- Email: [email protected]
Hydroxyapatite, fluor-hydroxyapatite and fluorapatite produced via the sol–gel method. Optimisation, characterisation and rheology
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
Available online at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Hydroxyapatite, fluor-hydroxyapatite and fluorapatite produced via the sol–gel method. Optimisation, characterisation and rheology Christopher J. Tredwin a , Anne M. Young b , George Georgiou b , Song-Hee Shin c , Hae-Won Kim c,d,e , Jonathan C. Knowles b,c,∗ a
Plymouth University Peninsula Schools of Medicine and Dentistry, Universities of Exeter & Plymouth, The John Bull Building, Tamar Science Park, Research Way, Plymouth PL6 8BU, United Kingdom b Division of Biomaterials and Tissue Engineering, UCL Eastman Dental Institute, 256 Grays Inn Road, London WC1X 8LD, United Kingdom c WCU Research Centre of Nanobiomedical Science, Dankook University, San#29, Anseo-dong, Dongnam-gu, Cheonan-si, Chungnam 330-714, South Korea d Institute of Tissue Regeneration Engineering (ITREN), Dankook University, Cheonan 330-714, South Korea e Department of Biomaterials Science, School of Dentistry, Dankook University, Cheonan 330-714, South Korea
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. Currently, most titanium implant coatings are made using hydroxyapatite and a
Received 6 February 2012
plasma spraying technique. There are however limitations associated with plasma spraying
Received in revised form
processes including poor adherence, high porosity and cost. An alternative method utilising
19 September 2012
the sol-gel technique offers many potential advantages but is currently lacking research
Accepted 7 November 2012
data for this application. It was the objective of this study to characterise and optimise the production of Hydroxyapatite (HA), fluorhydroxyapatite (FHA) and fluorapatite (FA) using a sol-gel technique and assess the rheological properties of these materials.
Keywords:
Methods. HA, FHA and FA were synthesised by a sol-gel method. Calcium nitrate and tri-
Hydroxyapatite
ethylphosphite were used as precursors under an ethanol-water based solution. Different
Fluor-hydroxyapatite
amounts of ammonium fluoride (NH4F) were incorporated for the preparation of the sol-
Fluorapatite
gel derived FHA and FA. Optimisation of the chemistry and subsequent characterisation of
Sol–gel
the sol-gel derived materials was carried out using X-ray Diffraction (XRD) and Differential
Optimisation
Thermal Analysis (DTA). Rheology of the sol-gels was investigated using a viscometer and
Characterisation
contact angle measurement.
Rheology
Results. A protocol was established that allowed synthesis of HA, FHA and FA that were at least 99% phase pure. The more fluoride incorporated into the apatite structure; the lower the crystallisation temperature, the smaller the unit cell size (changes in the a-axis), the higher the viscosity and contact angle of the sol-gel derived apatite. Significance. A technique has been developed for the production of HA, FHA and FA by the solgel technique. Increasing fluoride substitution in the apatite structure alters the potential coating properties. Crown Copyright © 2012 Published by Elsevier Ltd on behalf of Academy of Dental Materials. All rights reserved.
∗ Corresponding author at: Division of Biomaterials and Tissue Engineering, UCL Eastman Dental Institute, 256 Grays Inn Road, London WC1X 8LD, United Kingdom. Tel.: +44 203456 1189. E-mail address: [email protected] (J.C. Knowles).
0109-5641/$ – see front matter Crown Copyright © 2012 Published by Elsevier Ltd on behalf of Academy of Dental Materials. All rights reserved.
http://dx.doi.org/10.1016/j.dental.2012.11.008
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
1.
Introduction
Titanium (Ti) and its alloys have long been recognized and used as dental and orthopedic implant materials. To improve the implant-tissue osseointegration, considerable effort has been exerted to modify the Ti surface structure both physically and chemically [1–3]. Hydroxyapatite [HA, Ca10 (PO4 )6 (OH)2 ] coatings on Ti substrate have attracted significant attention for several years aimed at combining the excellent biocompatibility of HA and load-bearing ability of Ti [4,5]. Currently, most HA coatings are produced by a plasma-spraying technique [6,7]. In vivo reports on the system have shown good bone-bonding ability and fast osseointegration compared with pure Ti implants. This was mainly attributed to the osteoconductivity and chemical and biological similarities of HA with the human hard tissues [5,8,9]. There is some debate as to the actual requirements of a coating. There are two possibilities: (1) the coating has very low solubility and will remain in place for some considerable time (for the devices lifetime) and maintain a stable biological interface or (2) the coating is merely temporary and its role is to promote rapid growth of bone onto the surface of the coating but the coating will be replaced such that the bone will bond directly to the metal implant surface. This is of particular relevance to immediate loading implants. This is further complicated by the role of crystallinity. Lower crystallinity coatings showed faster resorption in a canine model and also higher levels of bone apposition onto the surface at 16 weeks, with mechanical fixation better in the lower crystallinity samples at 16 weeks but no difference at 32 weeks [10,11]. There are some parameters to be improved in the plasma spraying process, such as coating strength, chemical homogeneity, and residual porosity. These are related to the high fabrication temperature and coating thickness [7]. Recently, alternative methods have been developed to produce thin HA films. The sol–gel technique, being one of the thin film methods provides some benefits over the plasma spraying method, such as chemical homogeneity, fine grain structure, and low processing temperature [12]. Moreover, compared with other thin film methods, it is simple and cost efficient, as well as effective for the coating of complex-shaped implants. There are conflicting reports [13] which may be due to differing methodologies, but in general it is agreed that pure fluorapatite (FA, Ca10 (PO4 )6 F2 ) is known to have a much lower solubility than HA, because FA possesses a greater stability than HA, both chemically and structurally [13,14]. Moreover, the HA is able to form fluorhydroxyapatite (FHA, Ca10 (PO4 )6 F1 OH1 ) with the crystallographic substitution of OH− by F− . Moreover this substitution has been shown to occur in aqueous solutions and is also irreversible [15]. Hence, modulation of the extent of fluoride substitution provides an effective way of controlling the solubility of the apatite which thus allows us to produce a wide range of coatings with differing biological properties. In practice, the fluoride ion itself has been studied widely in dental restorative areas, due to its advantages over other ions in that it can reduce the formation of caries in bacterially contaminated environments and promotes mineralisation and crystallisation of calcium phosphates in the formation of bone [14,16]. There is some conflict in the literature when com-
167
paring hydroxyapatite with fluorhydroxyapatite coatings. In unloaded models, a fluorine containing hydroxyapatite coating was shown to be more stable [7]. However, when loaded, no significant difference was seen [17,18], although both types of implants showed good stability. Interestingly in a human trial, the hydroxyapatite coating was considerably thinner than the equivalent fluorhydroxyapatite coating [19]. Nevertheless, there have been few reports concerning the fabrication or characterisation of the HA, FA or FHA materials produced via the sol–gel route and it was the purpose of this study to undertake this. Furthermore with reference to the literature, while use of triethylphosphite and calcium nitrate as precursors have been the most successful method for the production of HA via a sol gel [20–22], the majority of the materials produced via the sol–gel reactions are -tricalcium phosphate (-TCP) and not HA/FA/FHA and this study sought to optimise the production of phase pure apatites and investigate their thermal and rheological properties.
2.
Materials and methods
2.1.
Preparation of HA sols
16.16 g of triethylphosphite (TEP [P(C2 H5 0)3 ], Aldrich, USA) was hydrolysed for 72 h in a mixture of 33.12 g of ethanol and 5.04 g of distilled water (P containing solution, VWR, UK). This mixture was then added to a solution of 39.36 g calcium nitrate [Ca(NO3 )2 ·4H2 0, Aldrich USA] in 15.12 g of distilled water. A 5% (w/v) solution of ammonium hydroxide (NH4 OH, VWR, UK) was added to the solution in order to improve gelation and subsequent formation of an apatite structure. The solution was allowed to react for 24 h and then age for a further 24 h at room temperature.
2.2.
Preparation of FHA and FA sols
The FHA sols were prepared using various amounts of ammonium fluoride (NH4 F, Aldrich, USA) in the P containing solution. The [P]/[F] molar ratios were 12, 6, 4 and 3 in order to have the corresponding compositions of Ca10 (PO4 )6 F0.5 OH1.5 , Ca10 (PO4 )6 F1 OH1 , Ca10 (PO4 )6 F1.5 OH0.5 and Ca10 (PO4 )6 F2 by replacing the OH group with F ions in molar ratios of 0.25, 0.5, 0.75 and 1 respectively. After stirring for 72 h, the solutions were added slowly to a solution containing a stoichiometric amount (Ca/P ∼1.67) of calcium nitrate [Ca(NO3 )2 ·4H2 0, Aldrich, USA] following the protocol developed for the HA sols.
2.3.
Powder preparation
Following successful preparation of apatites via the sol–gel reaction, samples of the sol–gel derived materials were heated to 500, 600, 700, 800, 900 and 1000 ◦ C for 2 h in air using a hot air oven (Lenton Furnace, Lenton Thermal Designs Limited, UK) with a ramp of 5 ◦ C/min, dwell time of 60 min and cool down rate of 10 ◦ C/min. Following heating in the hot air oven the samples were ground to a powdery form for 20 min using a vibrating agate ball mill (Fritsch, Germany) and to ensure similar particle sizes ≈20 m were passed through a sieve (Fritsch, Germany).
168
2.4.
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
Powder X-ray diffraction (XRD)
X-ray diffraction data were collected using an automated Bruker D8 Advance system equipped with a sample changer ˚ in a flat plane – geometry and Ni filtered CuK␣ ( = 1.5418 A) in the range 10–100◦ 2, in steps of 0.02◦ , with a count time of 12 s per step and using a Lynx Eye detector. From the data obtained, the unit cell size was calculated using the software TOPAS Academic V4.1. The kernel of this software is the same as that for the Bruker-AXS Topas software.
2.5.
Differential thermal analysis (DTA)
The sol–gel derived materials were evaluated using differential thermal analysis (DTA) to elucidate the thermal and weight changes as a function of time, using a Labsys TG/DTA 1600 ◦ C (Setaram Instruments, Caluire, France). Weighed (100 mg) sol–gel derived samples were placed into platinum sample holders and tested in an inert nitrogen atmosphere, and compared with a platinum pan only as a reference. Samples were run from 20 ◦ C to 1000 ◦ C at a rate of 20 ◦ C/min. The data was baseline corrected by carrying out a blank run and subtracting this from the plot obtained. The experiment was completed three times so a mean and standard deviation could be obtained for the crystallisation temperature.
2.6.
Rheology
A viscometer (Model DV-III, Brookfield, USA) was used to investigate the varying viscosity of the sol–gels after a set aging time (24 h) at various shear rates (from 20 to 200 s−1 with intervals of 10 s−1 ), each for 20 s at a constant temperature of 25 ◦ C. This was repeated 3 times. Changes in viscosity at a constant shear rate as the sample was aged were also measured. After mixing the hydrolysed Triethyl phosphite [P(C2 H5 0)3 ] Calcium Nitrate [Ca(NO3 )2 ·4H2 0] and ammonium hydroxide (NH4 OH) the viscosity at a constant shear rate 150 s−1 was measured every 6 h up to 72 h. This was repeated 3 times.
2.7.
Contact angle measurement
Using a standardised protocol 5 mL of the sol–gel samples were pipetted onto a commercially pure titanium disk and contact angle measurements were undertaken with a CAM 200 Optical Contact Light Meter (KSV, Instruments Limited, Finland) every second for 30 s. The measurements were repeated three times for each sol–gel derived material.
3.
Results
3.1.
HA/FHA/FA production
Using the protocol defined, HA derived from a sol–gel which was 99.7% HA and 0.3% -TCP purity was obtained. The Xray diffraction pattern associated with this is seen in Fig. 1a. All other sol–gels derived samples made (i.e. Ca10 (PO4 )6 F2 , Ca10 (PO4 )6 OH0.5 F1.5 , Ca10 (PO4 )6 OH1 F1 and Ca10 (PO4 )6 OH1.5 F0.5 ) exhibited a similar XRD pattern (data not shown) and were
thermally stable to 1000 ◦ C. The phases measured in the fluoride substituted apatites can be seen in Table 1.
3.2.
Differential thermal analysis
Fig. 1b shows an example combined TG/DTA trace for sample Ca10 (PO4 )6 OH0.5 F1.5 . As can be seen there are three main thermal events. The first is at around 100 ◦ C and is associated with unbound water loss as there is a thermal and associated large mass loss event. A second event occurs at around 300 ◦ C and is associated with bound organic loss, again signified by the large weight loss. A third sharp thermal event occurs at around 450 ◦ C and there is also a very slow weight change event. These are thought to be unrelated as they differ in rates and also the weight change remains in the same temperature region but the thermal event changes with composition. This third thermal event was assigned to crystallisation and the temperatures at which they occur are detailed in Table 2, which shows the mean and standard deviation obtained for the crystallisation temperatures of each of the sol gels. All of the materials were thermally stable to 1000 ◦ C.
3.3.
Unit cell measurements
Comparable graphs of the a-axis, c-axis and volume changes as a reflection of material type and temperature are shown in Figs. 2–4. Figs. 2 and 4 showed that increasing the fluoride content in the apatite structure decreased the a axis and volume of the unit cell. Furthermore as the heating temperature increased there was a trend for the a axis and volume to decrease in size. This was particularly marked with the Ca10 (PO4 )6 F0.5 OH1.5 . Fig. 3 shows that fluoride content and heating temperature had minimal effect on the c axis.
3.4.
Rheology
3.4.1. Varying viscosity of the sol–gels after 24 aging time at various shear rates Fig. 5 shows the viscosity changes (mean ± standard deviation) of the sol–gels after aging for 24 h, as represented with respect to shear rate. The viscosity of the sol–gels was as follows; Ca10 (PO4 )6 F2 > Ca10 (PO4 )6 OH0.5 F1.5 > Ca10 (PO4 )6 )OH1 F1 > Ca10 (PO4 )6 OH1.5 F0.5 > Ca10 (PO4 )6 (OH)2 . This pattern in viscosity was consistent throughout the entire time period tested. Up to a shear rate of 40 sec−1 there was a decrease in the viscosity of all materials. Increasing the fluoride content decreased the viscosity of the materials.
3.4.2. Measuring changes in viscosity at a constant shear rate as the sample is aged The mean viscosity change (±standard deviation) was plotted with aging time at a constant shear rate of 150 s−1 . This is shown in Fig. 6. All of the sol–gels showed a rapid increase in viscosity over the first 24 h, this subsequently stabilized and saw minimal changes over the next 48 h. At each time point the increasing fluoride content shows an increase in viscosity.
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
169
Fig. 1 – (a) XRD pattern of the of the HA sol–gel after heating to 1000 ◦ C. (b) Combined thermogravimetric and differential thermal analysis of Ca10 (PO4 )6 OH0.5 F1.5 sample.
3.5.
Contact angle measurement
The overall mean contact angles, plus and minus the standard deviation for each of the sol–gels is shown in Table 3. Fig. 7 graphically shows the contact measurement as a result of time of the HA, FA and FHA sol–gels. With reference to Fig. 7 it can be seen that initially on dropping the sol–gel, for approximately the first 4 s there is rapid decrease in contact angle, subsequently the reduction starts to slow. The contact angle is higher with the increasing fluoride concentration.
4.
Discussion
4.1.
Hydroxyapatite (HA) production
From the literature the most promising combination for the precursors to produce HA have been calcium nitrate and triethylphosphite (TEP) using alcohol at the hydrolysis stage [23].
However Kim et al. (2004) could not produce phase pure HA and were unable to reduce the significant amounts of beta tricalcium phosphate (-TCP) formed. Initial hydrolysation of the TEP is required to increase its reactivity so that it can react readily with the calcium nitrate. The first stage of optimizing the chemistry involved varying the amount of water used to hydrolyse the TEP in ethanol. Using the basic protocol from the literature [23] it was found that 100 mL of water was required to result in sol-gel production. The second stage of investigation found that the hydrolysation time of the TEP was optimal at 72 h and at this stage it can be proposed that total substitution has taken place and the TEP is able to react efficiently with the calcium nitrate. The ethanol produced as a side product from this reaction has a boiling point of 78 ◦ C and can be easily driven off once the sol–gel is heated to 80 ◦ C in the hot air oven. It was found that a mixing time of 24 h as used by Kim et al., 2004 was effective in the production of HA [23]. The addition of ammonium hydroxide can catalyse the reaction between TEP and calcium nitrate and this was confirmed in this study [23].
170
99.8% 0.2%
0.2%
Ca10 (PO4 )6 F2 -TCP
Fig. 3 – Graph to show c-axis changes with heating of the sol–gels (mean ± SD).
The sol–gel production protocols that had been established for HA sol–gel production were utilised as the basic methodology for the production of the fluoride substituted apatite sol–gels. With reference to the literature ammonium fluoride was chosen as the donor for the fluoride ions [23]. After preliminary pilot experiments it proved that this should initially be added to the TEP and mixed for 24 h and then both subsequently added to the calcium nitrate. With attention to the chemistry and controlling the amounts of ammonium fluoride initially added attempt was made to control the level of fluoride substitution in the apatite structure.
99.8% 0.3% 99.7%
0.3%
99.8%
0.2%
99.7%
Ca10 (PO4 )6 OH0.5 F1.5
Fig. 2 – Graph to show a-axis changes with heating of the sol–gels (mean ± SD).
-TCP Ca10 (PO4 )6 OH1 F1 -TCP Ca10 (PO4 )6 OH1.5 F0.5 -TCP Ca10 (PO4 )6 OH2
Table 1 – Constituents components of HA, FA and fluoride substituted apatites.
-TCP
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
Fig. 4 – Graph to show changes in unit cell volume with heating of the sol–gels from 500 ◦ C to 1000 ◦ C (mean ± SD).
171
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
7.5
Table 2 – Crystallisation temperature of fluoride substituted apatite compositions as determined by DTA.
Ca10 (PO4 )6 F2 Ca10 (PO4 )6 OH0.5 F1.5 Ca10 (PO4 )6 OH1 F1 Ca10 (PO4 )6 OH1.5 F0.5 Ca10 (PO4 )6 OH2
4.2.
7
Crystallisation temperature mean ± SD (◦ C) 467 470 492 498 499
± ± ± ± ±
6.5
Viscosity (MPa)
Composition
Shear rate = 150 s-1
0.21 0.43 1.22 0.23 0.65
6
5.5
5
Differential thermal analysis
There is a clear trend in the crystallisation temperatures seen. As the level of fluoride substitution increased the crystallisation temperature decreased. During substitution the F− occupies to varying degrees (depending on the level of substitution) the position of OH− in the apatite. It is also known that increasing the level of fluoride substitution in the apatite structure produces a smaller more compact crystal structure [24]. The results of this study suggest that the increasing levels of fluoride allow the structure to crystallise more easily and hence have a lower crystallisation temperature. This finding could be attributed to the fact that increasing fluoride substitution results in a more stable, compact structure that is capable of crystallising quicker than HA.
Fig. 5 – Mean viscosity changes in HA, FA, and FHA sol-gels after aging for 24 h versus shear rate (mean ± SD). With reference to figure 62 several clear patterns can be seen. The initial [2]viscosity of the sol–gels is such that the higher the concentration of fluoride in the.
4.5
0
24
48
72
AgingTime time (Hou rs) Aging (Hou rs)
Fig. 6 – Viscosity changes in HA, FA and FHA sol–gels measured at shear rate of 150 s−1 , as represented with respect to aging (mean ± SD).
4.3.
Unit cell measurements
All of the sol–gel produced fluoride substituted apatites had manifestly smaller a-axis than the HA produced from the sol–gel. This is in line with the published literature, which reported in the case of FHA powders, there were significant decreases in the a-axis compared to HA [13,25,26]. The more fluoride substitution the smaller the aaxis, such that the a-axis decreases in size as follows: Ca10 (PO4 )6 F0.5 OH1.5 > Ca10 (PO4 )6 F1 OH1 > Ca10 (PO4 )6 F1.5 OH0.5 > Ca10 (PO4 )6 F2 . The fluoride ion has a smaller size than OH− ,
Fig. 7 – Contact angle measurements for each of the sol–gels.
Table 3 – Mean contact angle of each of the sol–gels. Mean (±SD) contact angle (◦ ) apatite sol–gel (Ca10 (PO4 )6 ) (OH)2 55.34 ± 0.35
(F0.5 OH1.5 )2
(F1 OH1 )2
(F1.5 OH0.5 )2
F2
56.36 ± 0.48
57.32 ± 0.37
58.12 ± 0.44
60.34 ± 0.46
172
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
˚ compared to 1.37 A, ˚ and as such is able to pack being 1.28 A more closely into the apatite structure [27]. There is a downward trend in the size of the a-axis as the firing temperature is increased. Abrahams and Knowles (1994) investigated the effects of sintering conditions on hydroxyapatite using X-ray diffraction [28]. They were investigating temperatures in excess of 1000 ◦ C and found small but significant decreases in unit-cell volume as temperature increased. They proposed this to be due to a loss of small amounts of carbonate ion from channels within the structure. It is possible to suggest that the pattern seen in the unit cell volume here is similar and may be due to loss of small amounts of carbonate ion at the lower temperatures investigated. However this was not seen when examined via FTIR. Furthermore heating the apatite structure to temperatures above the crystallisation temperature, but below the decomposition temperature, allows structural rearrangement to a more optimal position. The more optimal position of the atoms the smaller the a-axis and overall unit cell volume will be. The a-axis values for the Ca10 (PO4 )6 F2 (FA) were similar to ˚ [25,29]. The the reported values for pure FA (a-axis 9.3684 A) similarity of the values obtained in this study to reported values provides further support for the purity of the samples that have been produced by the sol–gel method adopted. While there were manifest changes in the a-axis, there were little changes observed in the c-axis. This is well reported in the literature for fluoride substituted apatites [13,25,26]. Within the apatite structure there is no spare volume for the structure to change in dimensions along the c axis, therefore it is to be expected that little change would be seen. These findings give further confidence in the data. All of the sol–gel produced fluoride substituted apatites had manifestly smaller unit cell volumes than the HA produced from the sol–gel. The more fluoride substitution the smaller the a-axis, such that the volume (Å3 ) decreases in size as follows: Ca10 (PO4 )6 (OH)2 > Ca10 (PO4 )6 F0.5 OH1.5 > Ca10 (PO4 )6 F1 OH1 > Ca10 (PO4 )6 F1.5 OH0.5 > Ca10 (PO4 )6 F2 . In direct proportion to the changes seen in the a-axis, there is a downward trend in the unit cell volume seen as the firing temperature is increased. Because of the clear correlations above it can be proposed that the controlled incorporation of fluoride in the sol–gel protocol used leads to the successful formation of FHA solid solutions with differing levels of fluoride substitution.
4.4.
Rheology
4.4.1. Varying viscosity of the sol–gels after 24 h aging time at various shear rates With reference to Fig. 5 it can be seen that after aging for 24 h, the viscosity of all the sol–gels showed an exponential decrease at low shear rates and stabilised with the higher shear rates. This behaviour is characteristic of sol–gels. As the shear rate is initially increased the viscosity of the sol–gel falls as there is a breakdown in the sol structure [30,31]. As the shear rates become higher, this effect becomes reduced such that any additional increases in shear rate cause negligible effects on the viscosity [30,31].
There is a clear pattern that can be seen throughout; the higher the concentration of fluoride in the sol–gel the higher the viscosity i.e. the viscosity of the sol–gels was as follows; Ca10 (PO4 )6 F2 > Ca10 (PO4 )6 OH0.5 F1.5 > Ca10 (PO4 )6 )OH1 F1 > Ca10 (PO4 )6 OH1.5 F0.5 > Ca10 (PO4 )6 (OH)2 . A potential theory to explain this is that there could be stronger cross links within the sol–gel with higher fluoride content. As a result of these stronger links the viscosity is increased. Clearly the rheological results obtained here are novel and may have implications when coating the sol–gels on to titanium i.e. to achieve the same coating with increasing fluoride concentration, different coating speeds may have to be adopted.
4.4.2. Measuring changes in viscosity at a constant shear rate as the sample is aged There is a clear pattern in the viscosity of the sol–gels; the higher the concentration of fluoride in the sol–gel the higher the viscosity i.e. the viscosity of the sol–gels was as follows; Ca10 (PO4 )6 F2 > Ca10 (PO4 )6 OH0.5 F1.5 > Ca10 (PO4 )6 OH1 F1 > Ca10 (PO4 )6 OH1.5 F0.5 > Ca10 (PO4 )6 (OH)2 . Again a possible theory that can be suggested for the reason for the increasing viscosity as more fluoride is substituted in to the sol–gel structure could be that there could be stronger cross links within the sol–gel with higher fluoride content. As a result of these stronger links this may have an overall effect on the viscosity.
4.5.
Contact angle measurement
For the contact angle measurements, at the initial time, the contact angles were relatively high, between around 70◦ and 75◦ . With time, for all samples the contact angle values dropped sharply, such that after around nearly 30 s, the values had dropped to values around 50–55◦ . Generally, if surfaces have contact angles less than 90◦ , they are regarded as hydrophilic [32]. With time the values drop as the surface becomes wet by the applied drop and the drop spreads. Throughout all contact angle measurements, at all times, there was a pattern with the contact angle being higher with increasing fluoride concentration such that the contact angles were as follows; Ca10 (PO4 )6 F2 > Ca10 (PO4 )6 OH0.5 F1.5 > Ca10 (PO4 )6 )OH1 F1 > Ca10 (PO4 )6 OH1.5 F0.5 > Ca10 (PO4 )6 (OH)2 . This suggests that increasing the concentration of fluoride seems to increase the hydrophobicity (contact angle) of the sol–gel. However it should be borne in mind that the values for all the sol–gel solutions were relatively similar and all of them may be regarded as hydrophilic. This finding may have implications when coating the sol–gels on to titanium. It can be proposed, that because of the altered rheological characteristics with higher levels of fluoride substitution, in the apatite structure produced by the sol–gel technique, different coating speeds/protocols may have to be adopted when coating the sol–gels on to titanium.
5.
Conclusions
A protocol has been established for producing HA, FHA and FA with virtually 100% purity, using the sol–gel process and
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
increasing incorporation of fluoride into the apatite structure results in a decreased crystallisation temperature and a more compact unit cell structure. Addition of fluoride results in changes in the rheological properties of the sol gel and may have implications if used to coat titanium.
Acknowledgement This work was supported in part (JCK) by WCU Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (No. R31-10069).
references
[1] Adell R, Lekholm U, Rockler B, Branemark PI. A 15-year study of osseointegrated implants in the treatment of the edentulous jaw. International Journal of Oral Surgery 1981;10:387–416. [2] Ratner BD. New ideas in biomaterials science – a path to engineered biomaterials. Journal of Biomedical Materials Research 1993;27:837–50. [3] Nanci A, Wuest JD, Peru L, Brunet P, Sharma V, Zalzal S, et al. Chemical modification of titanium surfaces for covalent attachment of biological molecules. Journal of Biomedical Materials Research 1998;40:324–35. [4] Block MS, Finger IM, Fontenot MG, Kent JN. Loaded hydroxylapatite-coated and grit-blasted titanium implants in dogs. International Journal of Oral and Maxillofacial Implants 1989;4:219–25. [5] McPherson EJ, Dorr LD, Gruen TA, Saberi MT. Hydroxyapatite-coated proximal ingrowth femoral stems. A matched pair control study. Clinical Orthopaedics and Related Research 1995:223–30. [6] Bloebaum RD, Dupont JA. Osteolysis from a press-fit hydroxyapatite-coated implant. A case study. The Journal of Arthroplasty 1993;8:195–202. [7] Overgaard S, Soballe K, Josephsen K, Hansen ES, Bunger C. Role of different loading conditions on resorption of hydroxyapatite coating evaluated by histomorphometric and stereological methods. Journal of Orthopaedic Research: Official Publication of the Orthopaedic Research Society 1996;14:888–94. [8] Hench LL. Bioceramics – from concept to clinic. Journal of the American Ceramic Society 1991;74:1487–510. [9] Ducheyne P, Cuckler JM. Bioactive ceramic prosthetic coatings. Clinical Orthopaedics and Related Research 1992:102–14. [10] Overgaard S, Bromose U, Lind M, Bunger C, Soballe K. The influence of crystallinity of the hydroxyapatite coating on the fixation of implants. Mechanical and histomorphometric results. The Journal of Bone and Joint Surgery British 1999;81:725–31. [11] Soballe K, Overgaard S, Hansen ES, Brokstedt-Rasmussen H, Lind M, Bunger C. A review of ceramic coatings for implant fixation. Journal of Long-Term Effects of Medical Implants 1999;9:131–51. [12] Brinker CJ, Scherer GW. Sol–gel science: the physics and chemistry of sol–gel processing. Academic Press; 1990. [13] Larsen MJ, Jensen SJ. Solubility, unit cell dimensions and crystallinity of fluoridated human dental enamel. Archives of Oral Biology 1989;34:969–73.
173
[14] Moreno EC, Kresak M, Zahradnik RT. Fluoridated hydroxyapatite solubility and caries formation. Nature 1974;247:64–5. [15] Fulmer M, Brown PW. Low-temperature formation of fluorapatite in aqueous-solution. Journal of the American Ceramic Society 1992;75:3401–7. [16] Marie PJ, Hott M. Short-term effects of fluoride and strontium on bone formation and resorption in the mouse. Metabolism: Clinical and Experimental 1986;35:547–51. [17] Overgaard S, Lind M, Josephsen K, Maunsbach AB, Bunger C, Soballe K. Resorption of hydroxyapatite and fluorapatite ceramic coatings on weight-bearing implants: a quantitative and morphological study in dogs. Journal of Biomedical Materials Research 1998;39:141–52. [18] Overgaard S, Lind M, Glerup H, Grundvig S, Bunger C, Soballe K. Hydroxyapatite and fluorapatite coatings for fixation of weight loaded implants. Clinical Orthopaedics and Related Research 1997:286–96. [19] Overgaard S, Soballe K, Lind M, Bunger C. Resorption of hydroxyapatite and fluorapatite coatings in man. An experimental study in trabecular bone. The Journal of Bone and Joint Surgery British 1997;79:654–9. [20] Kim HW, Kim HE, Knowles JC. Sol–gel apatite films on titanium implant for hard tissue regeneration. Bioceramics 2004;16:254–62, 423-6. [21] Kim HW, Kim HE, Kim HW, Knowles JC. Improvement of hydroxyapatite sol–gel coating on titanium with ammonium hydroxide addition. Journal of the American Ceramic Society 2005;88:154–9. [22] Kim HW, Kim HE, Knowles JC. Fluor-hydroxyapatite sol–gel coating on titanium substrate for hard tissue implants. Biomaterials 2004;25:3351–8. [23] Kim HW, Knowles JC, Salih V, Kim HE. Hydroxyapatite and fluor-hydroxyapatite layered film on titanium processed by a sol–gel route for hard-tissue implants. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2004;71:66–76. [24] Song WH, Jun YK, Han Y, Hong SH. Biomimetic apatite coatings on micro-arc oxidized titania. Biomaterials 2004;25:3341–9. [25] Smith DK. Calcium phosphate apatites in nature. Hydroxyapatite and Related Materials 1994:29–45. [26] Kim HW, Koh YH, Yoon BH, Kim HE. Reaction sintering and mechanical properties of hydroxyapatite-zirconia composites with calcium fluoride additions. Journal of the American Ceramic Society 2002;85:1634–6. [27] Shannon RD. Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Crystallographica Section A 1976;32:751–67. [28] Abrahams I, Knowles JC. Effects of sintering conditions on hydroxyapatite for use in medical applications – a powder diffraction study. Journal of Materials Chemistry 1994;4:185–8. [29] Elliott JC. Structure and chemistry of the apatites and other calcium orthophosphates. Amsterdam: Elsevier; 1994. [30] Ben-Nissan B, Choi AH. Sol–gel production of bioactive nanocoatings for medical applications. Part 1: an introduction. Nanomedicine (London) 2006;1:311–9. [31] Choi AH, Ben-Nissan B. Sol–gel production of bioactive nanocoatings for medical applications. Part II: current research and development. Nanomedicine (London) 2007;2:51–61. [32] Förch R, Schönherr H, Jenkins ATA. Surface design: applications in bioscience and nanotechnology. Weinheim: Wiley-VCH; 2009.
Available online at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Hydroxyapatite, fluor-hydroxyapatite and fluorapatite produced via the sol–gel method. Optimisation, characterisation and rheology Christopher J. Tredwin a , Anne M. Young b , George Georgiou b , Song-Hee Shin c , Hae-Won Kim c,d,e , Jonathan C. Knowles b,c,∗ a
Plymouth University Peninsula Schools of Medicine and Dentistry, Universities of Exeter & Plymouth, The John Bull Building, Tamar Science Park, Research Way, Plymouth PL6 8BU, United Kingdom b Division of Biomaterials and Tissue Engineering, UCL Eastman Dental Institute, 256 Grays Inn Road, London WC1X 8LD, United Kingdom c WCU Research Centre of Nanobiomedical Science, Dankook University, San#29, Anseo-dong, Dongnam-gu, Cheonan-si, Chungnam 330-714, South Korea d Institute of Tissue Regeneration Engineering (ITREN), Dankook University, Cheonan 330-714, South Korea e Department of Biomaterials Science, School of Dentistry, Dankook University, Cheonan 330-714, South Korea
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. Currently, most titanium implant coatings are made using hydroxyapatite and a
Received 6 February 2012
plasma spraying technique. There are however limitations associated with plasma spraying
Received in revised form
processes including poor adherence, high porosity and cost. An alternative method utilising
19 September 2012
the sol-gel technique offers many potential advantages but is currently lacking research
Accepted 7 November 2012
data for this application. It was the objective of this study to characterise and optimise the production of Hydroxyapatite (HA), fluorhydroxyapatite (FHA) and fluorapatite (FA) using a sol-gel technique and assess the rheological properties of these materials.
Keywords:
Methods. HA, FHA and FA were synthesised by a sol-gel method. Calcium nitrate and tri-
Hydroxyapatite
ethylphosphite were used as precursors under an ethanol-water based solution. Different
Fluor-hydroxyapatite
amounts of ammonium fluoride (NH4F) were incorporated for the preparation of the sol-
Fluorapatite
gel derived FHA and FA. Optimisation of the chemistry and subsequent characterisation of
Sol–gel
the sol-gel derived materials was carried out using X-ray Diffraction (XRD) and Differential
Optimisation
Thermal Analysis (DTA). Rheology of the sol-gels was investigated using a viscometer and
Characterisation
contact angle measurement.
Rheology
Results. A protocol was established that allowed synthesis of HA, FHA and FA that were at least 99% phase pure. The more fluoride incorporated into the apatite structure; the lower the crystallisation temperature, the smaller the unit cell size (changes in the a-axis), the higher the viscosity and contact angle of the sol-gel derived apatite. Significance. A technique has been developed for the production of HA, FHA and FA by the solgel technique. Increasing fluoride substitution in the apatite structure alters the potential coating properties. Crown Copyright © 2012 Published by Elsevier Ltd on behalf of Academy of Dental Materials. All rights reserved.
∗ Corresponding author at: Division of Biomaterials and Tissue Engineering, UCL Eastman Dental Institute, 256 Grays Inn Road, London WC1X 8LD, United Kingdom. Tel.: +44 203456 1189. E-mail address: [email protected] (J.C. Knowles).
0109-5641/$ – see front matter Crown Copyright © 2012 Published by Elsevier Ltd on behalf of Academy of Dental Materials. All rights reserved.
http://dx.doi.org/10.1016/j.dental.2012.11.008
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
1.
Introduction
Titanium (Ti) and its alloys have long been recognized and used as dental and orthopedic implant materials. To improve the implant-tissue osseointegration, considerable effort has been exerted to modify the Ti surface structure both physically and chemically [1–3]. Hydroxyapatite [HA, Ca10 (PO4 )6 (OH)2 ] coatings on Ti substrate have attracted significant attention for several years aimed at combining the excellent biocompatibility of HA and load-bearing ability of Ti [4,5]. Currently, most HA coatings are produced by a plasma-spraying technique [6,7]. In vivo reports on the system have shown good bone-bonding ability and fast osseointegration compared with pure Ti implants. This was mainly attributed to the osteoconductivity and chemical and biological similarities of HA with the human hard tissues [5,8,9]. There is some debate as to the actual requirements of a coating. There are two possibilities: (1) the coating has very low solubility and will remain in place for some considerable time (for the devices lifetime) and maintain a stable biological interface or (2) the coating is merely temporary and its role is to promote rapid growth of bone onto the surface of the coating but the coating will be replaced such that the bone will bond directly to the metal implant surface. This is of particular relevance to immediate loading implants. This is further complicated by the role of crystallinity. Lower crystallinity coatings showed faster resorption in a canine model and also higher levels of bone apposition onto the surface at 16 weeks, with mechanical fixation better in the lower crystallinity samples at 16 weeks but no difference at 32 weeks [10,11]. There are some parameters to be improved in the plasma spraying process, such as coating strength, chemical homogeneity, and residual porosity. These are related to the high fabrication temperature and coating thickness [7]. Recently, alternative methods have been developed to produce thin HA films. The sol–gel technique, being one of the thin film methods provides some benefits over the plasma spraying method, such as chemical homogeneity, fine grain structure, and low processing temperature [12]. Moreover, compared with other thin film methods, it is simple and cost efficient, as well as effective for the coating of complex-shaped implants. There are conflicting reports [13] which may be due to differing methodologies, but in general it is agreed that pure fluorapatite (FA, Ca10 (PO4 )6 F2 ) is known to have a much lower solubility than HA, because FA possesses a greater stability than HA, both chemically and structurally [13,14]. Moreover, the HA is able to form fluorhydroxyapatite (FHA, Ca10 (PO4 )6 F1 OH1 ) with the crystallographic substitution of OH− by F− . Moreover this substitution has been shown to occur in aqueous solutions and is also irreversible [15]. Hence, modulation of the extent of fluoride substitution provides an effective way of controlling the solubility of the apatite which thus allows us to produce a wide range of coatings with differing biological properties. In practice, the fluoride ion itself has been studied widely in dental restorative areas, due to its advantages over other ions in that it can reduce the formation of caries in bacterially contaminated environments and promotes mineralisation and crystallisation of calcium phosphates in the formation of bone [14,16]. There is some conflict in the literature when com-
167
paring hydroxyapatite with fluorhydroxyapatite coatings. In unloaded models, a fluorine containing hydroxyapatite coating was shown to be more stable [7]. However, when loaded, no significant difference was seen [17,18], although both types of implants showed good stability. Interestingly in a human trial, the hydroxyapatite coating was considerably thinner than the equivalent fluorhydroxyapatite coating [19]. Nevertheless, there have been few reports concerning the fabrication or characterisation of the HA, FA or FHA materials produced via the sol–gel route and it was the purpose of this study to undertake this. Furthermore with reference to the literature, while use of triethylphosphite and calcium nitrate as precursors have been the most successful method for the production of HA via a sol gel [20–22], the majority of the materials produced via the sol–gel reactions are -tricalcium phosphate (-TCP) and not HA/FA/FHA and this study sought to optimise the production of phase pure apatites and investigate their thermal and rheological properties.
2.
Materials and methods
2.1.
Preparation of HA sols
16.16 g of triethylphosphite (TEP [P(C2 H5 0)3 ], Aldrich, USA) was hydrolysed for 72 h in a mixture of 33.12 g of ethanol and 5.04 g of distilled water (P containing solution, VWR, UK). This mixture was then added to a solution of 39.36 g calcium nitrate [Ca(NO3 )2 ·4H2 0, Aldrich USA] in 15.12 g of distilled water. A 5% (w/v) solution of ammonium hydroxide (NH4 OH, VWR, UK) was added to the solution in order to improve gelation and subsequent formation of an apatite structure. The solution was allowed to react for 24 h and then age for a further 24 h at room temperature.
2.2.
Preparation of FHA and FA sols
The FHA sols were prepared using various amounts of ammonium fluoride (NH4 F, Aldrich, USA) in the P containing solution. The [P]/[F] molar ratios were 12, 6, 4 and 3 in order to have the corresponding compositions of Ca10 (PO4 )6 F0.5 OH1.5 , Ca10 (PO4 )6 F1 OH1 , Ca10 (PO4 )6 F1.5 OH0.5 and Ca10 (PO4 )6 F2 by replacing the OH group with F ions in molar ratios of 0.25, 0.5, 0.75 and 1 respectively. After stirring for 72 h, the solutions were added slowly to a solution containing a stoichiometric amount (Ca/P ∼1.67) of calcium nitrate [Ca(NO3 )2 ·4H2 0, Aldrich, USA] following the protocol developed for the HA sols.
2.3.
Powder preparation
Following successful preparation of apatites via the sol–gel reaction, samples of the sol–gel derived materials were heated to 500, 600, 700, 800, 900 and 1000 ◦ C for 2 h in air using a hot air oven (Lenton Furnace, Lenton Thermal Designs Limited, UK) with a ramp of 5 ◦ C/min, dwell time of 60 min and cool down rate of 10 ◦ C/min. Following heating in the hot air oven the samples were ground to a powdery form for 20 min using a vibrating agate ball mill (Fritsch, Germany) and to ensure similar particle sizes ≈20 m were passed through a sieve (Fritsch, Germany).
168
2.4.
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
Powder X-ray diffraction (XRD)
X-ray diffraction data were collected using an automated Bruker D8 Advance system equipped with a sample changer ˚ in a flat plane – geometry and Ni filtered CuK␣ ( = 1.5418 A) in the range 10–100◦ 2, in steps of 0.02◦ , with a count time of 12 s per step and using a Lynx Eye detector. From the data obtained, the unit cell size was calculated using the software TOPAS Academic V4.1. The kernel of this software is the same as that for the Bruker-AXS Topas software.
2.5.
Differential thermal analysis (DTA)
The sol–gel derived materials were evaluated using differential thermal analysis (DTA) to elucidate the thermal and weight changes as a function of time, using a Labsys TG/DTA 1600 ◦ C (Setaram Instruments, Caluire, France). Weighed (100 mg) sol–gel derived samples were placed into platinum sample holders and tested in an inert nitrogen atmosphere, and compared with a platinum pan only as a reference. Samples were run from 20 ◦ C to 1000 ◦ C at a rate of 20 ◦ C/min. The data was baseline corrected by carrying out a blank run and subtracting this from the plot obtained. The experiment was completed three times so a mean and standard deviation could be obtained for the crystallisation temperature.
2.6.
Rheology
A viscometer (Model DV-III, Brookfield, USA) was used to investigate the varying viscosity of the sol–gels after a set aging time (24 h) at various shear rates (from 20 to 200 s−1 with intervals of 10 s−1 ), each for 20 s at a constant temperature of 25 ◦ C. This was repeated 3 times. Changes in viscosity at a constant shear rate as the sample was aged were also measured. After mixing the hydrolysed Triethyl phosphite [P(C2 H5 0)3 ] Calcium Nitrate [Ca(NO3 )2 ·4H2 0] and ammonium hydroxide (NH4 OH) the viscosity at a constant shear rate 150 s−1 was measured every 6 h up to 72 h. This was repeated 3 times.
2.7.
Contact angle measurement
Using a standardised protocol 5 mL of the sol–gel samples were pipetted onto a commercially pure titanium disk and contact angle measurements were undertaken with a CAM 200 Optical Contact Light Meter (KSV, Instruments Limited, Finland) every second for 30 s. The measurements were repeated three times for each sol–gel derived material.
3.
Results
3.1.
HA/FHA/FA production
Using the protocol defined, HA derived from a sol–gel which was 99.7% HA and 0.3% -TCP purity was obtained. The Xray diffraction pattern associated with this is seen in Fig. 1a. All other sol–gels derived samples made (i.e. Ca10 (PO4 )6 F2 , Ca10 (PO4 )6 OH0.5 F1.5 , Ca10 (PO4 )6 OH1 F1 and Ca10 (PO4 )6 OH1.5 F0.5 ) exhibited a similar XRD pattern (data not shown) and were
thermally stable to 1000 ◦ C. The phases measured in the fluoride substituted apatites can be seen in Table 1.
3.2.
Differential thermal analysis
Fig. 1b shows an example combined TG/DTA trace for sample Ca10 (PO4 )6 OH0.5 F1.5 . As can be seen there are three main thermal events. The first is at around 100 ◦ C and is associated with unbound water loss as there is a thermal and associated large mass loss event. A second event occurs at around 300 ◦ C and is associated with bound organic loss, again signified by the large weight loss. A third sharp thermal event occurs at around 450 ◦ C and there is also a very slow weight change event. These are thought to be unrelated as they differ in rates and also the weight change remains in the same temperature region but the thermal event changes with composition. This third thermal event was assigned to crystallisation and the temperatures at which they occur are detailed in Table 2, which shows the mean and standard deviation obtained for the crystallisation temperatures of each of the sol gels. All of the materials were thermally stable to 1000 ◦ C.
3.3.
Unit cell measurements
Comparable graphs of the a-axis, c-axis and volume changes as a reflection of material type and temperature are shown in Figs. 2–4. Figs. 2 and 4 showed that increasing the fluoride content in the apatite structure decreased the a axis and volume of the unit cell. Furthermore as the heating temperature increased there was a trend for the a axis and volume to decrease in size. This was particularly marked with the Ca10 (PO4 )6 F0.5 OH1.5 . Fig. 3 shows that fluoride content and heating temperature had minimal effect on the c axis.
3.4.
Rheology
3.4.1. Varying viscosity of the sol–gels after 24 aging time at various shear rates Fig. 5 shows the viscosity changes (mean ± standard deviation) of the sol–gels after aging for 24 h, as represented with respect to shear rate. The viscosity of the sol–gels was as follows; Ca10 (PO4 )6 F2 > Ca10 (PO4 )6 OH0.5 F1.5 > Ca10 (PO4 )6 )OH1 F1 > Ca10 (PO4 )6 OH1.5 F0.5 > Ca10 (PO4 )6 (OH)2 . This pattern in viscosity was consistent throughout the entire time period tested. Up to a shear rate of 40 sec−1 there was a decrease in the viscosity of all materials. Increasing the fluoride content decreased the viscosity of the materials.
3.4.2. Measuring changes in viscosity at a constant shear rate as the sample is aged The mean viscosity change (±standard deviation) was plotted with aging time at a constant shear rate of 150 s−1 . This is shown in Fig. 6. All of the sol–gels showed a rapid increase in viscosity over the first 24 h, this subsequently stabilized and saw minimal changes over the next 48 h. At each time point the increasing fluoride content shows an increase in viscosity.
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
169
Fig. 1 – (a) XRD pattern of the of the HA sol–gel after heating to 1000 ◦ C. (b) Combined thermogravimetric and differential thermal analysis of Ca10 (PO4 )6 OH0.5 F1.5 sample.
3.5.
Contact angle measurement
The overall mean contact angles, plus and minus the standard deviation for each of the sol–gels is shown in Table 3. Fig. 7 graphically shows the contact measurement as a result of time of the HA, FA and FHA sol–gels. With reference to Fig. 7 it can be seen that initially on dropping the sol–gel, for approximately the first 4 s there is rapid decrease in contact angle, subsequently the reduction starts to slow. The contact angle is higher with the increasing fluoride concentration.
4.
Discussion
4.1.
Hydroxyapatite (HA) production
From the literature the most promising combination for the precursors to produce HA have been calcium nitrate and triethylphosphite (TEP) using alcohol at the hydrolysis stage [23].
However Kim et al. (2004) could not produce phase pure HA and were unable to reduce the significant amounts of beta tricalcium phosphate (-TCP) formed. Initial hydrolysation of the TEP is required to increase its reactivity so that it can react readily with the calcium nitrate. The first stage of optimizing the chemistry involved varying the amount of water used to hydrolyse the TEP in ethanol. Using the basic protocol from the literature [23] it was found that 100 mL of water was required to result in sol-gel production. The second stage of investigation found that the hydrolysation time of the TEP was optimal at 72 h and at this stage it can be proposed that total substitution has taken place and the TEP is able to react efficiently with the calcium nitrate. The ethanol produced as a side product from this reaction has a boiling point of 78 ◦ C and can be easily driven off once the sol–gel is heated to 80 ◦ C in the hot air oven. It was found that a mixing time of 24 h as used by Kim et al., 2004 was effective in the production of HA [23]. The addition of ammonium hydroxide can catalyse the reaction between TEP and calcium nitrate and this was confirmed in this study [23].
170
99.8% 0.2%
0.2%
Ca10 (PO4 )6 F2 -TCP
Fig. 3 – Graph to show c-axis changes with heating of the sol–gels (mean ± SD).
The sol–gel production protocols that had been established for HA sol–gel production were utilised as the basic methodology for the production of the fluoride substituted apatite sol–gels. With reference to the literature ammonium fluoride was chosen as the donor for the fluoride ions [23]. After preliminary pilot experiments it proved that this should initially be added to the TEP and mixed for 24 h and then both subsequently added to the calcium nitrate. With attention to the chemistry and controlling the amounts of ammonium fluoride initially added attempt was made to control the level of fluoride substitution in the apatite structure.
99.8% 0.3% 99.7%
0.3%
99.8%
0.2%
99.7%
Ca10 (PO4 )6 OH0.5 F1.5
Fig. 2 – Graph to show a-axis changes with heating of the sol–gels (mean ± SD).
-TCP Ca10 (PO4 )6 OH1 F1 -TCP Ca10 (PO4 )6 OH1.5 F0.5 -TCP Ca10 (PO4 )6 OH2
Table 1 – Constituents components of HA, FA and fluoride substituted apatites.
-TCP
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
Fig. 4 – Graph to show changes in unit cell volume with heating of the sol–gels from 500 ◦ C to 1000 ◦ C (mean ± SD).
171
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
7.5
Table 2 – Crystallisation temperature of fluoride substituted apatite compositions as determined by DTA.
Ca10 (PO4 )6 F2 Ca10 (PO4 )6 OH0.5 F1.5 Ca10 (PO4 )6 OH1 F1 Ca10 (PO4 )6 OH1.5 F0.5 Ca10 (PO4 )6 OH2
4.2.
7
Crystallisation temperature mean ± SD (◦ C) 467 470 492 498 499
± ± ± ± ±
6.5
Viscosity (MPa)
Composition
Shear rate = 150 s-1
0.21 0.43 1.22 0.23 0.65
6
5.5
5
Differential thermal analysis
There is a clear trend in the crystallisation temperatures seen. As the level of fluoride substitution increased the crystallisation temperature decreased. During substitution the F− occupies to varying degrees (depending on the level of substitution) the position of OH− in the apatite. It is also known that increasing the level of fluoride substitution in the apatite structure produces a smaller more compact crystal structure [24]. The results of this study suggest that the increasing levels of fluoride allow the structure to crystallise more easily and hence have a lower crystallisation temperature. This finding could be attributed to the fact that increasing fluoride substitution results in a more stable, compact structure that is capable of crystallising quicker than HA.
Fig. 5 – Mean viscosity changes in HA, FA, and FHA sol-gels after aging for 24 h versus shear rate (mean ± SD). With reference to figure 62 several clear patterns can be seen. The initial [2]viscosity of the sol–gels is such that the higher the concentration of fluoride in the.
4.5
0
24
48
72
AgingTime time (Hou rs) Aging (Hou rs)
Fig. 6 – Viscosity changes in HA, FA and FHA sol–gels measured at shear rate of 150 s−1 , as represented with respect to aging (mean ± SD).
4.3.
Unit cell measurements
All of the sol–gel produced fluoride substituted apatites had manifestly smaller a-axis than the HA produced from the sol–gel. This is in line with the published literature, which reported in the case of FHA powders, there were significant decreases in the a-axis compared to HA [13,25,26]. The more fluoride substitution the smaller the aaxis, such that the a-axis decreases in size as follows: Ca10 (PO4 )6 F0.5 OH1.5 > Ca10 (PO4 )6 F1 OH1 > Ca10 (PO4 )6 F1.5 OH0.5 > Ca10 (PO4 )6 F2 . The fluoride ion has a smaller size than OH− ,
Fig. 7 – Contact angle measurements for each of the sol–gels.
Table 3 – Mean contact angle of each of the sol–gels. Mean (±SD) contact angle (◦ ) apatite sol–gel (Ca10 (PO4 )6 ) (OH)2 55.34 ± 0.35
(F0.5 OH1.5 )2
(F1 OH1 )2
(F1.5 OH0.5 )2
F2
56.36 ± 0.48
57.32 ± 0.37
58.12 ± 0.44
60.34 ± 0.46
172
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
˚ compared to 1.37 A, ˚ and as such is able to pack being 1.28 A more closely into the apatite structure [27]. There is a downward trend in the size of the a-axis as the firing temperature is increased. Abrahams and Knowles (1994) investigated the effects of sintering conditions on hydroxyapatite using X-ray diffraction [28]. They were investigating temperatures in excess of 1000 ◦ C and found small but significant decreases in unit-cell volume as temperature increased. They proposed this to be due to a loss of small amounts of carbonate ion from channels within the structure. It is possible to suggest that the pattern seen in the unit cell volume here is similar and may be due to loss of small amounts of carbonate ion at the lower temperatures investigated. However this was not seen when examined via FTIR. Furthermore heating the apatite structure to temperatures above the crystallisation temperature, but below the decomposition temperature, allows structural rearrangement to a more optimal position. The more optimal position of the atoms the smaller the a-axis and overall unit cell volume will be. The a-axis values for the Ca10 (PO4 )6 F2 (FA) were similar to ˚ [25,29]. The the reported values for pure FA (a-axis 9.3684 A) similarity of the values obtained in this study to reported values provides further support for the purity of the samples that have been produced by the sol–gel method adopted. While there were manifest changes in the a-axis, there were little changes observed in the c-axis. This is well reported in the literature for fluoride substituted apatites [13,25,26]. Within the apatite structure there is no spare volume for the structure to change in dimensions along the c axis, therefore it is to be expected that little change would be seen. These findings give further confidence in the data. All of the sol–gel produced fluoride substituted apatites had manifestly smaller unit cell volumes than the HA produced from the sol–gel. The more fluoride substitution the smaller the a-axis, such that the volume (Å3 ) decreases in size as follows: Ca10 (PO4 )6 (OH)2 > Ca10 (PO4 )6 F0.5 OH1.5 > Ca10 (PO4 )6 F1 OH1 > Ca10 (PO4 )6 F1.5 OH0.5 > Ca10 (PO4 )6 F2 . In direct proportion to the changes seen in the a-axis, there is a downward trend in the unit cell volume seen as the firing temperature is increased. Because of the clear correlations above it can be proposed that the controlled incorporation of fluoride in the sol–gel protocol used leads to the successful formation of FHA solid solutions with differing levels of fluoride substitution.
4.4.
Rheology
4.4.1. Varying viscosity of the sol–gels after 24 h aging time at various shear rates With reference to Fig. 5 it can be seen that after aging for 24 h, the viscosity of all the sol–gels showed an exponential decrease at low shear rates and stabilised with the higher shear rates. This behaviour is characteristic of sol–gels. As the shear rate is initially increased the viscosity of the sol–gel falls as there is a breakdown in the sol structure [30,31]. As the shear rates become higher, this effect becomes reduced such that any additional increases in shear rate cause negligible effects on the viscosity [30,31].
There is a clear pattern that can be seen throughout; the higher the concentration of fluoride in the sol–gel the higher the viscosity i.e. the viscosity of the sol–gels was as follows; Ca10 (PO4 )6 F2 > Ca10 (PO4 )6 OH0.5 F1.5 > Ca10 (PO4 )6 )OH1 F1 > Ca10 (PO4 )6 OH1.5 F0.5 > Ca10 (PO4 )6 (OH)2 . A potential theory to explain this is that there could be stronger cross links within the sol–gel with higher fluoride content. As a result of these stronger links the viscosity is increased. Clearly the rheological results obtained here are novel and may have implications when coating the sol–gels on to titanium i.e. to achieve the same coating with increasing fluoride concentration, different coating speeds may have to be adopted.
4.4.2. Measuring changes in viscosity at a constant shear rate as the sample is aged There is a clear pattern in the viscosity of the sol–gels; the higher the concentration of fluoride in the sol–gel the higher the viscosity i.e. the viscosity of the sol–gels was as follows; Ca10 (PO4 )6 F2 > Ca10 (PO4 )6 OH0.5 F1.5 > Ca10 (PO4 )6 OH1 F1 > Ca10 (PO4 )6 OH1.5 F0.5 > Ca10 (PO4 )6 (OH)2 . Again a possible theory that can be suggested for the reason for the increasing viscosity as more fluoride is substituted in to the sol–gel structure could be that there could be stronger cross links within the sol–gel with higher fluoride content. As a result of these stronger links this may have an overall effect on the viscosity.
4.5.
Contact angle measurement
For the contact angle measurements, at the initial time, the contact angles were relatively high, between around 70◦ and 75◦ . With time, for all samples the contact angle values dropped sharply, such that after around nearly 30 s, the values had dropped to values around 50–55◦ . Generally, if surfaces have contact angles less than 90◦ , they are regarded as hydrophilic [32]. With time the values drop as the surface becomes wet by the applied drop and the drop spreads. Throughout all contact angle measurements, at all times, there was a pattern with the contact angle being higher with increasing fluoride concentration such that the contact angles were as follows; Ca10 (PO4 )6 F2 > Ca10 (PO4 )6 OH0.5 F1.5 > Ca10 (PO4 )6 )OH1 F1 > Ca10 (PO4 )6 OH1.5 F0.5 > Ca10 (PO4 )6 (OH)2 . This suggests that increasing the concentration of fluoride seems to increase the hydrophobicity (contact angle) of the sol–gel. However it should be borne in mind that the values for all the sol–gel solutions were relatively similar and all of them may be regarded as hydrophilic. This finding may have implications when coating the sol–gels on to titanium. It can be proposed, that because of the altered rheological characteristics with higher levels of fluoride substitution, in the apatite structure produced by the sol–gel technique, different coating speeds/protocols may have to be adopted when coating the sol–gels on to titanium.
5.
Conclusions
A protocol has been established for producing HA, FHA and FA with virtually 100% purity, using the sol–gel process and
d e n t a l m a t e r i a l s 2 9 ( 2 0 1 3 ) 166–173
increasing incorporation of fluoride into the apatite structure results in a decreased crystallisation temperature and a more compact unit cell structure. Addition of fluoride results in changes in the rheological properties of the sol gel and may have implications if used to coat titanium.
Acknowledgement This work was supported in part (JCK) by WCU Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (No. R31-10069).
references
[1] Adell R, Lekholm U, Rockler B, Branemark PI. A 15-year study of osseointegrated implants in the treatment of the edentulous jaw. International Journal of Oral Surgery 1981;10:387–416. [2] Ratner BD. New ideas in biomaterials science – a path to engineered biomaterials. Journal of Biomedical Materials Research 1993;27:837–50. [3] Nanci A, Wuest JD, Peru L, Brunet P, Sharma V, Zalzal S, et al. Chemical modification of titanium surfaces for covalent attachment of biological molecules. Journal of Biomedical Materials Research 1998;40:324–35. [4] Block MS, Finger IM, Fontenot MG, Kent JN. Loaded hydroxylapatite-coated and grit-blasted titanium implants in dogs. International Journal of Oral and Maxillofacial Implants 1989;4:219–25. [5] McPherson EJ, Dorr LD, Gruen TA, Saberi MT. Hydroxyapatite-coated proximal ingrowth femoral stems. A matched pair control study. Clinical Orthopaedics and Related Research 1995:223–30. [6] Bloebaum RD, Dupont JA. Osteolysis from a press-fit hydroxyapatite-coated implant. A case study. The Journal of Arthroplasty 1993;8:195–202. [7] Overgaard S, Soballe K, Josephsen K, Hansen ES, Bunger C. Role of different loading conditions on resorption of hydroxyapatite coating evaluated by histomorphometric and stereological methods. Journal of Orthopaedic Research: Official Publication of the Orthopaedic Research Society 1996;14:888–94. [8] Hench LL. Bioceramics – from concept to clinic. Journal of the American Ceramic Society 1991;74:1487–510. [9] Ducheyne P, Cuckler JM. Bioactive ceramic prosthetic coatings. Clinical Orthopaedics and Related Research 1992:102–14. [10] Overgaard S, Bromose U, Lind M, Bunger C, Soballe K. The influence of crystallinity of the hydroxyapatite coating on the fixation of implants. Mechanical and histomorphometric results. The Journal of Bone and Joint Surgery British 1999;81:725–31. [11] Soballe K, Overgaard S, Hansen ES, Brokstedt-Rasmussen H, Lind M, Bunger C. A review of ceramic coatings for implant fixation. Journal of Long-Term Effects of Medical Implants 1999;9:131–51. [12] Brinker CJ, Scherer GW. Sol–gel science: the physics and chemistry of sol–gel processing. Academic Press; 1990. [13] Larsen MJ, Jensen SJ. Solubility, unit cell dimensions and crystallinity of fluoridated human dental enamel. Archives of Oral Biology 1989;34:969–73.
173
[14] Moreno EC, Kresak M, Zahradnik RT. Fluoridated hydroxyapatite solubility and caries formation. Nature 1974;247:64–5. [15] Fulmer M, Brown PW. Low-temperature formation of fluorapatite in aqueous-solution. Journal of the American Ceramic Society 1992;75:3401–7. [16] Marie PJ, Hott M. Short-term effects of fluoride and strontium on bone formation and resorption in the mouse. Metabolism: Clinical and Experimental 1986;35:547–51. [17] Overgaard S, Lind M, Josephsen K, Maunsbach AB, Bunger C, Soballe K. Resorption of hydroxyapatite and fluorapatite ceramic coatings on weight-bearing implants: a quantitative and morphological study in dogs. Journal of Biomedical Materials Research 1998;39:141–52. [18] Overgaard S, Lind M, Glerup H, Grundvig S, Bunger C, Soballe K. Hydroxyapatite and fluorapatite coatings for fixation of weight loaded implants. Clinical Orthopaedics and Related Research 1997:286–96. [19] Overgaard S, Soballe K, Lind M, Bunger C. Resorption of hydroxyapatite and fluorapatite coatings in man. An experimental study in trabecular bone. The Journal of Bone and Joint Surgery British 1997;79:654–9. [20] Kim HW, Kim HE, Knowles JC. Sol–gel apatite films on titanium implant for hard tissue regeneration. Bioceramics 2004;16:254–62, 423-6. [21] Kim HW, Kim HE, Kim HW, Knowles JC. Improvement of hydroxyapatite sol–gel coating on titanium with ammonium hydroxide addition. Journal of the American Ceramic Society 2005;88:154–9. [22] Kim HW, Kim HE, Knowles JC. Fluor-hydroxyapatite sol–gel coating on titanium substrate for hard tissue implants. Biomaterials 2004;25:3351–8. [23] Kim HW, Knowles JC, Salih V, Kim HE. Hydroxyapatite and fluor-hydroxyapatite layered film on titanium processed by a sol–gel route for hard-tissue implants. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2004;71:66–76. [24] Song WH, Jun YK, Han Y, Hong SH. Biomimetic apatite coatings on micro-arc oxidized titania. Biomaterials 2004;25:3341–9. [25] Smith DK. Calcium phosphate apatites in nature. Hydroxyapatite and Related Materials 1994:29–45. [26] Kim HW, Koh YH, Yoon BH, Kim HE. Reaction sintering and mechanical properties of hydroxyapatite-zirconia composites with calcium fluoride additions. Journal of the American Ceramic Society 2002;85:1634–6. [27] Shannon RD. Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Crystallographica Section A 1976;32:751–67. [28] Abrahams I, Knowles JC. Effects of sintering conditions on hydroxyapatite for use in medical applications – a powder diffraction study. Journal of Materials Chemistry 1994;4:185–8. [29] Elliott JC. Structure and chemistry of the apatites and other calcium orthophosphates. Amsterdam: Elsevier; 1994. [30] Ben-Nissan B, Choi AH. Sol–gel production of bioactive nanocoatings for medical applications. Part 1: an introduction. Nanomedicine (London) 2006;1:311–9. [31] Choi AH, Ben-Nissan B. Sol–gel production of bioactive nanocoatings for medical applications. Part II: current research and development. Nanomedicine (London) 2007;2:51–61. [32] Förch R, Schönherr H, Jenkins ATA. Surface design: applications in bioscience and nanotechnology. Weinheim: Wiley-VCH; 2009.