M A TE RI A L S CH A RACT ER IZ A TI O N 59 ( 20 0 8 ) 1 2 9 –1 3 3

In vitro precipitation of electrodeposited calcium-deficient

hydroxyapatite coatings on Ti6Al4V substrate

N. Dumelie a , H. Benhayoune a,⁎, D. Richard b , D. Laurent-Maquin b , G. Balossier a

a
INSERM ERM 0203, 21 rue Clément Ader, 51685 Reims BP 138 Cedex 02, France
b
INSERM ERM 0203, 1 rue Maréchal Juin 51096 Reims Cedex, France

AR TIC LE D ATA ABSTR ACT

Article history: In this study, electrodeposited calcium phosphate coatings were characterized by X-ray
Received 24 October 2006 diffraction, using a scanning electron microscope equipped with an EDAX detector, before and
Accepted 29 November 2006 after immersion in DMEM (Dulbecco's Modified Eagle Medium). After 1, 7, 14, and 21 days of
immersion, the calcium and phosphate contents in solution were measured by inductively
Keywords: coupled plasma atomic emission spectrometry (ICP-AES). The results indicated that
Electrodeposition precipitation of the coating occurred. Before immersion in DMEM, the electrodeposited
Hydroxyapatite coating was a mixed crystalline and amorphous calcium-deficient hydroxyapatite with a Ca/P
Calcium phosphate atomic ratio of about 1.5, but during the immersion period these phases rapidly disappeared
Dissolution and were followed by the precipitation of a crystalline apatite with a Ca/P atomic ratio near 1.65.
Precipitation On the basis of these results, we conclude that an electrodeposited calcium phosphate
X-ray microanalysis coating on roughened titanium alloy substrate may act as a precursor for newly precipitated
DMEM calcium phosphate in in vitro experiments independent of cellular activities.
© 2006 Elsevier Inc. All rights reserved.

1. Introduction tite (Ca/P with an atomic ratio ranging from 1.33 to 1.65) seems
to be more soluble and may induce precipitation of a new bone-
Calcium phosphates (CaP) bioceramics have received much like apatite after implantation [10–13]. Electrodeposition has
attention for clinical application in dentistry and orthopedics. recently attracted much attention [14–17] but, surprisingly,
Bulk or porous, the CaP bioceramics in the form of granules or there are few papers about electrodeposited calcium phosphate
tablets are generally used for maxillo-facial reconstruction. coatings and their behavior in cell culture media.
Another use of CaP is as a coating on metallic substrates which In this study, calcium-deficient hydroxyapatite coatings
are currently used as implants [1,2]. were prepared using an electrodeposition method; their
It is believed that CaP coatings show a favorable bone chemical behaviors (dissolution, precipitation, phase evolu-
response in a biological environment, but this characteristic is tion) were subsequently investigated after immersion in
strongly dependent on the surface chemistry and topography DMEM from 1 to 21 days.
as well as its dissolution behavior [3,4]. The most important
concerns are those factors affecting the implant fixation, such
as coating resorption, bone in-growth and mechanical 2. Materials and Methods
strength [5–9].
The most widely used calcium phosphate is hydroxyapatite 2.1. Preparation of Materials
[HA, Ca10(PO4)6(OH)2] but many studies have indicated that the
dissolution of HA in the human body after implantation is too Electrolytes for making calcium phosphate deposits were
low to achieve optimal results. Calcium-deficient hydroxyapa- prepared by mixing a solution of 0.042 M Ca2(NO3)2·4H2O and

⁎ Corresponding author. Tel.: +33 326057564; fax: +33 326051900.

E-mail address: hicham.benhayoune@univ-reims.fr (H. Benhayoune).

1044-5803/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.matchar.2006.11.030
130 MA TE RI A L S CH A R A CT ER IZ A TI O N 59 ( 20 0 8 ) 1 2 9–1 3 3

Table 1 – Main elements composing DMEM used for dissolution experiments

Element CaCl2, 2H2O Fe(NO3, 9H2O KCl MgSO4, 7H2O NaCl NaH2P, 2H2O NaHCO3
Concentration (mg/l) 264 0.10 400 200 6400 141 3700

0.125 M NH4(H2PO4) (MERCK, France). The pH of this solution IAN LIBERTY 2). The measured values were then compared to
was 4.4. Deposits were placed on Ti6Al4V substrate plates those of DMEM alone in the same experimental conditions.
(10 × 10 × 2 mm). The substrate surfaces were blasted with
alumina particles to create a surface with an average roughness 2.3. Physico-chemical and Structural Characterization
over 3 μm. Prior to coating, the substrates were ultrasonically
cleaned in acetone to remove residual grease. Deposition was The morphologies of the coatings were observed before and
carried out with a current density of 5 mA/cm2 at 60 °C for after immersion in DMEM using a LaB6 electron microscope
20 min. A graphite rod was used as a counter electrode. After (JEOL JSM-5400 LV, France) operating at 0–30 kV and a primary
the samples were coated with calcium phosphate, they were beam current of several nA. The specimens were coated with a
removed from the electrolytic solution, dried and annealed at conductive layer of carbon by sputter-coating to avoid
200 °C for 1 h in order to stabilize the coatings. charging effects. A scanning electron microscope (SEM),
equipped with an EDAX ultra-thin window Si(Li) detector,
2.2. Dissolution Experiments was used for X-ray measurements. X-ray spectra were
acquired at a primary beam energy of 30 kV and a primary
Calcium phosphate coatings were immersed in 1.5 ml of DMEM beam current of 5 nA with an acquisition time of 180 s.
(Table 1) supplemented with 10% FCS (fetal calf serum) and 0.5% For the quantitative analysis (concentration and thick-
antiobiotics (penicillin–streptomycin) at 37 °C for 1, 7, 14, and ness), commercial software (STRATAGem, SAMx France)
21 days. After these time periods, the coatings were collected based on the work of Pouchou et al. [18] was used. This
and dried for analysis. The medium was collected to determine software is based on an iterative procedure for the determi-
the amounts of Ca and P; analysis was done using inductively nation of coating concentrations and thickness by using bulk
coupled plasma atomic emission spectrometry (ICP-AES, VAR- standards in the same experimental conditions.
An X-ray diffractometer (Bruker D8 Advance, France) was
used for phase analysis of the coating before and after
immersion in DMEM. The X-ray diffraction (XRD) patterns
were collected from 2Θ = 2° to 60° using a monochromatic
CuKα radiation at steps of 0.02° with a count time of 20 s at
each step. The average degree of crystallinity was evaluated
from the XRD pattern (ratio between amorphous and identi-
fied characteristic peaks).

3. Results and Discussion

Fig. 1 shows SEM micrographs of the calcium phosphate

coating obtained at j = 5 mA/cm2 current density. These results
were obtained in a previous study conducted to study the
influence of current density on calcium phosphate formation
by electrodeposition [19]. Two different morphologies of the

Fig. 1 – SEM micrographs of coating obtained at 5 mA/cm2: (A)

Needles (magnification ×3500), (B) spheres (magnification Fig. 2 – XRD pattern of the calcium phosphate coating
×3500). obtained by electrodeposition at 5 mA/cm2.
 M A TE RI A L S CH A RACT ER IZ A TI O N 59 ( 20 0 8 ) 1 2 9 –1 3 3 131

Ca P
Fig. 3 – Castand
spec spec
(full line) and Pstand (dashed line) ratios measured by ICP-AES as a function of immersion time in DMEM.

coating are observed: a region composed of needles (Fig. 1A) of centrations in DMEM which were in contact with the electro-
approximately 1 μm size, and a region composed of 300– deposited coatings, and Castand and Pstand are, respectively, the
500 nm diameter calcium phosphate spheres (Fig. 1B). The Ca/ calcium and phosphorous concentrations in DMEM without
P ratios and the coating thickness (t) deduced from SEM-EDAX any added material (standard) measured in the same exper-
analysis are: Ca/P = 1.51 ± 0.1, t = 1.9 μm for the region of needles imental conditions as Caspec and Pspec.
and Ca/P = 1.51 ± 0.1, t = 8.2 μm for the region of spheres. These The results obtained are presented in Fig. 3 which clearly
Caspec
Ca/P ratio values correspond to a Ca-deficient hydroxyapatite: shows that Castand ratio decreases significantly from 1 to 14 days
Pspec
Ca-def HAP (needles) and its amorphous phase (spheres). This and stabilizes thereafter, but, the Pstand ratio is relatively stable
was confirmed by the XRD pattern (Fig. 2) which shows the for all immersion times.
typical diffraction peaks at 25.8°–31.6°–32.1°–32.7° and 34° The dissolution and precipitation of CaP can be controlled
corresponding to the Ca-def HAP as indicated in Joint by the diffusion of ions from the CaP surface into the bulk
Committee on Powder Diffraction Standards card no. 82– solution and vice versa. In our case, for the first 7 days of
1943; the presence of the amorphous phase was confirmed by immersion, precipitation occurred, as indicated by an increase
the presence of broadened peaks around 2Θ = 32°. The degree of the Ca/P ratio at this time period on Fig. 4. This figure
of crystallinity of the coating is about 42%. presents the change of the Ca/P ratio measured by X-ray
Next, these calcium phosphate coatings were immersed in microanalysis as a function of time period.
1.5 ml of DMEM for 1, 7, 14, and 21 days. In order to show the It also shows, in agreement with Fig. 3, that after a 14-day
evolution of calcium and phosphorous concentration in immersion in DMEM the precipitated coating seems to be a
DMEM collected after each time period, we determined by stable apatite phase with a Ca/P ratio about 1.65 (dashed line
Caspec Pspec
ICP-AES the ratios Castand and Pstand where: Caspec and Pspec are, in Fig. 4). The Ca/P ratio variation is also related to a change of
respectively, the measured calcium and phosphorous con- the coating surface morphologies. The SEM micrographs of

Fig. 4 – Ca/P atomic ratio of the precipitated apatite measured by X-ray microanalysis as a function of immersion time in DMEM.
132 MA TE RI A L S CH A R A CT ER IZ A TI O N 59 ( 20 0 8 ) 1 2 9–1 3 3

Fig. 6 – XRD pattern of the precipitated apatite: (A) 7 days after

immersion in DMEM; (B) 21 days after immersion in DMEM.

Fig. 5 – SEM micrographs of the precipitated apatite: (A) 7 days coating surface into DMEM would have taken place (dissolu-
after immersion in DMEM; (B) 21 days after immersion in Caspec
tion) and after 1 h of immersion the Castand ratio is slightly lower
DMEM. (magnification ×3500). than 1 which corresponds to the beginning of precipitation.
According to the chemical model for the dissolution of
calcium apatites [23], the initial stages of apatite dissolution
Fig. 5 clearly show that the morphology of the as-deposited consist of calcium detachment from the surface. When all (or
coating (needles and spheres, Fig. 1) is replaced by a structure almost all) neighboring cations of calcium have been removed,
mainly composed of small crystallites. The thickness of the phosphate groups also detach from the surface but immedi-
precipitated coating is about 4 μm as indicated in Table 2. ately precipitate to form apatite. This can explain the fact that
Pspec
The X-ray diffraction patterns of the precipitated coating the Pstand ratio remains stable. This high solubility of the
(Fig. 6) confirm the increased crystallinity with immersion time, electrodeposited CaP is mainly due to the presence of poorly
going from 42% to 47% at 7 days up to 51% at 21 days. Moreover, crystallized phases and the presence of small crystallites
one may notice that diffraction peaks corresponding to the which may represent a higher contact surface area with the
calcite (CaCO3) observed in the XRD pattern of Fig. 2 disappear solution (the specific area) as mentioned by Kwon et al. [24].
progressively with time in DMEM since calcite is converted into There may be some concern that this dissolution process is
calcium phosphate when it is in contact with HPO3− 4 groups as too fast. However, it can be decreased by an appropriate
recently shown by Yoshimura et al. [20]. treatment after the electrodeposition process in order to
However, before precipitation, one would expect to first reduce the amorphous phase in the coating. Indeed, Han
observe dissolution of the as-deposited CaP coating as et al. [25] showed that an hydrothermal treatment can be used
reported by several authors [21,22]. In our case, the precipita- to convert the Ca-def HAP to crystallized HAP.
tion occurs during the early stage of immersion as shown on Nevertheless, this high solubility of electrodeposited Ca-
Caspec
Table 3. This table represents the Castand ratio measured by ICP- def HAP does not affect the biocompatibility of these materials
AES from 15 min to 1 day of immersion. It is seen that at as we have shown in a recent study where the behavior of
Caspec
15 min of immersion, the Castand ratio is slightly higher than 1
which could mean that diffusion of calcium ions from the
Caspec
Table 3 –-
Ca ratio measured by ICP-AES at the early stage
Table 2 – Thickness of the precipitated coating measured
stand

of immersion in DMEM
by X-ray microanalysis
Time period immersion in DMEM 15 min 1h 1 day
Time period (days) 7 14 21
Caspec
Thickness (μm) 4.24 ± 0.22 3.99 ± 0.19 2.99 ± 0.25 Castand ratio 1.02 0.84 0.48
 M A TE RI A L S CH A RACT ER IZ A TI O N 59 ( 20 0 8 ) 1 2 9 –1 3 3 133

human osteoblast-like cells in contact with electrodeposited [12] Wong AT-C, Czernuszka J. Transformation behavior of
Ca-def HAP coatings was investigated [26]. calcium phosphate: 1. Theory and modelling. Colloids Surf A
Physicochem Eng Asp 1993;78:245–53.
[13] Dorozhkin SV. A review on the dissolution models of calcium
apatites. Prog Cryst Growth Charact Mater 2002;44:45–61.
4. Conclusion [14] Han Y, Tao F, Lu J, Xu K. Characterization and stability of
hydroxyapatite coatings prepared by an electrodeposition
The work reported in this paper concerns the dissolution and alkaline-treatment process. J Biomed Materi Res
2001;54:96–101.
behavior of electrodeposited Ca-def HAP coatings without any
[15] Benhayoune H, Laquerrière P, Jallot E, Perchet A, Kilian L,
post-treatment. The results obtained indicate that this
Bubendorff JL, et al. Micrometer level structural and chemical
material may be rapidly dissolved by immersion in DMEM, evaluation of electrodeposited calcium phosphate coatings
which practice induces the precipitation of crystalline apatite. on TA6V substrate by STEM-EDXS. J Mater Sci Mater Med
Consequently, we show that electrodeposition allows the 2002;13:1057–63.
preparation of biomaterials which seem to be suitable for the [16] Kuo MC, Yen SK. The process of electrochemical deposited
surface treatment of orthopedic and dental implants as a hydroxyapatite coatings on biomedical titanium at room
temperature. Mater Sci Eng C Biomim Mater Sens Syst
matrix for the growth of new bone during early stages of
2002;20:153–60.
implantation, thereby avoiding the need for post-treatment of
[17] Lin S, LeGeros RZ, LeGeros John P. Adherent
the implant. octacalciumphosphate coating on titanium alloy using
modulated electrochemical deposition method. J Biomed
Mater Res Part A 2003;68A:819–28.
REFERENCES [18] Pouchou JL. X-ray microanalysis of stratified specimens. Anal
Chim Acta 1993;283:81–97.
[19] Dumelie N, Benhayoune H, Rousse-Bertrand C, Bouthors S,
[1] De Groot K, Geesink RGT, Klein CPAT, Serekian P. Plasma Perchet A, Wortham L, et al. Characterization of
sprayed coatings of hydroxylapatite. J Biomed Mater Res electrodeposited calcium phosphate coatings by
1987;21:1375–81. complementary scanning electron microscopy and
[2] Havelin LI, Engesæter LB, Espehaug B, Furnes O, Lie SA, scanning-transmission electron microscopy associated to
Vollset SE. The Norwegian Arthroplasty Register 11 years and X-ray microanalysis. Thin Solid Films 2005;492:131–9.
73,000 arthroplasties. Acta Orthop Scand 2000; 71(4):337–353 [20] Yoshimura M, Sujaridworakun P, Koh F, Fujiwara T, Pongkao
337. D, Anwar A. Hydrothermal conversion of calcite crystals to
[3] Becker P, Neumann HG. Cellular investigations on hydroxyapatite. Mater Sci Eng C Biomim Mater Sens Syst
electrochemically deposited calcium phosphate composites. J 2004;24:521–5.
Mater Sci, Mater Med 2004;15:437–40. [21] Wolke JGC, van der Waerden JPCM, Schaeken HG, Jansen JA.
[4] Fazan F, Marquis PM. Dissolution behaviour of In vivo dissolution behavior of various RF
plasma-sprayed hydroxyapatite coatings. J Mater Sci, Mater magnetron-sputtered CaP coatings on roughened titanium
Med 2000;11:787–92. implants. Biomaterials 2003;24:2623–9.
[5] Suchanek W, Yoshimura M. Processing and properties of [22] Mavropoulos E, Rossi AM, Da Rocha NNC, Soares GA, Moreira
hydroxyapatite based biomaterials for use as hard tissue JC, Moure GT. Dissolution of calcium-deficient
replacement implants. J Mater Res 1998;13:94–117. hydroxyapatite synthesized at different conditions. Mater
[6] LeGeros RZ, LeGeros JP, Daculsi G, Kijkowska R. In: Wise DL, Charact 2003;50:203–7.
et al, editor. Calcium phosphate biomaterials: preparation, [23] Dorozhkin SV. A review on the dissolution models of calcium
properties and biodegradation. Encyclopedic handbook of apatites. Prog Cryst Growth Charact 2002;44:45–61.
biomaterials and bioengineeringNew York: Marcel Dekker; [24] Kwon SH, Jun YK, Hong SH, Kim HE. Synthesis and dissolution
1995. Part A:1429– 63. behavior of β-TCP and HA/β-TCP composite powders. J Eur
[7] Ducheyne P, Qiu Q. Bioactive ceramics, the effect of surface Ceram Soc 2003;23:1039–45.
reactivity on bone formation and bone cell function. [25] Han Y, Xu K, Lu J. Morphology and composition of
Biomaterials 1999;20:2287–303. hydroxyapatite coatings prepared by hydrothermal
[8] Dalton JE, Cook SD. In vivo mechanical and histological treatment on electrodeposited brushite coatings. J Mater Sci
characteristics of HA-coated implants vary with coating Mater Med 1999;10:243–8.
vendor. J Biomed Mater Res 1995;29:239–45. [26] Richard D, Dumelie N, Benhayoune H, Bouthors S, Guillaume
[9] Klein C, Patka P, Wolke J, de Blieck-Hogervorst J, De Groot K. C, Lalun N, et al. Human osteoblast-like cells behavior in
Long-term in vivo study of plasma-sprayed coatings on contact with electrodeposited calcium phosphate coatings.
titanium alloys of tetracalcium-phosphate, hydroxylapatite J Biomed Mater Res B 2006;79B:108–15.
and a-tricalciumphosphate. Biomaterials 1994;15:146–50.
[10] Monteiro MM, Rocha NCC, Rossi AM, Soares GA. Dissolution
properties of calcium phosphate granules with different
compositions in simulated body fluid. J Biomed Mater Res
2003;65:299–305.
[11] Narasaraju TSB, Phebe DE. Some physico-chemical aspects of
hydroxyapatite. J Mater Sci, Mater Med 1996;31:1–21.