of hy-
droxyapatite ceramics with zirconia dispersion prepared by post-sintering, Bioma-
terials 11 (1990) 57–61.
[28] L. Fu, K.A. Khor, J.P. Lim, Effects of yttria-stabilized zirconia on plasma-sprayed
hydroxyapatite/yttria-stabilized zirconia composite coatings, J. Am. Ceram. Soc. 85
(2002) 800–806.
[29] J.-Z. Yang, R. Sultana, X.-Z. Hu, Z.-H. Huang, Porous hydroxyapatite coating on strong
ceramic substrate fabricated by low density slip coating-deposition and coating–
substrate co-sintering, J. Eur. Ceram. Soc. 31 (2011) 2065–2071.
[30] H.-W. Kim, H.-E. Kim, V. Salih, J.C. Knowles, Dissolution control and cellular re-
sponses of calcium phosphate coatings on zirconia porous scaffold, J. Biomed.
Mater. Res. A. 68 (2004) 522–530.
[31] J. Zhang, M. Iwasa, N. Kotobuki, T. Tanaka, M. Hirose, H. Ohgushi, et al., Fabrication of
hydroxyapatite–zirconia composites for orthopedic applications, J. Am. Ceram. Soc.
89 (2006) 3348–3355.
[32] M. Stefanic, R. Milacic, G. Drazic, M. Skarabot, B. Budič, K. Krnel, et al., Synthesis of
bioactive β-TCP coatings with tailored physico-chemical properties on zirconia
bioceramics, J. Mater. Sci. Mater. Med. 25 (2014) 2333–2345.
[33] J. Weng, Q. Liu, J.G. Wolke, X. Zhang, K. de Groot, Formation and characteristics of
the apatite layer on plasma-sprayed hydroxyapatite coatings in simulated body
fluid, Biomaterials 18 (1997) 1027–1035.
[34] M.G. Holthaus, L. Treccani, K. Rezwan, Osteoblast viability on hydroxyapatite with
well-adjusted submicron and micron surface roughness as monitored by the prolif-
eration reagent WST-1, J. Biomater. Appl. 27 (2013) 791–800.
[35] Q. Tang, R. Brooks, N. Rushton, S. Best, Production and characterization of HA and
SiHA coatings, J. Mater. Sci. Mater. Med. 21 (2010) 173–181.
[36] I. Izquierdo-Barba, F. Conde, N. Olmo, M.a. Lizarbe, M.a. García, M. Vallet-Regí, Vitre-
ous SiO2–CaO coatings on Ti6Al4V alloys: reactivity in simulated body fluid versus
osteoblast cell culture, Acta Biomater. 2 (2006) 445–455.